The Creative Development of Johann Sebastian Bach, Volume II.pdf

The Creative Development of Johann Sebastian Bach Volume II: 1717–1750

This page intentionally left blank

The Creative Development of Johann Sebastian Bach volume ii: 1717–1750

Music to Delight the Spirit

R I CH AR D D. P. J ON ES

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries # Richard D. P. Jones 2013 The moral rights of the author have been asserted First Edition published in 2013 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2013948106 ISBN 978–0–19–969628–4 Printed in Great Britain by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

For my wife Anne Paul Jones

This page intentionally left blank

Preface This is the second volume of a two-volume study of the music of Johann Sebastian Bach. An analytical and (as far as possible) chronological approach is adopted with the object of gaining a clearer picture of the composer’s creative development. Whereas the first volume dealt with the music of Bach’s early and Weimar periods (c. 1695–1708 and 1708–17 respectively), the second volume is concerned with the Co¨then and Leipzig periods (1717–23 and 1723–50). The book is designed for music lovers in general, not just for students or scholars. The thirty-three years of Bach’s high maturity are subdivided as follows—Part I: 1717–29; Part II: 1729–39; and Part III: 1739–50. The first part bridges Bach’s move from Co¨then to Leipzig, mainly due to the very pronounced continuity in the sphere of keyboard and instrumental music. 1729 was deemed an appropriate date for the division between Parts I and II: it apparently marks the end of Bach’s regular composition of sacred cantatas, but at the same time it marks the beginning of his directorship of the Leipzig Collegium musicum, together with the intensified interest in secular music (both vocal and instrumental) that this entailed. Around 1739, the starting date of Part III, Bach began work on The Well-Tempered Clavier II and the Eighteen Chorales—collections whose retrospective character forms a significant feature of this last decade. Within each of the three parts a brief introduction sets the scene in terms of biographical details and compositional activity; three major chapters then deal with keyboard, instrumental, and vocal music in turn; and finally a brief conclusion attempts to draw all the strands together. Problems that hamper an enterprise of this kind are less acute than in Bach’s early works, but nonetheless considerable. Firstly, losses are very extensive. Two of Bach’s five church-year cycles of sacred cantatas—over 100 compositions—are lost, as are some twenty-three secular cantatas (as against nineteen extant and seven incomplete). The greatest single loss of all is the St Mark Passion of 1731, for there is no reason to suppose that it would have been inferior to the St John or St Matthew Passions. In addition, the Bach Obituary, after listing the works we know, mentions ‘a mass of other instrumental pieces of all sorts and for all kinds of instruments’, which can no longer be identified today and must therefore be presumed lost. Secondly, although dating problems have been to a considerable extent resolved, thanks to the work of Alfred Du¨rr, Georg von Dadelsen, and Yoshitake Kobayashi, problems remain in cases where no original source material survives (notably BWV 80, 209, 542 no. 1, 547, 590, 736, 903, 904, 918, 933–8, 1029, 1042, and 1044). Thirdly, there are still some problems of authenticity. Legitimate doubts, with which the present writer concurs, have been

viii

pref ace

expressed in relation to the following compositions: BWV 50, 230, 534, 546 no. 2, 1023, 1031, and 1033. For this reason they are excluded from the present study. In addition to the chronological studies mentioned above, the present writer is greatly indebted to the following: Kirsten Beißwenger: Werner Breig: Gregory Butler: Alfred Du¨rr: Martin Geck: David Ledbetter: Yo Tomita: Peter Williams: Christoph Wolff:

J. S. Bachs Notenbibliothek numerous studies of the organ music and harpsichord concertos studies of Clavieru¨bung III and the Canonic Variations studies of the cantatas, oratorios, and Passions editor of the invaluable series of Dortmund conference reports studies of The Well-Tempered Clavier and the unaccompanied solos editor of Understanding the B-minor Mass, conference report, Belfast, 2007 The Organ Music of J. S. Bach biography, numerous essays, Der Stile antico in der Musik J. S. Bachs

Further details, alongside other studies that have played a vital role in the preparation of this book, may be found in the footnotes and bibliography. The writer is greatly indebted to Martin Holmes and the other music staff of the Bodleian Library, Oxford, where much of the work involved in preparing this book was carried out, and to Jeanne Roberts for her expert setting of the music examples. Richard D. P. Jones

Contents List of Abbreviations

Part I: The Co¨then and early Leipzig years: 1717–1729 1. Introduction 2. The Well-Tempered Clavier I and other keyboard works Fantasia: BWV 903, 542/1 Prelude, fugue, and invention: Clavierbu¨chlein for W. F. Bach, Das Wohltemperierte Clavier I, Aufrichtige Anleitung (Inventions and Sinfonias) Suite: English Suites, French Suites and BWV 818–19, Clavieru¨bung I Prelude, fantasia, and fugue for organ: BWV 544, 548, 546/1, 537, 562/1, 540/2

3. The Brandenburg Concertos and other instrumental works Concertos and ouvertures: Brandenburg Concertos; BWV 1046–51, 1042, 1066, 1069 Violin, cello, and ﬂute solos: Sei Solo a violino senza basso; Six Suites for solo cello; BWV 1013 Sonatas with obbligato harpsichord or continuo: Sei Sonate a Cembalo certato e Violino solo (BWV 1014–19), BWV 1034, 1039, 1021

4. Sacred and secular: the vocal works Secular cantatas: BWV 134a, 173a, 249a, 36c, 205, 207, 198, 204, 210a Sacred cantatas: Leipzig Cycle I Magniﬁcat and Passion: Magniﬁcat in E♭ (BWV 243a), St John Passion, Version I Leipzig Cycle II: chorale cantatas St John Passion, Version II Leipzig Cycle II: non-chorale cantatas Leipzig Cycle III St Matthew Passion Cantatas of 1727–1728 and the Picander Cycle (1728–1729) The motets: BWV 225–9

5. Conclusion

3 12 12 15 36 58 65 65 86 97 106 106 115 132 142 157 162 168 181 192 198 206

Part II: The middle Leipzig years: 1729–1739 1. Introduction

221

2. Clavieru¨bung II–III and other harpsichord, organ, and lute works

228 228 238 243 244

Clavieru¨bung II–III Miscellaneous keyboard works: BWV 904, 906, 918, 933–8 Two lute works: BWV 997–8 Miscellaneous organ works: BWV 590, 547, 736

x contents 3. The harpsichord concertos and other instrumental works Two ouvertures: BWV 1068, 1067 Two violin concertos: BWV 1041, 1043 Concertos for three or four harpsichords: BWV 1063–5 Concertos for two harpsichords: BWV 1060–2 Concertos for solo harpsichord: BWV 1052–9 Sonatas in concerto style: Six Organ Sonatas (BWV 525–30), BWV 1029–30, 1032

4. Sacred and secular: vocal works II Secular cantatas: BWV 201, 213–15, 206, 211, 209, 30a Sacred cantatas: BWV 117, 192, 51, 112, 29, 140, 36, 177, 97, 100, 14, 80, 9, 197, 30 Passion and motet: St Mark Passion (BWV 247), BWV 118 Magniﬁcat and Missa: Magniﬁcat in D (BWV 243), Missa in B minor (BWV 2321), Missae (BWV 233–6), Kyrie-Christe (BWV 233a) Oratorio: Weihnachts-Oratorium, Himmelfahrts-Oratorium, Oster-Oratorium

5. Conclusion

248 248 251 252 254 257 261 271 271 279 284 287 307 316

Part III: The late Leipzig years: 1739–1750 1. Introduction

327

2. The Well-Tempered Clavier II and other keyboard/organ works

333 333 335 346 350 352 360 361

Eighteen Chorales (BWV 651–68) Das Wohltemperierte Clavier II Aria mit verschiedenen Vera¨nderungen (Goldberg Variations) Verschiedene Canones (Fourteen Canons) Die Kunst der Fuge Canonische Vera¨nderungen (Vom Himmel hoch) Sechs Chora¨le von verschiedener Art (Schu¨bler Chorales)

3. The Musical Offering and other instrumental works Sonatas BWV 1035, 1027–8 Concerto BWV 1044 Musicalisches Opfer BWV 1079

4. The B minor Mass and other vocal works Cantate burlesque (Peasant Cantata) (BWV 212) Tilge, Ho¨chster, meine Su¨nden (BWV 1083) Mass in B minor (BWV 232)

363 363 365 366 374 374 376 378

5. Conclusion

393

Bibliography

402

Index of Bach’s works

423

General Index

432

List of Abbreviations ABRSM Bach BD I

BD II

BD III

BD V

BD VII

Berlin Bethlehem BJ Brussels BWV

C/c Cambridge FM Darmstadt Dresden DTB Du¨rr Chr 2

Halle

Associated Board of the Royal Schools of Music, London Journal of the Riemenschneider Bach Institute, Baldwin-Wallace College, Berea, Ohio (Berea, 1970– ) Bach-Dokumente I: Schriftstu¨cke von der Hand Johann Sebastian Bachs, ed. W. Neumann and H.-J. Schulze (Kassel and Leipzig, 1963) Bach-Dokumente II: Fremdschriftliche und gedruckte Dokumente zur Lebensgeschichte Johann Sebastian Bachs 1685–1750, ed. W. Neumann and H.-J. Schulze (Kassel and Leipzig, 1969) Bach-Dokumente III: Dokumente zum Nachwirken Johann Sebastian Bachs 1750–1800, ed. H.-J. Schulze (Kassel and Leipzig, 1972) Bach-Dokumente V: Dokumente zu Leben, Werk und Nachwirken Johann Sebastian Bachs 1685–1800, ed. H.-J. Schulze and A. Glo¨ckner (Kassel, 2007) Bach-Dokumente VII: Johann Nikolaus Forkel, Ueber Johann Sebastian Bachs Leben, Kunst und Kunstwerke (Leipzig, 1802), ed. C. Wolff and M. Maul (Kassel, 2008) Staatsbibliothek zu Berlin, Preußischer Kulturbesitz Bach Choir, Bethlehem, Pa. Bach-Jahrbuch (Leipzig, 1904– ) Bibliothe`que du Conservatoire Royal de Musique, Brussels Bach-Werke-Verzeichnis: Thematisch-systematisches Verzeichnis der musikalischen Werke von Johann Sebastian Bach, ed. Wolfgang Schmieder (Leipzig, 1950; 2nd rev. and enlarged edn, Wiesbaden, 1990); Kleine Ausgabe, ed. A. Du¨rr and Y. Kobayashi (Wiesbaden, 1998) C major/C minor etc. Fitzwilliam Museum, Cambridge Hessische Landes- und Hochschulbibliothek, Darmstadt Sa¨chsische Landesbibliothek, Dresden Denkma¨ler der Tonkunst in Bayern Alfred Du¨rr, ‘Zur Chronologie der Leipziger Vokalwerke J. S. Bachs’, 2nd rev. edn (Kassel, 1976); orig. pub. in Bach-Jahrbuch 44 (1957), pp. 5–162 Universita¨ts- und Landesbibliothek Sachsen-Anhalt, Halle

xii l i s t o f a b b r e v i a t i o n s Kobayashi Chr

Krako´w Leipzig BA Leipzig Go.S. Leipzig MB Leipzig TS Leipzig UL London BL NBA

NBR

New York PL New York PML P Paris BN RV SATB Spitta I, II

St Stuttgart Tokyo VBN

Vienna Washington

Yoshitake Kobayashi, ‘Zur Chronologie der Spa¨twerke J. S. Bachs: Kompositions- und Auffu¨hrungsta¨tigkeit von 1736 bis 1750’, BachJahrbuch 74 (1988), pp. 7–72 Biblioteka Jagiellon´ska, Krako´w, Poland Bach-Archiv, Leipzig Gorke Sammlung, Leipzig Musikbibliothek der Stadt Leipzig Thomasschule, Leipzig (MSS formerly owned by the school, now in safe keeping of Bach-Archiv, Leipzig) Universita¨tsbibliothek, Leipzig British Library, London Neue Bach-Ausgabe: J. S. Bach, Neue Ausgabe sa¨mtlicher Werke, ed. Johann-Sebastian-Bach-Institut, Go¨ttingen and Bach-Archiv, Leipzig (Kassel and Leipzig, 1954– ) The New Bach Reader: A Life of J. S. Bach in Letters and Documents, rev. and enlarged edn by Christoph Wolff (New York and London, 1998) of documentary biography by Hans T. David and Arthur Mendel (London, 1945) Public Library, New York Pierpont Morgan Library, New York Partitur (score), as in shelfmark Berlin, Mus. ms. Bach P 42 etc. Bibliothe`que nationale, Paris Peter Ryom, Verzeichnis der Werke Antonio Vivaldis: kleine Ausgabe (Leipzig, 1974; 2nd edn 1979) soprano, alto, tenor, bass Philipp Spitta, Johann Sebastian Bach, 2 vols (Leipzig, 1873 and 1880); Eng. trans. by C. Bell and J. A. Fuller-Maitland, 3 vols (London, 1884–5; repr. 1952) Stimmen (parts), as in shelfmark Berlin, Mus. ms. Bach St 132 etc. Internationale Bachakademie Stuttgart Ueno Gakuen College, Tokyo Verzeichnis der Werke in J. S. Bachs Notenbibliothek, in Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992), pp. 223–400 ¨ sterreichische Nationalbibliothek, Vienna O The Library of Congress, Washington

PA R T I The Co¨then and early Leipzig years: 1717–1729

This page intentionally left blank

I.1 Introduction

In Prince Leopold of Anhalt-Co¨then Bach seems to have found in many ways the ideal patron. The prince was not only a good bass singer but also played the violin, viola da gamba, and harpsichord. And in 1713, after his return from a grand tour during which he acquired published copies of much French and Italian music, he took advantage of the dissolution of the Berlin court Capelle under Friedrich Wilhelm I by employing six musicians (and later a seventh) who had been made redundant. By late 1717, when Bach took up his appointment as Capellmeister, Leopold had increased the number of musicians at the Co¨then court to sixteen, of whom about half were players of the front rank. There were disadvantages for Bach at Co¨then, however. Since it was a Calvinist court, there was no opera—such an enterprise, had it existed, could hardly fail to have attracted Bach’s interest. Furthermore, the Calvinism of the ruling prince meant that there was no regular opportunity for Bach to compose and perform church music, though he did so at least once for the prince’s birthday1 and might have done occasionally at the Lutheran Agnuskirche, which Bach and his family attended.2 As for secular vocal music, one of Bach’s regular duties was to perform a cantata every year for the prince’s birthday and another for New Year’s Day, though very few of these works survive.3 In the ﬁeld of instrumental music Bach’s situation was considerably more advantageous. In moving from Weimar to Co¨then he had risen from the second-rank post of Concertmeister to the top-rank post of Capellmeister. And as such he directed an instrumental ensemble that few German courts could rival. In addition, the reigning prince was clearly passionate about music and no doubt gave his brilliant Capellmeister all the support he needed. In this favourable atmosphere Bach was able to compose some of his greatest keyboard and instrumental music, much of it never to be exceeded in later years:

1 The lost church cantata Lobet den Herrn, alle seine Heerscharen, BWV Anh. I 5, performed on 10 Dec. 1718. 2 Certain Weimar cantatas were evidently revived during the Co¨then years, including Erschallet, ihr Lieder, BWV 172, Mein Herze schwimmt im Blut, BWV 199, and Ich hatte viel Beku¨mmernis, BWV 21. 3 Only the New Year cantata Die Zeit, die Tag und Jahre macht, BWV 134a, and the birthday serenata Durchlauchtster Leopold, BWV 173a, survive complete.

4 par t i Title

Place of composition

Completion date

English Suites

Weimar and Co¨then

c. 1720

Cello Suites

Co¨then

c. 1720

Violin Solos

Co¨then

1720

Brandenburg Concertos

Co¨then

1721

Well-Tempered Clavier I

Co¨then

1722

Inventions and Sinfonias

Co¨then

1723

Sonatas for Violin and Harpsichord

Co¨then and Leipzig

1725

Clavierbu¨chlein for W. F. Bach, 1720

Co¨then and Leipzig

c. 1725

Clavierbu¨chlein for A. M. Bach, 1722

Co¨then and Leipzig

1725

French Suites

Co¨then and Leipzig

c. 1725

The place of composition shows that there were overlaps at both ends of the Co¨then period. Only the first of the English Suites can be securely dated within the Weimar period; the remainder most likely originated during the early Co¨then years. The small manuscript books Bach dedicated to his eldest son Wilhelm Friedemann and his second wife Anna Magdalena, though begun in Co¨then, continued to be filled in after the move to Leipzig in 1723. And two of Bach’s most important collections, the Violin and Harpsichord Sonatas and the French Suites, were left finished when he moved away from Co¨then in 1723, with the result that he had to return to them in the early Leipzig years. The keyboard collections were partly designed for teaching purposes. Tuition, which comprehended keyboard playing and composition alike, began in Bach’s own family circle and then spread outwards towards the private instruction of individual students. Thus the Clavierbu¨chlein of 1720 and 1722, representing the domestic phase, included drafts of preludes destined for The Well-Tempered Clavier, of the Inventions and Sinfonias (then called ‘praeambula’ and ‘fantasias’), and of the French Suites. Later, fair copies were made of the first two of these collections, representing the public phase, and, like the Orgelbu¨chlein, revived from the Weimar years, they were furnished with title pages that clarified their didactic purpose. These title pages reveal the holistic nature of Bach’s musical philosophy: he is concerned not only with education but with pure delectation. Thus The Well-Tempered Clavier and the Aufrichtige Anleitung, as the fair copy of the Inventions and Sinfonias is entitled, are written not only for ‘those desirous of learning’ (‘denen Lehrbegierigen’) but for ‘those already skilled’ (‘als auch derer in diesem studio schon habil seyenden besonderem ZeitVertreib’) and for ‘lovers of the clavier’ (‘denen Liebhabern des Clavires’). In addition, these title pages are concerned with issues of playing and composition alike. The Orgelbu¨chlein gives ‘instruction in developing a chorale in many different ways’ (‘Anleitung gegeben wird, auff allerhand Arth einen Choral durchzufu¨hren’),

introduction

but also ‘in acquiring facility in the study of the pedal’ (‘anbey auch sich im Pedal studio zu habilitiren’). And the Aufrichtige Anleitung on the one hand shows how ‘to play clearly in two [and three] voices’ (‘mit 2 Stimmen reine spielen zu lernen, sondern auch . . . mit dreyen obligaten Partien richtig und wohl zu verfahren’) and how ‘to arrive at a singing style of playing’ (‘eine cantable Art im Spielen zu erlangen’), but, on the other hand, how ‘to have good ideas [and] develop them well’ (‘gute inventiones nicht alleine zu bekommen, sondern auch selbige wohl durchzufu¨hren’) and how ‘to acquire a strong foretaste of composition’ (‘einen starcken Vorschmack von der Composition zu u¨berkommen’).4 In 1720 Bach suffered the heavy blow of the sudden death of his first wife Maria Barbara. It might have been partly for this reason that in November of that year he sought a new start in different surroundings, travelling to Hamburg as a candidate for the post of organist at the Jacobikirche. During the same visit, perhaps, Bach’s obituary (by C. P. E. Bach and J. F. Agricola) informs us that ließ sich daselbst, vor dem Magistrate, und vielen andern Vornehmen der Stadt, auf der scho¨nen Catharinenkirchen Orgel, mit allgemeiner Verwunderung mehr als 2 Stunden lang, ho¨ren. Der alte Organist an dieser Kirche, Johann Adam Reinken, der damals bey nahe hundert Jahre alt war, ho¨rete ihm mit besondern Vergnu¨gen zu, und machte ihm, absonderlich u¨ber den Choral: An Wasserflu¨ssen Babylon, welchen unser Bach, auf Verlangen der Anwesenden, aus dem Stegreife, sehr weitlaüftig, fast eine halbe Stunde lang, auf verschiedene Art, so wie es ehedem die braven unter den Hamburgischen Organisten in den Sonnabends Vespern gewohnt gewesen wahren, ausfu¨hrete, folgendes Compliment: Ich dachte, diese Kunst wa¨re gestorben, ich sehe aber, daß sie in Ihnen noch lebet. (he was heard for more than two hours on the fine organ of St. Catherine’s before the magistrate and many other distinguished persons of the town, to their general astonishment. The aged organist of this church, Johann Adam Reinken, who at that time was nearly a hundred years old, listened to him with particular pleasure. Bach, at the request of those present, performed extempore the chorale An Wasserflu¨ssen Babylon at great length (for almost half an hour) and in different ways, just as the better organists of Hamburg in the past had been used to do at the Saturday vespers. Particularly on this Reinken made Bach the following compliment: ‘I thought this art was dead, but I see that in you it still lives.’)5

In the event Bach decided not to take the Hamburg post; and circumstances at Co¨then in any case soon changed for the better. Bach hired the young soprano Anna Magdalena Wilcke for the court in the summer of 1721, and he and she were married later in the same year (on 3 December). Only about a week after the wedding Prince Leopold also married. His bride, Friederica Henrietta of Bernburg, was unfortunately quite uninterested in music. Bach described the situation nearly ten years later in a letter to his former school friend Georg Erdmann:

4 The quotations are drawn from Bach’s title pages for The Well-Tempered Clavier I, the Inventions and Sinfonias, and the Orgelbu¨chlein; see BD I, Nos. 152, 153, and 148; and NBR, Nos. 90, 92, and 69. 5 Bach obituary, 1750 (pub. 1754); BD III, No. 666; NBR, No. 306.

par t i

die mutation, so mich als Capellmeister nach Co¨then zohe. Daselbst hatte einen gna¨digen und Music so wohl liebenden als kennenden Fu¨rsten; bey welchem auch vermeinete meine Lebenszeit zu beschließen. Es muste sich aber fu¨gen, daß erwehnter Serenißimus sich mit einer Berenburgischen Princeßin verma¨hlete, da es den das Ansehen gewinnen wolte, als ob die musicalische Inclination bey besagtem Fu¨rsten in etwas laulicht werden wolte, zumahln da die neu¨e Fu¨rstin schiene eine amusa zu seyn ([a] change in my fortunes . . . took me to Co¨then as Capellmeister. There I had a gracious prince, who both knew and loved music, and in his service I intended to spend the rest of my life. It must happen, however, that the said serenissimus should marry a princess of Berenburg, and that then the impression should arise that the musical interests of the said prince had become somewhat lukewarm, especially as the new princess seemed to be unmusical)6

For this and other reasons Bach sought the post of Cantor and Music Director at Leipzig, which had become vacant upon the death of Johann Kuhnau on 5 June 1722. Telemann and Graupner in turn were both chosen to fill the post by the Leipzig authorities, but neither could gain release from their current employment. Meanwhile, Bach performed his audition cantatas Du wahrer Gott und Davids Sohn (BWV 23) and Jesus nahm zu sich die Zwo¨lfe (BWV 22) in the Thomaskirche on Quinquagesima (Estomihi) Sunday, 7 February 1723. According to the local press, Bach’s music was ‘amply praised . . . by all knowledgeable persons’.7 After Graupner had declined the post, it was offered to Bach, who was elected on 22 April 1723. Bach and his family moved to Leipzig on 22 May, and his official duties began on the First Sunday after Trinity (30 May), when he performed his inaugural cantata before the Leipzig public, Die Elenden sollen essen, BWV 75. According to a Leipzig chronicle,8 its performance was regarded as a ‘great success’. Leipzig, the second city of Saxony after the capital Dresden, had long been renowned for its trade and commerce, for its fairs, which took place three times a year, at New Year, Easter, and Michaelmas, and for its university, which had been founded in 1409. At this lively, thriving city Bach had a prominent post as ‘Cantor et Director Musices’—above all, he was responsible for music at the four main Leipzig churches. The pupils at the Thomasschule, where Bach taught, were divided into four cantorates, which provided the music at the four churches. At those with modest musical provision, the Neue Kirche and the Peterskirche, the two less able cantorates sang, and Bach was able to delegate their direction to others. The two leading cantorates, however, alternated on Sundays between the two principal churches, the Thomaskirche and the Nicolaikirche. The second cantorate had to sing relatively simple cantatas by composers other than Bach. The first cantorate, on the other hand, which had long been celebrated throughout Lutheran Germany—Schu¨tz’s Geistliche Chormusik had been dedicated to it—was given the task by Bach of

6 7 8

BD I, No. 23; NBR, No. 152. BD II, No. 124; NBR, No. 95. The university chronicle Acta Lipsiensium Academica, 1723; BD II, No. 139; NBR, No. 103.

introduction

performing only his own exceptionally demanding compositions (this was not in his contract—compositions by others would have sufﬁced). As a result, during much of his ﬁrst three or four years in Leipzig, while he was engaged in building up a new repertoire of church music, Bach composed a new cantata virtually every week, not to mention the task of having the performing parts copied and undertaking the necessary rehearsal. Only occasionally did the revival of an older composition from the Weimar years give him some respite. The reward for such diligence was the regular performance of his church works on Sundays and feast days at the Thomaskirche or the Nicolaikirche before a congregation of well over 2,000 people.9 The services concerned were without question the biggest musical events in Leipzig at the time. Like his predecessors Schelle and Kuhnau, Bach was also responsible for the Old Service at the Paulinerkirche, the university church, which involved performing a cantata on the three High Feasts—Christmas, Easter, and Whit—as well as at the Reformation Festival (31 October). On these occasions Bach gave a repeat performance of the cantata that had already been performed that day in the Thomaskirche or the Nicolaikirche. The Kirchenstu¨ck, or cantata, as cultivated by Bach, was usually based on a biblical dictum or chorale text (most often from the Reformation period), whose theme, related to the Gospel or Epistle of the day, was then expounded in free verse. Generally, Bach would set the biblical or chorale text as an opening chorus of large dimensions, whereas the free verse would be set, in accordance with Neumeister’s reforms,10 as alternating recitative and arias. This ‘modern’ Italianate element, derived from opera and secular cantata, was thus wedded to the old German ecclesiastical element of dictum and chorale. The latter provided a foundation of sermon-like authority, whereas the more subjective, free-verse element allowed individual members of the congregation to relate the overall theme, or aspects thereof, to their own personal experience. Bach’s setting of the ecclesiastical texts would no doubt appeal to the church authorities; to what extent his pseudo-operatic treatment of the free verse did is a moot point,11 though it is interesting to note that on the occasion of his election, one of the councillors, Dr Steger, while voting for Bach, added that ‘he should make compositions that were not theatrical’.12 Furthermore, it was a condition of Bach’s appointment that in church he should ‘die Music dergestalt einrichten, daß . . . sie nicht opernhafftig herauskommen, sondern die Zuho¨rer vielmehr zur Andacht aufmuntere’ (‘so design the music that it should not create an operatic impression, but rather incite the listeners to devotion’).13

See Tanya Kevorkian, Baroque Piety: Religion, Society, and Music in Leipzig, 1650–1750 (Aldershot, 2007), p. 30. 10 See Vol. I of this study, p. 243. 11 On the other hand, it is very likely that Bach’s librettists were not infrequently drawn from the ranks of the clergy. 12 BD II, No. 129; NBR, No. 98. 13 BD I, No. 92; NBR, No. 100.

pa r t i

According to the obituary by C. P. E. Bach and Agricola,14 Bach wrote five cycles of cantatas for the whole church year, each of which would have numbered some fiftynine compositions. Only three cycles survive in a virtually complete state, however, and all three originated during Bach’s first few years in Leipzig, when his enthusiasm for the project must have been at its height. They are: Cycle I, 1723–4: dictum cantatas (based primarily on a biblical text) Cycle II, 1724–5: chorale cantatas (based on a chorale text with its associated melody) Cycle III, 1725–7: cantatas with a large input of instrumental music Cycle IV might have originated in 1727–8 (see Part I Ch. 4), but very few cantatas from this period have been transmitted. Of Cycle V (1728–9) only eight cantatas survive.15 The texts are drawn from a complete set for the whole church year by Bach’s regular librettist Picander, who stated in his preface of 24 June 1728 that they were to be set to music by Bach. The fate of the remaining settings is not known. In general Bach’s lost cantatas, which might have numbered over 100, were probably for the most part inherited by W. F. Bach, who according to Forkel16 later had to sell them off. Occasionally, for various reasons, Bach resorted to the performance of cantatas by respected contemporaries. In the period 1724–5 he performed Telemann’s cantata Der Herr ist Ko¨nig (TVWV8:6); and in the early Trinity period of 1725 (Third to Sixth Sunday, 17 June to 8 July) a series of five Telemann cantatas might have been performed in the two main Leipzig churches, perhaps during Bach’s absence.17 In the following year Bach’s Meiningen cousin Johann Ludwig Bach provided him with a printed cycle of cantata texts, Sonntags- und Fest-Andachten (Meiningen, 1704) and with the scores of at least some of his own settings of these texts. Bach and assistants wrote out the parts and performed no fewer than eighteen of Johann Ludwig’s settings between 2 February (Feast of the Purification) and 15 September (Thirteenth Sunday after Trinity).18 Bach seems to have been so impressed with the librettos (and perhaps with Johann Ludwig’s settings) that he set seven of them himself during the latter half of this period, from Ascension Day, 30 May, onwards: BWV 43, 39, 88, 187, 45, 102, and 17. On Good Friday of the same year, 19 April 1726, Bach revived an anonymous setting of the St Mark Passion (Hamburg, 1707) that he attributed, perhaps wrongly, to Reinhard Keiser. Bach had already performed this work in 1713, during his Weimar period. Its 1726 revival was the first of several Bach performances of Passions by other composers during the Leipzig years (see Part II Ch. 1 and Part III Ch. 1). Not long afterwards he might have performed Telemann’s setting of the Brockes Passion, for a 14

See n. 5. Plus a few bars of a ninth, BWV Anh. I 190. 16 BD VII, pp. 81–2; NBR, pp. 472–3. 17 BJ 59 (1973), pp. 5–32 (W. Hobohm) and BJ 78 (1992), pp. 73–6 (A. Glo¨ckner). 18 BJ 46 (1959), 48 (1961), and 49 (1962) (W. H. Scheide); BJ 63 (1977), pp. 7–25 (W. Blankenburg); BJ 73 (1987), pp. 159–64 (K. Ku¨ster). 15

introduction

copy from the 1720s was apparently in the library of the Thomasschule, Leipzig till the end of the Second World War.19 As for his own settings, the obituary informs us that he wrote five Passions, but only two survive: the St John and the St Matthew (of the St Mark Passion, first performed in 1731, only Picander’s libretto is extant). In Leipzig, according to J. C. Rost, sexton of the Thomaskirche, ‘on Good Friday of the year 1721, in the Vespers service, the Passion was performed for the first time in concerted style’,20 in a setting by Bach’s predecessor Johann Kuhnau. Bach continued this practice, and in musical terms the performance became the biggest event in the entire church calendar. The St John Passion was first performed on 7 April 1724, in the context of Cycle I, and then revived in a modified form—significantly including several elaborate chorale arrangements—on 30 March 1725, during the chorale-cantata cycle (Cycle II). The St Matthew Passion was first performed at Good Friday Vespers (11 April) 1727 and revived in 1729, though it did not acquire its definitive form till 1736. There is much in Bach’s two great oratorio-Passions—the seventeenth-century Lutheran genre to which he adhered—that could be described as dramatic or even theatrical, though we do not hear of objections raised by the clergy or members of the congregation. However, it is clear from the following account, published in Leipzig only a few years after the first performance of the St Matthew Passion, that strongly antagonistic feelings were raised by Passion music in an operatic style: When in a large town [such] Passion music was done for the first time . . . many people were astonished and did not know what to make of it. In the church pew of a noble family, many ministers and noble ladies were present, who sang the first Passion chorale out of their books with great devotion. But when this theatrical music began, all these people were thrown into the greatest bewilderment, looked at each other, and said, ‘What will come of this?’ And an old widow of the nobility said, ‘God save us, my children! It’s just as if one were at an opera comedy!’21

The Leipzig opera had closed in 1720, before Bach’s arrival in the city, but other forms of secular music were frequently heard, in some cases performed by the Collegium musicum (music society) that had been founded by Telemann in 1701. Bach took over the directorship of this organization in 1729, but it is not unlikely that he was able to avail himself of its resources even before then.22 At any rate during the 1720s he was already composing and performing a good deal of secular music that anticipates the Collegium musicum period: drammi per musica (the equivalent of one-act operas),

19 See Andreas Glo¨ckner, ‘Bach and the Passion Music of His Contemporaries’, Musical Times, 116 (1975), pp. 613–16, and his ‘J. S. Bachs Auffu¨hrungen zeitgeno¨ssischer Passionsmusiken’, BJ 63 (1977), pp. 75–119 (esp. 76–89 and 99–101). 20 BD II, No. 180; NBR, No. 114. 21 Christian Gerber, Historie der Kirchen-Ceremonien in Sachsen (Dresden and Leipzig, 1732), pp. 283–4; Eng. trans. in NBR, No. 324. 22 See Andreas Glo¨ckner, ‘Bachs Leipziger Collegium Musicum und seine Vorgeschichte’, in C. Wolff (ed.), Die Welt der Bach-Kantaten, vol. ii (Stuttgart, 1997), pp. 105–17, and his ‘Zur Vorgeschichte des “Bachischen” Collegium musicum’, in M. Geck (ed.), Bachs Orchesterwerke (Witten, 1997), pp. 293–303 (esp. 299).

10 par t i such as Der zufriedengestellte Aeolus (Aeolus Placated), BWV 205 (1725), Vereinigte Zwietracht, BWV 207 (1726), and Die Feier des Genius (The Celebration of Genius), BWV 249b (1726); the Trauer-Ode (Mourning Ode), or Tombeau de S. M. la Reine de Pologne, BWV 198 (1727); the solo cantata Von der Vergnu¨gsamkeit (On Contentedness), BWV 204 (1727/8); and the wedding cantata Vergnu¨gte Pleißenstadt, BWV 216 (1728). He also composed birthday cantatas for courts with which he had strong connections from of old: the pastoral cantata Entfliehet, verschwindet, entweichet, ihr Sorgen, BWV 249a, for Weißenfels (1725) and Steigt freudig in die Luft, BWV 36a, for Co¨then (1726). During these early Leipzig years, despite the huge demands made upon him by church music, Bach also engaged in concert activities outside the church. This is clear from an account by Ernst Ludwig Gerber, who informs us that in 1724 his father Heinrich Nicolaus ‘hatte . . . manche vortrefliche Kirchenmusik und manches Conzert unter Bachs Direktion mit angeho¨rt’ (‘had heard much excellent church music and many a concert under Bach’s direction’.)23 Music that he might have performed at this time includes the ouverture-suites in C and D, BWV 1066 and 1069, the Violin Concerto in E, BWV 1042, the Brandenburg Concertos, and perhaps the lost originals of some of the harpsichord concertos. At the same time Bach maintained contact with the court of Co¨then and the Saxon capital Dresden. He gave two extremely wellreceived organ recitals at the Sophienkirche, Dresden, in 1725. And in 1724, 1725, and 1728, alongside his second wife Anna Magdalena who was an able soprano, he gave guest performances in Co¨then in his capacity as Honorary Capellmeister. His first keyboard Partita (BWV 825) was dedicated to Prince Leopold’s newborn son in 1726; and finally he undertook the sad duty of composing and performing the prince’s funeral music in March 1729. As we have seen, alongside his teaching duties at the Thomasschule, Bach undertook much private tuition in keyboard playing and composition—it is clear that he regarded the two as inseparable. For this purpose he made use of the two great collections that had been completed at Co¨then, The Well-Tempered Clavier I and the Aufrichtige Anleitung (the Inventions and Sinfonias). In addition, the French Suites were completed in the early Leipzig years and became popular among Bach’s pupils, and the English Suites now became available for teaching purposes. Prominent pupils, such as Bernhard Christian Kayser,24 Heinrich Nicolaus Gerber, and Johann Caspar Vogler, made their own copies of these works or selections from them. The very act of copying might have given them insight into the compositional techniques involved in their creation, while they no doubt gained practical knowledge of the music by learning to play it at the keyboard from their own copies. According to E. L. Gerber,

BD III, No. 950 (p. 476); NBR, No. 315. See BD V, No. B240a, and Andrew Talle, ‘Nu¨rnberg, Darmstadt, Ko¨then—Neuerkenntnisse zur Bach¨ berlieferung’, BJ 89 (2003), pp. 143–72. U 24

introduction

his father Heinrich Nicolaus studied Bach’s music under the composer in the order: Inventions, suites, Well-Tempered Clavier.25 In 1725 Bach began a new Clavierbu¨chlein for his wife Anna Magdalena, entering two new keyboard suites at the start as a form of dedication. In revised versions these two compositions were later included in the set of six keyboard partitas that Bach published in separate instalments between 1726 and 1730, and then reissued in a collected edition as the First Part of the Clavieru¨bung (Leipzig, 1731). These partitas return to the large scale and considerable technical demands of the English Suites; and, like them, they were not primarily intended for teaching purposes. Instead, they were composed, according to the title page, ‘denen Liebhabern zur Gemu¨ths Ergoetzung’ (‘for music lovers, to delight their spirit’);26 in other words, for the skilled amateur or connoisseur. By publishing these works one by one in the late 1720s, Bach made tentative steps towards one of the great projects of his later Leipzig years—the dissemination of his keyboard works in print in order to bring them to a far wider audience than he had hitherto been able to command.

25 26

BD III, No. 950; NBR, No. 315. Hence the subtitle of the present book, ‘Music to Delight the Spirit’.

I.2 The Well-Tempered Clavier I and other keyboard works

Fantasia Title

Earliest source/s

Scribe, date

Fantaisie chromatique in D minor, BWV 903

Darmstadt, Mus. ms. 69 Berlin, P 803 Berlin, P 803 Berlin, P 421 Berlin, P 651 Berlin, P 288/9 Berlin, Am.B.531

Anon., early 1730s J. T. Krebs, post-1714 S. G. Heder, c. 1730? Anon., 6 Dec. 1730 J. F. Agricola, 1738–41 Anon., post-1750 Anon., post-1750

Fantasia in G minor, BWV 542/1

These two works represent the summit of Bach’s achievement in the free-fantasy style that he cultivated mainly in his earlier years. In the D minor composition, fugue might have been present from the outset; in the G minor, it was added at some later stage.1 But in both cases the fantasy element forms the main content of the work and defines its character. While neither work can be securely dated, an origin in the Co¨then (D minor) and Leipzig years (G minor) seems most likely.2 The Chromatic Fantasia and Fugue in D minor, BWV 903, one of Bach’s most extraordinary keyboard works, exists in three versions, though substantial changes are confined to the fantasia—the fugue seems to have remained largely unaltered. Notational and stylistic features of the early version, BWV 903a, suggest an origin in the Co¨then period, around 1720.3 An intermediate version, transmitted by J. T. Krebs and S. G. Heder, is of uncertain date. The final, definitive version, copied by Agricola while 1

The two G minor compositions, fantasia and fugue, are rarely found in the same source. The earliest source of the fugue (P 803) probably dates from the late Weimar years (1714–17), whereas the fantasia is transmitted only in posthumous sources. 2 Peter Williams (following Spitta) suggests a post-Weimar date for the G minor Fantasia; see his The Organ Music of J. S. Bach (rev. 2nd edn, Cambridge, 2003), p. 85. Certain features of the decorative writing, including the many multiple suspensions/appoggiaturas at phrase-ends, link it with Clavieru¨bung I and hence suggest a date in the mid-to-late 1720s. 3 See George B. Stauffer, ‘ “This Fantasia . . . never had its like”: On the Enigma and Chronology of Bach’s Chromatic Fantasia and Fugue in D minor, BWV 903’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 160–82, esp. 175ff. The early version is transmitted only in Darmstadt, Mus. ms. 69.

fantas ia

he was a student of Bach’s, perhaps dates from the 1730s. It has been observed that the work is not found in copies by Bach’s early Leipzig pupils, which suggests that the composer might have kept it to himself at first and not used it regularly for teaching purposes till the 1730s; or else he might have returned to it then after a long interval.4 If so, a possible use of the work might have been as a virtuoso showpiece in Collegium musicum concerts. In spite of its singular qualities, highlighted in Forkel’s oft-quoted remark that ‘this fantasia is unique and never had its like’,5 the Chromatic Fantasia may be viewed as the culmination of Bach’s writing in pseudo-improvisatory style for the harpsichord (according to the title pages of some of the chief sources,6 it is specifically written ‘pour le clavecin’). Not only is it a brilliant, virtuoso showpiece, with which Bach must have dazzled his first audiences, but it is also a chromatic and enharmonic tour de force. It exhibits a fully chromatic command of the keyboard and of the resources of tonality, of the kind that Bach is said to have displayed in his improvised fantasies. ‘When he played from his fancy,’ Forkel informs us, ‘all the 24 keys were in his power; he did with them what he pleased.’7 The chromatic element in the great fantasia is progressively intensified in the course of its three paragraphs and coda. The first paragraph is an extended passaggio that half-closes in the tonic at b. 20. The second (bb. 21–49) introduces arpeggiando chords in alternation with further passaggi. The third (bb. 49–74) modulates to the furthest reaches of the tonal system and back within the context of an instrumental recitative (so designated)—a style of writing that Bach had attempted before (in BWV 912a and 922) and would also have encountered in the slow movement of Vivaldi’s ‘Grosso Mogul’ Concerto (RV 208), which he transcribed for organ (BWV 594). The modulations of the fantasia’s recitative produce the effect of astounding, spontaneous strokes of genius, despite the careful tonal planning that clearly underlies the passage. We encounter here the contradiction that lies at the heart of the pseudoimprovisatory style from Frescobaldi to Bach—that great art has to be deployed in order to conjure up the impression of spontaneity. In the coda (b. 75), the treble descends chromatically through an octave, while the rich, full chords of the accompaniment simultaneously undergo their own fully chromatic descent—total chromaticism prevails. Yet the entire coda is underpinned by a tonic pedal. We thus meet the further contradiction here that, at the point in the fantasia where Bach’s chromaticism is most explicit, it is also most firmly grounded in the home key. By its very nature, the great fantasia is quite athematic. In comparable earlier cases, however (BWV 912a, 922, etc.), Bach had often introduced music structured around a

4 See the preface to Ulrich Leisinger’s edition J. S. Bach: Chromatische Fantasie und Fuge BWV 903 ¨ berlieferung der chromatischen Fantasie BWV (Vienna, 1999), and Uwe Wolf, ‘Fassungsgeschichte und U 903/1’, in M. Geck (ed.), Bachs Musik fu¨r Tasteninstrumente (Dortmund, 2003), pp. 145–58 (esp. 154). 5 J. N. Forkel, Ueber Johann Sebastian Bachs Leben, Kunst und Kunstwerke (Leipzig, 1802); repr. in BD VII, pp. 9–89 (see p. 69); Eng. trans. in NBR, pp. 417–82 (see 468). 6 Including the ﬁrst three of the sources listed here. 7 BD VII, p. 30; NBR, p. 436.

the well-te mpered clavier i etc.

definite theme as a counterbalance to the improvisatory freedom that otherwise prevailed. This was no doubt the raison d’eˆtre for the fugue that follows the fantasia, whether or not it was part of the original conception—the different key signatures of fantasia and fugue (the one without flat, the other with) might be signs of separate origin. Certain aspects of the fugue seem to represent the opposite pole from free fantasy: the clearly articulated subject, with its sequential headmotive (a recurring feature of Bach’s Weimar fugues) and its inversion halfway; the use of a well-defined, regular countersubject; the substantial element of reprise, an import from concerto form; and the clear division of the modulatory phase into sharp-side and flat-side zones (as in the fugues from BWV 542 and 894, etc.). On the other hand, certain other features of the fugue make it seem a perfectly natural and consistent outcome of the fantasia that precedes it. The most obvious of these is, of course, the chromatic nature of the sequential headmotive. No less significant, however, is the tonally unstable character of the subject, its refusal to settle into a clearly defined key till after the halfway point. To this we must add the bold, unprepared 7th that bursts upon the scene at the answering entry of the subject (b. 9) and the inexactness of that answer due to the dotted rhythm that opens it, which is later taken up in an episodic sequence (bb. 72–5). Finally, during an E minor subject entry (b. 90) the three-part fugal texture suddenly explodes into an eight-part dominant-9th chord (b. 94)—among the sharpest dissonances known to Bach—which recurs with climactic effect during the last stages of the fugue (bb. 135–9 and 158). If the Chromatic Fantasia in D minor represents the ne plus ultra of Bach’s freefantasy works for harpsichord, the Fantasia in G minor (BWV 542 no. 1) occupies a similar position among his organ works. The two works differ, however, in the role played by fugue. Whereas that of the D minor composition was either present from the outset or else added at a very early stage, there is no incontrovertible evidence that the pairing of fantasia and fugue in the G minor work goes back to Bach at all.8 The G minor Fantasia is very clearly articulated into five paragraphs as follows:

Bars: Material: Key:

1–9 a g

9–14 b c–g–d

14–25 a1 –g

25–31 b1 f–c–g

31–49 a2 –g

Unlike the Chromatic Fantasia, this composition incorporates a fully structured element within itself, namely the triple-counterpoint episodes b and b1, which alternate with writing in improvisatory style. The overall form is rondeau-like, not only in its contrasting episodes and (admittedly, very free) returns, a1 and a2, but also in its key structure, returning repeatedly to the tonic. In addition, whereas the Chromatic

8 The two movements clearly originated independently of each other (see n. 1); whether they were subsequently united by Bach himself is not known. Dietrich Kilian doubted the authenticity of the twomovement version; see his Krit. Bericht, NBA IV/5–6 (1978–9), p. 456.

p r e l u d e , f u g u e , a nd i n v e n t i o n

Fantasia is entirely through-composed, this work incorporates significant elements of reprise, even within its free passages. Thus paragraph 5 recapitulates much of paragraph 3 in reverse order and in different keys (bb. 36–8 = 21–3; 44–6 = 15–17). To a far greater extent than in the Chromatic Fantasia, then, improvisatory freedom is here checked and modified by structural restraint, which would be consistent with a later dating for this composition. The links that can be established with the harpsichord work are, however, no less obvious than the differences. Among them are the totally athematic character of the improvisatory-style passages, and the recurring ‘sigh’ figure, whether it takes the form of single appoggiaturas (D minor Fantasia) or multiple suspensions (G minor). Again, while only the harpsichord work is actually termed ‘chromatic’, the term might have been quite aptly applied to the organ work too: some of its most intense and mysterious passages of all are built on the basis of a chromatic ascent (pedals: bb. 20–3, 36–8; manuals: bb. 31–4). Among the most obvious resemblances between the two pieces are the startlingly abrupt, seemingly spontaneous modulations to unrelated keys, brought about by enharmonic change or by the changing tonal function of pivot notes. In stylistic terms the organ fantasia shares with its harpsichord counterpart two different species of improvisatory-style writing, namely passaggio (para. 1) and instrumental recitative (paras. 3 and 5). But the incorporation of these rhapsodic elements within the highly structured overall framework of the G minor Fantasia suggests, as has already been noted, a later date of origin than that of the D minor—perhaps Leipzig (1720s?) rather than Co¨then. Improvisatory freedom is now no longer possible except in conjunction with tight control.

Prelude, fugue, and invention Title

Earliest source/s

Scribe, date

Clavierbu¨chlein for W. F. Bach, 1720 New Haven, Yale Univ. J. S. and W. F. Bach, 1720–5/6 Das Wohltemperierte Clavier I BWV 846–69 Berlin, P 415 Autograph, 1722 Aufrichtige Anleitung (Inventions and Sinfonias), Berlin, P 610 Autograph, 1723 BWV 772–801

A new approach to keyboard music is clearly evident in Bach’s Co¨then and early Leipzig years. In their initial or early stages new compositions were often entered into small manuscript keyboard books called ‘Clavierbu¨chlein’ (equivalent, in name at least, to ‘Orgelbu¨chlein’), dedicated to Bach’s eldest son Wilhelm Friedemann or to his second wife Anna Magdalena. Family members were thus the first to benefit from Bach’s newest ideas. As his compositions developed, they would be copied by pupils from Bach’s immediate circle, such as Heinrich Nicolaus Gerber, who could then profit from them in their keyboard and composition studies. Finally, a definitive autograph fair copy would be produced, from which (at least indirectly) large

16 the we ll-temper ed clavier i etc. numbers of copies could be made, allowing dissemination of the work over a more extensive area. In its final form the work would consist of a standard set of six compositions, as in the suites or violin solos, or a multiple of six, as in the twentyfour Preludes and Fugues or the thirty Inventions and Sinfonias. Collecting together compositions in sets of six or more was, of course, customary at the time, but for Bach around 1720 it had a special significance: a new desire—no doubt linked to the arrival of full creative maturity—to be fully comprehensive and exhaustive in his approach to any keyboard form, whether it be dance suite, prelude and fugue, or the newly devised ‘invention’. The early exposure of family and pupils to these compositions is closely bound up with their very conception: they are designed not simply for pure delectation but as composition models and keyboard studies. These aims cannot be dissociated from each other: they are entirely integrated within the fabric of each composition. Among the first items in the Clavierbu¨chlein for W. F. Bach are five praeludia or praeambula, BWV 924, 926–8, and 930, all but one of which were entered by J. S. Bach in 1720,9 the year in which the book was dedicated to his son (the exception is BWV 927, which was entered by W. F. Bach in 1722/6). The first two preludes are numbered 1 and 2, and are in the keys of C major and D minor, which suggests that Bach might originally have planned a set of preludes in ascending key order. The five existing preludes, clearly designed specifically for the musical education of the young Wilhelm Friedemann, proceed from the simplest type, the arpeggiation of a chord sequence; hence the first two pieces, BWV 924 and 926, may be described as arpeggiated preludes.10 The other three preludes are also built on arpeggiated chords, but this material is now used thematically in sequence (BWV 927), motivically in imitation (BWV 930), or as the thematic material of a miniature ritornello design within an overall ABA1 structure (BWV 928). The preludes thus offer an instructive course of progressively increasing difficulty. That it was partly intended as a composition course is suggested by the three praeludia in the key order C, D, e (BWV 924a, 925, and 932) that W. F. Bach entered in the book in 1725/6 in imitation of his father. J. S. Bach’s preludes were clearly intended for keyboard instruction too, however, hence the ornamentation, which refers back to the ‘Explication’, a table of ornaments that Bach wrote out in imitation of D’Anglebert and Dieupart; hence, too, the four-bar cadenza in the D minor Prelude (BWV 926, bb. 39–42) and the fingering that Bach supplied throughout the G minor (BWV 930).

The chronology of the Clavierbu¨chlein, on which the dates given here are based, is the work of Wolfgang Plath; see his Krit. Bericht, NBA V/5 (1963), pp. 58–63. 10 A type cultivated not only by the middle-German composers Bo¨hm, Kuhnau, and Zachow but also by the South German J. C. F. Fischer, who termed it ‘praeludium harpeggiato’. Its historical development is traced by Dominik Sackmann in his ‘ “A la recherche du Pre´lude perdu”: Die Pra¨ludien im Wohltemperierten Klavier I und ihre Stellung in der Geschichte der Gattung’, in S. Rampe (ed.), Bach: Das Wohltemperierte Klavier I: Tradition, Entstehung, Funktion, Analyse (Munich and Salzburg, 2002), pp. 161–80.

p r el u d e , f u gu e , a n d i n v en t i o n

Some months later, probably in 1721, W. F. Bach, with the help of his father, started copying into the Clavierbu¨chlein a series of preludes that would eventually be incorporated in The Well-Tempered Clavier I (henceforth WTC I). A second series followed in 1722–3. The two series are as follows: Series I Praeludium no. 1 in C

WTC I, no. 1 in C

Praeludium no. 2 in C minor

WTC I, no. 2 in C minor

BWV 846a BWV 847

Praeludium no. 3 in D minor

WTC I, no. 6 in D minor

BWV 851

Praeludium no. 4 in D

WTC I, no. 5 in D

BWV 850

Praeludium no. 5 in E minor

WTC I, no. 10 in E minor

BWV 855a

Praeludium no. 6 in E

WTC I, no. 9 in E

BWV 854

Praeludium no. 7 in F

WTC I, no. 11 in F

BWV 856

Series II Praeludium no. [1] in C♯

WTC I, no. 3 in C♯

BWV 848

Praeludium no. [2] in C♯ minor

WTC I, no. 4 in C♯ minor

BWV 849

Praeludium no. [3] in E♭ minor

WTC I, no. 8 in E♭ minor

BWV 853

Praeludium no. [4] in F minor

WTC I, no. 12 in F minor

BWV 857

As shown, the keys of the first series are those of the diatonic tetrachord C–F, while those of the second series fill the chromatic gaps (except for E♭, which is missing). The versions of the first series are similar to those of the Forkel manuscripts, the chief sources of the early version of the WTC I, though slightly revised. The versions of the second series are similar to those of Bernhard Christian Kayser’s copy of the WTC I (Berlin, Mus. ms. Bach P 401), made in 1722–3, which represents the stage immediately before the autograph fair copy.11 The first series, like the five preludes of 1720, represent a progressive course of instruction in composition and keyboard technique. Again, Bach begins with the arpeggiated prelude, first in simple form (no. 1 in C), then with figured arpeggios (the so-called arpe`gement figure´ ; no. 2 in c); triplet arpeggios against a quaver bass (no. 3 in d); broken chords decorated by an ostinato motive in two-part texture with running treble and spaced-quaver bass (no. 4 in D), then the same with interchanged 11

The Forkel MSS are the complete copy from the estate of Franz Konwitschny, Leipzig, and the selection in Berlin, P 212. Regarding the early stage represented by these MSS, see Alfred Du¨rr, ‘Zur Fru¨hgeschichte des Wohltemperierten Klaviers I von J. S. Bach’, in Nachrichten der Akademie der Wissenschaften in Go¨ttingen, i. Philologisch-historische Klasse (Go¨ttingen, 1984), pp. 19ff. Anon. 5, the scribe of P 401, has recently been identiﬁed as the Bach pupil Bernhard Christian Kayser (1705–58). See Andrew Talle, ‘Nu¨rnberg, Darmstadt, ¨ berlieferung in der ersten Ha¨lfte des 18. Jahrhunderts’, BJ 89 (2003), Ko¨then— Neuerkenntnisse zur Bach-U pp. 143–72 (see esp. 155–72). See also BD V (2007), No. B240 a.

18 the well-tempere d clavier i etc. parts (no. 5 in e); thematic use of an arpeggio figure as the basis of a cantabile piece in three-part texture (no. 6 in E); and finally, motivic use of an arpeggio figure in exchanges between treble and bass (no. 7 in F). The first series of preludes, then, are not only numbered 1–7 and arranged in a logical key order, but they are also technically graded, both as keyboard and composition studies, and closely interrelated in style, theme, and motive. All this suggests that they might have been composed as a group. This is not to say that they were necessarily composed with W. F. Bach’s musical education in mind, as the earlier set of preludes (BWV 924, 926–8, and 930) obviously were. It is more likely that, around 1720–1, Bach was working on the beginnings of the WTC I and simultaneously assisting his young son, and that there was a substantial overlap between the two tasks. The second series of WTC preludes were entered in the Clavierbu¨chlein at a later stage (1722–3), when the WTC I was nearing completion. Here the child would learn to play in double counterpoint, with perfect equality of the two hands (no. 1 in C♯), and in cantabile style within a freistimmig (free-voiced) texture (nos. 2–4). It is notable that three of the four preludes have tonics on the black keys, in accordance with Werckmeister’s prediction that eventually musicians would be able to play equally ‘aus dem c oder cis’,12 which was certainly part of Bach’s achievement, if not part of his intention. In the end the Clavierbu¨chlein contained all twelve preludes from the first half of the WTC I, copied out by the son with his father’s assistance, with the exception of no. 7 in E♭, which was no doubt felt to be excessively long and hard for the young Wilhelm Friedemann. Alongside other manuscripts the Clavierbu¨chlein offers certain hints as to how the WTC I might have evolved in its early stages. The Clavierbu¨chlein, B. C. Kayser’s copy (P 401), and J. G. Walther’s copy (P 1074), taken together, suggest that Bach might have originally composed a series of preludes (and fugues?) in the diatonic key order C c, d D, e E, f F, g G, a A,13 subsequently filling in the gaps to create a fully chromatic series. This theory is supported by the later date of the second series of WTC preludes in the Clavierbu¨chlein and by the observation that there would have been room there for between seven and ten additional preludes.14 The early versions of Preludes 1–15 in the Forkel manuscripts, seven of which recur in the Clavierbu¨chlein, are mostly a good deal shorter than the definitive versions, whereas the earlier and later versions of the fugues differ only in matters of detail. It is possible, then, that the WTC I might have been compiled from a collection of preludes in the keys C–G (or C–a) and that, in the first place, the fugues might have formed a separate collection.15 Gaps might have been filled not only by composing new pieces ad hoc but by adapting existing pieces. There

Quoted by David Ledbetter, Bach’s Well-Tempered Clavier (New Haven and London, 2002), p. 156. Klaus Hofmann posits an original series of eight arpeggiated preludes in the key series C–c–d–D–e–F– ¨ berlegungen zur Fru¨hG–a; see his ‘Die Klangﬂa¨chenpra¨ludien des “Wohltemperierten Klaviers” I: U geschichte der Sammlung’, in M. Staehelin (ed.), Die Zeit, die Tag und Jahre macht: Zur Chronologie des Schaffens von J. S. Bach (Go¨ttingen, 2001), pp. 157–68 (esp. 159–60). 14 See Plath, Krit. Bericht, NBA V/5, pp. 60–1. 15 Hofmann also holds this view; see ‘Die Klangﬂa¨chenpra¨ludien’, p. 168. 13

p r el u d e , f u gu e , a nd i n v en t i o n

is some evidence in the sources that the more remote keys might have been catered for by transposition. Praeludium et Fuga 8 in e♭/d♯, for example, might have been transposed from e/d, which would imply separate origin for prelude and fugue; no. 18 might have been transposed from g to g♯, and no. 24 from c or g to b. There are signs in the sources that some preludes and fugues originally had the old modal key signatures (one flat or sharp fewer than today’s signatures); for example, in Kayser’s copy Fuga 2 in C minor has a key signature of two flats. In addition, among the early sources the title ‘Pre´lude’ is found, as well as ‘Praeludium’, and ‘Fughetta’ in place of ‘Fuga’. On the evidence of Kayser’s copy the WTC I was probably compiled by collecting together separate bifolios, each of which would have contained a single prelude-andfugue pair under its own title. These bifolios would have been combined to form a composite manuscript (now lost), within which folios could be inserted or replaced at will, serving as a vehicle for the compilation process in much the same way as the British Library autograph did for the WTC II about twenty years later. The existing autograph fair copy of Part I (P 415) was probably begun in late 1722, when the compilation process and the main revision of the text were complete. The object of this manuscript was clearly to bring the work into its definitive form. A further revision of the text was undertaken. The key order became fully chromatic, with major preceding minor throughout. Modern key signatures were invariably used. Individual titles throughout took the form ‘Praeludium 1’ or ‘Fuga 1’ and so on. The word ‘fine’ now occurs only at the end of the whole collection, not after each prelude-and-fugue pair, as it did originally. Finally, after ‘fine’ Bach writes ‘SDG’, soli Deo gloria, his customary sign of completion. The elaborate ornamental title page of the autograph fair copy reads: Das Wohltemperirte Clavier oder Praeludia und Fugen durch alle Tone und Semitonia, so wohl tertiam majorem oder Ut Re Mi anlangend, als auch tertiam minorem oder Re Mi Fa betreffend. Zum Nutzen und Gebrauch der Lehr-begierigen Musicalischen Jugend, als auch derer in diesem studio schon habil seyenden besonderem Zeit Vertreib auffgesetzet und verfertiget von Johann Sebastian Bach p. t. Hoch Fu¨rstlich Anhalt-Co¨thenischen Capellmeistern und Directore derer Cammer Musiquen. Anno 1722. (The Well-Tempered Clavier, or Preludes and Fugues through all the tones and semitones, both as regards the tertia major or Ut Re Mi and as concerns the tertia minor or Re Mi Fa. For the use and profit of the musical youth desirous of learning, as well as for the pastime of those already skilled in this study, drawn up and written by Johann Sebastian Bach, p. t. Capellmeister to His Serene Highness the Prince of Anhalt-Co¨then and director of his chamber music. In the year 1722.)

Bach’s circumlocutory terminology for the major and minor modes is borrowed from Kuhnau.16 ‘Clavier’ in this context most likely means simply ‘keyboard’, the

16 ¨ bung, Part I (1689) consists of seven partitas ‘aus dem Ut, Re, Mi, oder Tertia His Neuer Clavier U majore’; Part II (1692), of seven partitas ‘aus dem Re, Mi, Fa, oder Tertia minore’.

20 the well-tempere d clavier i etc. commonest meaning of the word at the time.17 In other words, Bach is deliberately non-prescriptive as to the type of instrument that should be used. This is in keeping with the restriction of the work to C–c3, the standard keyboard compass at the time, which made it as widely playable as possible on the instruments then in use. The epithet ‘wohltemperirte’ (well-tempered) was clearly borrowed from the leading contemporary authority Andreas Werckmeister, who frequently made use of it in his theoretical works. His treatise of 1691, for example, is entitled Musicalische Temperatur, oder deutlich und warer mathematischer Unterricht, wie man . . . ein Clavier . . . wohl temperirt stimmen ko¨nne (Musical Temperament, or clear and true mathematical instruction how to tune a keyboard well-tempered). Here, as elsewhere in Werckmeister’s writings, ‘wohl temperirt’ evidently means ‘appropriately tuned’. But in his later writings he increasingly advocated equal temperament on account of its unlimited possibilities of modulation, transposition, and enharmonic change.18 It is not necessary, however, to adopt entirely equal temperament in order to play in all keys. And many theorists of Bach’s day, such as Neidhardt, Mattheson, Sorge, Marpurg, and Kirnberger, advocated a slight deviation from absolute equality in order that the distinctive colourings of different keys could be maintained. It may well be that something approaching equal temperament, but subtly nuanced in this way, was what Bach had in mind.19 Or else he might have meant by ‘wohltemperirte’: use whatever temperament you ﬁnd appropriate for playing music in all keys. The didactic purpose of the work is clear from the words ‘for the use and proﬁt of the musical youth desirous of learning’. And indeed for Bach’s pupils it became the prime vehicle for advanced study in both keyboard playing and composition. The work was also intended for pure delectation, however, as is clear from the words ‘as well as for the pastime of those already skilled in this study’. By including a prelude and fugue in every one of the twenty-four major and minor keys, Bach gave a practical demonstration of the full range of the tonal system. There were at least partial precedents, of course, of which only the most prominent can be mentioned here. Since the traditional function of the prelude was to establish the mode of the work that followed, each prelude in sixteenth- and early seventeenthcentury published collections tended to be in a different mode. By the late seventeenth century this procedure was applied to keys rather than modes. For example, the Tabulatura 12 Praeambulorum . . . durch alle Claves und Tonos auff Clavichordien und Spinetten zu gebrauchen (Tablature of 12 Praeambula through all the keys and tones, to be used on clavichords and spinets) of 1682 by the Dresden court organist Johann

17 It could, however, refer to stringed keyboard instruments as opposed to the organ. Bach, for example, in recommending G. G. Wagner as cantor at Plauen, declared that ‘he plays the organ and clavier well’ (‘spielet er eine gute Orgel und Clavier’); see BD I, No. 15, and NBR, No. 124. 18 See Rudolf Rasch, ‘Does “Well-Tempered” Mean “Equal-Tempered”?’, in P. Williams (ed.), Bach, Handel, Scarlatti: Tercentenary Essays (Cambridge, 1985), pp. 293–308. 19 See Ledbetter’s balanced discussion of these issues in his Bach’s Well-Tempered Clavier, pp. 30–50.

p r e l u d e , f u g u e , a nd in v e nt i o n

Heinrich Kittel20 contains one prelude in each of the twelve most common major and minor keys, those with up to three sharps or flats. An even greater range of keys was occasionally in use at that time—for example, in the seventeen anonymous suites of 1683, formerly attributed to Pachelbel,21 whose keys include a major and/or a minor on every degree of the chromatic scale. By the turn of the century it was possible to list all twenty-four keys, with modern key signatures and in fully chromatic order, as did the organist T. B. Janovka in his influential treatise Clavis ad thesaurum magnae artis musicae (Prague, 1701).22 Since this work was known to Johann Bernhard Bach and Johann Gottfried Walther, it is quite possible that Bach was acquainted with it. The nearest precedent to the WTC I23 was published in the following year, 1702: Ariadne musica, a collection of twenty preludes and fugues in nineteen different keys by the South German composer Johann Caspar Ferdinand Fischer. Here only five remote keys are missing from the complete cycle of twenty-four: g♯, b♭, e♭, F♯, and C♯. Certain rather conservative, seventeenth-century features of the collection, however, distance it from the WTC I. Fischer’s preludes and fugues are tiny miniatures, reflecting the South German verset tradition; and the frequent ‘modal’ key signatures24 show that he was often thinking in terms of transposed modes rather than modern keys. Despite these antiquated features, there is no doubt that Bach was acquainted with the work and that it exerted a powerful influence upon the conception and composition of the WTC I. Although Ariadne first appeared in 1702, as already noted, it is known today only from a later edition (Augsburg, 1715). Bach, too, might have known only this 1715 edition. That would account for the absence of any trace of the influence of Ariadne on Bach’s keyboard music prior to 1715. In addition, it would square with the likely date of the WTC’s conception—some time within the period 1715–20. The overall structure of the two works is remarkably similar: a ‘Praeludium et Fuga’ in all the major and minor keys (or nearly all, in Fischer’s case), chromatically ordered from C to b. Fischer places minor before major throughout, a relic of modal theory that recurs in the early stages of Bach’s work on the WTC I. Later, we shall have occasion to notice how often even the substance of Fischer’s preludes and fugues resonates in the later work. During the period when Bach became acquainted with Fischer’s Ariadne musica, he must have heard of the public, protracted, and heated dispute between Johann Mattheson of Hamburg and Johann Heinrich Buttstedt of Erfurt over the relative

See Willi Apel, Geschichte der Orgel- und Klaviermusik bis 1700 (Kassel, 1967); trans. and rev. by H. Tischler as The History of Keyboard Music to 1700 (Bloomington and Indianapolis, 1972), p. 649. 21 See M. Seiffert (ed.), Johann Pachelbel: Klavierwerke, DTB II.1 (1901). 22 See Ledbetter, Bach’s Well-Tempered Clavier, pp. 110 and 357 n. 15. 23 Thomas Synofzik lists further precedents in German prelude collections of 1639–1722, whose composers include C. Michael, J. E. Kindermann, W. Fabricius, G. E. Pestel, J. Krieger, and J. Pachelbel. The last-named composer’s Fugen und Praeambulen u¨ber die gewo¨hnlichsten tonos ﬁguratos of 1704 is unfortunately lost. See T. Synofzik, ‘ “Fili Ariadnaei”: Entwicklungslinien zum Wohltemperierten Klavier’, in Rampe (ed.), Bach: Das Wohltemperierte Klavier I, pp. 109–46. 24 As already noted, traces of these are still found in the early developmental stages of the WTC I.

the well-t empered clavier i et c.

virtues of conservative and progressive approaches to music and its education. In his treatise Das neu-Ero¨ffnete Orchestre (Hamburg, 1713) Mattheson dismissed the old system of solmization, based on modes and hexachords, in favour of the modern tonal system, with its twenty-four major and minor keys. Buttstedt replied in the treatise Ut, mi, sol, re, fa, la: tota musica et harmonia aeterna (Erfurt, 1715), in which he subscribed to the opposing, conservative viewpoint, advocating the continued validity of the old system. In the WTC I Bach makes a resounding statement in favour of the progressive approach, embracing the ‘modern’ tonal system in its totality, complete with chromatic key order and modern key signatures. To that extent he can be seen as siding with Mattheson in the dispute. On the other hand, in the title page he uses solmization syllables—‘Ut, re, mi’ and ‘Re, mi, fa’—for major and minor keys instead of the ‘dur’ and ‘moll’ that were already in widespread use. Moreover, in Mattheson’s Orchestre of 1713, fugue is regarded as an honourable but antiquated genre, of little interest to the ‘galant homme’. Bach, for his part, shows no inclination whatsoever to reject learned counterpoint—the very mainstay of his art—in favour of the galant, ‘natural’, singable, treble-dominated style that Mattheson advocated. As far as tonality is concerned, however, Mattheson and Bach were at that time thinking alike, as is clear from the Hamburg composer’s Exemplarische Organisten-Probe (Hamburg, 1719), the first publication to contain pieces in all twenty-four keys, even though they are only figured-bass exercises. Before that, in 1711, another prominent theorist-composer, Johann David Heinichen, had already given the first description and illustration of the circle of 5ths, showing the relationship between all keys, major and minor, in his Neu erfundene Anweisung des General-Basses (Hamburg, 1711).25 In 1722, the very year in which the WTC I was completed, two highly relevant, though quite independent, publications appeared: the Labyrinthus Musicus, bestehend in einer Fantasia durch alle Tonos, nehmlich durch 12 duros und 12 molles, zusammen 24 Tonos by the Dresden composer Friedrich Suppig;26 and Jean-Philippe Rameau’s Traite´ de l’harmonie, which for the first time gave a fully systematic account of the functional harmony that lies at the heart of the new tonal system. There is no question of direct influence here, but these works illustrate how an exhaustive approach to harmony and tonality was in the air at the time, as does L’ABC musical (lost; date unknown) by the Halle organist Gottfried Kirchhoff, which is said to have contained preludes and partimento fugues in all keys.27 In the early version of the WTC I, as represented by the Konwitschny manuscript, there is an obvious mismatch between the short, relatively straightforward preludes— many of them used for the young Wilhelm Friedemann’s studies—and the extended, often complex fugues that follow. At the very beginning of the collection, for example, the elementary prelude in C is coupled with an advanced, difficult stretto fugue, and

25 26 27

Subsequently reworked in his celebrated treatise Der General-Bass in der Composition (Dresden, 1728). Facsimile edn, ed. R. Rasch (Utrecht, 1990). It is described by F. W. Marpurg in his Abhandlung von der Fuge, vol. i (Berlin, 1753), pp. 149–50.

p r e l u d e , f u g u e , a n d i n ve n t i o n

such discrepancies are frequent thereafter. Observations of this kind are one of the main arguments for the theory, broached earlier, that the preludes and fugues originated as independent collections. It might have been Bach’s realization of their disproportionate size and weight that led to his expansion of the first fifteen preludes. Whether or not he also intended to expand the remaining nine preludes but ran out of time,28 perhaps due to preparations for the move to Leipzig, we do not know. At any rate, Bach’s objective seems to have been to bring the preludes into line with the fugues in terms of weight, substance, and dimensions. Historically, of course, the preparatory function of the prelude rendered it subsidiary to the movement that followed; and this is, to some extent, reflected in Bach’s earlier keyboard works and in his initial conception of the WTC preludes. Moreover, in many cases the preludes had to be elementary enough to serve as studies for the young Wilhelm Friedemann. But when bringing the work into its definitive form, Bach was no longer guided by such constraints. What he sought now was something largely new: a substantial measure of equality between the relatively free preludial movement and the stricter, more rigorous fugal movement that followed. Bach’s mode of procedure in these prelude expansions is instructive. Typically, he inserts a new period shortly before the final cadence, often by interrupting an existing cadence (c♯, e♭). In several cases the new period is a tonic or subdominant reprise of the main theme, rounding the movement off and thereby giving it a more finished shape (D, f). Often Bach incorporates into the prelude an entirely new, toccata-like mode of treatment, characterized by idiomatic figuration, sudden tempo changes to Adagio or Presto, rapid runs intermingled with spread chords, and so forth (c, C♯, e). In these cases Bach enhances the brilliant, virtuoso aspect of the preludes, raising them to the level of the fugues in terms of technical demands, as well as in weight and dimensions. The preludes of the WTC I may be conveniently divided into three overall types: arpeggiated/figural, cantabile, and contrapuntal. The simplest and closest to improvisation are the arpeggiated (C, c) and figural preludes (D, d, e, G, B♭)—the latter may be viewed as a rather more elaborate derivative of the arpeggiated type. Both have been discussed, to some extent, in connection with the Clavierbu¨chlein for Wilhelm Friedemann Bach. The cantabile preludes (c♯, e♭, f, g, b♭) are pathos-laden pieces in minor keys, mostly with four to six sharps or flats. The first three were, as we have seen, used in the later stages of the young Wilhelm Friedemann’s musical education. All five have in common a dense, free-voiced texture. Although the treble is the leading part, it often engages in dialogue with the tenor or bass, and occasionally even the alto takes the chief thematic role (b♭, bb. 10–12). The texture is thus a highly flexible one, in which all parts contribute to the motivic/thematic development, as well as helping to provide harmonic support. Although all five preludes merit the 28 See Alfred Du¨rr’s discussion of this issue in Krit. Bericht, NBA V/6.1 (1989), pp. 191–2, and in his Johann Sebastian Bach: Das Wohltemperierte Klavier (Kassel, 1998), p. 68.

the well-tempered clavier i etc.

epithet ‘cantabile’, only those in c♯ and e♭ are truly melodic in the nature of their thematic material: in both cases, melodic phrases in a slow dance rhythm are built up into large melodic periods. The thematic material of the other three preludes is essentially harmonic. That in f might be viewed as a cantabile equivalent of the arpeggiated prelude: figured arpeggios of great beauty are treated motivically and furnish the decorative surface of the chord progressions. In the g prelude, chord sequences are decorated using the well-known diminution technique of filling in 3rds; and the sequential theme of the b♭ prelude has the function, at each occurrence, of leading a richly dissonant series of harmonies. Six of the preludes (C♯, F, F♯, f♯, A♭, a) are structured to a large extent in two-part counterpoint, though most of them contain a certain measure of free-voiced writing too. The shortest and simplest are those in F and a, which are throughout built out of a single half-bar or whole-bar motive. This dominant motive is worked into a theme, which is heard several times with interchanged parts and in different keys—I, vi, I (F, bb. 1, 6b, 16b) or i, v, III, i (a, bb. 1, 5, 13, 22). And the same motive forms the substance of recurring, modulatory episodes. The structure that arises—fixed-key thematic statements alternating with modulating episodes—is common to ritornello and fugue, and by this time had become second nature to Bach. Double counterpoint, an essential technique of two-part writing, forms the mainstay of the much larger Praeludium in C♯. The main theme is repeatedly heard in double counterpoint (bb. 1–31, 47–62), as is the derived subsidiary theme (bb. 32–46). The toccata-style paragraph at bb. 63–98 was, as we have seen, interpolated at an intermediate stage of revision.29 Its broken-chordal figuration, clearly derived from the original theme, reminds us that, for all the contrapuntal treatment of the theme, it is at root no more than a series of arpeggiated chords. No less elaborate contrapuntally is the Praeludium in F♯ minor, which takes the form of a miniature fugue. The brief, rolling subject is accompanied by three different counterpoints in turn (bb. 1, 2, and 4), each of which plays a prominent role later on. Bach’s habitual thinking in terms of broad paragraphs, articulated by strongly marked cadences, can by this stage even break the continuities of fugue (cf. bb. 12–14 and 21b–24). The remaining preludes in two-part texture, those in F♯ and A♭, owe more to ritornello form than to fugue. That in F♯ takes the form of a series of variations on the opening ritornello-like period, which consists of a broken-chordal headmotive, or Vordersatz (b. 1), a sequential continuation, or Fortspinnung (bb. 2–4), in continuous syncopation, and a charming, decorative cadential theme, or Epilog (bb. 5–6). The Praeludium in A♭ is concerto-like in a rather different manner. The triadic opening theme, underpinned by a I–V–I progression, recalls those of two ritornello-based organ pieces from Bach’s Weimar years, the Praeludium in G, BWV 541 no. 1, and the

29 The MS copies of B. C. Kayser and H. N. Gerber indicate that the 36 new bars had already been added to the lost autograph draft.

p r e l u d e , f u g u e , a n d i n ve n t i o n

Toccata in F, BWV 540 no. 1.30 Accordingly, it functions like a ritornello, alternating with a modulatory episode. The overall form could hardly be more lucid (Rit. = Ritornello): Bars: Section: Key:

1–8 Rit. I

9–17 Epis. I–V

18–21 Rit. V

22–34 Epis. V–IV–I

35–44 Rit. I

Five preludes are structured in three-part counterpoint (E, g♯, A, B, b), though only in one case (A) is a pure three-part texture maintained throughout. The Praeludium in E is monomotivic (like those in F and a), but here the dominant motive—b. 1, first six treble notes—forms the chief constituent of a cantabile theme in the rhythm of a pastorale. In keeping with its lyrical character, the movement falls into a tripartite reprise form (ABA1) with subdominant reprise. In the figural Praeludium in B, the dominant motive already plays a sequential role within the main theme itself, which retains its integrity, recurring several times with interchanged parts (bb. 6 and 17). Again, the movement is articulated by clear cadences (in keys V, vi, and I at bb. 5– 6, 10, and 14–15), the third rhyming with the first. No less monomotivic is the Praeludium in G♯ minor, whose prevailing motive opens the brief subject (b. 1). The term ‘subject’ is here used advisedly, since the structure of the piece is pseudofugal, with four expositions (bb. 1, 5, 13, 27) alternating with three episodes (bb. 3, 11, and 19). The expositions, however, open with imitation at the octave rather than at the 5th, which, strictly speaking, places them outside fugue. In the Praeludium in B minor, a specific point is made of the three-part texture. The parts are no longer equal but form two clearly distinct layers—a pair of contrapuntally interacting upper parts and a bass in continuous quavers. The analogy with the Corellian trio sonata is clear; and the prelude’s binary dance form with repeats (the only such case in the WTC I) suggests that Bach had in mind the sonata da camera rather than da chiesa.31 The various imitative points are all interrelated. The Praeludium in A is still closer to fugue than that in G♯ minor, since the subject is answered at the 5th (b. 4). Virtually the entire piece is constructed in triple counterpoint: the subject is combined with two regular countersubjects (bb. 1–3 a), and this combination is stated six times (bb. 1, 4, 8b, 12, 17b, 20) and in four of its six possible permutations. If the Praeludium in A imitates fugue in miniature, that in E♭ does so on a massive scale, exceeding all other preludes in the collection in length and complexity, and threatening to overwhelm the concise fugue that follows—the balance between prelude and fugue that Bach sought elsewhere is here seriously undermined. The movement may be viewed as a late offshoot of the multi-sectional praeludium that

See Vol. I of the present study, p. 182, Ex. 1. A possible model might have been the Preludio from Corelli’s Sonate a tre, Op. 4 No. 2 (Rome, 1694). The movement is quoted extensively in connection with Bach’s Praeludium in B minor by Ledbetter, Bach’s Well-Tempered Clavier, p. 230. 31

the well-te mpered clavier i et c.

Bach cultivated in his youthful days under the influence of Buxtehude.32 This is not necessarily a sign of early date, however, for the Toccata in E minor of 1725 (BWV 830 no. 1) belongs to the same tradition. Bach takes a standard four-voice doublefugue scheme—exposition of S[ubject] I, exposition of S II, exposition of S I + II combined—and applies it with an informality appropriate to the prelude. Thus the texture, though essentially four-part, is often free-voiced. At the outset, the old preludial principle of building up from improvisatory beginnings to consolidated thematic statements is brought into play. The two opening paragraphs (bb. 1 and 10) are not fugal expositions but freely imitative passages, based on brief motives that will eventually form the chief constituents of the combined fugue subjects at b. 25. Despite the abundant use of contrapuntal devices after the initial exposition (bb. 25–34)— stretto (bb. 35–49a) and double counterpoint at the 10th and 12th (bb. 49b–70)—an unmistakable air of informality hangs over the entire movement. And the manner in which the opening semiquaver motive saturates the entire texture creates a monomotivic impression similar to that of numerous other preludes in the collection. Altogether, the movement represents a radical revaluation of the old multi-sectional praeludium and a new and remarkable fusion of the prelude and fugue genres. The fugues of the WTC I are greatly varied in terms of the contrapuntal procedures employed: stretto fugue (C, b♭), inversion fugue with stretto (d, d♯, a); double counterpoint (E♭, E, e, A♭, b); the same with stretto (F, g), with inversion (f♯, B), or with both stretto and inversion (G); triple counterpoint (c, C♯, g♯, B♭, F♯); quadruple counterpoint (f); and finally double and triple fugue (A and c♯ respectively). Only the French overture-style Fuga in D is entirely devoid of such artifice. Although Bach’s pupils must have learnt much from these various modes of fugal treatment, there is no sense in which the collection is designed as a fugal textbook. Each individual composition, in its finished form at least, is not a fugue or prelude but a ‘Praeludium et Fuga’. Moreover, no attempt is made to grade the fugues technically (unlike the preludes); for example, as already noted, the first fugue (C) is a complex and difficult stretto fugue, and the second (c) a study in triple counterpoint. Unlike The Art of Fugue, the WTC I is completely unsystematic in its employment of contrapuntal devices. In all probability, Bach first invented the subject and countersubjects for a fugue in a particular key. The nature of these themes would then determine which contrapuntal procedures were to be employed, as well as the overall character of the fugue. Thus contrapuntal artifice is not an end in itself but merely one aspect of the exhaustive treatment of a specific theme and its counter-themes. Due to this strictly thematic quality, in conjunction with the contrapuntal procedures employed, all the fugues of the WTC I (with the possible exception of the relatively informal Fuga in D) qualify for inclusion in Friedrich Wilhelm Marpurg’s category of ‘strict fugue’—a category that he particularly associates with Bach—as opposed to the

See Vol. I, pp. 51–4.

prelude, fugue, and invention

‘free fugue’ cultivated by Handel.33 Strict fugue was by this time Bach’s standard mode of thought in contrapuntal writing. For further elucidation in any particular case, we have to ask what implications of the subject are realized and what form that realization takes. In the Fuga in D, the second of the two paragraphs (b. 17) contains no further entries of the French-style subject, but rather much imitative and sequential play on its headmotive. Bach’s overriding concern here seems to have been the recreation of a particular style—that of the French overture. Something similar might be said of the five-part fugues in c♯ and b♭, both of which recreate the traditional alla breve style of the seventeenth century. But that style had long been associated with strict counterpoint, whereas in the French overture only the fast middle section was (freely) fugal. Bach, however, paradoxically invents a fugue subject in the style of the dottedrhythm introduction, which was never fugal in the overtures of Lully or his followers. The fugues in C and b♭—significantly one from each half (Bach seems to have taken care to balance the two halves of the collection)—may be described as ‘stretto fugues’, since the device of stretto, or overlapping subject entries—one of the most potent tools in Bach’s fugal armoury—is employed throughout. The seven strettos of the Fuga in C, which succeed one another without intervening episode (bb. 7, 10b, 14, 16, 19, 21, 24), render it the most densely thematic fugue in the entire collection. It is, however, articulated by clear cadences in different keys (V, vi, ii, I) between the stretto expositions. In the Fuga in B♭ minor, the entries of the striking question–answer subject, with its pathos-filled minor 9th at the watershed, already overlap in stretto from the outset. After much modulation to related keys, the tonic return is marked by subject entries in three octaves (bb. 46–50). Finally, a half-bar stretto descending through all five parts (bb. 67–72), already adumbrated earlier (bb. 50–4), forms a magnificent peroration. A third stretto fugue, that in d, is diversified by two additional features: inversion of the subject (from b. 12) and the use of a regular countersubject (treble, bb. 3–5). The overall form is bipartite, as in the fugues in C and b♭, but on this occasion there is a strong element of reprise (A A1; bb. 21–42 = 1–20): the second half is a varied repeat of the first with interchanged parts, tonally adjusted so as to remain in the tonic. It can hardly be mere chance that a similar reprise form—a strikingly novel and ‘modern’ way of handling fugue—occurs in the fugues in c and e. A possible explanation might be that one of Bach’s first tasks in compiling the fugues of the WTC I was to compose a diatonically ordered series in the minor keys of c, d, and e. The fugues in c and e belong to a series of compact fugues in which one or two regular countersubjects are combined with the subject in double or triple counterpoint: those in the keys of c, E♭, E, e, F♯, A♭, B♭. In the Fuga in E minor, the only two-part fugue in the whole collection, the counterpoint of subject entries and episodes alike is inverted, so that the second half of the bipartite structure 33 Marpurg makes this fundamental distinction at the outset of his Abhandlung von der Fuge, vol. i; see BD III, No. 655 (p. 25), and NBR, No. 351.

the well-tem pe red clavie r i etc.

(bb. 20–38) represents a tonally modified double-counterpoint inversion of the first half. The Fuga in C minor has a clear counterpart in that in B♭ from the second half of the collection. Not only are both in three voices, but in both cases the main substance of the discourse is the triple counterpoint of three themes—subject and two countersubjects—which is heard in four (B♭) or five (c) of its six possible permutations. Another pair of fugues from either side of the central divide, those in F and g, not only possess a single countersubject each but are both enhanced by stretto. In the three-part Fuga in F, an octave stretto at two bars is heard on three occasions and in three different keys, I, vi, and ii (bb. 26, 37, and 47). It takes over from the countersubject, which drops out after b. 30, as the chief contrapuntal interest of a fugue that is greatly enlivened by its dance rhythm—that of a passepied. The subject appears to be indebted to that in the same key from Fischer’s Ariadne musica, which is in a different sort of dance rhythm (Ex. 1). Bach’s Fuga in G minor perhaps owes still more to another fugue from Fischer’s collection, that in E♭. Here the resemblance is not confined to the subject itself (Ex. 2) but extends to the contrapuntal structure: Bach’s culminating strettos at the octave and half-bar, the first in two voices (bb. 17–18), the second in three (bb. 28–9), are clearly modelled on Fischer’s fugue, as is the tenor entry that leads directly into the final cadence. Two four-part fugues, those in f♯ and B—again, one from each half of the collection—are concerned not only with the combination of subject and countersubject but with the melodic inversion of the subject itself. In the middle paragraph of the Fuga in F♯ minor, the inverted subject (alto, b. 20) is answered by the direct subject (treble, b. 25). The procedure is then reversed—direct tenor entry (b. 29) answered by inverted bass (b. 32). The inverted subject is clearly anticipated by the countersubject (cf. bb. 4b–5a and 20b–21), which itself forms a free inversion of the subject. The three-note headmotive of the subject, twice intensified in a freely sequential continuation (bb. 1–2), is then inverted in the countersubject and clothed with a characteristic anticipatory-note figure (bb. 4b–5 a), often used by Bach in moods of great pathos. Hence this fugue might justly be described as a fuga pathetica. A similar description might be applied to a series of large-format fugues in the minor mode that form the culmination of each group of four preludes and fugues—

Ex. 1

a) Subject of Fuga in F, BWV 856 no. 2

b) J. C. F. Fischer, subject of Fuga in F from Ariadne musica

prelude, fugue, and invention

Ex. 2

a) Subject of Fuga in G minor, BWV 861 no. 2

b) J. C. F. Fischer, subject of Fuga in E♭ from Ariadne musica those in c♯, d♯, f, [g], a, and b. It is possible that these fugues were deliberately composed on a large scale with their position in mind. The Fuga in C♯ minor corresponds to a movement outside this scheme, however, the B♭ minor fugue, in its five-part texture and traditional alla breve style. In each case, moreover, the rising interval at the midpoint of the subject—a diminished 4th (c♯) or minor 9th (b♭)— lends great emotive power to the otherwise formal subject and generates much of the pathos of the ensuing discourse.34 Unlike the B♭ minor fugue, however, the C♯ minor is designed as a triple fugue—the only one in the collection—and its tripartite form (ABC) is dictated by the fugal structure. The triple counterpoint of the three subjects is heard in five of its six possible permutations. One of the traditional formalities of triple fugue, however—the separate exposition of each new subject before its combination with the original subject—is not observed here, presumably because it would have made the fugue inordinately long. The large-format fugues in d♯ and a, which occupy equivalent positions in the two halves of the collection (nos. 8 and 20), are both stretto-inversion fugues, a type that Bach employed on a much smaller scale in the D minor fugue. In both cases an overall three-phase form is dictated by the fugal structure (as in the C♯ minor fugue). The Fuga in D♯ minor owes its highly expressive, cantabile character to the stepwise quavers of the subject, which, in one form or another, furnish most of the incidental figure-work throughout. The theme is subjected not only to stretto and inversion but also to variation and augmentation. A significant role is played by a rhythmic variant of the subject, which occurs once within each of the three phases (b. 24, middle part; b. 48, treble; b. 77, middle). Among Bach’s fugal resources, subject variation of this kind is among the most frequently overlooked. The concluding phase takes the form of a stretto exposition on the augmented subject, which is heard three times (bb. 62, 67, and 77), once in each voice (bass, alto, soprano), combined with two or more entries of the subject in standard note-values. The four-part Fuga in A minor is not dissimilar in structure, though treatment of the subject is here confined to stretto and

34 Ledbetter, in Bach’s Well-Tempered Clavier, pp. 165–7, makes illuminating comparisons between Bach’s C♯ minor subject and Corelli’s Concerti grossi, Op. 6 No. 3, Grave (Amsterdam, 1714), and Georg Muffat’s Armonico tributo (Salzburg, 1682), Sonata No. 2, Grave.

30 the we ll-temper ed clavier i etc. inversion. Despite the great strength of the material, however, there is something pedantic—a fault from which Bach was not always entirely free—about the sheer thoroughness and exhaustiveness of the stretto expositions; and the reprise of an old stretto at bb. 73–5 (cf. bb. 53–5) seems redundant. As a result, the sense of impetus is sometimes lost—something that never happens in the sister-fugue in D♯ minor. Each half of the work culminates in a large-format, four-part fugue—those in f and b—based on a chromatic subject. The chromaticism of the F minor subject is incomplete, as befits the halfway point; that of the B minor subject, containing all twelve semitones, is complete. It is quite possible that Bach intended these subjects to symbolize the chromatic key order of the whole collection. A different aspect of the F minor subject links it with that of the G♯ minor fugue. The subjects of both fugues are ‘interrupted’ in the middle by a rising semitone at a higher pitch, which has the effect of an interpolation from a different voice, breaking the otherwise even, stepwise flow of its surroundings (Ex. 3). The Fuga in G♯ minor, however, is in Bach’s most refined, up-to-date contrapuntal style—the unusually regular phrase structure is not the least ‘modern’ of its attributes—whereas the Fuga in F minor is characterized by relatively antiquated features: the severe cantus firmus-like soggetto, in the style of the seventeenth-century ricercar or fantasia, and the complementary semiquaver figures of the countersubjects, a common feature of the partitas of Bach’s predecessors. The grand, but intricately wrought Fuga in B minor possesses only one regular countersubject and no special contrapuntal device beyond invertible counterpoint. Yet it stands as one of the most richly inventive and powerfully expressive fugues in the entire collection. This is due partly to the fertile harmonic implications of the chromatic subject, but also to the exceptionally rich figure-work. This is derived from the sequential, semitonal figure of the subject and from two semiquaver figures, both extracted from the countersubject (bb. 4 and 6b). The second of these forms the chief material of the episodes, together with an unrelated, diatonic trio formulation (bb. 17–18)—an oasis of calm that recalls the trio-sonata texture of the preceding prelude. The collection also includes three large-format fugues in a lighter tone, partly due to their three-part texture and major mode—those in C♯, G, and A. In the Fuga in C♯, which falls into an overall ABA1 reprise form, it is the middle paragraph (bb. 23–42) that is chiefly responsible for the movement’s large dimensions. It consists, in the main, of a very long episode in concerto style, not dissimilar to some of the harpsichord writing in the first movement of Brandenburg Concerto No. 5. The triple counterpoint of the exposition belongs to a type that became classic for Bach: lively, figural subject; first countersubject in sequential semiquavers (derived from the turn figure of the subject); and second countersubject in long notes that form suspensions. The brilliant episodes of the Fuga in G, with their pseudo-cross-string figures, often combined with rising or falling scales, are still more reminiscent of the fifth Brandenburg Concerto. Both in terms of inventive profusion and fugal artifice, this is undoubtedly one of the most remarkable fugues in the collection. Like those in d♯

p r e l u d e , f u g u e , a nd i n v e n t i o n

Ex. 3

a) Subject of Fuga in F minor, BWV 857 no. 2

b) Subject of Fuga in G♯ minor, BWV 863 no. 2 and a, it is a stretto-inversion fugue, but a substantial contribution is also made by a regular countersubject. The Fuga in A, a ‘subtle and complex scherzo’,35 might be described as a very free double fugue. It has certain features in common with the triple fugue in C♯ minor: the tripartite design (ABC); the entry of a new countersubject in flowing quavers (or semiquavers) in the middle paragraph; and the dropping-out of this theme in the concluding paragraph. But it is as free as the C♯ minor fugue is strict. The new countersubject is heard in full only twice (bb. 23–4 and 27–8), after which Bach is content with mere informal allusions to it. And the subject itself is absent from all but the first four bars of the concluding paragraph, which in this respect recalls the similarly free conclusion of the Fuga in D. The Inventions and Sinfonias, collectively entitled Aufrichtige Anleiting (Sincere Instruction) in the autograph fair copy, come from the same ‘stable’ as the WTC I: like many of the preludes from that collection, they were first entered in the Clavierbu¨chlein for W. F. Bach, evidently being intended in the first place for his musical education. All but five of the thirty pieces are in the composer’s hand in the Clavierbu¨chlein and appear to be mostly first drafts, entered between late 1722 and early 1723.36 The exceptions are nos. 3–7 of the two-part pieces, which were written out by W. F. Bach (no doubt copying from lost drafts) during the same period. At this initial stage, the fifteen two-part pieces were each entitled ‘Praeambulum’, a synonym for prelude, and the fifteen three-part pieces ‘Fantasia’. Bach used fifteen of the sixteen ‘primary’ keys described by Niedt (Musikalische Handleitung, 1710) and Mattheson (Das neu-Ero¨ffnete Orchestre, 1713)—the keys in the most common use at the time.37 In the Clavierbu¨chlein, he arranged each set of fifteen pieces in an ascending and descending key order. The ascending series outlines the diatonic scale of C major: C d e F G a b (key signatures of up to two sharps); the descending series includes the

35 According to Donald Francis Tovey, commentary to ABRSM edn of WTC I, ed. Richard Jones (London, 1994), p. 158. 36 According to Plath, Krit. Bericht, NBA V/5, pp. 58–63. 37 See Don O. Franklin, ‘Reconstructing the Urpartitur for WTC II: A Study of the “London Autograph” (BL Add. MS 35021)’, in Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 240–78 (see 255).

the well-tempered clavier i etc.

opposite modes (except B) in the reverse order and inserts two commonly used flat keys: B♭ A g f E E♭ D c (key signatures of two to four sharps or flats). In the early months of 1723, before the move to Leipzig that April, Bach made an autograph fair copy of the thirty pieces in which their titles, key order, and music text were all revised. The two-part pieces were now each entitled ‘inventio’38 and the threepart pieces ‘sinfonia’. A new key order is established in which each set of fifteen pieces falls into a single ascending series, partly diatonic and partly chromatic, and alternating major and minor modes: C c D d E♭ E e F f G g A a B♭ b. In addition, the individual pieces are carefully revised to produce the definitive version of the text. The only subsequent changes of note are ornamentation schemes, partly recorded in pupils’ copies. The informative title page of the fair copy reads:39 Auffrichtige Anleitung, wormit denen Liebhabern des Clavires, besonders aber denen Lehrbegierigen, eine deu¨tliche Art gezeiget wird, nicht alleine (1) mit 2 Stimmen reine spielen zu lernen, sondern auch bey weiteren progreßen (2) mit dreyen obligaten Partien richtig und wohl zu verfahren, anbey auch zugleich gute inventiones nicht alleine zu bekommen, sondern auch selbige wohl durchzufu¨hren, am allermeisten aber eine cantable Art im Spielen zu erlangen, und darneben einen starcken Vorschmack von der Composition zu u¨berkommen. Verfertiget von Joh: Seb: Bach. Hochfu¨rstlich Anhalt-Co¨thenischen Capellmeister. Anno Christi 1723. (Sincere Instruction, wherein lovers of the clavier, especially those desirous of learning, are shown a clear way not only 1. of learning to play clearly in two voices, but also, after further progress, 2. of dealing correctly and well with three obbligato parts; furthermore, at the same time, not only to have good inventions but to develop them well, and above all to arrive at a singing style in playing and, at the same time, to acquire a strong foretaste of composition. Produced by Joh. Seb. Bach, Capellmeister to His Serene Highness, the Prince of Anhalt-Co¨then, in the Year of our Lord, 1723.)

It is clear, then, that in the intended use of this collection keyboard and composition studies are of equal importance. As far as keyboard playing is concerned, the pieces help to develop the art of realizing two- and three-part contrapuntal textures. In addition, they foster a cantabile style of playing. Not only would Bach have inherited such a style from Pachelbel via Johann Christoph Bach, Sebastian’s elder brother, but the melodic style that Bach was developing in his keyboard music in the early 1720s has pronounced cantabile qualities and would therefore call for that style of execution. One further aspect of playing technique, furthered by the collection, is not mentioned in the title page but is nonetheless clear from the sources, namely the tasteful application and execution of ornaments. Inventions Nos. 3, 5, 7, and 9–12 and Sinfonia

38 Bach might have borrowed the term ‘inventio’ from F. A. Bonporti’s Inventioni da camera, Op. 10 (Bologna, 1712), for in 1723 he assisted his pupil B. C. Kayser in copying out four of these pieces; see Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992), VBN I/B/50. They were copied after Bach’s move to Leipzig (see Krit. Bericht, NBAV/7, p. 189), whereas the Inventions were completed in Co¨then, but that does not preclude the possibility that Bach was already acquainted with them. 39 BD I, No. 153; NBR, No. 92.

p r e l u d e , f u g u e , a n d i n ve n t i o n

No. 5 are more or less fully ornamented by Bach in the fair copy; and Sinfonias Nos. 4, 5, 7, 9, 11, and 13 are profusely embellished in the copies of Bach’s pupils B. C. Kayser and H. N. Gerber. No doubt many of these ornamentation schemes were worked out with the composer in the course of tuition. As far as composition is concerned, the thirty pieces provide models of good ‘inventiones’, or musical ideas, and illustrate effective ways of developing them—all this in the hope that the student will ‘acquire a strong foretaste of composition’. Since the Inventions and Sinfonias were composed in Co¨then not long after the WTC I and to a large extent with a similar educational purpose in mind, it is necessary to view them in the light of the earlier work. They might be seen as a bridge between the preludes and fugues of the WTC I, with their different technical demands, or else as a stage preparatory to the fugues. Perhaps with these purposes in mind Bach creates a new genre, identical with neither prelude nor fugue but combining elements of both. As usual, he operates within certain restricted, self-imposed parameters, articulating the contrapuntal, motivic, and melodic dimensions of each piece with different emphasis. All three of these dimensions are referred to in Bach’s title page: ‘motivic’ in the reference to developing ideas (‘inventiones’); ‘contrapuntal’ in the reference to learning to play in two or three obbligato parts; and ‘melodic’ in the reference to a cantabile style of playing. The underlying ‘inventio’ is the initial theme, which might be present within a single part, in a leading part with accompaniment, or in two equal parts in counterpoint (as in the Invention in G minor). Either the ‘inventio’ itself or motives derived from it are subjected to various types of contrapuntal manipulation in the course of the piece. At the same time, however, they are disposed in melodic periods, whose dynamic lies in their goal-oriented key structure. The true precursors of the Inventions and Sinfonias are the two-part preludes of the WTC I in C♯, F, F♯, and G, the two-part fugue in e, the three-part preludes in E, g, g♯, A, B, and b, and some of the three-part fugues, especially perhaps those in c and B♭. The WTC preludes in c♯, g♯, and A♭ all open with octave imitation between treble and bass or between right and left hand. This establishes an essential right/left or treble/bass equality from the outset, even in a three-part piece (g♯) or in pieces that do not adhere to a set number of parts (c♯, A♭). This type of opening is a feature of seven Inventions (C, c, D, d, e, F, a) and one Sinfonia (c). As an integral feature of the opening theme, it often recurs at a later restatement thereof; for example, in the Invention in A minor, bb. 6b–8a and 18. In two of the Inventions (c and F) the initial octave imitation is spun out canonically for ten bars. The overriding contrapuntal principle of the Inventions and Sinfonias, however—as of the WTC preludes on which they were no doubt modelled—is invertible counterpoint. Typically, the parts of the opening theme are inverted at its dominant or relative-major counterstatement (Inventions in C, a, B♭; Sinfonia in c). Or if the piece opens with a double theme, its parts are regularly inverted in double counterpoint, just like a fugue subject and its

the well-tempered clavier i etc.

countersubject (Inventions in E♭ and f; Sinfonia in b). In extreme cases, a whole period is subjected to double-counterpoint inversion. For example, the second period of the Invention in C minor is a dominant counterstatement of the first with interchanged parts; and the second strain of the Invention in E, the only one of the fifteen cast in binary dance form, is (except for bb. 29–42) a reprise of the first in double-counterpoint inversion. An exceptional and particularly subtle case is the Invention in G minor, in which the parts of the double subject, when inverted (bb. 3–4), are combined in a different manner: the trailing, chromatic subject is melodically inverted and enters after a whole bar rather than half a bar. Three of the Inventions (G, A, b) and the majority of the Sinfonias (C, D, d, E, e, F, f, G, A, a, B♭) might be described as informal fugues. As far as fugal technique is concerned, however, they exhibit a number of aberrant features that distance them from the fugues of the WTC I. First, the themes are in many cases not orthodox fugue subjects, nor are they so treated: a single bar of keyboard figuration (Invention in G; Sinfonias in C and E) or a brief sequence (Sinfonia in d) might serve as a subject. Second, the initial entry of the subject is, more often than not, accompanied by a free part (Invention in b; Sinfonias in C, D, d, E, e, F, G, A, a, B♭). Third, the opening exposition might include an additional treble entry (for example, Sinfonias in e and G). Fourth, several entries often follow one another in sequence within the same part (for example, Invention in G, bb. 27–8; Sinfonia in C, bb. 3–5, 14–15). Invertible counterpoint plays as vital a part in the fugal as in the non-fugal pieces. The Inventions in A and b are both based primarily on the double counterpoint of two subjects, which renders them successors to the WTC pieces that make a special feature of double counterpoint, namely the preludes in C♯ and A♭, and above all the fugue in E minor. Similarly the Sinfonias in D and f, like the WTC fugues in c and B♭, are based on the combination of their subject with two regular countersubjects in triple counterpoint, which is heard in four (f) or all (D) of its six possible permutations. Other fugal techniques, standard in the WTC I, recur here too. The Sinfonias in C (from b. 4) and E (from b. 17) each present a play on the direct and inverted forms of their subject. In the Sinfonias E minor (b. 14) and A minor (b. 21) a new, highly distinctive theme enters after the initial exposition, acting as a fresh countersubject and contributing much to the character of the piece thereafter. In addition, a significant role is played by stretto in the Sinfonias in C, F, and B♭. Indeed, in the B♭ Sinfonia stretto becomes the dominant contrapuntal device after the cadence in b. 11, so that the piece might reasonably be described as a stretto fugue. It might be thought that the ‘cantabile style of playing’ to which Bach refers in the title page would be more effectively demonstrated by a teacher than by a book of keyboard pieces. That Bach mentions it in this context suggests that he saw a strong cantabile element in the music itself that needed to be nurtured by the player.

p r e l u d e , f u g u e , a n d i n ve n t i o n

Such a cantabile style goes hand in hand with a galant mode of composition. For it has been observed that ‘many of the pieces seem to reflect Bach’s acceptance of galant elements—‘sigh’ figures, expressive ‘singing’ melodies—into his keyboard polyphony.40 And the chief rule of composition for the galant homme, according to Johann Mattheson’s first book, Das Neu-Ero¨ffnete Orchestre (Hamburg, 1713), was ‘daß man Cantable setze’—that is, to compose cantabile.41 This galant, cantabile style is perhaps evident, above all, in a group of pieces with certain characteristics in common, the Inventions in D, E, f, and g, and the Sinfonias in d, E♭, e, f, g, a, and B♭. All are notable for their predominantly stepwise motion, which naturally tends to invite a legato touch; and, as if to underline this point, several of them are furnished with a substantial amount of slurring in the autographs. On occasion, ornamental figures that include demisemiquavers are built into the themes (Invention in E, Sinfonia in f); and rich schemes of French-style ornamentation were subsequently added to six of the Sinfonias (d, E♭, e, f, g, a), largely in the copies of Bach’s pupils. Finally, a number of the cantabile pieces are conspicuous for their dance-style rhythms (Inventions in D and E; Sinfonias in E♭, g, and a). All but one of these pieces are in 3/8 time. The exception is the Sinfonia in E♭, which is cast in a sarabande-like 3/4 with dotted rhythms; this dance rhythm goes hand in hand with a graceful, elegant dialogue between the upper parts over an ostinato-bass accompaniment. Rightly considered, the melodic aspect of these cantabile pieces is no less significant than the motivic, thematic, or contrapuntal aspects. The melodic lines are built up, phrase by phrase, into large periods, articulated by clear-cut cadences in different keys. The only exceptions are fugues of the more traditional kind, where cadences are avoided in the interests of continuity (Inventions in E♭ and A; Sinfonias in C, A, and a). Elsewhere, even the fugues, like those of the WTC I, partake in this large-scale, melodic period structure. It is noticeable, however, that the fugal and period structures of a piece do not necessarily coincide. The Invention in b and the Sinfonias in D and B♭, for example, are clearly divided into two periods (AB), but their fugal structure is rather ABA1. The tonally oriented melodic scheme of a piece often takes precedence over the motivic-contrapuntal structure. In the Inventions in C and e, for example, the twofold imitation of the headmotive at the lower octave is entirely subordinate to the opening two-bar melodic phrase in the treble. Other instances may be cited in which contrapuntal considerations take second place to melodic writing. In the central g♯ episode from the Invention in E (bb. 29–42), for example, contrapuntal equality of parts is temporarily set aside in favour of a florid, lyrical treble with purely accompanying bass.

To quote David Schulenberg, The Keyboard Music of J. S. Bach (London, 1993), p. 149. The link between the Inventions and Mattheson’s treatise and between cantabile and galant styles was made by Martin Geck, ‘Bachs Inventionen und Sinfonien im galanten Diskurs’, in Geck (ed.), Bachs Musik fu¨r Tasteninstrumente (Dortmund, 2003), pp. 159–80. 41

the well-t empered clavier i et c.

Suite Title English Suites, BWV 806–11 No. 1 (early version) Nos. 1–6 Nos. 1, 3, 5, 6 Nos. 2, 6 French Suites, BWV 812–17 Clavierbu¨chlein for A. M. Bach, 1722 (lacks No. 6) Nos. 1–4 Nos. 1–6 Nos. 1–2, 4–6 Clavierbu¨chlein for A. M. Bach, 1725 (Nos. 1–2) Nos. 1–6 Suite in A minor, BWV 818 Suite in E♭, BWV 819

Clavieru¨bung I (Six Partitas), BWV 825–30 Clavierbu¨chlein for A. M. Bach, 1725 (Nos. 3 and 6) Nos. 1–6

Earliest source/s

Scribe, date

Berlin, P 803 Berlin, P 1072 Halle, 12 C 14–17 Berlin, P 803

J. G. Walther, pre-1717 B. C. Kayser, c. 1719–25 H. N. Gerber, 1725 J. T. Krebs, post-1725

Berlin, P 224

Autograph, 1722 (completed 1724)

Berlin, P 418 Berlin, P 1221 Berlin, P 420 Berlin, P 225

B. C. Kayser, 1722–5 H. N. Gerber, 1725 J. C. Vogler, c. 1725 A. M. Bach, post-1725

Washington, ML 96.B187 Berlin, P 418 Leipzig, Go. S. 9 Leipzig, Go. S. 10 Berlin, P 418 Berlin, P 420

J. C. Altnickol, post-1744 B. C. Kayser, 1722 H. N. Gerber, 1725 H. N. Gerber, 1725 B. C. Kayser, post-1725 J. C. Vogler, c. 1725

Berlin, P 225

A. M. Bach, 1725

Original edition

Leipzig, 1726–31

Unlike Bach’s other major keyboard works of the Co¨then and early Leipzig years, the ‘Six Suittes avec leurs Pre´ludes pour le clavecin, composeés par Jean: Sebast: Bach’, as the English Suites were perhaps originally entitled (according to the manuscript copies of B. C. Kayser and C. F. Penzel), have left no trace at all in the Clavierbu¨chlein for W. F. and A. M. Bach of 1720, 1722, and 1725. The obvious conclusion to draw is that they were not composed during this period. Their precise date of origin cannot be established in the absence of an autograph manuscript. The earliest pupils’ copies of all but the first suite date from 1725, but by then the suites might have been in existence for at least five years. An early dating of this kind—Co¨then, c. 1717/19—is suggested by comparison with Bach’s other suites. Like the Cello Suites, the English Suites, especially Nos. 2–6, are remarkably unified and consistent in style and design. Moreover, the two sets have in common many of the features that unite their constituent suites.42 In particular, they share a certain overall movement order, which is maintained through all six suites of both sets (with the partial exception of English Suite No. 1): Pre´lude—Allemande—Courante—Sarabande—Bourreé I,

42 As shown by Hans Eppstein, ‘Chronologieprobleme in Johann Sebastian Bachs Suiten fu¨r Soloinstrument’, BJ 62 (1976), pp. 35–57 (esp. 41–6).

suite

II—Gigue. In this scheme the bourreés are interchangeable with gavottes, menuets, or passepieds. All told, the links between the English and Cello Suites are close enough to suggest that the two sets might have originated in close temporal proximity; and since the Cello Suites are likely to have been in existence by about 1720 (since their origin seems to have been closely bound up with that of the Violin Solos of that year), the same is probably true of the English Suites. Generally speaking, their movement order might have been modelled on that of Charles Dieupart’s Six Suittes de clavessin (Amsterdam, 1701), which Bach had copied out during the Weimar years (1709–16).43 Dieupart’s scheme differs from Bach’s only in that the introductory movement is entitled ‘Ouverture’, rather than ‘Pre´lude’, and that the two optional dances are of contrasting type, rather than an alternativement pair. The question arises why we lose sight of the English Suites altogether between their presumed date of origin (c. 1717/19) and the first pupils’ copies in 1725. A possible explanation relates to their likely origin as commissioned works: an inscription in J. C. Bach’s copy reads ‘Fait pour les Anglois’; and J. N. Forkel reported that ‘They are known under the name of English Suites because the composer wrote them for a prominent Englishman’.44 It is possible that for the first few years of their existence Bach felt obliged to keep them for his own and the dedicatee’s use. The highly unified series of suites Nos. 2–6 were no doubt newly composed in fulfilment of the commission. No. 1, however, already existed: an early version in J. G. Walther’s hand dates from Bach’s Weimar period (before 1717). If Bach’s conception of the suites had begun with this work, one would expect the later suites to be modelled on it. In fact, however, the reverse seems to be the case: Suite No. 1 in A was revised with the apparent purpose of bringing it more into line with the design of the other five suites. This view is supported by the key order of the set: A a g F e d. In this scheme, the key A is redundant and appears to be tagged on to the beginning. It is not unlikely, then, that the intended key order of the English Suites was originally a g F e d C, but that Bach never completed the proposed sixth suite in C—perhaps because time ran out before the commission had to be fulfilled—and therefore revived an old suite instead, placing it first because it had the same tonic as the original no. 1.45 In accordance with its earlier, independent origin, Suite No. 1 differs in numerous aspects of style and design from Nos. 2–6.46 It is undoubtedly among the most French of all Bach’s keyboard suites. The Pre´lude is a pastorale that resembles gigues by Dieupart and le Roux. The Allemande, unlike its equivalents in Suites Nos. 2–6, makes ample use of the style luthe´. Multiple courantes follow, as in much French keyboard music of the seventeenth and early eighteenth centuries. This element is not only maintained but accentuated in the revised version: Courante 2, which at first had a

43 44 45 46

See Beißwenger, J. S. Bachs Notenbibliothek, VBN I/D/1. BD VII, p. 70; NBR, p. 468. Suite No. 2 was originally entitled ‘Suite 1’ in Kayser’s copy. See Vol. I, pp. 170–2.

the well-tem pered clavier i etc.

single variation—‘Courante precedent avec la basse simple’—is now presented ‘avec deux doubles’. The Sarabande alone is not unlike the dances of this type in the later suites of the set. Upon revision, the original Bourreé became the first of an alternativement pair, bringing the suite, in this respect, in line with the others. The concluding Gigue has the character of a double to the opening gigue-pastorale, as a comparison of their first few bars illustrates (Ex. 4). The later suites, Nos. 2–6, differ from No. 1 above all in their huge expansion of the pre´lude. Whereas Dieupart introduced his suites with an ouverture in the traditional French manner, Bach employed for this purpose a vast, Italianate concerto-Allegro. His aim was clearly to juxtapose and even integrate the French and Italian styles of the day, as well as to augment greatly the virtuoso aspect of the keyboard writing. These concertante pre´ludes embody a fusion of ritornello and da capo form such as Bach also achieved in contemporaneous instrumental music—in certain movements from the Brandenburg Concertos and in the opening Allegros of the Sonatas for violin and harpsichord, BWV 1014–19.47 In the pre´ludes, the A-paragraph of the ABA da capo form may be regarded as the framing ritornello, whereas the central B-paragraph alternates modulatory episodes, often on a contrasting theme, with free and/or partial ritornello returns in related keys. As in the sonatas, the crucial moment occurs when the more or less tutti-like ritornello reaches a tonic full-close, upon which a restart is

Ex. 4

a) Pre´lude to English Suite No. 1 in A, BWV 806, bb. 3–5

b) Opening of Gigue from the same suite

47 Views differ on the precise form of these preludes; see Alfred Du¨rr, ‘Zur Form der Pra¨ludien in Bachs Englischen Suiten’, in P. Ansehl et al. (eds.), Beitra¨ge zum Konzertschaffen J. S. Bachs, Bach-Studien 6 (Leipzig, 1981), pp. 101–8; repr. in A. Du¨rr, Im Mittelpunkt Bach (Kassel, 1988), pp. 232–8; and Gregory Butler, ‘The Prelude to the Third English Suite, BWV 808: An Allegro Concerto Movement in Ritornello Form’, in A. Leahy and Y. Tomita (eds.), Bach Studies from Dublin (Dublin, 2004), pp. 93–101.

suite

made with utterly contrasting material, akin to that of a concertino episode. The imitation of ensemble music is often unmistakable. The ritornello theme of the Pre´lude from Suite No. 3 in G minor sounds very much like the stretto entries from the central fugato of a French ouverture. The ‘tutti’ sequence in the Pre´lude from Suite No. 4 in F (bb. 7–8) recalls that of the Ouverture in C, BWV 1066 (bb. 50–2; Ex. 5 a), whereas the ‘concertino’ episode in the same pre´lude (bb. 20–3) closely resembles that of the first movement from Brandenburg Concerto No. 5 (bb. 9–13; Ex. 5b). Fused with these concerto and ouverture features throughout the pre´ludes are the strongly Germanic elements of motivic structuring and imitative counterpoint. The openings of the Pre´ludes in a and F resemble nothing so much as the two-part Inventions in their themes, octave imitation, and treble-bass equality. And much of the B-paragraphs of the Pre´ludes in g and F are taken up with the contrapuntal combination of two themes, one derived from the ritornello, the other episodic. The contrapuntal element is strongest in the Pre´ludes in e and d, whose ritornellos are designed as fugues, as often in Bach’s Co¨then instrumental music. In the Suite in D minor, this fugue is preceded by a 37-bar introduction, which is exploratory in traditional preludial mode, not settling into a consolidated theme or motive. The massive display of keyboard and compositional virtuosity in this movement threatens to swamp the following dances and burst the bounds of the suite form altogether. The style of the dances in these suites seems to be partially determined by that of the ´ preludes. Thus Bach tends to move away from the French brise´ texture of Suite No. 1 in favour of Italianate line drawing. Thematic/motivic working of the material is already present in No. 1, but in the later suites it is allied to Germanic contrapuntal textures rather than to the intricate French lute style. A single-bar ‘invention’, largely in continuous semiquavers, similar to certain Weimar organ themes or Co¨then Invention/Sinfonia subjects,48 forms the basis of the Allemandes in Suites Nos. 2, 3, and 5. In each case, the theme is immediately imitated at the octave (as in many of the Inventions), establishing a texture that, though free-voiced, involves frequent imitation between treble and bass. The theme is freely inverted after the double bar—an old convention of fugal gigues—which often yields rather awkward, angular results. In Suites Nos. 3 and 5 (but not in No. 2) the direct form of the theme returns at the end. The Allemandes of Suites Nos. 4 and 6 are similar to these, but fuller in texture and less even in flow, diversifying the semiquavers with shorter note-values in a manner characteristic of the ‘mixed style’ of the time. The Courantes are perhaps the least successful dances in the set. They are tied to the conventional French type, with its ambiguity between 3/2 and 6/4 time, and as a result less room is found for individuality than in the other dance types. In addition, they tend to rely on stock figures rather than on freely invented ideas. The Courantes of Suites Nos. 3 and 6 have no clearly identifiable theme, only a few recurring figures.

BWV 540 no. 1, 545 a no. 1, and 651 a (see Vol. I, p. 183); BWV 773, 782, and 787.

40 t he w ell-tempe red clav ie r i etc.

Ex. 5

a) 1. Pre´lude to English Suite No. 4 in F, BWV 809, bb. 7–8 2. 1st movement of Ouverture in C, BWV 1066, bb. 50–2 (outer parts only)

b) 1. Pre´lude to English Suite No. 4 in F, bb. 20–1 2. 1st movement of Brandenburg Concerto No. 5 in D, BWV 1050, bb. 9–10, ﬂute, violin, and harpsichord (ripieno omitted) Those of Nos. 2, 4, and 5 have a single-bar theme, inverted halfway, as in the Allemandes; but only occasional use is made of this theme, except in No. 5, whose Courante is as strictly thematic as its Allemande. The Sarabandes, unsurpassed in later years, possess strong, simple melodies with a very pronounced second-beat stress, characteristic of the French dance type, and are richly harmonized in four-part texture. The agre´ments and doubles were perhaps added subsequently for the beneﬁt of pupils.49 Only the grand Sarabande of Suite No. 6 belongs to the older 3/2 variety. That of No. 3 ranks among the most original that Bach ever wrote, with its chromatic harmony, pedal points, and unexpected 49

The sarabandes followed by a variation are those of Suites Nos. 2, 3 (agre´ments), and 6 (double).

suite

modulations, often involving enharmonic change. The Sarabande of No. 5 is quite unlike the others, with its three-part texture and polonaise-like rhythm. The dance, like the Courante that precedes it, is here handled with true individuality. In both cases Bach turns away from a relatively conventional treatment of the dance type in favour of a dance-like character piece—a step in the direction of the keyboard Partitas. The optional dances, or intermezzi, are also highly melodious but in a simpler texture and at a faster tempo. In Bach’s hands they are expanded well beyond the normal length for Lullian ballet dances. Binary dance form is on occasion diversified by rounded binary (Suite No. 6) or rondeau (No. 5). In each suite the second dance of the pair, to be played alternativement with the first (this is not indicated, but to be assumed), is thematically related to it but provides contrast of mode and texture. The concluding gigues vary: those of Suites Nos. 2 and 4 belong to the Italian giga type, being lightweight pieces in flowing compound time and in two-part non-fugal or very freely fugal texture. Suites Nos. 3, 5, and 6, on the other hand, conclude with Germanic, strictly fugal gigues in three-part texture with a regular countersubject and with inversion of the subject at the halfway point. Consequently these gigue-finales exhibit a weight and substance comparable with that of the opening pre´lude. The gigues of Suites Nos. 5 and 6 also possess a powerful chromatic element; and in No. 6 simultaneous melodic and contrapuntal inversion is so employed as to render the second strain a mirror version of the first. Thus not only Suite No. 6 but the entire set culminates in a remarkable contrapuntal tour de force. The French Suites, unlike the English Suites, originated within Bach’s own domestic environment: all but No. 6 were dedicated to his wife Anna Magdalena, forming the opening items in her Clavierbu¨chlein of 1722. These autographs are first drafts, preceded at most by sketches (possibly more extensive sketches in the case of Suites Nos. 1 and 4). Of Suite No. 5 only the beginning of the Allemande was entered in 1722/3 (alongside Nos. 1–4); the work was not completed till after the move to Leipzig, around mid-1724.50 Anna Magdalena’s personal interest in these compositions is illustrated by the (unfinished) fair copy she made of a revised version of them in her Clavierbu¨chlein of 1725. These might have been Anna’s first entries in the book, dating from 1725 or soon afterwards. Anna’s copy breaks off towards the end of the Sarabande from Suite No. 2, but originally she might have intended to include fair copies of all the five suites dedicated to her, since she had the right number of pages at her disposal.51 During the period between the two Clavierbu¨chlein entries, 1722–5, Bach opened up the suites for the use of his pupils. It is clear that at this stage the final order and contents of the set had not been established, since the manuscript copies of Bach’s pupils B. C. Kayser, H. N. Gerber, and J. C. Vogler contain not only some or all of the

50 According to Georg von Dadelsen, Beitra¨ge zur Chronologie der Werke Johann Sebastian Bachs (Trossingen, 1958), pp. 99–100. 51 See G. von Dadelsen, Krit. Bericht, NBA V/4 (1957), p. 25.

the well-tempered clavier i etc.

five suites dedicated to Anna Magdalena but also one or more of three additional suites, those in E, a, and E♭ (BWV 817, 818, and 819 respectively). Each copy contains a selection of these eight suites in a different order, perhaps reflecting the order in which they were studied. It is not known exactly when the definitive set of six suites—with the E major suite as No. 6 and those in a and E♭ excluded—was established, but the most likely date is late 1724. Suite No. 5 had been completed by about the middle of that year, and No. 6 had been composed by 1725, the probable date of Gerber’s and Vogler’s copies. The earliest manuscript to present all six suites in their final order, J. C. Altnickol’s copy, dates from after 1744, but in general it reflects a stage of the text earlier than that of the 1725 copies (Anna Magdalena, Vogler, and Gerber), which points to 1724 as the likeliest date of its autograph source. In the establishment of the final text there was clearly no single process of revision, but rather incidental improvements carried out in the years immediately following the composition of the suites. This was the period when Bach was making maximum use of them for teaching purposes, hence the lavish additional ornamentation and notated over-holding in some of the pupils’ copies. Many movements were apparently revised on several different occasions, with the result that the later readings are recorded variously in the pupils’ copies. It appears, then, that Bach’s revision was carried out in a rather haphazard, unsystematic manner, presumably as a by-product of tuition. The revision is not merely concerned with details, however, but embodies far-reaching changes—above all, the incorporation of five new movements (all minuets) in Suites Nos. 2–4 and the substitution of radically revised endings for four movements (Suite No. 2, Allemande and Courante; No. 3, Menuet I; and No. 5, Sarabande). The four standard dances of the classical suite—allemande, courante, sarabande, and gigue—are fixed components of all eight suites (except that BWV 819 lacks a gigue), as they were of the English Suites. But the role played by dances of optional type, the intermezzi, gradually increased. All are Lullian ballet dances, as in the earlier set, but only in Suite No. 1 are they restricted to an alternativement pair—standard in the English Suites. Suites Nos. 2 and 3 originally contained only a single intermezzo (Air and Anglaise respectively); but probably in early 1723 a minuet for No. 2 in Anna’s hand and a minuet and trio for No. 3 in Bach’s hand were appended at the back of the 1722 Clavierbu¨chlein. By 1725 a second minuet had been inserted in Suite No. 2 (as reflected in Vogler’s copy), so that by then the first three suites all included an alternativement pair of minuets. Suite No. 4 originally contained two intermezzi, Gavotte and Air, but a third (a minuet) was added subsequently. This brought it in line with Nos. 5–6: the last three suites of the set now all included three intermezzi before the gigue, of which the first was in every case a gavotte. By 1762 the six suites of the definitive set, BWV 812–17, were already known as the ‘French Suites’. This is clear from a reference to Bach’s allemandes published in that year by Friedrich Wilhelm Marpurg: ‘Authentic models of the allemande style of writing are found chiefly in the six French Suites by the late Herr Capellmeister

suite

Bach’.52 J. N. Forkel, in his Bach biography of 1802, informs us that ‘They are generally called French Suites because they are written in the French taste.’53 This explanation does not seem altogether convincing, however. It is true that French ballet dances play a prominent part, as does French ornamentation and the style luthe´. But Germanic and Italianate elements are also present, and the overall impression is of a rich stylistic mix—Bach’s personal version of the ‘mixed style’ that was widespread at the time. The traditional dances, which had been more uniformly stylized in the English Suites, are here diversified by the application of a richer, more colourful range of style. Of course, the virtuoso, concertante element of the pre´ludes from the English Suites is absent, but the French Suites more than compensate by virtue of their intimacy and subtle expressivity. It has been suggested that after the English Suites Bach might have made a conscious decision here, in keeping with the times, to adopt a simpler, more popular, and more accessible style.54 In the absence of a prelude, the allemande sets the tone of the whole suite—Johann Gottfried Walther labelled it the ‘proposition from which the remaining movements of the suite—the courante, sarabande, and gigue, etc.—flow as constituent parts’.55 The Allemande of Suite No. 1 (D minor, BWV 812) resembles those of English Suites Nos. 2–6 in its motivic counterpoint, and yet it is largely unthematic and free-voiced, with much recourse to the style luthe´. In these respects it is closer to the French style of English Suite No. 1, exhibiting something of the random, indeterminate quality of the French allemande. This movement sets the tone for a suite that is closer than its successors to the pattern of the English Suites in overall design, in its stylization of the dances, and in its contrapuntal textures. The Courante is somewhat tighter and more thematic than those of the English Suites: the clear-cut theme and (after the double bar) its inversion are constantly interchanged between the hands. The Sarabande is especially close to its equivalents in the English Suites: richly chordal with a highly expressive treble melody. A more florid version of the theme surfaced in the following year (1723) in the duet ‘Et misericordia’ from Bach’s Magnificat, BWV 243a (Ex. 6). The Menuets, to be played alternativement, are both in the tonic and in a pure threepart texture—invertible in Menuet I. Menuet II is cast in a simple rondeau form (ABA), the return being indicated by a da capo. The concluding Gigue is a three-part fugue with subject inversion at the halfway point, as in English Suites Nos. 3, 5, and 6. It is singular, however, in its cut i time with dotted rhythms—a species of gigue rhythm encountered frequently in the German tradition from Froberger onwards. Suite No. 2 (C minor, BWV 813) shows the composer attempting to diversify the traditional suite movements by resorting to a variety of different styles, as well as by 52 ‘Aechte Muster von der Allemanden-Schreibart findet man hauptsa¨chlich in den VI. franz. Suiten vom seel. Herrn Capellm. Bach’; BD III, No. 715. 53 BD VII, p. 71; NBR, pp. 468–9. 54 See Schulenberg, The Keyboard Music of J. S. Bach, p. 254. 55 ‘Proposition, woraus die u¨brigen Suiten, als die Courante, Sarabande, und Gigue als Partes fliessen’; Musicalisches Lexicon (Leipzig, 1732).

44 the well-tempere d clavier i etc.

Ex. 6

a) Opening of Sarabande from French Suite No. 1 in D minor, BWV 812

b) Opening of ‘Et misericordia’ from Magnificat in E♭, BWV 243a, violins and continuo (viola omitted) introducing a new type, the Air. Again, the Allemande leads the way: a highly florid treble in Bach’s personal cantabile manner is broken up into pseudo-parts in luthe´ style. The Courante adopts the style of the Italian corrente (though not yet its name), unknown in the English Suites—a lively, flowing 3/4 time and, in this case, a mainly two-part texture of treble and bass. The florid style of the Sarabande recalls the Allemande. Bach’s normally rich sarabande texture is here reduced to two supporting parts, freeing the treble for florid cantabile treatment. This texture would henceforth become a prime resource in Bach’s slow movements for solo keyboard. The Air, though a usual constituent of the French ballet suite, was new to Bach’s mature suites. The alternativement minuets share both key (tonic) and texture (two parts), thereby following the example of the first suite. The Gigue, with its dotted rhythms in 3/8, belongs to the French canarie type. Like almost everything else in this suite, for Bach it represents a new departure. Much the same might be said of Suite No. 3 (B minor, BWV 814). By diversifying his stylistic resources to the utmost Bach is now able to stamp each suite with its own individuality. In the third suite this is immediately evident from the very densely motivic two-part counterpoint of the Allemande, whose opening four-note motive, freely inverted after the double bar, is employed throughout in sequential imitation. The Courante is no less tightly motivic on the basis of its initial six-note figure, which is spun out by sequence into continuous quaver motion, reminiscent of a double. The metre is 6/4—as in several of Rameau’s courantes—rather than the standard 3/2 of the French courante; but the admixture of 3/2 bars results in Bach’s rhythmically subtlest courante so far, anticipating those of the second and fourth keyboard Partitas. The Sarabande, like that of the second suite, illustrates Bach’s florid cantabile style, and the texture of its first strain is similar—florid treble with two left-hand supporting parts. After the double bar, however, the theme moves to the left hand, after which the two

suite

Ex. 7

a) Opening of Sarabande from French Suite No. 3 in B minor, BWV 814 (treble only)

b) Sinfonia No. 13 in A minor, BWV 799, bb. 49–50 (treble only) hands alternate as carriers of the thematic line. The theme itself closely resembles a striking theme from the roughly contemporaneous Sinfonia in A minor, BWV 799 (Ex. 7). The Anglaise, like the Air from the second suite, represents a new departure for Bach—an English dance type cultivated at the court of Louis XIV. The texture is similar to that of the cantabile movements already mentioned: thematic treble with two supporting parts. As in the first and second suites, the alternativement Menuets share their key (tonic), but they differ in texture (no. 1: two-part; no. 2: three-part). Bach’s gigues are already capable of immense variety of metre and rhythmic movement. The 3/8 semiquavers of this B minor Gigue are unique in the French Suites, though they had already occurred in the fugal gigue-finale of the fifth English Suite. The B minor Gigue is non-fugal, however, and its two-part imitative counterpoint, with opening octave imitation, most clearly resembles the gigue-finale of the first English Suite. The opening movement of Suite No. 4 (E♭, BWV 815) once again brings a new style to the Allemande: that of Bach’s arpeggiated preludes. The second movement is a corrente (though not so called), as in the second suite, though quite different in rhythmic movement: flowing quavers against crotchets are here replaced by triplets against dotted rhythms. In the Sarabande, the traditional second-beat stress is more pervasive than in the two preceding suites, but the constant thematic exchange between the two hands breaks new ground for this dance type. The Gavotte bears no relation at all to those of the English Suites. It is written in two-part counterpoint on the basis of a theme made up of quaver couplets, of which strictly motivic use is made throughout. The Air, a type already employed in the second suite, is here treated in a similar fashion: cut i time with running semiquavers in a two-part texture. The paired-quavers theme of the brief Menuet recalls the minuets of the two previous suites. The Gigue is the only movement of its type so far written in the Italian giga style. As a free two-part fugue, it resembles the gigue-finale of the fourth English Suite, but its subject is closer to that of the gigue from Telemann’s Suite in A, which W. F. Bach entered in his Clavierbu¨chlein around the time when the French Suites were drafted.56

See Schulenberg, The Keyboard Music of J. S. Bach, p. 271.

the well-tempered clavier i etc.

In his conception of the French Suites Bach seems to have sought a balance between the opposing demands of originality and convention, between inventive freedom and the received characteristics of the various dance types. From this point of view, the highest level of perfection is attained in the fifth and sixth suites. Both unite Bach’s most melodious cantabile style, a highly personal innovation, with the most faithful stylization of the traditional dances. The Allemande of Suite No. 5 (G major, BWV 816), which as usual sets the tone of the whole suite, is Bach’s most singing, melodious allemande so far, following the lead of the second suite. It is treble-dominated but within a highly flexible, free-voiced texture. Characteristically, the pseudo-cross-string motive of the theme (bb. 2a, 3a) later turns into an accompaniment figure (bb. 8b–9). The Courante is in reality a 3/4 corrente, as in the second and fourth suites, but it takes a further step in the progressive shortening of note-values: from quavers, via triplet quavers, to semiquavers. The Sarabande, like that of the second suite (and the first strain of the third), is an early illustration of what later became Bach’s favoured slowmovement texture: florid cantabile treble against two left-hand supporting parts. The dotted rhythms of the ornamented treble are new to Bach’s sarabandes, whether vocal or instrumental, but would be much exploited thereafter. The Gavotte is handled in a highly individual manner, as in the fourth suite, whereas the pervasive anapaest rhythm of the Bourreé was often associated with this dance from Lully onwards. The loure was new to Bach’s keyboard suites, though it had been included in the solo violin Partia in E (BWV 1006) a few years earlier (1720). Both loures are in 6/4 with dotted rhythms and their upbeat figures are similar. The absence of the Loure from Vogler’s and Gerber’s copies suggests that Bach might have intended to omit it from the revised version of the suite. The Gigue, like that of the sixth English Suite, is a three-part fugue in 12/16 with running triplet semiquavers. The subject is of a highly individual, playful character, and its fugal treatment brings the suite to a brilliant conclusion. The Allemande of Suite No. 6 (E major, BWV 817) is in much the same singing style as that of No. 5 and both employ ‘cross-string’ figuration, but its texture is two-part rather than free-voiced. A tight motivic structure is built around the opening half-bar figure (and its inversion), which is subject to ‘kinetic recurrence’,57 as in a concerto theme. The Courante is a 3/4 corrente with running semiquavers, as in the fifth suite. On this occasion, however, Bach employs a brilliant, flexible, fleet-footed, free-voiced texture, moving freely from luthe´ effects to two-part writing or to a single part divided between the hands. The Sarabande is close to that of the previous suite, both in its dotted rhythms and in the shaping of its opening theme. The texture, however, resembles that of the English-Suite sarabandes: florid, ornamented treble with richly chordal accompaniment. The Gavotte is highly individual but resembles that of the fifth suite in its three-part texture, with the upper parts often moving in parallel over a supporting bass. The Polonaise, with its simple two-part texture of melodic treble and

A term used by Arthur Hutchings, The Baroque Concerto (London, 1959; 3rd edn 1973), pp. 43–4.

suite

bass, is new to Bach’s keyboard suites. The Bourreé, like that of the fifth suite, is pervaded by the anapaest rhythm that had become so characteristic of the dance. The Gigue is not fugal but in imitative two-part counterpoint, like the finale of the third suite. The Menuet, notable for its unusual (in this context) ostinato bass, takes last place in the copies of Bach’s pupils (Gerber, Vogler, and Altnickol). Perhaps it was an afterthought, appended at the end of the lost autograph without any indication of its correct position. From many of the remarks made here it is clear that Suite No. 6 is a sister-work to No. 5, which strongly suggests that they might have been composed one after the other in the summer of 1724. Where do the Suites in A minor and E♭ (BWV 818–19) belong in this picture of the French Suites? They are intermingled with those suites in the copies of Bach’s pupils Kayser, Gerber, and Vogler from 1722 to 1725. And it must be remembered that for much of that period—till summer 1724 at the earliest—only four of the set of six existed, with the result that the two miscellaneous suites enjoyed equal status with them, as reflected in the pupils’ copies. Moreover, these copies contain both earlier and later versions of the two suites, just as they do of the French Suites. We may conclude that the eight suites are all intimately connected, perhaps in origin and certainly in their use as teaching material and in the revision process that took place at the same time. Accordingly, the two miscellaneous suites are close to the French Suites in style and design. Their movement order is as follows: BWV 818: Allemande, Courante, Sarabande, Gigue BWV 818a: Pre´lude, Allemande, Courante, Sarabande, Menuet, Gigue BWV 819/a: Allemande, Courante, Sarabande, Bourreé, Menuet I, II The A minor Suite, then, originally had no intermezzo at all—an even simpler scheme than that of French Suites Nos. 1–4 in their original state. This suggests that it might date from before 1722 (the earliest copy, that of B. C. Kayser, dates from 1722 at the latest). At a later stage, a single intermezzo (a minuet) was added, bringing the suite into line with the original versions of French Suites Nos. 2 and 3. This later version, BWV 818a, also possesses a prelude. Bach seems to have toyed with the idea of a prelude for some of the French Suites too, hence the use of the E major Praeludium from the WTC I as prelude to Suite No. 6 in Gerber’s copy; hence also the prelude to No. 4 in the variant version BWV 815a (which, however, is of doubtful authenticity). The Suite in E♭, BWV 819, remained unchanged in movement order. The curious feature of this suite is the absence of a concluding gigue—singular in Bach,58 though common in Mattheson and other composers of the time. The group of intermezzi here—a French ballet dance (bourreé) plus an alternativement pair of minuets—is also found in the later version of French Suites Nos. 2 and 3. This is in keeping with the date of the earliest copies of BWV 819, namely 1725.

With the exception of Partita No. 2 where, however, the concluding Capriccio acts in lieu of gigue.

the well-te mpered clavier i et c.

The relatively early origin of the A minor Suite, already indicated, is clearly reflected in its compositional style. The Allemande, with its single-bar theme in flowing semiquavers, imitated at the octave, is remarkable similar to those of the English Suites, especially No. 2, which is in the same key. The Courante belongs to the rhythmically subtle 3/2 type that Bach had employed in all six English Suites and in the first French Suite. It is strictly thematic and makes copious use of the inverted theme. The Sarabande belongs with the freely invented sarabandes of the French Suites, especially that of No. 4—the two movements have in common their use of an ostinato rhythm. The fugal Gigue is quite remote from those of the English Suites, but again it has a clear affinity with that of French Suite No. 4. The Pre´lude, added to the later version, is the sort of piece that might easily have been improvised. The Sarabande of this version was newly composed on the basis of the old harmonic and phrase structure. The original ostinato rhythm has vanished in favour of more varied rhythms and articulation, creating an improvisatory effect closer to the French style. The dotted rhythms recall the sarabandes of French Suites Nos. 5 and 6. The newly inserted Menuet is perhaps the most remarkable piece among all the intermezzi of the eight suites under discussion. With its mix of 3/8 semiquavers and triplets, it sounds decidedly modish59 and represents an almost complete departure from the conventional rhythmic movement of a minuet. It conveys the impression not so much of a traditional dance as of a hauntingly lyrical character-piece. Judging by its compositional style, alongside source findings, the Suite in E♭, BWV 819, appears to be of somewhat later origin. After the opening gambit of the Allemande, a new figure enters in the third bar and then proceeds to dominate the remainder of the movement. This figure is almost identical with the leading motive of the Allemande from the sixth French Suite. Indeed, so alike are the two movements that the E♭ reads like an earlier adumbration of the E major. This might explain why Bach later completely remodelled the movement (in the context of the revised version BWV 819a) while retaining the original harmonic and phrase structure. The new Allemande is highly motivic and contrapuntal—rather like that of the third French Suite. The Courante belongs to the French type, but the prevailing metre is 6/4 rather than 3/2—again as in the third French Suite. Both courantes include occasional bars in 3/2, so that the subtle, indeterminate quality of the French courante is still to some extent discernible. The Sarabande shares the dotted rhythms of its equivalents in the fifth and sixth French Suites. It is written in a three-part ‘trio-sonata’ texture— right-hand duet over bass—to which Bach occasionally resorted in Co¨then, as in Praeludium No. 24 in B minor (WTC I), Sinfonia No. 5 in E♭, and, most relevantly, the polonaise-like Sarabande from the fifth English Suite. The Bourreé is characterized by a pervasive anapaest rhythm, as in the bourreés from the fifth and sixth French Suites.

59 Schulenberg, in The Keyboard Music of J. S. Bach, p. 259, links it with the Menuet from Partita No. 4: both mix duplet and triplet semiquavers in an up-to-date galant manner.

suite

The alternativement Menuets are in the key relation of tonic major and minor (E♭/e♭), like the bourreés of the first English Suite. Just as the first five French Suites were dedicated to Bach’s new young wife Anna Magdalena in the Clavierbu¨chlein of 1722, so two of the keyboard Partitas—those later published within Clavieru¨bung I as Nos. 3 and 6—opened her second Clavierbu¨chlein of 1725 as a form of dedication. Perhaps Bach was trying to tell her that, now she had mastered a series of small, intimate suites, it was time to try her hand at something altogether larger and more demanding. For in the partitas Bach apparently wished to carry further the inventive freedom and stylistic range of the French Suites, while at the same time reverting to the large scale and virtuoso demands of the English Suites. The later stages in the composition of the French Suites and the early stages of the partitas might have been very close in time. French Suites Nos. 5 and 6 probably date from mid-1724; Partitas Nos. 3 and 6 were entered in Anna’s manuscript book in 1725; and Partita No. 1 was composed by 1726, the year of its dedication and publication. This composition is in many ways the closest of the partitas in style to the last two French Suites. In all three cases a bright major mode is allied to limpid, melodious cantabile writing, and the texture is mainly figurative rather than contrapuntal, with smooth implied lines within violinistic figurations. It is possible, then, that Partita No. 1 pre-dated Nos. 3 and 6, and originated in late 1724 or early 1725, almost immediately after the last two French Suites. As for the internal dating of the set as a whole, it has been suggested that the six partitas were conceived in complementary pairs in the order: Nos. 1 and 2 (late 1724?), Nos. 3 and 6 (1725), and Nos. 4 and 5 (after 1725).60 The crucial factor that distinguishes the partitas from Bach’s earlier keyboard suites is publication. They were published ‘in Verlegung des Autoris’ (by the author) in six separate instalments over a period of four years (Leipzig, 1726–30). This tentative approach allowed Bach to test the water in a venture that was not only new to him but carried an element of financial risk. A collected edition of the six partitas followed in 1731, printed from the same engraved plates as the single editions. The title page, whose wording is virtually identical with that of the separate instalments, reads: Clavir Ubung bestehend in Praeludien, Allemanden, Couranten, Sarabanden, Giguen, Menuetten, und andern Galanterien; Denen Liebhabern zur Gemu¨ths Ergoetzung verfertiget von Johann Sebastian Bach, Hochfu¨rstl: Sa¨chsisch-Weisenfelsischen wu¨rcklichen Capellmeistern und Directore Chori Musici Lipsiensis. Opus 1. In Verlegung des Autoris. 1731. Leipzig, in Comission bey Boetii Seel: hinderlassene Tochter, unter den Rath:hause. (Keyboard Practice, consisting of Preludes, Allemandes, Courantes, Sarabandes, Gigues, Minuets, and other galanteries; composed for music lovers to delight their spirits by Johann Sebastian Bach, actual Capellmeister to the court of Saxe-Weissenfels and Director Chori Musici in Leipzig. Opus 1. Published by the author. 1731. Leipzig, [sold] on commission by the daughter of the late Boethius, under the town hall.)

60 See Richard D. P. Jones, ‘The History and Text of Bach’s Clavieru¨bung I’, diss., University of Oxford, 1988, pp. 79–83.

50 the we ll-temper ed clavier i etc. Bach evidently modelled this, his first keyboard publication, in externals on that of ¨ bung (Parts I and II) his Leipzig predecessor Johann Kuhnau, whose Neuer Clavier U had been published there in 1689 and 1692. From that work Bach borrowed not only ¨ bung’ (Keyboard Practice)61 and the specific title ‘Partita’, the generic title ‘Clavier U but also the overall conception of a series of suites whose keynotes represent the seven degrees of the diatonic scale—Bach’s scheme is B♭ c a D G e [F]. For according to a Leipzig press announcement of 1 May 1730,62 Bach intended to publish not only a sixth but a seventh partita at the Michaelmas Fair that autumn. No one knows why the planned seventh partita never materialized, but it might be significant that the next work Bach published, the Italian Concerto (Clavieru¨bung II, 1735), was in the same key, F major. It is even conceivable that the opening Allegro of the concerto might originally have been intended as the introductory prelude to the planned seventh partita. The keyboard Partitas, then, might have been conceived as a tribute to Johann Kuhnau, whose keyboard works had exerted a significant influence on Bach in his youth, and whom Bach had succeeded in 1723 as Cantor and Music Director in Leipzig. Market considerations might also have played a part in Bach’s debt to Kuhnau. For the older composer’s four engraved books of keyboard music, published from 1689 to 1700, had met with enormous success and popularity. Not since Frescobaldi had any European composer so dominated the music market. This was no doubt partly connected with the status of Leipzig as the most important fair and trade centre in East Germany. But it also reflected the late seventeenth-century emergence of a new middle class with the time and inclination to enjoy cultural leisure activities. Kuhnau’s keyboard works met perfectly the demand for entertaining, easily playable music for middle-class amateurs. Some thirty years later Bach evidently had an eye on the very same market of cultivated amateurs to whom Kuhnau had appealed so successfully. This is clear from his announcement in the title page that the partitas were ‘composed for music lovers, to delight their spirits’ (‘Denen Liebhabern zur Gemu¨ths Ergoetzung’). No less significant than this German precedent is the Italian sonata da camera tradition which, in comparison with Bach’s earlier keyboard suites, steers the partitas of Clavieru¨bung I in a fundamentally new direction. He had already cultivated this Italian genre in the three solo violin ‘partias’ (a synonym for ‘partita’) of 1720, whose movements are as follows: No. 1 (b): Allemanda, Corrente, Sarabande, Tempo di Borea (all with doubles) No. 2 (d): Allemanda, Corrente, Sarabanda, Giga, Ciaccona No. 3 (E): Preludio, Loure, Gavotte en rondeau, Menuet I, II, Bourreé, Gigue In the Clavieru¨bung [Part I] of 1726–31, Bach returns to the suite title ‘partita’, to the use of Italian titles for many of the dance movements, and to highlighting the

61 Which was in turn derived from the 17th-century Italian term essercizi, later revived by Domenico Scarlatti and Telemann. 62 In the Leipziger Post-Zeitungen; BD II, No. 276.

suite

juxtaposition of Italian and French styles by means of clear linguistic differentiation. None of this applied to the English or French Suites, regardless of their strong Italianate input, presumably because Bach regarded the keyboard suite as a French genre. In the keyboard Partitas, however, he moves decisively in the direction of the Italian sonata tradition, which greatly enriches and diversiﬁes the dance suite, injecting it with a new vitality. The starting point of the E minor Partita, for example, was a pair of Italianate instrumental dance movements, the ‘Cembalo solo’ and ‘Violino solo e basso l’accompagnato’ from an early version of the Sonata in G, BWV 1019a (1725).63 In a revised form these movements, later in the same year, became the Corrente and Tempo di Gavotta (third and sixth movements) of the partita. The movement order of the two partitas (BWV 827 and 830) in Anna Magdalena’s 1725 Clavierbu¨chlein is as follows: BWV 827: Pre´lude, Allemande, Corrente, Sarabande, Menuet, Gigue BWV 830: Pre´lude, Allemande, Corrente, Sarabande, Tempo di Gavotta, Gigue Both, then, are cast in a six-movement form with a single intermezzo. The introductory movement has the non-committal French title ‘Pre´lude’, as in the English Suites. The violin-sonata movements appropriately receive Italian titles, as does the courante from BWV 827. The six partitas in their published form display the following movement order: No. 1 (B♭): No. 2 (c): No. 3 (a): No. 4 (D): No. 5 (G): No. 6 (e):

Praeludium, Allemande, Corrente, Sarabande, Menuet I, II, Giga Sinfonia, Allemande, Courante, Sarabande, Rondeaux, Capriccio Fantasia, Allemande, Corrente, Sarabande, Burlesca, Scherzo, Gigue Ouverture, Allemande, Courante, Aria, Sarabande, Menuet, Gigue Praeambulum, Allemande, Corrente, Sarabande, Tempo di Minuetto, Passepied, Gigue Toccata, Allemanda, Corrente, Air, Sarabande, Tempo di Gavotta, Gigue

Thus, while Nos. 1 and 2 retain the six-movement scheme of the two Clavierbu¨chlein partitas (if the alternativement Menuets of No. 1 are viewed as a single movement), Nos. 3 and 6 in their revised form and Nos. 4 and 5 revert to the seven-movement structure (with two intermezzi) of Dieupart, which Bach had adopted as a norm in the English and Cello Suites. Stylistic diversity, to some extent anticipated in the French Suites, now becomes a governing principle, reﬂected in the colourful movement titles which are no doubt intended to appeal to the market. The preludes are all differently titled in accordance with their formal structure. One is French (Ouverture), three are Italian (Sinfonia, Fantasia, and Toccata), and two Latin (Praeludium and Praeambulum). The

See Jones, ‘The History and Text of Bach’s Clavieru¨bung I’, pp. 63–7.

the well-tempered clavier i etc.

courantes alternate between the French courante and the Italian corrente. The other main dances are mostly French in name, in accordance with long-standing German tradition, but Partita No. 1 concludes with an Italian ‘Giga’, and the first dance of No. 6 was at some stage renamed ‘Allemanda’.64 The contrast of national styles, highly fashionable in German music of the time, is particularly clear in the intermezzi. Partita No. 1 has a French pair of Menuets, no doubt to be played alternativement; No. 3, an Italian pair with fanciful titles, Burlesca and Scherzo. The other four partitas all include a contrasted pair of French and Italian intermezzi: Rondeaux and Capriccio (No. 2), Aria and Menuet (No. 4), Tempo di Minuetto and Passepied (No. 5), and Air and Tempo di Gavotta (No. 6). Of course, the French and Italian styles, emblazoned on the movements by their titles, are to a certain extent a fiction. By this stage Bach was cultivating his own personal version of the ‘mixed style’ that was prevalent in German music of the time, and for this reason much of the music cannot be allocated to any one national style. A clear example is the Sarabande from Partita No. 6, in which the Italian Adagio is integrated with the French sarabande grave. Where the Italian style is uppermost, it has to be remembered that Bach’s models, above all Corelli, were substantially freer and less hidebound by tradition in their characterization of the various dance types than the French clavecinistes. In Corelli the infiltration of sonata da chiesa style often dilutes the dance element, and the same is true of many of the dance movements in Bach’s partitas. In some cases the dance element is modified or marginalized to the extent that it is barely recognizable (Partita No. 1: Giga; No. 3: Sarabande; No. 4: Menuet; No. 6: Corrente). This explains Bach’s use of the Corellian formula ‘Tempo di . . .’ in Partitas Nos. 5 and 6. It also explains the use of titles that make no reference to dance: Capriccio (No. 2), Burlesca, and Scherzo (No. 3). Many of the movements discussed here might be described as character-pieces—the highly individual quality of the invention takes precedence over received dance elements. The partitas are often given a weighty frame by Bach’s trademark imitative counterpoint, as in the fugal sections of the preludes (Partitas Nos. 2, 4, and 6), the Invention-like Fantasia (No. 3), and the fugal gigue-finales (Nos. 3–6). Many movements, however, show Bach moving in the direction of the progressive style galant, with its lighter touch and freedom from polyphonic complexity. The Giga from Partita No. 1 is as simple and homophonic as Mattheson, or any other enthusiast for the new style, could wish, consisting of little more than a thematic line and accompaniment figure, a texture to which Bach returns in the cross-rhythm Tempo di Minuetto from No. 5. The first three of the main dances, Allemande, Corrente, and Sarabande, whether two-part or free-voiced, are often largely treble-dominated and non- or semi-contrapuntal, particularly in Partitas Nos. 1, 3, and 6. The trio-sonata texture of the Sarabandes from Nos. 3 and 5 has a directly melodious appeal, with the

See R. D. P. Jones, Krit. Bericht, NBA V/1 (1978), p. 65.

suite

right-hand parts moving much in parallel 3rds and 6ths. The florid cantabile of the Allemande from Partita No. 4 or the Sarabande from No. 6 may be viewed as a personal contribution of Bach’s to the new style. The written-out multiple appoggiaturas of Nos. 3 (Allemande) and 6 (Toccata and Sarabande), whether thematic or cadential, add a certain galant grace and elegance to the discourse. Finally, a dance-like style of writing in 2/4 time that must have sounded fashionable at the time was increasingly cultivated by Bach from 1726/7 onwards,65 hence the Capriccio (Partita No. 2), with its giant 10th leaps, the Scherzo (No. 3), with its Scarlattian acciaccatura, and the Aria (No. 4), with its modish syncopation and broken-off cadences. Whereas in the French Suites the Allemande sets the tone of the following movements, in the partitas it is the prelude that takes that function. And since the preludes all differ in form and style, the partitas as a whole are each unique individuals in their own right—far more so than Bach’s earlier suites. The Praeludium from Partita No. 1 in B♭ is not unlike a three-part sinfonia in its inversion of the parts at each statement of the two themes (bb. 1 and 3). In its violinistic figuration one can already detect the emphasis on smooth cantabile melody that permeates the entire suite. The treble of the Allemande is similarly violinistic but in the context of a highly flexible texture—light and airy, predominantly two-part but constantly changing, like that of the Courante from the sixth French Suite. The Corrente moves in triplet quavers against dotted rhythms, just like the Courante from the fourth French Suite. The Sarabande recalls that of the Suite in A minor, BWV 818a, in its repeated-note, dotted-rhythm motive, but the partita movement introduces far more ornamental, decorative writing between the returns of this motive. Menuet I is very close in conception to its equivalent in the third French Suite; and as there, Menuet II contrasts with it in texture but not in key. The Giga is an original conception, without precedent among Bach’s earlier gigues. It is a cross-hands piece with the theme in crotchets and its accompaniment in triplet quavers, creating the effect of a gigue-like 12/8 in a purely homophonic texture. The mixture of figuration and cantabile, a hallmark of this suite, is maintained to the last. Partita No. 2 in C minor, on the other hand, might be described as motiviccontrapuntal. Within an Italianate frame (Sinfonia and Capriccio) the dances incline towards the French style—note, in particular, the Courante and Rondeaux—as opposed to the Italian sonata style of their equivalents in the first partita. The tripartite form of the Sinfonia, with its progressive increase of tempo—Grave adagio, Andante, and [Allegro]—has no equivalent elsewhere in Bach’s keyboard music. It is possible that Bach was here imitating a certain type of ensemble sinfonia that Mattheson described in a note to Niedt’s Musicalische Handleitung, Part II (Hamburg, 1706; second edition 1721).66 The Allemande is written in strictly motivic counterpoint, as

65 Examples of this style in Bach’s vocal works are listed by Doris Finke-Hecklinger, Tanzcharaktere in Johann Sebastian Bachs Vokalmusik (Trossingen, 1970), pp. 142–3. 66 Quoted in Andreas Jacob, Studien zu Kompositionsart und Kompositionsbegriff in Bachs Klavieru¨bungen (Stuttgart, 1997), p. 98.

the well-tempered clavier i etc.

in the third French Suite and the Suite in E♭, BWV 819a. The French Courante, in 3/2 time, is particularly close to that of the first French Suite, as a comparison of their opening bars illustrates (Ex. 8). The harmony of the Sarabande is dissolved into flowing lines in a two-part texture; the effect is more like a double than a sarabande proper. The quick 3/8 time of the Rondeaux creates the impression of dance rhythm, though no specific dance is evoked. The movement is clearly modelled on the French rondeau (here ABACADA) and even reproduces some of its details of style, such as the cadence form at the end of the rondeau theme (bb. 15–16). The Capriccio stands in lieu of the standard gigue-finale. The texture and form of the gigue are maintained—three-part fugue in binary form with thematic inversion in the second strain—but the time and rhythm are quite different, hence the title. ‘Capriccio’, according to Mattheson, means ‘composed in accordance with one’s caprice’;67 here it presumably refers in particular to the giant strides of a 10th in bar 3 of the subject, thereafter much used sequentially in the episodes. The dances of Partita No. 3 in A minor tend to be treble-dominated, with smoothly flowing parts. Like No. 1, it inclines towards the Italian violin-sonata style. The prelude title ‘Fantasia’ no doubt refers back to the polyphonic fantasia of the seventeenth century, cultivated by Sweelinck, Frescobaldi, Froberger, and others.68 It is also worth noting that Bach’s Sinfonias were originally entitled Fantasias (in the Clavierbu¨chlein for W. F. Bach of 1720), for the Fantasia that opens the third partita resembles nothing so much as a very extended (120-bar) two-part invention. The second

Ex. 8

a) Opening of Courante from Partita No. 2 in C minor, BWV 826 (inner parts to ﬁrst chord omitted)

b) Opening of Courante from French Suite No. 1 in D minor

67 In a note to Niedt’s Handleitung II, Mattheson described capriccios as pieces ‘darinn einer seinem Sinn folget und nach seiner caprice etwas hinsetzet oder herspielet’; quoted in Jacob, Studien, p. 103. 68 See Vol. I of the present study, p. 65.

suite

movement is perhaps Bach’s most florid allemande so far, carrying further the intricate free-voiced texture and diminution rhythms of its equivalent from the second French Suite. The Corrente, with its fluent two-part texture, might almost have been written for violin and bass.69 The rhythmic movement goes a step beyond the triplets/dotted rhythms of the first partita to flowing semiquavers against dotted rhythms. The Sarabande is written in the ‘trio-sonata’ texture (two matching treble parts against bass) that Bach had already employed in the sarabandes from the fifth English Suite and the Suite in E♭, BWV 819. In this case, however, no link is established with the original dance rhythm: not only is the customary second-beat stress absent, but the anacrusis (three quavers) and the florid triplet semiquavers lie outside the dance type. The result is a charming character-piece, without parallel elsewhere. The Burlesca was, in the 1725 Clavierbu¨chlein, entitled ‘Menuet’. Its capricious changes of rhythm, texture, and material, however, render it another character-piece rather than a minuet proper, hence the change of name to ‘Burlesca’, which implies something comic and playful. Similar characteristics apply to the Scherzo, which was not in the original 1725 version of the suite but added for the 1727 print as a duple-time partner to the triple-time Burlesca. The title ‘scherzo’ was also used in F. A. Bonporti’s Inventioni da camera, Op. 10 (Bologna, 1712) and in C. F. Hurlebusch’s Compositioni musicali per il cembalo (Hamburg, n.d.).70 The Gigue, like that of the third English Suite, is a three-part fugue in flowing 12/8 quavers. The subject is accompanied by two regular countersubjects, so that the partita both begins and ends with invertible counterpoint. Partita No. 4 in D is lyrical and ornamental with intricate textures; and like No. 2, it inclines to the French style—note particularly the Ouverture and Courante. The placing of this partita in fourth place ensures that the second half of Clavieru¨bung I opens with a French overture, signifying a new beginning, as in Clavieru¨bung II and [IV]. Both framing movements, Ouverture and Gigue, contain fugues in compoundtriple time. That of the Ouverture, like the pre´ludes to the last two English Suites, amalgamates fugue and ritornello form. The crucial moment when the full-close at the end of the ritornello prefaces a new start with entirely different material has clear precedents in the pre´ludes to English Suites Nos. 2–6. The Allemande is made up of a florid cantabile treble with accompaniment (free-voiced but largely two-part), a texture that Bach made his own in Co¨then and in numerous slow movements thereafter. This Allemande exceeds all its precedents in spaciousness and in the florid written-out embellishment of its melodic lines. In these respects only the Allemande from the Cello Suite in the same key (No. 6, BWV 1012) approaches it. The intermingling in the keyboard Allemande of the most diverse rhythms, including triplets, is characteristic of the ‘mixed style’ of the time. The Courante, like that of the second

As Schulenberg remarks in The Keyboard Music of J. S. Bach, p. 285. Both collections were in Bach’s music library; see Beißwenger, J. S. Bachs Notenbibliothek, VBN I/B/50, II/B/7, and II/H/9. 70

the well-t empered clavier i et c.

partita, belongs to the French type. Despite the opening octave imitation, the texture largely consists of florid treble and accompaniment, as in the allemande. We encounter in this Courante the most subtle interaction of 3/2 and 6/4 time in all Bach’s French courantes. Similar textures—treble-dominated, often florid, with free-voiced but largely two-part accompaniment—pervade the next three movements, Aria, Sarabande, and Menuet. The Sarabande, like the Courante, is notable for its galant long appoggiaturas, notated in full rather than as ornaments. The Menuet has a clear precedent in that of the Suite in A minor, BWV 818a: triplet motion is predominant in both cases, so that the two movements might be considered character-pieces rather than minuets proper. The fugal Gigue also breaks the bounds of its type in its compound-triple time (9/16), since compound duple or quadruple was the norm both for Bach and for his contemporaries. Another innovation of note is the introduction of a new subject after the double bar (as opposed to the traditional inversion of the original subject), which is then combined with the first subject in the manner of a double fugue. Partita No. 5 in G is also treble-dominated, but with its lighter, swifter flow it inclines more to the Italian style, like No. 3, hence the Corrente and Tempo di Minuetto. The Praeambulum, despite its non-committal title, is an Italianate concerto-Allegro with short ritornellos such as Bach sometimes employed in Weimar under the influence of Torelli.71 The texture is an exceptionally brilliant and flexible one, constantly changing from a single part divided between the hands to block chords or two-part writing. A striking galant feature is the sharp cutting-off by rests of the two sub-phrases of the ritornello (bb. 1–4). The largely stepwise triplet semiquavers of the Allemande give it a smooth, even, Italianate flow, while the dotted rhythms inject a measure of variety. The Corrente likewise runs along in even semiquavers, which impart a lightness and athleticism typical of this partita. The three-voice cantabile Sarabande in dotted-quaver rhythms recalls that in the same key in the fifth French Suite, which shares the same features. The crucial difference is that the latter exhibits Bach’s typical slow-movement texture of treble solo with twopart accompaniment, whereas the partita sarabande illustrates the ‘trio-sonata’ texture of matching right-hand parts and bass, which Bach had employed in other sarabandes (BWV 819 and 827). Like the sarabandes of the Clavierbu¨chlein partitas (A minor and E minor), the G major deviates from the dance type proper in opening with an anacrusis. The Tempo di Minuetto, like the Giga from the first partita, with which it has a certain affinity, moves altogether beyond the bounds of the normal rhythm and texture of its dance type, hence the formula ‘tempo di . . .’. The continuous cross-rhythm (6/8 in 3/4) has no parallel in Bach’s minuets, though a certain resemblance has often been noted with the Menuet Le Lardon from Rameau’s Pie`ces de

71 See Jean-Claude Zehnder, ‘Giuseppe Torelli und J. S. Bach: Zu Bachs Weimarer Konzertform’, BJ 77 (1991), pp. 33–95.

suite

clavecin of 1724.72 In sharp contrast, the Passepied that follows keeps relatively close to the traditional characterization of the dance. The three-part fugal Gigue exhibits the same structure as its counterpart in the fourth partita, suggesting close temporal origin: the second strain opens with a new subject, which is later combined in invertible counterpoint with the original subject. Partita No. 6 in E minor is florid and ornamental with intricate textures—rather like No. 4, except that it inclines more to the Italian style than the French, as is clear from the movement titles Toccata, Allemanda, Corrente, and Tempo di Gavotta. The toccata-fugue-toccata form of the prelude is clearly a revival from Bach’s early, North German-influenced organ music.73 Yet the style is completely up to date—note, for example, the galant ‘sighing’ appoggiatura figures that knit together the framing toccata and the central fugue. The Allemanda, whose Italian title recalls the soloviolin partitas, is written in a florid style with diminution figures and dotted-semiquaver rhythms—a style close to that of the equivalent movement from the third partita. The Corrente is Bach’s most elaborate so far, going well beyond that of the other Clavierbu¨chlein partita. It is immensely spacious (116 bb., 54 + 62), highly motivic—at least five distinct figures take on a motivic role—and extremely inventive in rhythm, with its continuous syncopation and dotted/demisemiquaver figures against even quavers. Like the intermezzi of the fourth partita, the Air (absent from the 1725 Clavierbu¨chlein and inserted anew in the 1730 print) and Tempo di Gavotta are placed either side of the Sarabande. Perhaps Bach considered them too alike to be placed together, for they represent two different views of the gavotte, one French and the other Italian. Little of the dance type is retained, however, beyond the characteristic half-bar openings and endings, hence Bach’s titles which leave plenty of scope for freedom of interpretation. The Sarabande is quite the most florid and deeply felt (in Bach the two attributes often go hand in hand) of all his sarabandes. The theme unites repeated notes in dotted rhythm (cf. the sarabandes from BWV 818a and 825) with the galant sighing appoggiatura figure from the opening Toccata. This theme and its variants form a powerful underpinning of the texture beneath the profusely decorated surface (Ex. 9 a). No less remarkable is the fugal Gigue, a worthy successor to the chromatic fugal gigues from the fifth and sixth English Suites. As is clear from the 1725 (Clavierbu¨chlein) version, the time and rhythm are essentially those of the Gigue from the first French Suite (Ex. 9b)—dotted rhythms in cut i time, as are often encountered in the German tradition from Froberger onwards—but the partita gigue is rhythmically a great deal more complex, hence the later (1730) doubling of notevalues for the sake of greater clarity. The subject, invariably accompanied by its regular

72 See e.g., Schulenberg, The Keyboard Music of J. S. Bach, p. 293. Cuthbert Girdlestone, however, ﬁnds that ‘The skittish minuet in [Bach’s] G major Partita’ is very like Rameau’s Les Tricotets; see his Jean-Philippe Rameau: His Life and Work (London, 1957), p. 29. 73 See Vol. I, pp. 49–60.

the well-tem pered clavier i etc.

Ex. 9

a) Sarabande from Partita No. 6 in E minor, BWV 830, bb. 31–3 (LH parts only)

b) 1 Subject of Gigue from the same partita (early version) 2 Subject and answer of Gigue from French Suite No. 1 in D minor countersubject or a variant thereof, is inverted after the double bar, but the direct subject and countersubject return together at the end (bb. 49–50).

Prelude, fantasia, and fugue for organ Title

Earliest source/s

Scribe, date

Praeludium et Fuga in B minor, BWV 544 Praeludium et Fuga in E minor, BWV 548 Praeludium in C minor, BWV 546/1 Fantasia et Fuga in C minor, BWV 537 Fantasia in C minor, BWV 562/1 Fuga in F, BWV 540/2

Private poss., Zu¨rich Berlin, P 274/2 Berlin, P 286/10 Berlin, P 803 Berlin, P 490 Berlin, P 803

Autograph, 1727/31 J. S. Bach, J. P. Kellner, 1727/31 J. P. Kellner, post-1730 J. T. Krebs, date unknown Autograph, 1743/5 J. L. Krebs, by 1731

These pieces, like their predecessors from Bach’s earlier years, seem to have been written at different times and for various occasions. Accordingly, they are miscellaneous in character and, unlike the preludes and fugues for manuals only, were never collected by the composer into an organized set. The preludes and fugues in B minor and E minor, BWV 544 and 548, survive in autograph fair copies that display the same watermark and style of handwriting, pointing to the period 1727–31. There is good reason to believe that they were composed not long before that.74 Both are likely to have been conceived as prelude–fugue pairs from the outset. This is also true of the Fantasia et Fuga in C minor, BWV 537, whose two movements are joined by a tonic half-close. In the case of BWV 546, 562, and 540, however, only the freely composed movement—praeludium, fantasia, or toccata—seems to have existed in the ﬁrst place; the fugue was apparently added at a later date. In the absence of an autograph, the 74

See Werner Breig, ‘Freie Orgelwerke’, in K. Ku¨ster (ed.), Bach Handbuch (Kassel and Stuttgart, 1999), pp. 692–3.

p r e l u d e , f a n t a s i a , a n d f u gu e f o r o r g a n

chief source of the Praeludium et Fuga in C minor, BWV 546, is a manuscript copy in the hand of Bach’s colleague Johann Peter Kellner dating from some time after 1730. The fugue appears to be not by Bach at all but by an associate or pupil of his—possibly Kellner himself.75 The autograph fair copy of the Fantasia in C minor, BWV 562 no. 1, perhaps dates from 1743/5, but the unfinished fugue was added around 1747/8.76 An earlier version of the fantasia must have existed before the 1740s, for J. P. Kellner copied it out some time between 1727 and 1740.77 The Toccata et Fuga in F, BWV 540, survives only in pupils’ copies: the toccata was copied by Johann Tobias Krebs around 1714, and the fugue by his son Johann Ludwig by 1731 at the latest. It is likely that this reflects a widely separate origin for the two movements.78 The preludes in the minor keys of B, E, and C (BWV 544, 548, and 546) show Bach applying massive ritornello structures to the organ prelude. Over and above its ritornello structure, the magnificent Praeludium in B minor may be heard as bipartite, for it is articulated by dominant and tonic cadences at the middle and end, immediately preceded by a substantial rhyming-close (bb. 38–43 = 79–85). The dominant cadence occurs at the exact halfway point (bb. 42–3): Part I Bar: Para: Key:

1 A b

Part II 17 B b–f♯

27 A1 f♯

43 B1 f♯–D

56 C D–e

69 D e–b

The tonally closed paragraph A, which is largely pedaliter, possesses the character of a tutti ritornello; the manualiter paragraph B, that of a concertino episode. C acts as a further development and D ties all the ends together. It is noteworthy that the chief episode B, which is heard three times in various guises, is a three-part fugal exposition. Bach thus reverses his usual procedure in which the ritornellos are fugal and the episodes homophonic. No less unusual in the context of an organ prelude is the 6/8 metre with dotted rhythms, which is more often found in sonata or concerto movements of the siciliana type. The closest parallel with this prelude is found in pieces that clothe this dance rhythm with florid patterns in short note-values, such as the slow movements of Organ Sonatas Nos. 3, 5, and 6 (BWV 527, 529, and 530). Indeed, the florid expressivity of the prelude’s leading voices, allied to the grandeur of the four- and five-part organ textures, is the main source of its immense emotive power.

75 See W. Breig, ‘Versuch eine Theorie der Bachschen Orgelfuge’, Musikforschung, 48 (1995), pp. 14–52 (esp. 16–18), and his ‘Freie Orgelwerke’, p. 702. 76 According to Kobayashi Chr, p. 59. 77 All dates of Kellner sources are drawn from Russell Stinson, The Bach Manuscripts of Johann Peter Kellner and His Circle (Durham and London, 1989). 78 See Breig, ‘Freie Orgelwerke’, p. 696.

60 the well-tempere d clavier i etc. The Praeludium in E minor, which might have been composed around the same time, is not bipartite, but its structure is nonetheless lucid owing to the very clear distinction that is made between ritornello and episodic material: Bar: Para: Key:

1 A e

19 B e–b

33 A b

51 C b–G

81 A1 G–a

90 B1 a

103 C1 + A2 a–e

The ritornellos are built on a turn figure followed by a repeated-note plus appoggiatura figure. The latter plays a prominent part in several roughly contemporaneous pieces in the same key of E minor (Ex. 10): the Toccata and Sarabande from Partita No. 6 (BWV 830) and the finale of Organ Sonata No. 4 (BWV 528). This style of writing must have seemed a` la mode to Bach and his contemporaries in the 1720s, and it hardly occurs in his music before that period. In the E minor prelude, the theme thus described, plus two sequences of great strength (bb. 7 and 12), constitute one of Bach’s most powerful ritornellos. The chief episodic formulations are a characteristic invertible trio combination (b. 19), heard at greater length later on (bb. 55, 65, 111, and 121), and a dotted-rhythm sequence (bb. 51, 61, 90, and 114), the only case in this prelude where the resting of the pedals is used to define an episodic theme. In the Praeludium in C minor, BWV 546 no. 1, a da capo of the opening ritornello and a reverse-order subdominant reprise create a palindromic structure thus: Bar: Para: Key:

1 A c

25 B c–g

49 C (A + B) g–f

97 B1 f–c

Ex. 10

a) Incipit of Praeludium in E minor, BWV 548 no. 1 (treble only)

b) 1st bar of Toccata from Partita No. 6 in E minor (RH only)

c) Opening of Sarabande from same partita (treble only)

d) Incipit of ﬁnale from Organ Sonata No. 4 in E minor, BWV 528

120 A c

prelude, fa ntas ia, and fugue for organ

Ex. 11

a) Opening of Praeludium in C minor, BWV 546 no. 1 Allegro

b) Opening of 1st movement from Wer sich selbst erho¨het, BWV 47 (strings, oboes, and continuo) The opening ritornello theme, in which the repeated-note plus appoggiatura figure already encountered in the E minor prelude is treated antiphonally, so closely resembles that of the cantata Wer sich selbst erho¨het, der soll erniedriget werden, BWV 47, of October 1726 as to raise the possibility that one theme might have been invented with the other in mind (Ex. 11). As in the B minor prelude, the episodes are fugal, in this case on the basis of a double subject: long-note scale figures against triplet figures developed out of the ritornello. This combination is subsequently enriched by doubling the longnote theme in 3rds or 6ths (bb. 53, 97, and 117) and, at the climactic return of the tonic key in the central paragraph C, by doubling both themes in 10ths (b. 82). The Fantasia in C minor, BWV 562 no. 1, is quite unlike the organ preludes considered so far. Its texture of intricate five-part counterpoint, adorned by profuse French ornamentation, and even the theme itself are often thought to be indebted to the Gloria fugue from de Grigny’s Premier livre d’orgue (Ex. 12), which Bach had copied out in Weimar around 1709/12.79 The fantasia possesses none of the thematic or textural contrasts of the ritornello-based preludes. On the contrary, it is pseudo-fugal throughout on the basis of its single-bar subject which saturates the texture, occurring in virtually every bar. The movement is not continuous, however, like many fugal movements, but is articulated by cadences into distinct periods: exposition, dominant

79 See Victoria Horn, ‘French Inﬂuence in Bach’s Organ Works’, in G. Stauffer and E. May (eds.), J. S. Bach as Organist (London, 1986), pp. 256–73 (esp. 266–7); and George Stauffer, ‘Boyvin, Grigny, d’Anglebert, and Bach’s Assimilation of French Classical Organ Music’, Early Music, 21 (1993), pp. 83–96 (see 90).

the well-te mpered clavier i et c.

Ex. 12

a) Opening of Fantasia in C minor, BWV 562 no. 1 (pedal bass omitted)

b) Nicolas de Grigny, opening of Gloria from Premier livre d’orgue counter-exposition (b. 13), three further expositions (bb. 23, 38, and 51), and coda (b. 71). Reminiscent of traditional preludial style are the long pedal points that deﬁne the keynote of every period but one. The other Fantasia in C minor, BWV 537 no. 1, is similar enough to suggest a common period of origin. Particularly close are the opening periods of the two fantasias, with their fugal entries descending through the parts over a tonic pedal. Again, it has been suggested that French music might lie behind this conception, especially that of de Grigny.80 The major difference between the two fantasias is that, whereas BWV 562 is monothematic, BWV 537 alternates between two contrasting themes according to the bipartite scheme A (a b) + A1 (a1 b1).The secondary theme is a highly characteristic quaver-couplet formulation (b. 11), again treated imitatively in four voices. Later treatment of this theme is greatly extended and varied: it involves ten imitative entries in place of the original four and the inverted as well as the direct form of the theme. All the fugues display the tripartite form that became standard in Bach’s organ fugues of the Leipzig period.81 The manner in which it is carried out differs widely, but one element remains constant: a search for the maximum possible contrast between the central and framing paragraphs. In the reprise-form (ABA1) Fuga in B minor, the framing A-sections are in effect ‘tuttis’, with their full four-part texture including pedals, whereas the middle B-section (b. 28) is for manuals only, creating something of the effect of a concertino episode. The contrast extends to thematic material. Much of the character of section A derives from its regular countersubject in progressively longer note-values. Section B dismisses it, however, in favour of a new countersubject in smoothly ﬂowing semiquavers. It also includes somewhat modish homophonic episodes (bb. 37–40 and 50–3), quite unlike anything in the strictly fugal framing sections. At the reprise of A (b. 59), not only does the original countersubject return, but it is combined with a new countersubject in the form of a decorated, sequential,

80 81

See P. Williams, The Organ Music of J. S. Bach (Cambridge, 1980), vol. i, p. 84. See Breig, ‘Versuch eine Theorie’, pp. 38–41, and ‘Freie Orgelwerke’, pp. 688–90.

prelude, fantasia, and fugue for organ

descending arpeggio figure. The combination of the three themes, S + CS I + CS II, provides a highly effective concluding enhancement. In the Fuga in E minor, an even greater contrast is created within an overall ABA da capo form: essentially, the framing paragraphs are fugal, whereas the central one has the character of a toccata—the reverse of Bach’s procedure in the similarly tripartite Toccata from the partita in the same key (BWV 830). The character of paragraph A is defined by the wedge-shaped, chromatic fugue subject and its (at first) somewhat angular countersubject. The central B-paragraph differs not only in its manuals-only texture (for the first 21 bars) but in its shorter note-values (semiquavers) and its toccata-style figuration, based on a decorated broken triad. This toccata writing alternates with pedal entries of the fugue subject (bb. 69 and 81), so that fugue and toccata are intertwined (as they are by other means in the partita’s Toccata). Paragraph B and the return of A are effectively dovetailed at bb. 173–7: the first fugal entry of the da capo is built into the last four bars of B. The Fuga in F (BWV 540 no. 2) is also tripartite, but this time in the context of a triple fugue. The three sections are A: double exposition of S I (first subject); B: double exposition of S II; C (= A + B): exposition of S I + II. Bach had employed much the same scheme in his early years (before 1710) in the Fuga in C minor on a theme of Legrenzi, BWV 574b.82 In section A, the plain alla breve subject is combined at every entry with a countersubject made up of conjunct crotchets, which (whether direct or inverted) pervades the episodes too. Section B is for manuals only (as in the B minor fugue) for maximum contrast. Furthermore, the new subject S II is disjunct and thus utterly different from the smooth themes of section A. The end of B is somewhat problematic: only two bars (bb. 126–7) are left for the modulation (c–d) that will bring about the tonic return, and the chromatic writing that Bach uses to cover the join is not altogether convincing. The first statement of the combined subjects, S I + II (b. 134), persuasively coincides with the re-entry of the pedals. In the ABA1 reprise structure of the Fuga in C minor, BWV 537 no. 2, the contrast between the central and framing sections is as great as that of the E minor fugue, albeit of a very different kind. Unusually for Bach, however, the middle and outer sections remain largely unrelated and little attempt is made to establish a connection between them. Section A is built on a concise four-bar subject with a repercussion headmotive and a falling diminished 7th at the central watershed—a theme quoted in a variant form by Johann Mattheson in his treatise Der vollkommene Capellmeister of 1739 (Ex. 13).83 There is no regular countersubject and the texture is largely unmotivic. Section B (b. 57), however, is as strict as A is free. It is built on a white-note soggetto made up of a rising chromatic 4th. The regular countersubject consists of a quaver motive in threefold sequence. This theme is, in fact, anticipated in section A (bb. 24–7,

See Vol. I of the present study, pp. 196–7. Mattheson’s version of the subject is quoted by Williams, The Organ Music (1980), vol. i, p. 86, and by Breig, ‘Freie Orgelwerke’, p. 691. 83

the well-tempered clavier i etc.

Ex. 13

a) Subject of Fuga in C minor, BWV 537 no. 2

b) Similar theme quoted by Johann Mattheson, Der vollkommene Capellmeister, p. 209 36, 50–1, and 54), but the two subjects themselves, S I and S II, remain entirely separate. How far the repise represents Bach’s intentions remains unclear. In the chief source (P 803), the fantasia and bars 1–89 of the fugue are written out by Bach’s pupil Johann Tobias Krebs. The last 41 bars of the fugue, however, are in the hand of Krebs’s son Johann Ludwig and are dated at the end 10 January 1751. It has been suggested that the elder Krebs might have broken off because his Vorlage was incomplete. The younger Krebs then perhaps undertook his own completion, which might explain the somewhat problematic nature of the transition between the central paragraph and the reprise.84

See John O’Donnell, ‘Mattheson, Bach, Krebs and the Fantasie & Fugue in C minor, BWV 537’, Organ Yearbook, 20 (1989), pp. 88–95.

I.3 The Brandenburg Concertos and other instrumental works

Concertos and ouvertures Title

Earliest source/s

Scribe, date

Brandenburg Concertos, BWV 1046–51 Violin Concerto in E, BWV 1042 Ouverture in C, BWV 1066 Ouverture in D, BWV 1069

Berlin, Am. B. 78 Berlin, P 252 Berlin, St 152 Berlin, St 160

Autograph, 1721 J. F. Hering, c. 1760 Anon., c. 1724 C. F. Penzel, c. 1755

In March 1719 Bach was paid by the Co¨then court for travelling up to Berlin to collect a new ‘grand harpsichord with two manuals by Michael Mietke’.1 Almost exactly two years later, on 24 March 1721, Bach dedicated Six Concerts avec plusieurs instruments, now known as the ‘Brandenburg Concertos’, to Christian Ludwig, Margrave of Brandenburg, who at that time of year resided in the royal palace, the Berliner Schloss.2 From the dedication we learn that Bach had played before the Margrave and received a commission from him two years before, presumably on the occasion of Bach’s 1719 visit to Berlin: Monseigneur. Comme j’eus il y a une couple d’anne´es, le bonheur de me faire entendre a` Votre Alteße Royalle, en vertu de ses ordres, & que je remarquai alors, qu’Elle prennoit quelque plaisir aux petits talents que le Ciel m’a donne´s pour la Musique, & qu’en prennant Conge de Votre Alteße Royalle, Elle voulut bien me faire l’honneur de me commander de Lui envoyer quelques pieces de ma Composition: j’ai donc selon Ses tres gracieux ordres, pris la liberte´ de rendre mes tres-humbles devoirs a` Votre Alteße Royalle, par les presents Concerts, que j’ai accommode´s a` plusieurs Instruments. (As I had, a couple of years ago, the pleasure of appearing before Your Royal Highness, by virtue of Your Royal Highness’s commands, and as I noticed then that Your Highness took some pleasure in the small talents that Heaven has given me for Music, and as in taking leave of

BD II, No. 95; NBR, No. 77. For a vivid account of the surrounding circumstances, see Sheridan Germann, ‘The Mietkes, the Margrave and Bach’, in P. Williams (ed.), Bach, Handel, Scarlatti: Tercentenary Essays (Cambridge, 1985), pp. 119–48. 2

t h e b r a nd en b u r g c o nc e r t o s e t c .

Your Royal Highness, Your Highness deigned to honour me with the command to send Your Highness some pieces of my composition: I have then, in accordance with Your Highness’s most gracious orders, taken the liberty of rendering my most humble duty to Your Royal Highness with the present concertos, which I have adapted to several instruments.3

It stands to reason, then, that the Brandenburg Concertos were composed, or at least compiled, within the two-year period from March 1719 to March 1721. By that time Bach had at least ten years of experience of the concerto genre—a reflection that elevates the six concertos to the status of culmination, which is entirely in keeping with the maturity and sophistication of the music. The fact that no original concertos by Bach survive before the Brandenburgs is surely misleading: many concertos from that period must be lost or survive only in later adaptations for harpsichord. None of the surviving concertos, however (with the single exception of BWV 1044), displays, even in its presumed original form, anything approaching the same highly diverse choice and disposition of instruments as the Brandenburg Concertos. While one or two of these concertos, then, might have originated in some form before the years 1719–21, the frequently heard theory that the six works were selected from an existing stock of concertos, perhaps mainly composed in Weimar, seems highly unlikely. It is true that Concertos Nos. 1 and 5 survive in early versions, but that of No. 5, BWV 1050a, might have originated after Bach’s commission of 1719.4 Only in the case of the first concerto is there good reason to believe that the early version might have existed before the Brandenburg set was conceived. The early version of Concerto No. 1 in F, BWV 1046a, which comprises three movements—[Allegro], Adagio, and Menuet (with two Trios)5—is quite unlike any other concerto by Bach. It is scored for ripieno only (that is, a relatively large, normally accompanying ensemble), which here consists of three contrasting instrumental groups, often deployed in antiphonal exchanges: two horns, three oboes and bassoon, and the usual four-part strings. It thus lacks the solo or concertino (that is, solo group) element found in all Bach’s other concertos. The opening Allegro is far removed from the Venetian concerto style, so clearly evident in the other Brandenburgs and in the majority of Bach’s other concertos. And the minuet-finale is unique. The key to all these anomalies surely lies in the title, for this early version is not a concerto at all but

BD I, No. 150; NBR, Nos. 83–4. This version survives in a late copy of the original performing parts, Berlin St 132. See Alfred Du¨rr, ‘Zur Entstehungsgeschichte des 5. Brandenburgischen Konzerts’, BJ 61 (1975), pp. 63–9, and his supplement to NBA VII/2 (1975), pp. 243–84. The chief scribe of the parts has been identified as Johann Christoph Farlau by ¨ berlieferung in MittelPeter Wollny, ‘Tennstedt, Leipzig, Naumburg, Halle—Neuerkenntnisse zur Bach-U deutschland’, BJ 88 (2002), pp. 29–60 (see 36–47). Du¨rr suggests that this original version might date from early 1719 as a demonstration of ‘das zu Berlin gefertigte Clavessin’ (BD II, No. 95). The autograph parts, Berlin St 130, originated around the same time as the autograph score, Berlin Am. B. 78, according to Georg von Dadelsen, Beitra¨ge zur Chronologie der Werke Johann Sebastian Bachs (Trossingen, 1958), pp. 82–5, and essentially represent the same version. 5 Movements nos. 1, 2, and 4 of the definitive version (but without the Polonaise in the finale). 4

c o n ce r t o s a n d ouvertures

a ‘sinfonia’, which for Bach and his contemporaries signified primarily the instrumental introduction to a vocal work, whether sacred or secular. A clue to the nature of this lost vocal work—assuming that is where the sinfonia originated—might lie in the startling use of horns in the outer movements (maintained in the later concerto). In the opening tutti of the first movement, the strings and oboes are accompanied by arresting horn calls in imitation, whose rhythm, in conflict with the other instruments, gives them a subversive character. It has been shown that Bach here quotes literally a contemporary greeting call that would have been familiar to the huntsmen of Saxony.6 Furthermore, the ‘Trio pour les Cors de chasse’ (Trio II from the minuetfinale) has been aptly described as a piece of Germanic hunt music.7 Due to its use of hunting horns, coupled with the identity of key and instrumentation, the sinfonia was formerly linked to the Hunt Cantata, BWV 208, composed by Bach in 1713 for the birthday of Duke Christian of Weißenfels, a passionate devotee of hunting. This connection has recently been discredited,8 but we cannot exclude the possibility that at some later date, presumably between 1713 and 1719, Bach composed a second ‘hunt’ cantata for Duke Christian that might have been introduced by the sinfonia under discussion. Bach’s cantata sinfonias are normally made up of a single movement only, but the lost pastoral cantata composed for Duke Christian in 1725, Entfliehet, verschwindet, entweichet, ihr Sorgen, BWV 249a, whose music mostly survives in the Easter Oratorio, was introduced by a two-movement sinfonia— [Allegro], Adagio—equivalent to the first two movements of the Brandenburg sinfonia. Moreover, three-movement sinfonias were standard in Italian vocal music of the time; and it is perfectly possible that in the case of the lost ‘hunt’ cantata for Duke Christian (assuming that the theory just outlined is correct) Bach adopted the full three-movement Italian form of sinfonia, adding the minuet and its trios as finale. The Menuet, played alternativement with two trios, would, of course, have been at home in a French overture-suite, and Trio I is overtly Lullian, with its scoring for two oboes and bassoon. In this context, it is worth noting that the ‘polychoral’ writing for contrasting choirs of instruments in the opening Allegro of the Sinfonia is paralleled by that of the quick fugal sections in Bach’s ensemble Ouvertures in C and D, BWV 1066 and 1069 respectively. This style of writing also occurs in several Weimar and preWeimar cantatas, however—most significantly, in the finale of the Hunt Cantata, BWV 208, where the same three instrumental choirs (two horns, three oboes and bassoon, and four-part strings) come into play.

6 See Horace Fitzpatrick, The Horn and Horn Playing and the Austro-Bohemian Tradition from 1680 to 1830 (London, 1970), pp. 20–1. 7 By Michael Marissen, The Social and Religious Designs of J. S. Bach’s Brandenburg Concertos (Princeton, 1995), p. 32. 8 See M. Marissen, ‘On Linking Bach’s F-major Sinfonia and His Hunt Cantata, BWV 208’, Bach, 23/2 (1992), pp. 31–46.

68 t he b rand enburg concertos et c. The origin of the first Brandenburg Concerto as a cantata sinfonia, if correct, would explain the curious discrepancy between its concertante forms and the relative paucity within it of the characteristic idioms of the Italian concerto. For in the concertante works of the Weimar period—cantata sinfonias and choruses; keyboard preludes, toccatas, and fugues; organ chorales—key features of the genre concerned are left recognizably intact; only circumscribed aspects of the concerto are brought to bear upon it. The chief concertante elements in the Brandenburg sinfonia are the ritornello structuring of the opening Allegro and the florid oboe/violin solos and duets of the following Adagio, which are strongly reminiscent of the Sinfonia to Ich hatte viel Beku¨mmernis, BWV 21 (Weimar, 1713). When the Brandenburg sinfonia was revised to form a concerto, presumably in 1719 or shortly afterwards, further concertante elements were added. A third-movement Allegro in gigue rhythm, very much of the type of a concerto-finale, was inserted in order that the first three movements should reflect the standard fast–slow–fast scheme of the Italian concerto. And this new movement has a prominent solo part for violino piccolo, which also takes over the florid Violin I part in the Adagio, though it plays virtually no independent part in the outer movements. The suite element in the original sinfonia is also reinforced, however, by the insertion of an additional trio in the form of a polonaise, so that the minuet is now played in alternation with French, Polish, and German dances,9 and reeds, strings, and horns are each brought into the limelight in succession. Thus, far from suppressing the suite element of the work when he revised it, Bach made a virtue of the sinfonia’s hybrid concerto-suite character. Five years after the dedication of the Brandenburg Concertos, he performed a variant version of the interpolated Allegro third movement as the opening chorus of the secular cantata Vereinigte Zwietracht der wechselnden Saiten, BWV 207 (Leipzig, 1726). It has been argued convincingly that the vocal version is probably (in an earlier incarnation) the original and the instrumental version a derivative, and that both might have a common source in a lost secular cantata of the Weimar or early Co¨then years. Detailed formal and thematic links between this Allegro and the first movement of the sinfonia/concerto suggest that the original choral movement and the sinfonia might have belonged to one and the same secular cantata—the sinfonia might have been followed by the movement in question as opening chorus.10 The relationship between Brandenburg Concerto No. 1 and the remainder of the set finds an illuminating parallel in the roughly contemporaneous English Suites. There, too, a Weimar work of mixed French and German stylistic provenance, but with relatively little Italian input, namely the Suite in A, BWV 806a, was adapted and enlarged in Co¨then to become the opening item in a set whose following items are

As pointed out by Marissen, The Social and Religious Designs, p. 31. See Malcolm Boyd, Bach: The Brandenburg Concertos (Cambridge, 1993), pp. 60–70, and Michael Talbot, ‘Purpose and Peculiarities of the Brandenburg Concertos’, in M. Geck and K. Hofmann (eds.), Bach und die Stile (Dortmund, 1999), pp. 255–89. 10

c o n ce r t o s a n d ouverture s

more closely interlinked and more clearly indebted to the Italian style.11 In the case of the Brandenburg Concertos, what might have induced Bach to open the set with an adapted composition? The theory has been put forward that the musical aim of the set was to construct a cycle of varied works out of a common fund of thematic material, presented in the opening Allegro of Concerto No. 1.12 This view rests on the premise that all six concertos are based on the same circumscribed group of motives. However, while certain motivic interconnections are perceptible, many of the so-called motives cited are, strictly speaking, not discrete motives at all but rather ﬁgural fragments. Nor are they peculiar to the Brandenburg Concertos: they belong to Bach’s concerto style in general and hence turn up frequently in other concertante compositions, such as the Violin Concerto in E, BWV 1042, the Harpsichord Concerto in D minor, BWV 1052, or the Sonata in G minor for viola da gamba and obbligato harpsichord, BWV 1029 (Ex. 1). Bach’s decision to open the Brandenburg set with the sinfonia-concerto

Ex. 1

a) 1. Incipit of 1st movement from Brandenburg Concerto No. 2 in F, BWV 1047 (top part only) 2. 1st movement of Violin Concerto in E, BWV 1042, b. 2 (top part only)

b) 1. Incipit of 1st movement from Brandenburg Concerto No. 3 in G, BWV 1048 (violins’ part only) 2. Finale of Harpsichord Concerto No. 1 in D minor, BWV 1052, bb. 13–14 (top part only) 3. Incipit of 1st movement from Sonata in G minor, BWV 1029 (gamba only)

11 Cf. the Courante & doubles of English Suite No. 1 and the Menuet & Trios of Brandenburg Concerto No. 1. 12 See Talbot, ‘Purpose and Peculiarities’, pp. 258–64.

70 t he b rand enburg concertos et c. thus probably has little to do with the establishment of a motivic fund from which to draw in subsequent concertos. It is more likely to be connected with the identity of the adapted work as, in essentials, an ensemble concerto, one without concertino–ripieno differentiation (albeit incorporating a solo-violin third movement); a concerto, moreover, scored for the largest and most diverse instrumental ensemble of all, comprising two brass instruments, four woodwind, a minimum of six strings, and harpsichord. Vivaldi employed a similar combination (though with two oboes rather than three) in four concertos that were evidently performed in Dresden by the violin virtuoso Johann Georg Pisendel, a pupil of Torelli and Vivaldi, and an acquaintance of Bach’s since 1709.13 The musical aim of the Brandenburg Concertos as a set, then, is unlikely to be connected with motivic unity. The key to it surely lies rather in the title of the autograph dedication score: ‘Six Concerts avec plusieurs instruments’. The word ‘plusieurs’ here may be understood to mean not just ‘several’ but ‘diverse’; for the equivalent Italian phrase ‘a piu` stromenti’ was habitually applied to contemporaneous Italian concerto sets whose instrumentation varies from work to work.14 Undoubtedly the most arresting feature of the Brandenburg set as a whole is the quite different scoring of each concerto. The minimum number of players or instruments required for the first half of the set (Concertos Nos. 1–3: 13, 10, 11) is significantly greater than that of the second half (Nos. 4–6: 9, 7, 7). Thus a progressive reduction takes place after the amply scored Concerto No. 1. The last concerto of each half (Nos. 3 and 6) is scored for strings only, whereas the first two concertos of each half (Nos. 1–2 and 4–5) are scored for diverse mixtures of strings and wind, with a gradual reduction in the number of participating wind instruments (Nos. 1 and 2: 6, 3; Nos. 4 and 5: 2, 1). Only the first two of these mixed-ensemble concertos involve brass instruments. On the other hand, all four require solo violin. The richly varied instrumentation of the Brandenburg Concertos is not without precedent: a rough parallel may be found in the Concerti da camera, Op. 1 (Amsterdam, c. 1713), by Francesco Venturini, Concertmeister and later Capellmeister at the court of Hanover.15 Moreover, German and Italian composers such as Telemann and Vivaldi wrote concertos for similar combinations of instruments. The Saxon capital Dresden, with which Bach had frequent dealings from 1717 onwards, was at that time a notable centre for the performance and composition of such richly and variously scored concertos. If Brandenburg Concerto No. 1 may be described as a concerto ripieno (that is, without concertino),16 there are clear signs that No. 2 might have originated as a

RV 568, 569, 571, and 574. As Boyd points out in The Brandenburg Concertos, p. 24. 15 As pointed out by Talbot, ‘Purpose and Peculiarities’, p. 255. 16 Precedents may be found in Torelli’s Sinfonie a tre e concerti a Quattro, Op. 5 (1692) and Concerti musicali, Op. 6 (1698). 14

concertos and ouvert ures

concerto da camera, scored for a concertino of trumpet, recorder, oboe, and violin (plus continuo) only—that is, without ripieno strings.17 Vivaldi composed eleven such concerti da camera for four solo instruments and continuo, the most usual combination being flute (transverse or recorder), oboe, violin, and bassoon.18 The first two Brandenburg Concertos, then, might have been conceived as a complementary pair: both are in the same key (F) and both require a mixture of brass, woodwind, and string instruments; but the first is scored for ripieno only (large ensemble), and the second originally for concertino only (small ensemble). The subsequent addition of a string band to Concerto No. 2 might have been designed to bring it more in line with No. 1 or else to achieve an equal balance within the set between concerti grossi (Nos. 2, 4, and 5) and ensemble concertos (Nos. 1, 3, and 6)—that is, between those with and without ripieno–concertino differentiation. However that may be, Concerto No. 1 is built on the principle of contrasting groups of instruments, No. 2 on the contrast between the very different solo instruments themselves—representatives of the string, reed, flute, and brass families. It is a very bold combination in which blend is eschewed in favour of colour contrast. There is a clear hierarchy within the concertino, with the trumpet acting as primus inter pares in the outer movements (the other solo instruments are compensated by having the slow movement to themselves). As with the horns of the first concerto, some extra-musical factor might have determined the virtuoso use of solo trumpet, perhaps connected with the Margrave or his Berlin Capelle; for example, the availability of a celebrated trumpeter, or the symbolic significance of the trumpet as the instrument of royalty par excellence. The distinctive character of the concerto is determined in large part by the radiant tone and commanding presence of the solo trumpet. It is also, however, a product of the kaleidoscopic changes of tone colour that are heard as the themes pass from one solo instrument to another within the concertino. It has been said that Bach writes for the four solo instruments as if they were interchangeable.19 Attention has been drawn, in particular, to the first episode of the opening Allegro (bb. 9–22), whose solo violin theme is soon passed to the oboe, to the accompaniment of a cross-string figure for the violin. When the theme recurs in the recorder and trumpet parts in turn, the accompaniment figure is still present in the oboe and recorder parts respectively, even though it is idiomatic to the violin only (Ex. 2). The truth is, however, that there is no such thing as a mere accompaniment figure in Bach: with his essentially polyphonic outlook, he views it as a regular countersubject to the theme. Elsewhere, the parts for the four concertino instruments are not truly interchangeable but take account of the distinctive qualities of each instrument; for example, double-stops and cross-string

17 See Klaus Hofmann, ‘Zur Fassungsgeschichte des zweiten Brandenburgischen Konzerts’, in M. Geck (ed.), Bachs Orchesterwerke (Witten, 1997), pp. 185–92. There is a clear parallel with Bach’s Concerto in C, BWV 1061, which also originated as a concerto senza ripieno (BWV 1061a). 18 RV 87–8, 90, 94–5, 98–9, 101, 104–5, and 107. 19 See Boyd, The Brandenburg Concertos, pp. 36–7.

72 t h e br a n d e n b u r g c o n c e r t o s e t c .

Ex. 2

1st movement from Brandenburg Concerto No. 2 in F, bb. 17–18, recorder and oboe (continuo omitted) figures for the violin; an exceptionally high, agile part for the recorder; and triadic writing for the natural trumpet. It is possible that the earlier and later versions of Concerto No. 2 might have influenced or even determined the instrumental disposition of Nos. 3–6. The original concerto da camera for four solo instruments and continuo might have led to the composition of Nos. 3 and 6 for concertino only; and the revised version, with added ripieno strings, to that of the two concerti grossi (with clear ripieno–concertino differentiation), Nos. 4 and 5. Among the concertino-only compositions, the connection between Concerto No. 2, in its original version, and No. 3 is reinforced by other considerations. The first-movement ritornellos of these two concertos are markedly similar in style and perhaps adhere more closely than the other Brandenburgs to contemporary Venetian concerto style. Moreover, certain specific figural or motivic links can be established: between the anapaest (cf. Exx. 1 a and 1b) and cadential figures (Ex. 3) of the opening ritornellos and between episodic tuttis that occur later in the first movements of the two concertos. Concertos Nos. 3 and 6 are interlinked in other ways: both are to be regarded as concerti senza ripieno, a type of chamber concerto much cultivated by Vivaldi, by whom no fewer than twenty-two such concertos ‘without orchestra’ are known.20 Bach’s string ensembles in the two Brandenburg Concertos prove infinitely flexible, functioning variously as solo or accompanying instruments, as equal antiphonal groups, or as full tuttis. The three groups of strings in Concerto No. 3—three each of violins, violas, and cellos—contribute to the texture on equal terms; only the violone takes a purely accompanimental role, playing within the continuo group alongside the harpsichord. In Concerto No. 6, on the other hand, a clear though unstated hierarchy is maintained among the six string instruments:21 the two violas and, to a lesser extent, the cello are allotted virtuoso solo parts and thus might be said to constitute an ‘unofficial’ concertino, whereas the two gambas and the violone take a secondary, often accompanying role and thus stand in lieu of ripieno.

20 21

RV 87–108. It is possible that number symbolism played a part here.

c o nc e r t o s a n d ouvertures

Ex. 3

a) 1st movement from Brandenburg Concerto No. 2 in F, b. 8 (top part only)

b) 1st movement from Brandenburg Concerto No. 3 in G, b. 7 (top part only) Certain key features of Concerto No. 2, in its definitive concerto grosso version, are reproduced in the presumably later concerti grossi Nos. 4 and 5: above all, the concept of a concertino with mixed strings and wind, and with an indisputable primus inter pares—the trumpet in No. 2, the violin in No. 4, and the harpsichord in No. 5. This leading solo instrument has the lion’s share of the virtuoso writing, though it is interesting to note that the other concertino members are often accorded a greater share of the thematic work. The fugal finale is another prominent feature of Concerto No. 2 that is reproduced in Nos. 4 and 5, though it is handled with considerably greater sophistication in the two last-named compositions—probably a sign of their later origin. In addition, the slow movement of Concerto No. 2 seems conceptually almost like an earlier adumbration of that of No. 5: in both cases, the ensemble is reduced to concertino only and a brief but clearly profiled rhythmic motive is constantly interchanged between the solo instruments in a four-voice texture. Once again, the concept is realized in a more sophisticated manner in the presumably later Concerto No. 5. Of the three concerti grossi, Nos. 4 and 5 are particularly closely linked: in their concertino trio with violin and flute/s; in their brilliantly idiomatic writing for the leading soloist, which often gives them the character of solo violin or harpsichord concertos; in the concitato string writing of their opening Allegros; and in the fusion of fugue, ritornello, and da capo or reprise form that characterizes their finales. This close relationship strengthens the impression that the six concertos might have been conceived in three complementary pairs: Nos. 1 and 2, Nos. 3 and 6, and Nos. 4 and 5.22 In compiling the Six Concerts avec plusieurs instruments, it appears that Bach sought the maximum variety not only of scoring and instrumental disposition but of concerto forms and styles. The great diversity among the concertos in this respect has often led to the view that they originated at widely different times, but it is surely an essential part of their conception. They reflect the many extremely diverse modes of treatment that Bach had at his disposal by this time, when he had at least ten years of

22 Some years later, the six keyboard Partitas seem to have been conceived in identical pairings. See Richard D. P. Jones, ‘The History and Text of Bach’s Clavieru¨bung I’, diss., University of Oxford, 1988, pp. 79–83.

74 the brandenburg concertos etc. experience of the concerto genre behind him, as performer, arranger, and composer, and had reached the very height of his powers. His conception of the concerto was built up from his knowledge of works in the genre by Albinoni, Torelli, Telemann, early Vivaldi—especially his L’estro armonico, Op. 3 (Amsterdam, 1711)—and others during the period 1707–14. This explains why any facile analysis of the Brandenburg Concertos, or indeed of other Bach concertos, in terms of ritornello–episode alternation or a fixed correlation between tutti and ritornello or solo and episode can be highly misleading. In the first place, ritornellos and episodes are not usually the largest building blocks in Bach’s concerto-Allegros, and neither were they in many of the concertos that Bach transcribed for organ or harpsichord during the Weimar years. In certain Vivaldi–Bach concertos, for example (BWV 972, 976, 978, and 593), an initial ritornello–episode complex, which might modulate to the dominant key, recurs in modified form towards the end of the movement, now in the tonic only. Some sort of reprise structure (ABA1) is thus superimposed on the ritornello scheme. Such a reprise form would become Bach’s most favoured option, alongside da capo form, for the overall structuring of his vocal arias. For the present purpose it is more relevant to note that all of the fast ritornello-form movements in the Brandenburg Concertos— that is, all outer movements but the finales of Nos. 2 and 3 (a fugue and a binary-dance movement respectively)—are either cast in pure da capo form (ABA) or may be construed as falling into some kind of overall ABA1 reprise form. The reciprocal relationship between aria- and concerto-ritornello form, despite the very real differences between them, is thus no less evident in Bach than in the Venetians Albinoni and Vivaldi, who were equally at home as composers in the operatic world and in that of the instrumental concerto. In the reprise-form movements from the Brandenburg Concertos, as in the VivaldiBach concertos cited earlier, the opening paragraph (A) often ends in the dominant with material that will be recapitulated in the tonic in the concluding paragraph (A1), so that the overall key scheme of the framing paragraphs is A: I–V; A1: I–I (third movement of Concerto No. 1; first of Nos. 2, 5, and 6). In some cases, clear cadences separate the three paragraphs, as in da capo form, but elsewhere the joins are bridged by modulatory transitions. The most subtle of these occur at the key articulation points in the opening Allegro of Concerto No. 5, where the dominant-key ritornello (bb. 19–39), much expanded by episodic development of its three phrases, bridges the join to the middle paragraph, and a ‘false’ dominant-key reprise of the first ritornello phrase (bb. 101–2), episodically diverted, anticipates the ‘true’ reverse-order tonic recapitulation (episode–ritornello in I–I, as opposed to the original ritornello– episode in I–V). Typically, the middle paragraph (B) is responsible not only for the most far-reaching tonal excursions but for introducing new episodic material of a highly contrasting nature, at least some of which returns, tonally adapted, in the concluding paragraph (A1), so that all the ends are tied together. This last paragraph functions chiefly, however, as a tonic reprise of the first, varied by reverse-order return

c o nc e r t o s a n d ouvertures

of its constituents (first movement of Nos. 1 and 5), by omission of the opening ritornello (third movement of No. 1), or by more radical restructuring (third movement of No. 4, first of Nos. 5 and 6). Only in the opening Allegro of Concerto No. 2 does the tonic reprise consist of the ritornello only, albeit in the episodically expanded version of the middle paragraph. The pure da capo forms among the Brandenburg Concertos are the first movement of No. 4 (that of No. 3 may be considered a pseudo-da capo form) and the finales of Nos. 5 and 6. The opening Allegro of Concerto No. 4 reproduces standard tonal features of the da capo aria as cultivated in Italian operas and cantatas from Alessandro Scarlatti onwards: the opening paragraph (A; bb. 1–82) in I–V–I; and the middle paragraph (B; bb. 83–344) modulating to vi, then to iii, with a tonal hiatus (here bridged by a brief bass figure) before the tonic return at the start of the concluding paragraph (A; bb. 345–427). All these features it has in common with the equivalent movement from Concerto No. 3. The crucial moment at the join between the first two paragraphs, when the tonic cadence that closes the tutti ritornello gives way to an episodic cross-string figure for solo violin over a tonic pedal, is so alike in the two movements, even down to the exact harmonic progression, as to cast doubt on the oft-stated view that the two concertos originated years apart, with No. 3 as much the earlier of the two (Ex. 4). On the contrary, this comparative analysis vindicates a

Ex. 4

a) 1st movement from Brandenburg Concerto No. 3 in G, bb. 46–50 (Violin I and continuo only)

b) 1st movement from Brandenburg Concerto No. 4 in G, BWV 1049, bb. 82–8 (solo violin and continuo only)

76 t h e b r a n d e n b u r g c o n c e r t o s e t c . recent assessment of the opening movement of Concerto No. 323 as being among the most mature, subtle, and complex of all Bach’s concerto movements. Its aria features set up a clear expectation of da capo form, which is then thwarted in the remarkable phenomenon of the concluding paragraph (b. 78). Here the main ritornello theme returns in a fugal exposition—with a bold triadic countersubject, derived from the original viola part—which is then peremptorily brushed aside by the richest, most sonorous tutti of all in a movement full of marvels of string-writing. The whole paragraph turns out to be a further development of already-stated themes, involving a new tonal excursion (to key ii), two minor-mode ritornellos in a powerful variant form (bb. 97 and 119), and two returns of the chief solo episode from the central paragraph (bb. 91 and 108), the second greatly extended to form a perfidia-like buildup to the final tonic minor-major double ritornello (b. 119). In one respect the opening Allegro of Concerto No. 3 accords better with da capo aria form than that of No. 4, namely in the relatively modest dimensions of its middle paragraph (B). For the equivalent paragraph in Concerto No. 4 (bb. 83–344) is a vast entity of 262 bars (compared with the 83 bars of paragraph A), containing ritornellos in five different keys (vi, ii, IV, I, iii) and incorporating solo violin episodes of the utmost brilliance (against thematic work in the accompanying instruments) in alternation with imitative episodes for the two fiauti d’echo. The entire first paragraph may be construed as a compound, superordinate ritornello, made up of three brief subordinate ‘ritornelli’ (in reality, the headmotive or Vordersatz of the larger structure) in keys I, V, and I, joined by modulating ‘episodes’ (actually sequential continuations or Fortspinnungen), and closing with a cadential phrase or Epilog. In the fugal finale of Concerto No. 5, paragraph A of the ABA da capo structure (bb. 1–78) is again a superordinate ritornello, this time made up of two internal, subordinate ‘ritornelli’: a solo exposition in keys I–V (five subject entries) and a tutti exposition in keys I–I (six entries), plus a stretto coda. In keeping with the relatively light style of this giguefinale, paragraph B (bb. 79–232) brings a sharp contrast: a bewitching cantabile melody (ultimately derived from the fugue subject), passed from one concertino instrument to another, with a motivic, ostinato accompaniment. In its continuation the episode is gradually assimilated to the fugal material of paragraph A, so that the culminating abridged fugal exposition, or ritornello, seems a natural outcome. This entire episode–ritornello process takes place twice, much varied the second time, before the da capo. The finale of Concerto No. 6 is another relatively straightforward, light-hearted da capo structure in gigue rhythm, though it is no longer fugal in texture. The key scheme is that of the textbook da capo aria, A: I–V, V–I; B: vi–iii. As in the other da capo movements, the integrity of paragraph A as pure exposition is

23 ¨ berlieferung und Chronologie’, in By Hans-Joachim Schulze, ‘J. S. Bachs Konzerte: Fragen der U P. Ahnsehl et al. (eds.), Beitra¨ge zum Konzertschaffen J. S. Bachs, Bach-Studien 6 (Leipzig, 1981), pp. 9–26 (at 18): ‘Der erste Satz des dritten Konzerts ist einer der kompliziertesten und ho¨chstentwickelten seiner Art’; quoted from Rudolf Eller, Bach-Fest-Buch (Leipzig, 1962), pp. 78ff.

c o nc e r t o s a n d ouve rtures

maintained—here by treating almost all of the episodic formulations as variants of the three ritornello phrases. This allows paragraph B (bb. 46–65) to function primarily as a diversion, introducing a brilliant minor-mode, cross-string episode for viola duet, which culminates in a brief ritornello in the same minor key. As in the finale of No. 5, this whole process takes place twice, the second time in a varied form. Bach’s large paragraphs are typically articulated into clear-cut periods by prominent cadences in different keys. In some cases these periods constitute ritornellos or episodes, giving rise to the ‘classic’ ritornello–episode alternation. More often a period begins as a modulatory episode and concludes as a ritornello, confirming the last key reached. It is also common for ritornellos and episodes to be grouped together in large complexes, as in some of the transcribed concertos (notably Vivaldi’s Op. 7 No. 8/BWV 973). This may involve the technique of troping (as in Vivaldi’s Op. 3 No. 8/BWV 593). In the opening Allegro of Brandenburg Concerto No. 2, for example, every ritornello after the opening paragraph is troped by one of two sequential formulations that develop ritornello material—a powerful key-affirming tutti (bb. 32–5) or a chromatically shifting, quietly accompanied concertino passage (bb. 50–5). In other cases a ritornello, far from being a self-contained unit, might be a mere tutti interruption or appendage to an episodic period. Thus, in the opening Allegro of Concerto No. 1, the ritornello that follows the first episode (bb. 21–3) is no more than a brief concluding tutti phrase that unexpectedly takes the discourse into a new tonal direction—towards the subdominant B♭. And at the end of the tonally discursive middle paragraph, a ‘ritornello’ of only one bar (b. 52) forms an intermediate stage in an episodic sequence of keys that modulates downwards by step to the tonic, a–g–F, in preparation for the reprise. As always, Bach’s overriding concern is integration, a crucial characteristic that often elevates his music to an altogether higher level than that of his contemporaries. Thus his concerto-episodes, unlike those of Vivaldi and others, are no less highly organized than his ritornellos. Indeed, ‘variation’ or even ‘development’ is often a more apt term than ‘episode’. In the opening Allegro of Concerto No. 1, for example, as in the finale of No. 6, the episodes present not new themes but variants of ritornello material. And in the first movement of Concerto No. 3, the extended central complex of the opening paragraph (bb. 9–38), made up of four periods, constitutes primarily a development of themes stated in the framing tonic ritornellos (new material is purely subsidiary). The glorious solo-harpsichord ‘cadenza’ that leads into the final ritornello in the opening Allegro of Concerto No. 5 is in reality one of Bach’s most highly organized episodes. The original version of this cadenza was only 18 bars long and entirely devoted to virtuoso display.24 Bach’s subsequent revision produced a version of three times the length (65 bars), retaining the bravura but prefacing it with 40 bars of genuine thematic substance. The concertino theme of the first episode (bb. 9–18) is

This represents the chief difference between the early and deﬁnitive versions; see n. 4.

t h e b r a n d en b u r g c o nc e r t o s e t c .

here developed at great length in alternation with an ostinato theme from the middle paragraph. This theme began life as an accompaniment to the haunting, lyrical theme exchanged between flute and violin in the F♯ minor episode (b. 71). But it is no mere accompaniment figure, for the flute/violin melody is marked pianissimo in the autograph score in order that the perfidia (passage-work characterized by persistent repetition of the same motive) of the solo harpsichord may be highlighted. As usual, Bach employs perfidia here as a dramatic form of preparation for an important thematic return, namely the dominant-key reprise of the ritornello’s headmotive, which heralds the concluding paragraph. Upon its second return within the ‘cadenza’, the ostinato figure once more takes the form of perfidia, settling on a dominant pedal and issuing in the great bravura passage of the original cadenza. Only in the opening Allegro of Concerto No. 6 is there a clear distinction throughout between the ritornello theme—a canonic duet for the two violas, quite remarkable material for a concerto-ritornello—and the unified episodic material. Elsewhere episodic and ritornello themes might be presented in rapid alternation, as in the opening Allegro of Concerto No. 2; or else they might be presented simultaneously, as in the equivalent movement of No. 4, where on two occasions the brilliant figuration of the solo violin is combined with extracts from the chief ritornello theme. The same conception recurs in the fugal finale of No. 4, where it gains special significance. This movement is written in alla breve metre, hence the long-note suspensions characteristic of the associated style. And one of the most extreme contrasts in the whole set occurs at the end of the opening ritornello or fugal exposition, at the moment when this traditional polyphonic style gives way to the bravura of the solo violin (b. 41). Bach, however, characteristically unites the opposing elements by introducing quotations of the fugue subject or ritornello theme in the accompaniment of the bravura solo violin episodes. Finally, in the opening Allegro of Concerto No. 2, where there are no discrete ritornellos or episodes, apart from the framing periods, the different functions of the various themes take on a special value in clarifying the structure: every one of the chief thematic formulations has a clearly defined role within the overall form. The ritornello headmotive and the main episodic theme (bb. 9–10) both have a leading function—every period is introduced by one of them; the second phrase of the ritornello (bb. 3–4) has a ‘continuation’ role; two episodic derivatives of the ritornello (bb. 33–5 and 50–5) act as ‘tropes’ or insertions within it; and the cadential phrase of the ritornello (bb. 5–8) furnishes a ‘rhyming’ close for every period in a different key (I, V, vi, IV, ii, iii, I). The ‘classic’ procedure of contrasting fixed-key ritornellos with modulatory episodes is the norm, but it is not uncommon for one of the intermediate ritornellos to modulate, usually during the tonally discursive middle paragraph. In one case, however—Concerto No. 6, first movement—the dominant-key ritornello in the first paragraph (bb. 25–8) unexpectedly returns to the tonic; and exceptionally, in the fugal finale of No. 4 both of the intermediate ritornellos modulate (bb. 67–86: V–vi; 127–51: vi–iii). Typically, only in the third paragraph of the ABA1 reprise structure, where the

c o nc e r t o s a n d ouv ertures

tonic key is fully re-established, are the episodes often tonally closed. In the central B-paragraph of both reprise and da capo movements, however, a sense of calm stasis amidst dynamic tonal movement can be established by a tonally stable, melodic or virtuoso episode (as in the finale of Concerto No. 6 and both outer movements of No. 5), in extreme contrast with the surrounding ritornello themes. The textbook equation of tutti and ritornello on the one hand and solo (or concertino) and episode on the other is flouted so often by Bach that the many exceptions may be viewed as a form of deliberate subversion—one of the most potent dramatic weapons in Bach’s concerto armoury, no doubt learnt in large part from Vivaldi. In Concerto No. 1, first movement, the tutti is already broken into antiphonal groups within the opening ritornello; and the central episode (b. 48), which opens with antiphonal exchanges, flows into the most imposing tutti of all, outdoing even the ritornello itself. This great episodic tutti has an important tonal function, preparing the dominant key C at the central point of the movement and then the tonic F towards the close, immediately before the concluding ritornello. In the third movement of the same concerto, the dominant-key ritornello (b. 35) is accompanied by violino piccolo, which then continues solo (with continuo accompaniment) towards a cadence before the ensemble returns for the remainder of the ritornello. The central A minor ritornello of Concerto No. 4, first movement (b. 185), is likewise accompanied by solo violin, which is here engaged in brilliant fireworks that continue unbroken into the ensuing episode. In the equivalent movement of Concerto No. 3, the massive first episode (bb. 9–33) alternates between solo-group and tutti clothings of the ritornello themes and other material. And the examples could readily be multiplied. The finales of Concertos Nos. 2 and 3 are the only fast movements in the set not constructed in ritornello form. The fugal finale of No. 2 nonetheless falls into a threeparagraph form analogous to that of the reprise movements (bb. 1–56: I–V; 57–106: V–vi–ii–IV; 107–39: IV–I–I), except that the third paragraph forms an abridged reprise of the second, so that the overall form is not ABA1 but rather ABB1. The main content of each paragraph is a fugal exposition, whose subject is constantly passed from one to another of the four concertino instruments; but all three paragraphs culminate in the same tutti (bb. 47–56), a powerful concertante episode in close sequential imitation, which thus functions as a sort of cadential ritornello. Taking account of its additional entry in the middle of paragraph B (b. 79), it is heard in four different keys: V, vi, IV, and I. The finale of Concerto No. 3 is a gigue-like movement in a greatly expanded binary dance form (AA, BB): the entire contents of paragraph A (I–V) return both in the middle (b. 17: vi–iii) and at the end (b. 37: IV–I) of paragraph B, so that it may be said to function as a ritornello. Among the slow movements of the Brandenburg Concertos, there is a certain affinity between those of Nos. 2 and 5, as already noted, but the others are each sui generis—another sign of Bach’s search for the maximum diversity of forms and styles as well as of instrumentation. In the Andante of Concerto No. 2, a brief, lightly

80 t h e b r a n d e n b u r g c o n c e r t o s e t c . decorated theme and two related ‘sighing’ counter-themes are interchanged among the concertino over a Corellian quaver bass. The resulting permutation scheme recalls that of the third movement of the Weimar cantata Christen, a¨tzet diesen Tag, BWV 63 (Christmas, 1714), a duet with obbligato oboe, especially in view of the identity of one of the countersubjects (Ex. 5).25 A triple-counterpoint cadential phrase articulates the concerto movement clearly into three melodic paragraphs, cadencing in keys v (bb. 10–15), iv (38–43), and i (58–65). The Affettuoso of Concerto No. 5 likewise treats a brief, lightly decorated motive in an imitative texture for concertino only, but the motive is woven into longer melodic phrases, the bass is thematic, and the movement is constructed in a miniature ritornello scheme—ritornellos in keys i, III, v, VI, and i alternate with interrelated episodes.

Ex. 5

a) Andante from Brandenburg Concerto No. 2 in F, bb. 4–7 (recorder, oboe, violin, and continuo)

b) 3rd movement from Christen, a¨tzet diesen Tag, BWV 63, bb. 8–10 (oboe, SB, continuo; words omitted)

See Vol. I of the present study, p. 268.

c o nc e r t o s a n d ouvertures

The Adagio of Concerto No. 1 draws nearest to the characteristic style of the slow movements of contemporary Venetian concertos, with its florid solo-oboe melody accompanied by repeated string chords. The profuse elaboration of the solo oboe and violin parts recalls the manner in which Bach decorated slow movements by Vivaldi and Alessandro Marcello in some of the Weimar concerto transcriptions (especially BWV 973–5). It was a style that Bach made thoroughly his own in various cantata sinfonias and organ chorales of that period.26 The highly embellished four-bar melody is treated in binary form with varied repeats (A–A1, B–B1, plus coda), in which not only are the accompanying instruments interchanged but the whole texture is transposed to different keys. In the B-section (b. 9), the original decorated melody is first heard in the bass, accompanied by antiphonal oboe and string chords, then in a canonic duet for the two solo instruments. Thus florid embellishment, still close to its improvisatory roots in Corelli or Vivaldi, here becomes an essential part of the principal idea. The Andante of Concerto No. 4 is the only Brandenburg slow movement with no reduction in scoring. This is due to its fundamental echo conception: forte tutti phrases are repeatedly echoed by piano phrases for concertino only. In the latter, the solo violin, the virtuoso protagonist of the framing movements, takes a back seat, merely providing a bassett to the thematic duet of the fiauti d’echo. The movement is constructed symmetrically in five melodic periods, articulated by cadences in keys i, iv, v, i, and i, and with clear elements of reverse-order recapitulation (ABCB1A1). Whereas all the other Brandenburg slow movements are tonally closed (ending in the key in which they began) and in the overall relative minor, that of Concerto No. 6 is tonally open, and the first two paragraphs (bb. 1 and 20) of its Bar structure (AA1B) have the overall subdominant E♭ as their tonic key. Only the concluding paragraph (b. 40) modulates to and establishes the overall relative minor G as its new tonic; and (like the slow movements of Concertos Nos. 1 and 4) it ends with the traditional link to the finale, a Corellian stepwise descent through a 4th from tonic to dominant in the bass, leading to a Phrygian half-close. This slow movement is also unique in its internal structure: it is designed as a fugue, whose eloquent subject features the falling-7th figure that occurs frequently not only in the music of Bach but in that of his German predecessors and contemporaries (Ex. 6; cf. Part I Ch. 2, Ex. 2).

Ex. 6

Theme of Adagio ma non tanto from Brandenburg Concerto No. 6 in B♭, BWV 1051 (Viola II only; cello and continuo omitted) 26

See Vol. I, pp. 151–2.

the bra nd enburg concertos et c.

The Violin Concerto in E, BWV 1042, cannot be dated on the basis of the sources, since all manuscript scores and parts from Bach’s lifetime are lost—the earliest sources date from around 1760.27 Instead, we have to turn to internal formal or stylistic evidence and comparison with datable Bach instrumental works. The ritornello of the opening Allegro has certain afﬁnities with those of the second and third Brandenburg Concertos (Co¨then, 1721), both in general style and in motivic invention (Ex. 7; cf. Exx. 1 a and 1b). Moreover, the dance-style ﬁnale is a rondeau with four couplets (ABACADAEA), just like the Gavotte en rondeau from the solo violin Partita in the same key (E), BWV 1006 (Co¨then, 1720). These observations are reinforced by a further consideration. Bach’s other two surviving violin concertos, the A minor, BWV 1041, and the D minor, BWV 1043, are clearly interlinked structurally and were both probably composed in Leipzig around 1730 (see Part II Ch. 3). The E major Concerto, however, does not share these formal characteristics and seems rather to belong to the Co¨then environment of the Brandenburg Concertos and the solo violin Sonatas and Partitas.

Ex. 7

Incipit of 1st movement from Violin Concerto in E, BWV 1042 (violins only) The opening Allegro, like that of Brandenburg Concerto No. 4 (as well as the finales of Nos. 5 and 6) is cast in an overall ABA da capo form, and accordingly has much in common with the da capo aria. Thus a solo ‘motto’ (b. 12), leading into a brief tutti ritornello (b. 15), anticipates the first solo episode, which is built into a full ritornello reprise in accordance with the Vokaleinbau, or ‘vocal insertion’, technique so common in Bach’s arias. The same technique is employed in the finale of Brandenburg Concerto No. 6, which also shares with the solo violin movement two features of its B-paragraph: the key scheme vi–iii, a long-standing convention of the da capo aria; and the concept of a discrete, sharply contrasting, tonally stable episode in the relative minor, culminating in a ritornello. Since Brandenburg Concerto No. 4 might almost be described as a solo violin concerto, it is hardly surprising that it, too, has certain features in common with the E major Violin Concerto. These include (in the opening Allegro) the very striking motivic, developmental use of the initial ritornello theme or

27 These are the score P 252 and the parts St 146, both of Berlin provenance. According to Peter Wollny, they are perhaps in the hand of F. A. Klu¨gling and were revised by J. F. Hering, who belonged to the Berlin circle of C. P. E. Bach. See P. Wollny, ‘Ein “musikalischer Veteran Berlins”: Der Schreiber Anonymus 300 ¨ berlieferung’, in G. Wagner (ed.), Jahrbuch des Staatlichen und seine Bedeutung fu¨r die Berliner Bach-U Instituts fu¨rs Musikforschung Preußischer Kulturbesitz 1995 (Stuttgart and Weimar, 1996), pp. 80–133 (esp. 102 and 106).

c o n c e r t o s a n d ouvert ures

its headmotive in the ensemble accompaniment of the solo episodes, and the tonic return of the opening theme towards the end of the B-paragraph, foreshadowing the reprise proper in the da capo, which is, however, delayed by a further tonal, episodic diversion. Details of this kind lend further support to the view that the E major Concerto might have originated during roughly the same period as the Brandenburg Concertos. The ritornello of the Adagio—a basso ostinato theme with quasi-continuo string accompaniment and unisono ending—forms a frame around an extended solo, within which its returns underpin the free, florid, lyrical flights of the solo violin by virtue of its function as a variable ground bass. The locus classicus of such a conception for Bach might have been the Adagio of Vivaldi’s Concerto in A minor, Op. 3 No. 8 (RV 522),28 which he transcribed for organ in Weimar around 1713 or 1714 (BWV 593). There the ritornello is unisono throughout, an idea that Bach took up in the framing ritornellos of the sinfonias to two Weimar cantatas, Gleichwie der Regen und Schnee vom Himmel fa¨llt, BWV 18 (about 1715), and Der Himmel lacht! Die Erde jubilieret, BWV 31 (also 1715). The same applies to the framing ritornello of the Adagio from the Violin Concerto in D minor, presumably of Weimar origin, that has been reconstructed from the harpsichord concerto in the same key (BWV 1052)29—a slow movement otherwise similar in conception to that of the E major Violin Concerto. The French-dance finale of this concerto is related to the solo violin Gavotte in the same key (BWV 1006 no. 3) not only by virtue of their common four-episode rondeau form (as already noted) but in the tonal structure of their episodes: both have, in first or second place, a relative-minor episode and one that modulates to the dominant, and in both cases the last episode modulates first to the submediant c♯ and then to the mediant g♯ and is considerably more extended and elaborate than those that precede it. Furthermore, there is a slight but unmistakable resemblance between their rondeau themes (Ex. 8). Observations of this kind strengthen the view that the E major Concerto might have originated in Co¨then around 1720.

Ex. 8

a) Incipit of ﬁnale from Violin Concerto in E (violins only)

b) Incipit of Gavotte en rondeau from Partia No. 3 in E, BWV 1006 28 According to Rebecca Kan, ‘Vivaldi, Bach and their Concerto Slow Movements’, in A. Leahy and Y. Tomita (eds.), Bach Studies from Dublin (Dublin, 2004), pp. 65–91, esp. 88–91. 29 By Wilfried Fischer, NBA VII/7 (1970), p. 3.

t h e b r a nd en b u r g c o nc e r t o s e t c .

By 1717 Bach must have been acquainted for many years not only with numerous Italian concertos but with at least some of the many collections of Ouvertures avec tous les airs—ensemble suites selected from Lully’s operas and ballets—which were published in Amsterdam from 1697 onwards by Estienne Roger and disseminated throughout Germany.30 In Weimar, we gather that much fine French music (as well as Italian) was heard, including ouvertures (that is, overture-suites),31 and it may be assumed that Bach participated in the performance of such music—before 1714 in the role of Cammermusicus, playing violin, viola, or harpsichord; and after his appointment as Concertmeister in March 1714, presumably leading the court Capelle from the first violins. We may also assume that he knew not only ensemble ouvertures by Lully, Marais, and Steffani, but also some of the many German imitations, notably those of J. S. Kusser (Composition de musique, 1682), Georg Muffat (Florilegium, 1695), J. C. F. Fischer (Journal de Printemps, 1695), and above all Georg Philipp Telemann, the most prolific of all eighteenth-century German exponents of the genre, with no fewer than 135 surviving contributions to his name, not to mention many more presumed lost.32 These experiences would have formed the necessary background to Bach’s composition of his own ensemble Ouvertures (popularly known as ‘Orchestral Suites’), BWV 1066–9. Of these four works, only those in C major, BWV 1066, and D major, BWV 1069, are known to have been in existence by the early 1720s. A surviving set of parts for the Ouverture in C dates from 1724–5;33 and at Christmas 1725 Bach extracted the first movement from the Ouverture in D and adapted it to form the opening chorus of the cantata Unser Mund sei voll lachens, BWV 110. Since he was occupied with the composition of church music, almost to the exclusion of all else, in his first Leipzig years (1723–5), a Co¨then origin for the two ouvertures (1717–23) seems far more likely. Bach’s other two ouvertures, on the other hand, the D major, BWV 1068, and the B minor, BWV 1067, exist in no sources that can be dated before the 1730s, which is likely to have some bearing on their date of composition. For both illustrate a later variant of the genre, namely the concert en ouverture, or concerto-overture, to which J. A. Scheibe referred in the 1740 issue of Der critische Musikus.34 These two ouvertures are consequently considered in a subsequent chapter (Part II Ch. 3).

30 See Vol. I, pp. 17–18, and H. Schneider, ‘The Amsterdam Editions of Lully’s Orchestral Suites’, in J. H. Heyer (ed.), Jean-Baptiste Lully and the Music of the French Baroque: Essays in Honour of James R. Anthony (Cambridge, 1989), pp. 113–30. 31 According to Bach’s pupil P. D. Kra¨uter; BD III, No. 53b (p. 649); NBR, No. 312c. 32 See Vol. I, pp. 17–18; regarding Steffani, see also Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992), VBN I/S/3. With the Collegium musicum in Leipzig Bach would later perform ﬁve ouvertures by his relative Johann Bernhard Bach; see Beißwenger, J. S. Bachs Notenbibliothek, VBN I/B/4–8. 33 Regarding this source (Berlin St 152) see Joshua Rifkin, ‘Verlorene Quellen, verlorene Werke: Miszellen zu Bachs Instrumentalkomposition’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke (Witten, 1997), pp. 59–75 (esp. 59–61). 34 See Steven Zohn, ‘Bach and the Concert en ouverture’, in G. G. Butler (ed.), Bach Perspectives 6 (Urbana and Chicago, 2007), pp. 137–56, esp. 137–8.

c o n c e r t o s a n d ouvert ures

By the second decade of the eighteenth century, some of the less parochial, more internationally-minded German composers, such as Bach and Telemann, were already writing in the so-called ‘mixed style’, a subtle blend of native German characteristics with elements imported from France and Italy (and sometimes from Poland and England too). By its very nature the ensemble ouverture, of course, involved for the German composer a conscious imitation of the French style. Yet even here we encounter a certain admixture of German and Italian elements. German, in Bach’s ensemble ouvertures, are the ineradicably polyphonic thinking that informs the textures and the tendency to make motivic use out of material that a French composer might have regarded as mere passing, incidental figures, such as the upbeat semiquaver figure in the otherwise very French dotted-rhythm introduction to the Ouverture in C. Italianate is, above all, the concerto-ritornello structure of the fast section that follows. Yet even here the concertino episodes are scored for the decidedly Lullian trio of two oboes and bassoon, as is the second of the two bourreés. The Ouverture in C might well have been Bach’s first ensemble work of this kind. If so, this might explain why much of the thematic invention strikes one as somewhat colourless and relatively conventional. Imitation of the French style takes precedence over inventive originality and prevents the composer’s own personality from being clearly stamped on the music, as it is, for example, in the Brandenburg Concertos. Bach amply compensates for this deficiency, however, by repeatedly employing colourful, imaginative scoring to diversify the textures. In the episodes of the first movement, for example, the woodwind duets and trios are accompanied by the fugue subject unisono in upper strings. The oboe duet of Gavotte II is accompanied by a string imitation of a trumpet fanfare—identical with that which opens the Weimar cantata Wachet, betet, BWV 70a (1716). And the melody of Passepied I returns in Passepied II, transposed down an octave and played on unison strings against a florid descant for unison oboes. The Ouverture in D, BWV 1069, gives the impression that it represents a later, rather more sophisticated contribution to the genre. The traditional French woodwind trio of the C major Ouverture gives way to a novel (in this context) double-choir conception: a woodwind choir of three oboes and bassoon engages in antiphonal exchanges with the usual four-part strings (the trumpets and drums were added in Leipzig for the Christmas-cantata version).35 Again, we encounter highly imaginative effects of scoring. In Bourreé II, for example, the tune is presented in a homophony of three oboes and continuo, against which the bassoon has an obbligato in flowing quavers and unison upper strings interject a decorative turn figure. And the interaction of woodwind and strings in the Gavotte is delicate, refined, and highly differentiated. The first movement is on similar lines to that of the C major

35 According to Heinrich Besseler and Hans Gru¨ss, Krit. Bericht, NBA VII/1 (1967), pp. 88–92; see also Joshua Rifkin, ‘Klangpracht und Stilauffassung: Zu den Trompeten der Ouvertu¨re BWV 1069’, in Geck and Hofmann (eds.), Bach und die Stile, pp. 255–89 (esp. 271–6).

86 t he b rand enburg concertos et c. composition—a dotted-rhythm French overture with a fast central fugue in ritornello-reprise form (in this case with reverse-order recapitulation). The majestic introduction, however, is almost twice as long as that of the C major Ouverture; and the fugue subject—in that work, a brief phrase of conventional type—here consists of three signiﬁcant ﬁgures in gigue rhythm, each of which is subjected to much motivic work in the ensuing discourse. This more varied, individual shaping of the thematic material also applies to the following dances. The theme of the Gavotte, for example, is made up of three distinct phrases, sharply contrasting in rhythm, texture, and scoring. Only the Menuets are relatively conventional, falling somewhat below the standard of the other dances. Elsewhere in this composition Bach succeeds in bringing his own powerful musical personality to bear on the French stylistic requirements of the ouverture genre.

Violin, cello, and flute solos Title

Earliest source/s

Scribe, date

Sei Solo a Violino senza basso accompagnato [Sonatas and Partitas for solo violin], BWV 1001–6 Six Suites for cello solo, BWV 1007–12 Solo in A minor for ﬂute, BWV 1013

Berlin, P 967

Autograph, 1720

Berlin, P 269 Berlin, P 968

A. M. Bach, 1727/31 B. C. Kayser, anon., post-1725

In his two major collections of sonatas from the Co¨then period—those for solo violin ‘senza basso accompagnato’ and those for violin and obbligato harpsichord—Bach elected to cultivate modes of instrumentation which, while not unknown to his contemporaries, were at that time decidedly peripheral to the main developments in chamber music. Whether Bach had earlier composed sonatas of the more usual type—that is, with basso continuo—we do not know, but it seems likely. At any rate, the wish to explore less common types is in keeping with the relatively rare instrumentation of the roughly contemporaneous Brandenburg Concertos—in particular, Bach’s cultivation of the chamber concerto senza ripieno in Nos. 3 and 6, and his cross between concerto grosso and solo concerto in Nos. 4 and 5. By 1720 Bach was already highly regarded as a virtuoso of the organ and harpsichord. And in numerous keyboard works of the Weimar period, including the sonata and concerto transcriptions and the concertante toccatas or preludes and fugues (BWV 916, 564, 540; 944 and 894), the keyboard had become a microcosm, taking upon itself all the parts of a real or imaginary ensemble work. Much the same applies to the solo violin in the unaccompanied Sonatas and Partitas of 1720: the entire texture, which would otherwise be shared by an instrumental ensemble, is concentrated within the hands of a single player. How did Bach come to demand of the violin a degree of virtuosity comparable with that which he had long expected of the keyboard? There must have been several contributing factors. Evidently Bach was

v i o l i n, c el l o , a n d f l u t e s o l o s

himself an accomplished violinist36 and, as Concertmaster, presumably acted as leader of the Weimar court orchestra from 1714 to 1717. At that court he would have encountered two of the leading German violin virtuosos of the time in 1703 and 1709 respectively, namely Johann Paul von Westhoff and Johann Georg Pisendel. His violin parts at Weimar and Co¨then, particularly in the Fuga in G minor for violin and continuo, BWV 1026 (see Vol. 1, pp. 202–4), in Brandenburg Concerto No. 4, and in the lost D minor violin concerto that was later adapted for harpsichord (BWV 1052), increasingly demand an exceptional technique. Above all, however, Bach would have been acquainted with the rich German and Austrian tradition of virtuoso violin music, to which not only Westhoff and Pisendel subscribed but also Johann Heinrich Schmelzer, Heinrich Ignaz Franz von Biber, and Johann Jakob Walther. The solo violin music of these composers frequently requires multiple-stopping, pseudo-polyphonic writing, high positions, and brilliant passage-work, all of which are in much demand throughout Bach’s Violin Solos of 1720. Most of the solo violin music of this great tradition is accompanied by basso continuo, but on occasion works were written ‘a violino solo senza basso’ (for unaccompanied violin), notably a lost Ciaccona by Schmelzer of 1670, the Passacaglia that crowns Biber’s Mystery Sonatas of c. 1676, a Sonata in A minor (of unknown date) by Pisendel, and several pieces in Pisendel’s Dresden collection.37 Most signiﬁcant, however, are Westhoff ’s Six Suites for unaccompanied violin (Dresden, 1696), the only known collection of its kind before Bach’s Violin Solos. Westhoff adopts the movement order Allemande—Courante— Sarabande—Gigue, which essentially corresponds with that of the ﬁrst two partitas from Bach’s collection. In formal terms, however, Bach’s partitas subscribe to the Italian tradition, whereas Westhoff ’s suites incline to the French, hence his description of them as ‘pie`ces en musique’. Some years earlier, Westhoff had published a Suite pour le violon seul sans basse (Paris, 1683), which he had played before Louis XIV during the previous year.38 Bach’s Violin Solos, then, must have been motivated at least in part by his wish to subscribe to the exceptionally rich German tradition of virtuoso violin writing. At the same time, it is clear that he wanted to attempt a reconciliation between this tradition and elements of the contemporary French and Italian styles—styles that he was at that time adopting in his instrumental music for larger ensemble (BWV 1046–51, 1066, 1069, etc.) in accordance with the so-called vermischte Geschmack (mixed taste), widely cultivated in German music of the period. The overall forms of the Violin Solos are, for the most part, Italianate: three ‘Sonatas’ da chiesa, each in the standard four-movement form, slow–fast–slow–fast, alternate with three ‘Partien’, of which the

36 According to his son Carl Philipp Emanuel—letter to J. N. Forkel of Dec. 1774; BD III, No. 801; NBR, No. 394. 37 A full account of the background to Bach’s Violin Solos is given by David Ledbetter, Unaccompanied Bach (New Haven and London, 2009), pp. 18–35. 38 See Ledbetter, Unaccompanied Bach, p. 27.

88 the brandenburg concertos etc. first two are essentially sonate da camera, the Italian equivalent of the French dance suite. Corelli’s Sonate a violino e violone o cimbalo, Op. 5 (Rome, 1700), had likewise fallen into two parts, containing respectively six sonatas da chiesa and six sonatas da camera. And Telemann’s Six Sonates for violin and continuo (Frankfurt, 1715), dedicated to Bach’s patron Prince Johann Ernst of Weimar, juxtapose three da chiesa and three da camera sonatas. Intermingled within Bach’s Partien, however, are certain key elements of contemporary French dance style: doubles (so designated), employing division technique, in Partia No. 1 (B minor); a Lullian chaconne (despite its Italian title ‘Ciaccona’) in Partia No. 2 (D minor); and a Lullian suite of dances—Loure, Gavotte en rondeau, Menuet I, II, Bourreé, and Gigue—in Partia No. 3 (E major). Bach’s three da chiesa Sonatas open with an Adagio–Fuga sequence analogous to the common prelude–fugue pairings of his keyboard music. In the opening Adagio or Grave, cantabile periods are built up into two or three overall paragraphs according to bipartite (AB) or reprise schemes (ABA1). The crucial articulating moment is the central dominant cadence, after which further modulation leads to a varied subdominant reprise of the initial theme. In the texture we hear not only purposive, goaldirected harmonic progressions but also a strong element of implied polyphony. Both dimensions are realized on the solo violin to a large extent by multiple-stopping. In addition, the texture is enriched by a highly elaborate melodic clothing, within which notes essential to the harmonic/polyphonic discourse are spread out horizontally and freely intermingled with ornamental notes. In the rather sarabande-like Adagio that opens Sonata No. 3, the melodic clothing is essentially made up of an ostinato figure in dotted rhythm. The equivalent movements in Sonatas Nos. 1 and 2, however, are clothed in a florid violin coloratura rich in ornamental melismas, a style of embellishment that has been linked with ‘Corelli’s graces’, as published in the 1710 Amsterdam edition of his Op. 5 violin sonatas.39 Ex. 9 shows that the Bach movements, like those of Corelli, can be reduced to a plain version devoid of embellishment, though whether this has any real significance beyond that of a mere reductive skeleton seems doubtful. The main weight of each sonata lies in the fugal Allegro, entitled ‘Fuga’, that forms the second movement. The element of implied polyphony, already evident in the preceding ‘prelude’, here becomes of paramount importance. All three fugues have terse, epigrammatic subjects (Ex. 10) upon which structures of immense proportions are built. It is interesting to note that two of these subjects are quoted in the writings of the well-known Hamburg theorist Johann Mattheson. Of the subject from Sonata No. 2 (A minor) Mattheson wrote: Who would believe that these eight short notes would be so fruitful as to bring forth a counterpoint of more than a whole sheet of music paper, without unusual extension, and

39 See Dominik Sackmann, Bach und Corelli: Studien zu Bachs Rezeption von Corellis Violinsonaten Op. 5 (Munich and Salzburg, 2000).

v i o l i n, c el l o , a n d f l u t e s o l o s

quite naturally? And yet the skilled and, in this species, particularly fortunate Bach has set just this before the world; indeed, he has in addition introduced the subject here and there in inversion.40

Elsewhere, Mattheson quoted both subject and countersubject from Sonata No. 3 (C major), albeit in a different key (G) and without citing Bach as composer. He informs us that in 1727 the two themes were presented to candidates for the post of

Ex. 9

a) Opening Adagio from Sonata No. 1 in G minor, BWV 1001, bb. 1–4

b) The same reduced to basic outline by the author

Ex. 10

a) Subject of Fuga from Sonata No. 1 in G minor

b) Subject of Fuga from Sonata No. 2 in A minor

c) Subject of Fuga from Sonata No. 3 in C

BD II, No. 408; NBR, No. 326.

90 t he b randenburg concertos et c. organist at the Domkirche, Hamburg as materials upon which they were to improvise a fugue.41 There is a clear concertante element in Bach’s solo violin fugues. Fugal expositions act as ‘tutti’ ritornellos, with multiple-stops standing in lieu of an instrumental ensemble; and passage-work, with its running quavers or semiquavers, takes the place of ‘solo’ episodes. The two forms of texture are often starkly juxtaposed. The canzona-style theme of the Fuga from Sonata No. 1 (G minor) is not subjected to such contrapuntal devices as stretto or inversion, but merely to variation. In No. 2, on the other hand, the overall structure is largely dictated by fugal procedures: three periods are based on the direct subject, three on direct and inverted forms thereof, and finally two on the inverted subject only. The movement seems somewhat pedantic and over-extended in its contrapuntal treatment, not unlike the fugue in the same key (A minor) from the WTC I. A similarly pedantic impression is conveyed by the even more gigantic (354-bar) stretto-inversion fugue from Sonata No. 3. Two lengthy paragraphs are devoted to stretto treatment of the alla breve subject; and by the time a dominant full-close is reached at b. 200, listeners might expect a final rounding-off in the tonic. Instead, Bach introduces a lengthy paragraph based on the inverted subject. It is worth noting that the progression from simple fugue (Sonata No. 1), via inversion fugue (No. 2), to stretto-inversion fugue (No. 3) is evidently purposeful and witnesses to Bach’s careful planning of the set as a whole. The slow third movements of these sonatas, by contrast, belong amongst Bach’s most exquisite creations. They are cast in simple Bar (AA1B; Sonata No. 3) or binary forms (AB, with or without repeats; Nos. 1 and 2). Only in the first sonata do we encounter a clear reference to dance rhythm: the movement in question is entitled ‘Siciliana’, a dance type often used in the slow movements of Bach’s sonatas and concertos. In all three of these slow movements, multiple-stopping conjures up the effect of a cantabile melody with accompaniment. In the Siciliana (Sonata No. 1) the impression is created of a trio, whose thematic bass (or, more accurately, bassett) is repeatedly answered by a soprano–alto duet. Only in the last eight bars does the ‘treble’ largely take over the leading role. In the equivalent movements from Sonatas Nos. 2 and 3 (Andante and Largo respectively), double-stopping conveys the effect of a two-part texture, made up of cantabile treble and supporting bass, though occasional triple-stops allow the texture to be filled out with fuller harmony or implied counterpoint. In the Andante (No. 2) it has been pointed out that clear stylistic reference is made to the typical solo accompanied by repeated string chords of certain Venetian concerto slow movements, notably the Larghetto from Vivaldi’s Op. 3 No. 9 (transcribed by Bach as BWV 972) or the Adagio from Alessandro Marcello’s Oboe Concerto in D minor, D 935 (BWV 974).42

41 Johann Mattheson, Grosse General-Baß-Schule (Hamburg, 1731), pp. 36–9. The relevant music examples are quoted by Ledbetter, Unaccompanied Bach, pp. 150–1. 42 See Ledbetter, Unaccompanied Bach, p. 127.

v i o li n , c el lo , an d f l ute s ol os

The Allegro or Presto finale of these sonatas is in each case a headlong perpetuum mobile, providing an opportunity for a brilliant display of virtuosity. The absence of multiple-stops creates a welcome contrast with the preceding movements. All three finales, in keeping with their light, airy style, are cast in binary dance form with repeats. Within this structure, each finale has special features of interest to the listener: in Sonata No. 1, the deliberate cross-rhythm ambiguity between the apparent 6/16 of the opening bars of each strain and the real 3/8 that follows; and in Nos. 2 and 3, the immediate, exact repeat of short phrases, which in No. 2 are treated as echo effects. The headlong brilliance of the C major finale (No. 3) is underlined by the absence of internal cadences: both strains preserve an unbroken continuity from beginning to end. On the whole Bach appears to have used the title ‘Partia’ or ‘Partita’ for dance sequences that are more indebted to the Italian sonata da camera tradition than to that of the French suite, though there is rarely a clear demarcation between the two styles, such as he attempted later in Clavieru¨bung II (1735). This understanding of the title ‘Partita’ would explain the many Italian movement titles in Clavieru¨bung I (1726–31). Moreover, Partien Nos. 1 and 2 from the Violin Solos of 1720 have Italian movement titles almost throughout: No. 1: Allemanda—Corrente—Sarabande—Tempo di Borea No. 2: Allemanda—Corrente—Sarabanda—Giga—Ciaccona The Allemanda of Partia No. 1 in B minor, despite its Italian title, possesses something of the intricacy and freedom from motivic/sequential restraint of the typical French allemande. Yet the style is more florid and rhythmically diverse than that of Bach’s own keyboard allemandes of the same period (those of the English Suites) and has close affinities with his sonata style, as is clear from a comparison with the opening Grave of solo violin Sonata No. 2. The Corrente and the doubles that follow each movement, despite their French name, exhibit the fluency and even flow of much contemporary Italian violin writing. The Sarabande is the only movement with a French title, though there is nothing specifically French about it; and the second-beat stress, characteristic of the French dance, occurs only at the final cadence of each strain. However, the movement has much in common with the solemn, richly chordal sarabandes of the English Suites. For the finale Bach borrows the formula ‘Tempo di . . .’ from Corelli (as he did later in the keyboard Partitas) to indicate only a loose connection with the original dance—here a borea, the Italian equivalent of the bourreé—as is habitual in the Italian sonata da camera. The second strain is far more extended and highly developed than usual. On the other hand, the dance rhythms and light-hearted tunefulness of the movement recall the French-titled bourreés of the English and Cello Suites. The Allemanda, Corrente, and Giga of Partia No. 2 in D minor, in keeping with their Italian titles, display an Italianate fluency of violin writing, hence perhaps the absence of doubles. The rhythms are highly diversified, however, in accordance with

92 t h e b r a n d e n b u r g c o n c e r t o s e t c . the ‘mixed style’ of the period. The Sarabanda and Ciaccona, on the other hand, despite their Italian titles, are based on related themes with a pronounced French flavour: richly chordal, with dotted-crotchet rhythms, second-beat stress, and in the Sarabanda profuse, unpredictable elaboration. The variation principle, present in the doubles of Partia No. 1, is here focused on thematic links between the dances, reminiscent of the old German variation-suite, and on the variation form of the massive Ciaccona (257 bars) that forms the finale, the only partita movement that vies with the fugues from the companion sonatas in dimensions and exhaustive diversity of treatment. As a series of continuous variations on a ground bass, it has a clear precedent in the Passacaglia in C minor for organ, BWV 582, from the Weimar period.43 There are other German precedents, of course: in the chaconnes and passacaglias of Biber, Georg Muffat, J. C. F. Fischer, Bo¨hm, Buxtehude, and so on. But like his German predecessors and contemporaries, Bach is indebted above all to the chaconnes of Lully’s operas, hence the second-beat anacrusis, the varied ground bass, and the standard rhythmic formulas that Bach adopts in the first few variations.44 The varied ground bass, which descends from tonic to dominant within a four-bar phrase, either in diatonic or chromatic steps, or else in some patterned form, is of course the main theme. But the melodic line above, together with its attendant harmony in the multiple-stopped phrases, is also to some extent thematic. The opening theme recurs in modified form at two key points in the structure: at the conclusion of the massive opening paragraph (A) in the tonic D minor; and at the conclusion of the much shorter third paragraph (A1), which, being in the same key, is clearly intended to create the effect of a varied, abridged reprise. Within this outer frame the ear is treated to the welcome respite of the tonic major in paragraph B, which opens with a new major-mode theme, without a melodic connection to the original theme, though with clear rhythmic and structural links to it. Even in this very French chaconne some of the faster, more fluent variations recall contemporary Italian violin writing. But the distinction between the French and Italian styles does not become overt till the last item in the set, Partia No. 3 in E major. Here the stylistic orientation is clear from the movement titles: a brilliant, Italianate, concertante Preludio introduces a series of dances ultimately derived in their essential character from the Lullian ballet suite—Loure, Gavotte en rondeaux, Menuet I and II, Bourreé, and Gigue. Thus elements of the two chief instrumental genres of the day, the Italian concerto and the French ouverture-suite, are consciously juxtaposed, as they would be again later in Clavieru¨bung II. The Preludio, though linked to several late seventeenth-century sonata movements,45 is closely related to concerto-ritornello form. Three large ritornello-like paragraphs, in keys I, IV, and I,

See Vol. I of the present study, pp. 173–8. These Lullian formulae are described by Ledbetter, Unaccompanied Bach, pp. 141–2. 45 Corelli’s Op. 3 No. 12 and Op. 5 No. 1; also the Sonata in A from Biber’s Sonatae of 1681. See Ledbetter, Unaccompanied Bach, p. 166. 44

v i ol in , ce ll o, an d f lut e so lo s

Ex. 11

a) Preludio from Partia No. 3 in E, bb. 13–14

b) Finale of lost violin concerto in D minor, reconstructed from BWV 1052, bb. 90–1 (solo violin only) are joined by related episodic periods that modulate to and cadence in the minor keys vi and ii. Regardless of this overall structure, however, we encounter throughout the cross-string, bariolage, and perfidia figurations typical of concerto episodes, as a comparison with the finale of Bach’s lost violin concerto in D minor (reconstructed from BWV 1052) illustrates (Ex. 11). The dances that follow are invariably imbued with the light, melodious style of the intermezzi from the roughly contemporaneous English and Cello Suites, presumably due to Bach’s wish to provide a relatively light ending for the set as a whole. Of the dance types employed, the Loure would not be transferred to the keyboard till later (French Suite No. 5; 1724). The Gavotte en rondeaux, with its four couplets (episodes), as already noted, is virtually identical in structure with the rondeau-finale of the Violin Concerto in the same key (BWV 1042), another clear case of the intermingling of French and Italian styles. The alternativement form of the Menuets is standard in the English and Cello Suites but occurs comparatively rarely in Bach’s later keyboard suites. Each strain of the Bourreé and Gigue opens with French dance rhythms but then proceeds with an Italianate, sonatalike fluency, to some extent diversified by echo effects. There is some evidence that Bach might have intended his violin and cello solos to form two parts of a single magnum opus. The autograph of the violin solos is entitled ‘Sei Solo . . . Libro Primo’; that of the cello solos is not extant, but G. H. L. Schwanenberger, a Bach pupil in 1727–8, referred to the violin solos as Part I and the cello solos as Part II in his title page to Anna Magdalena’s copies of the two sets. His separate title page for the Cello Suites reads: ‘6 Suites a Violoncello Solo senza Basso composeés par Sr J. S. Bach. Maıˆtre de Chapelle.’ It is not known exactly when they were written, but the Co¨then period, c. 1720, seems most likely, both in view of this close connection with the Violin Solos of 1720 and on internal, stylistic grounds. Both Schwanenberger and Anna Magdalena use the term ‘suite’ for the cello solos, and the constituent dances are invariably given French titles. This seems significant in view of Bach’s use of the term ‘partia’ and of Italian dance titles in the companion Violin Solos. For there the primary stylistic reference is to the Italian sonata da camera; the cello solos, on the other hand, refer primarily to the French dance suite. A comparison

94 the brandenburg concertos etc. between the Cello Suites and the English Suites makes this clear. In both cases Bach adopts a standard movement structure, derived from Charles Dieupart’s Six Suites de clavessin (Amsterdam, 1701), which he had copied out during the Weimar years (1709– 16): Pre´lude—Allemande—Courante—Sarabande—[dance pair of optional type]— Gigue. Moreover, in both cases the optional dances are a pair of menuets, bourreés, or gavottes (passepieds too in the English Suites), to be played alternativement. Since the English and Cello Suites are the most standardized and closely related of all Bach’s suites, they are likely to have originated in close temporal proximity, presumably during the early Co¨then years (c. 1717–20). There is also one interesting detail that the two sets have in common: the keynotes of the suites are in a meaningful order, but that of the first suite in each set lies outside that order; thus, English Suites: A; a–g–F–e–d; Cello Suites: G; d–C–E♭–c–D. It is possible, then, that the first suite of each set was composed independently at an earlier date and that only later did Bach conceive the idea of building upon it to form a complete set of six suites. There are a few indications that the English Suites might have been the earlier of the two sets. The first of them is anomalous in movement structure, whereas the first of the Cello Suites already accords fully with the Dieupart design used for suites nos. 2–5 from both sets. Also, the tonic major/minor key relationship between the alternativement dances of the English Suites recurs in Cello Suites Nos. 1–3, whereas those of Cello Suites Nos. 4–6 preserve complete unity of key, in accordance with Bach’s later conception.46 One object of the Cello Suites, then, might have been to compose for the cello within the stylistic and structural parameters that had been adopted for the harpsichord in the recently composed English Suites. But why the cello? There was at that time no German tradition of demanding music for this relatively new string instrument, which had originated in Bologna in the 1660s. However, Bach must have been aware of the rich seventeenth-century German tradition of virtuoso music for the viola da gamba, in which case another object of the Cello Suites might have been to apply the old virtuoso gamba style, with its frequent multiple-stopping and pseudopolyphonic textures, to the relatively modern violoncello.47 It seems very likely that Bach was inspired to make this attempt by the playing of an outstanding cello virtuoso. The most likely candidate, often mentioned in this regard, is the Co¨then cellist Christian Bernhard Linigke. He was one of eight former members of the Berlin court Capelle who had been appointed to the Co¨then court in 1713–16, not long before Bach took up his post as Capellmeister there in 1717. Whether or not Bach had Linigke’s playing in mind, a further aim of the Cello Suites might have been to enliven the old German tradition of virtuoso string playing with elements of the up-to-date vermischte Geschmack (‘mixed taste’). For, while the stylistic and structural framework is undeniably French, hence the French titles and headings, there is a very strong

46 See Hans Eppstein, ‘Chronologieprobleme in Johann Sebastian Bachs Suiten fu¨r Soloinstrument’, BJ 62 (1976), pp. 35–57 (esp. 42–3). 47 As suggested by Ledbetter, Unaccompanied Bach, pp. 37–43.

v i ol in , ce ll o, an d f lut e so lo s

admixture of Italianate string writing and of elements imported from the Italian sonata and concerto. Each half of the set opens with an arpeggiated prelude (No. 1 in G, No. 4 in E♭), a type very much at home in string and lute music but transferred by Bach to the keyboard (following the example of Bo¨hm, Kuhnau, Zachow, and others) in the 1720 Clavierbu¨chlein and in the WTC I (1722). At the end of each half of the set is a figural prelude (No. 3 in C, No. 6 in D), built on broken-chordal patterns with scale or bariolage figures. The middle prelude of each half (No. 2 in d, No. 5 in c) is in each case sui generis. The D minor movement is a cantabile prelude, whose opening four-bar theme is extended into three large melodic paragraphs (bb. 1, 13, and 40). The C minor prelude, which (like the rest of the suite) employs a common form of scordatura—top string tuned down from a to g—is cast in the form of a French ouverture, whose very free fugue is in the 3/8 dance rhythms of a passepied. The Allemandes of Suites Nos. 1 and 2 are, as often in the English Suites, Italianate in their continuous flow of semiquavers. Yet their minimal use of theme, motive, or sequence creates an impulsive, improvisatory impression more akin to the French style. The other four Allemandes each possess a more sharply defined individuality, mainly due to their clearly profiled rhythmic figures that recur throughout in various melodic forms: the decorative demisemiquaver figures of No. 3, the intermingled staccato quavers of No. 4, and the French-style dotted rhythms of No. 5. The freely rhapsodic, melismatic demisemiquaver figures of No. 6 render it the most elaborate of all Bach’s allemandes, comparable only with its equivalent (in the same key) in Partita No. 4 (BWV 828). In both of these cases the allemande is transformed into a meditative, cantabile, sonata slow movement (the cello piece is in some sources marked ‘Molto Adagio’). The courantes all possess the French title, presumably because the context is that of the French suite (as opposed to the sonata da camera genre of the solo violin Partitas). Yet only in Suite No. 5, the most French of the whole set, do we encounter the French 3/2 type, relatively indeterminate in theme and motive, and similar to the courantes of the English Suites. All the other cello courantes are, in effect, Italian correnti—in 3/4 time with running quaver or semiquaver motion. They are built on well-defined themes or motives and display a pronounced wit and vitality. In addition, they exceed the allemandes in their employment of idiomatic string figuration—cross-string patterns and the like. The sarabandes mostly belong to the French type of the English Suites: grave, richly chordal themes—here employing multiple-stops—strongly emphasizing the second beat of the bar. The last sarabande (No. 6), like that of the last English Suite, is the most traditional of all: a French sarabande of the older 3/2 type, with an archetypal opening in classic dotted rhythm. Even here, however, Bach injects an element of individuality, namely the slurred crotchet couplets at cadences, developed at length in the second strain. It cannot be purely accidental that this very French sarabande immediately follows a decidedly Italianate corrente. The Sarabande of Suite No. 4

96 t h e b r a n d e n b u r g c o n c e r t o s e t c . derives individual character from its ubiquitous dotted-rhythm figure, which receives strictly motivic treatment throughout. That of No. 5 is exceptional in that its harmonies are from the outset broken into slurred quaver figures, giving the movement the character of a double, like its equivalent in Partita No. 2 (BWV 826). The quaver figures are strictly motivic and the French second-beat stress is nowhere in evidence. Is it possible, then, that this movement was conceived as an Italian sarabanda to contrast with the very French 3/2 courante that precedes it? The intermezzi, placed between sarabande and gigue, are in every case a pair of French ballet dances, to be played alternativement, as in the English Suites. The set is strictly organized according to dance type: Suites Nos. 1 and 2: Menuet I, II; Nos. 3 and 4: Bourreé I, II; Nos. 5 and 6: Gavotte I, II. Menuet I in each case has a clearly profiled, rhythmic theme, to which Menuet II contrasts not only in mode but in phrase structure and expressive character. In the bourreés, traditional anapaest or dactyl figures that can be traced back to Lully are built into characteristically tuneful, lighthearted themes, as in the bourreés from the English Suites. The gavottes exhibit the traditional rhythm with its half-bar anacrusis, highlighted in No. 6 by repeated notes and multiple-stops. Gavotte II forms a sharp contrast—in No. 5 due to its continuous triplets, ironing out the dance character, and in No. 6 due to the ‘pastoral’ open-string pedals that imitate the drones of the French bagpipe la musette, as in the ‘Gavotte ou la musette’ from English Suite No. 3. The gigues essentially belong to the French gigue type rather than the Italian giga, though they vary much in rhythm and metre. Those of Nos. 2, 3, and 5 are all in a fast, lively 3/8 time, though only No. 5 belongs to the French canarie type, with its constant dotted-quaver rhythms. Outstanding, perhaps, are the gigues of Nos. 3 and 6, in which Bach gives a masterly display of the most richly varied rhythms and figure-work. In the idiomatic string figurations of No. 6, with its five-string instrument, the emphasis is above all on resonant, euphonious string sonority. Both of these gigues (like that of No. 1) have something of an outdoor, rustic flavour. The theme of No. 6 seems to imitate hunting horns, confirming the pastoral impression conveyed by the prelude and second gavotte. The Solo in A minor for unaccompanied flute, BWV 1013, often called ‘Partita’, though without authority, is the sole evidence that Bach attempted for the transverse flute something similar to what he had achieved in the Violin Solos and Cello Suites. The only surviving source, entitled ‘Solo pour la Flute traversiere par J. S. Bach’ and partly in the hand of Bach’s pupil B. C. Kayser, dates from after 1725—considerably later than the likely date of origin of the string solos.48 Nevertheless, the work clearly reproduces the sonata da camera form of the solo violin Partitas. Thus the movement order is, in essentials, identical with that of Partia No. 1 in B minor (BWV 1002):

48 See Yoshitake Kobayashi, ‘Noch einmal zu J. S. Bachs “Solo pour la ﬂuˆte traversie`re”, BWV 1013’, Tibia, 16 (1991), pp. 379–82.

s o n a t a s w i t h o b b l i g a t o ha rps i c hord

BWV 1002: Allemanda—Corrente—Sarabande—Tempo di Borea BWV 1013: Allemande—Corrente—Sarabande—Bourreé anglaise The chief difference is that only in the violin partita is each movement followed by a double, or variation. The flute Allemande takes a very free approach to the dance, lacking, for example, the traditional upbeat of its equivalents in the string solos. Its moto perpetuo belongs to the sonata style, and its half-bar or whole-bar repeats render it closer to the finale of solo violin Sonata No. 2 (BWV 1003) than to Bach’s other allemandes. Much of the movement displays the brilliance of the solo episodes from a concerto, and the clear thematic returns in different keys (C, e, d, G; bb. 9, 20, 25, and 28) have the effect of brief ritornellos. The Corrente is also a virtuoso showpiece, full of brilliant concertante figuration. Unlike the Allemande, however, it opens with a character-theme, whose varied rhythms in mixed note-values recall the courantes of several Cello Suites, while its dazzling parallel-10th leaps are reminiscent of cross-string figuration (Bach also transferred them to the keyboard in the Capriccio from Partita No. 2, BWV 826). The rounded-binary Sarabande is a melodious sarabande tendre, largely unrelated to the more common French type with second-beat stress. The Sarabande of Cello Suite No. 5, with its highly expressive single melodic line, provides a clear counterpart. The term ‘anglaise’, applied to the Bourreé-finale, occurs in none of Bach’s keyboard music before 1722 (‘Gavotte Anglaise’, French Suite No. 3). Like the Corrente, it opens with a character-theme, with prominent anapaests and repeated notes. However, whereas the Corrente reverts to sonata-style writing in running semiquavers, the Bourreé anglaise, in accordance with its French pedigree, maintains its dance rhythms throughout.

Sonatas with obbligato harpsichord or continuo Title

Earliest source/s

Scribe, date

Sei Sonate a Cembalo certato e Violino solo, BWV 1014–19 Sonata in E minor, BWV 1034 Sonata in G major, BWV 1039 Sonata in G major, BWV 1021

Berlin, St 162 Berlin, P 229 Berlin, P 804 Berlin, St 431 Leipzig, Go. S. 3

J. H. and J. S. Bach, 1725 J. C. Altnickol, post-1747 J. P. Kellner, 1726/7 Anon., c. 1726 A. M. and J. S. Bach, c. 1730/4

Bach’s Six Sonatas for violin and obbligato harpsichord, BWV 1014–19, probably originated during the later Co¨then years, around 1720/3. Forkel’s assertion that they were composed in Co¨then is presumably based on information supplied by C. P. E. Bach who, in a letter to Forkel of 7 October 1774, remarked that they were then over 50 years old.49 The implied date of origin, before 1724, is in keeping with the

BD III, No. 795.

t h e b r a nd en b u r g c o nc e r t o s e t c .

evidence of the earliest source (Berlin, St 162), which indicates that by 1725 the first five sonatas and the first two movements of the sixth were already in existence, since Bach was able to entrust their copying to an assistant, his Ohrdruf nephew Johann Heinrich Bach. However, composition of the sonatas immediately before 1725, during Bach’s first two years at Leipzig, is effectively ruled out by the immense amount of church music he was busy composing at that time. Therefore it is reasonable to accept the C. P. E. Bach–Forkel account of their origin at face value. There are isolated examples of the sonata with obbligato harpsichord in seventeenth-century Italy, as well as in France and Germany around 1700,50 but Bach seems to have been the first to cultivate it in a systematic fashion. What made it appeal to him more than the conventional trio sonata with basso continuo? The answer presumably lies partly in his own keyboard virtuosity, which perhaps induced him to emancipate the harpsichord from its subordinate role as a continuo instrument and elevate it to a leading, solo role. Bach took this momentous step not only in the sonatas under discussion but in the roughly contemporaneous fifth Brandenburg Concerto, where the solo harpsichord not only joins the violin and flute in the concertino but takes first place, as if in a solo concerto, particularly in the massive solo episode or cadenza towards the end of the first movement. The title page of the sonatas in the part-autograph source suggests that the harpsichord is intended to take an analogous role here too. It reads: ‘Sei Sonate a Cembalo [con]certato e Violino solo, col Basso per Viola da gamba accompagnato sei piace, composte a Giov: Sebast: Bach’ (‘Six Sonatas for concertato harpsichord and solo violin, with bass optionally accompanied by viola da gamba, composed by Joh. Sebast. Bach’). Thus the harpsichord takes precedence over the violin in the wording of the title and is to be deployed in the manner of a concerto. The first five sonatas share the Corellian four-movement (slow–fast–slow–fast) sonata da chiesa form that Bach employed in the solo violin sonatas of 1720 (the singular features of the sixth sonata will be discussed later). In both sets slow, lyrical, cantabile movements occupy first and third place, the third being the only movement in a key other than the tonic (it is usually in the relative major/minor). In both sets, too, the main weight of each work lies in the concertante fugue that forms the second movement. The finales differ, however: those of the solo violin sonatas are written in a quick, light perpetuum mobile style, whereas the finales of the obbligato-harpsichord sonatas are fugal like the second movements, albeit in a somewhat lighter vein as befits their position at the end of the work. C. P. E. Bach referred to the six sonatas as ‘die 6 Clavirtrio’; and in a set of parts that originated in his Hamburg circle, each sonata is entitled ‘Trio fu¨rs obligato Clavier

50 See William S. Newman, ‘Concerning the Accompanied Clavier Sonata’, Musical Quarterly, 33 (1947), pp. 327–49, and Hans Eppstein, Studien u¨ber J. S. Bachs Sonaten fu¨r ein Melodieinstrument und obligates Cembalo (Uppsala, 1966), pp. 26–7.

s o n a t a s w i t h o b b l i g at o ha rps i c hord

und eine Violine’.51 ‘Trio’ here denotes not the number of instruments but the number of parts: it refers to the characteristic pseudo-trio-sonata texture of two treble parts (violin and harpsichord right hand) and bass (harpsichord left hand, optionally doubled by viola da gamba). In the fugal second and fourth movements of the sonatas da chiesa (Nos. 1–5), this pure three-part texture comes into its own. Among the opening ‘Adagios’, however, it applies only to the Dolce of Sonata No. 2 (A major), whose three-part thematic imitation establishes an equality of parts that recalls the keyboard Sinfonias of the same period (BWV 787–801; Co¨then, 1722–3). All the other opening Adagios exhibit the pure duo texture (in the modern sense of two instruments rather than two parts) that is rightly regarded as one of Bach’s major achievements in this set and gives rise to some of his finest inspirations. The key factor here is the independence of the violin and harpsichord, alongside the emphasis that is laid on their individual capabilities. In the simplest case, the opening Largo of Sonata No. 4 (C minor), the violin has a highly characteristic and affecting melody in dotted rhythms, perfectly suited to the instrument; and the harpsichord, an idiomatic, arpeggiated accompaniment. Though not so titled, the movement is a siciliana in binary dance form, a dance type employed in countless sonata and concerto slow movements by Bach and his contemporaries. The opening Adagios of Sonatas Nos. 1, 3, and 5 are akin to a concerto slow movement or vocal aria, in which the solo part is allotted to the violin, while the harpsichord takes on the role of instrumental ensemble. The solo violin part unfolds freely and rhapsodically, suffused (in Nos. 1 and 3) with Corellian florid elaboration, as in the first two solo violin sonatas. In Sonata No. 3 (E major), the harpsichord takes an accompanying role, and the recurrence within its part of the same motive in every bar gives it the character of a motivic accompagnato. In Nos. 1 (B minor) and 5 (F minor), on the other hand, the main thematic content of the movement is concentrated within the harpsichord part, in lieu of instrumental ensemble, which allows the violin scope to roam freely, unfettered by thematic constraints. The trio texture of three equal parts is a sine qua non of the concertante fugue that forms the second movement and bears the main weight of the sonatas da chiesa (Nos. 1–5). In these fugal Allegros, three or four large paragraphs, clearly set off from each other by prominent cadences, together form da capo (ABA; Nos. 1 and 2), reprise (ABA1; No. 3), binary (AABB; No. 5), or alternating structures (ABA1B1; No. 4). A clear link with concerto-ritornello form is established, for paragraph A may be heard as a self-enclosed fugal ritornello, ending with a tonic full-close. Paragraph B then opens with contrasting material in the manner of a concerto episode. In its course, however, it revisits the fugal material of A alongside its own episodic material. In essence, then, opposites are first starkly juxtaposed and then increasingly reconciled in the further course of the movement. The reprise of paragraph A, whether exact or varied, is the

For the ﬁrst title, see BD III, No. 795; the second is found in Berlin, St 463–8.

100 the brandenburg concertos etc. only full, substantial ritornello return, but fugal entries or brief expositions in various keys within B will inevitably be heard as passing mini-ritornellos. Paragraph B is, however, perhaps better viewed as a development section rather than as an alternation of ritornellos and episodes, especially in view of its essential continuity and lack of articulation by internal cadences. The concertante element is particularly strong in Sonatas (Nos. 1–3, but relatively weak in Nos. 4 and 5, in which the fugal element is predominant. No. 6 is a special case due to the concerto form (fast–slow–fast) of its outer and second movements. In the opening ‘Allegro’, the concertante element naturally takes precedence and determines the entire structure. However, it exhibits unmistakable formal affinities with the Allegro second movements of the da chiesa sonatas Nos. 1–5): the overall ABA da capo form; the crucial moment when the opening ritornello reaches a tonic full-close, after which a restart is made in a quite new manner; and the presence of fugal writing. Here, however, the fugal and concertante elements are reversed: a concertante ritornello (A) gives way to a fugal exposition (start of B), which then alternates with ritornello elements. This design, far from being of relatively early origin, as was formerly thought,52 might have arisen as a late, conscious variant of the second-movement design of the da chiesa sonatas. The intermediate slow movements of the da chiesa sonatas are, except in one case (No. 3), laid out in cantabile periods according to reprise (ABA1; Nos. 1 and 4) or bipartite schemes (AA1; Nos. 2 and 5). All but No. 4 are constructed over a basso quasi ostinato, a mode of structuring common in Bach’s cantata arias and concerto slow movements. In Nos. 1 and 2, a three-part texture is made up of a lyrical duet for violin and harpsichord right hand over a supporting ostinato bass. In No. 1, however, the violin takes the leading role—unusually for Bach the parts are not interchanged— whereas in No. 2 the two treble parts are canonic and thus equal throughout. The inner slow movements of Nos. 4 and 5 adhere to the pure duo texture of their first movements. In No. 4 a cantabile violin melody in dotted rhythms is accompanied by an idiomatic broken-chordal figuration for harpsichord, as in the opening siciliana. In No. 5 violin double-stops are accompanied by an ostinato harpsichord figuration. The overall effect somewhat resembles a soprano–alto duet (violin double-stops) with motivic accompagnato (harpsichord). In the Adagio third movement of Sonata No. 3, ostinato is elevated to the role of a fully-fledged four-bar variable ground bass, reiterated fifteen times in the course of the movement. Over it Bach builds variations on a florid eight-bar cantabile melody in C♯ minor—one of his finest inspirations. Duo texture is maintained, but roles are constantly interchanged: at first, the eloquent violin solo is accompanied by a harpsichord imitation of repeated string chords, but each instrument takes over the other’s role at the relative-major counterstatement (b. 13).

52 Notably by Hans Eppstein; see especially his ‘Zur Problematik von J. S. Bachs Sonate fu¨r Violine und Cembalo G-dur (BWV 1019)’, Archiv fu¨r Musikwissenschaft, 21 (1964), pp. 217ff.

sonatas with obbligato harpsichord

101

The fugal structure of all six finales links them closely with the Allegro second movements. Indeed, the finales of Nos. 3 and 6 (referring to the latter in its final version) are cast in an overall ABA da capo form and thus formally indistinguishable from the second movements. The finales, however, are in every case written in a noticeably lighter style. Only in No. 2 does the fugue subject of the finale resemble those of the second movements. In Nos. 1 and 3 the double subjects of the finales are in quick triple time with rapid semiquaver movement, and the double subject of No. 6 is in 6/8 gigue rhythm. Moreover, three of the finales, Nos. 1, 2, and 4, are cast in a relatively light binary dance form with repeats. In every case, as in the second movements, a fresh start is made with a new subject (albeit related in some way to the original one) after the full-close that ends the first paragraph. But since the form is binary rather than ternary, that cadence is in the dominant rather than the tonic. The original subject returns thereafter in order to preserve unity. As already intimated, Sonata No. 6 in G (BWV 1019) is quite different in conception from the other five sonatas. It is clear that Bach experienced some difficulty in settling upon the final form of this sonata, since he revised it radically on several occasions during the Leipzig period.53 Only the first two movements remained constant throughout, being subjected merely to revision of detail. Since J. H. Bach copied them out in 1725, they must have been present in the lost Co¨then original version of the sonata. The two movements in question are a concertante Allegro and a brief Largo in the relative minor. Since Bach must have envisaged the finale as a fast movement from the outset, what we have here in essence—in all versions of the work, regardless of insertions—is the three-movement, fast–slow–fast form of the Sonate auf Concertenart (‘sonata in the manner of a concerto’), an important sub-genre that Bach was to cultivate during the Leipzig period. Whether the original Co¨then version already incorporated additional inner movements between Largo and finale we do not know. For Bach’s purpose in taking over the copying from J. H. Bach after the Largo must have been to replace the original movements from no. 3 onwards with new ones. In the 1725 version that resulted from this revision, the opening Allegro was to be repeated at the end as finale, and a suite-like sequence of three movements was inserted between Largo and finale: ‘Cembalo solo’ in e (vi); Adagio in b–g (iii–i, half-close); and ‘Violino solo e Basso l’accompagnato’ in g (i). This version, then, is striking for its diversity of key, style, and instrumentation. Far from being restricted to tonic and relative major/minor, like the first five sonatas, it presents a succession of six movements in various keys thus: 1. Vivace G

2. Largo e (half-close)

3. Cembalo solo e

4. Adagio b–g (half-close)

5. Violino solo g

6. Vivace G

53 ¨ berlegungen zur Chronologie der drei The latest reconstruction of its history is by Frieder Rempp, ‘U Fassungen der Sonate G-Dur fu¨r Violine und konzertierenden Cembalo (BWV 1019)’, in M. Staehelin (ed.), Die Zeit, die Tag und Jahre macht: Zur Chronologie des Schaffens von J. S. Bach (Go¨ttingen, 2001), pp. 169–83.

102

the brandenburg concertos etc.

Moreover, balancing harpsichord and violin solos are introduced as a foil to the duo writing that obtains in the other movements. These harpsichord and violin solos are dance movements—later in the same year Bach entered them in Anna Magdalena’s 1725 Clavierbu¨chlein as the Corrente and Tempo di Gavotta, respectively, of the Partita in E minor (BWV 830). The solos, dances, and variety of key help to create a lighter tone in the last sonata of the set, just as Bach lightened the last of the Violin Solos of 1720 (the Partita in E, BWV 1006) by stufﬁng it full of French ballet dances. The 1725 version of the sonata must have lost its validity when the two dance movements found their permanent home in the E minor keyboard Partita. Therefore, probably in the late 1720s,54 Bach radically revised the work again, replacing the two dance solos with a single expansive ‘Cantabile’, an instrumental arrangement of a vocal aria, perhaps drawn from a lost Co¨then secular cantata. The violin here retains its original ﬂorid, idiomatic obbligato part, while the harpsichord is given a highly ornamented version of the soprano part in the right hand and the continuo bass part in the left. The Adagio fourth movement of the 1725 version is retained, so that the new version takes the form: 1. Presto G

2. Largo e (half-close)

3. Cantabile G

4. Adagio b–g (half-close)

Clearly this intermediate version remained a fragment, since it lacks the necessary quick finale in the tonic.55 Bach might have left it unfinished on the grounds that he was unable to find a suitable conclusion; or else he might have been unhappy with an interior made up of three slow movements in succession. Moreover, the Adagio, which successfully mediated between the keys of the two dance solos in the 1725 version (e; b–g, half-close; g), has here lost its tonal raison d’eˆtre. Some time between about 1729 and 1741 (the last possible date of the copy by Bach’s pupil J. F. Agricola),56 Bach radically revised the work for one last time, returning to some of the concepts of the 1725 version: 1. Allegro G

2. Largo e (half-close)

3. [Cembalo solo] e

4. Adagio b–D

5. Allegro G

As in 1725, the new third movement is a harpsichord solo in the relative minor and in binary dance form, though no longer in a specific dance rhythm. The Adagio fourth movement is also new—more florid and substantial than its predecessor, but with the same tonal function of mediating between relative minor and tonic. The finale is now no longer merely a repeat of the opening movement, but a light-hearted, concertante,

¨ berlegungen’, p. 179. According to Rempp, ‘U ¨ berlegungen’, p. 171; previous scholars assumed a reprise of the ﬁrst As pointed out by Rempp, ‘U movement, as in the 1725 version, but this is not veriﬁed by the sources: the repeat indication in Berlin Am. B.61 is a late addition in another hand; Copenhagen, Weyses Samling has no repeat indication. 56 ¨ berlegungen’, p. 172. Regarding this source and its dating implications, see Rempp, ‘U 54 55

sonata s w ith obbligato harpsichord

103

fugal Allegro in gigue rhythm, fully comparable with the finales of the first five sonatas, on which it was presumably modelled.57 If obbligato-harpsichord texture is regarded as an improvement on the traditional continuo texture, it is easy to assume that the continuo sonatas originated at an earlier period.58 However, this view is not borne out by dates of the extant sources. The earliest source of the violin and harpsichord sonatas (BWV 1014–19) dates from 1725, but it was copied from a lost original that probably dates from the late Co¨then years (1720–3). The earliest sources of the continuo sonatas, on the other hand, date from c. 1726 (BWV 1039), 1726/7 (BWV 1034), and 1730/4 (BWV 1021).59 The dates of the sources reflect the maturity of these compositions, in which the opposing demands of convention and innovation are subtly interwoven. All exhibit the Corellian sonata da chiesa pattern of four movements arranged in two slow–fast pairs, but the relation between flute/violin and continuo is handled with great resource and variety. The Sonata in E minor, BWV 1034, copied out by Bach’s colleague Johann Peter Kellner in 1726/7,60 is not just a solo with accompaniment but often a genuine duo between flute and continuo bass. The opening theme of the first movement is an accompanied flute solo, but the answering theme (b. 5) is led by the continuo, which is then imitated by the flute at the upper octave. Such continuo-led writing recurs several times later on (bb. 11 and 24). The following Allegro is a concertante fugue with big perfidia-style episodes. Flute and continuo here form duo partners in the ritornello-like fugal expositions, but the episodes are brilliant accompanied flute solos. In the Andante we encounter a clear division of responsibilities. Within an overall ABA1 reprise structure, the florid, cantabile flute part to a large extent unfolds freely over the variable ground bass of the continuo. The Allegro-finale is cast in binary form with repeats. The closest parallel is found in solo violin Sonatas Nos. 2 and 3, for all three finales have lively, rhythmic themes with internal repetition and (in two cases) echo dynamics. A concertante effect is created by the manner in which the tonic full-close at the end of the first period (b. 12) is followed by an entirely new, contrasting theme. The two halves of the binary structure are bound together by an extended, 20-bar reprise (bb. 23–42 = 69–88).

57 Whether the oft-mentioned thematic link with the aria ‘Phoebus eilt mit schnellen Pferden’ from the secular wedding cantata Weichet nur, betru¨bte Schatten, BWV 202 (no. 3), has any special signiﬁcance it is impossible to say. The cantata probably originated during the Weimar period (see Vol. I, pp. 270–4), but it is not impossible that Bach revived it for a spring wedding in or around 1730, the date of the MS copy by Johannes Ringk, and consequently had the aria theme and its counterpoint in mind when he came to write a new ﬁnale for the G major Sonata. However, this is mere conjecture. 58 This is a constant assumption throughout Eppstein’s Studien and in many of his other publications. 59 Even later are BWV 1035 (probably 1741) and 1079 no. 3 (1747). BWV 1023 has been left out of consideration, being of doubtful authenticity. 60 According to Russell Stinson, The Bach Manuscripts of Johann Peter Kellner and His Circle (Durham and London, 1989), p. 23.

104 t h e br a n d e n b u r g c o n c e r t o s e t c . The Sonata in G, BWV 1039, whose original performing parts perhaps date from around 1726, lacks the strong concertante element of the E minor Sonata, perhaps on account of its scoring for two flutes and continuo.61 This instrumentation also explains the different role of the continuo: in the E minor Sonata, the flute and continuo were partners; in the G major, the partnership belongs to the two flutes, with the result that the continuo takes a subordinate role. Its thematic contribution is restricted to two entries of the subject in each of the two fugal movements (no. 2, bb. 37 and 64; no. 4, bb. 17 and 98); elsewhere it does little more than accompany. The opening pastorale-like Adagio, a lyrical movement in bipartite form (AA1), immediately establishes the equality of the two flutes. Florid main theme and plain counter-theme are first stated in the tonic, then restated in the dominant with interchanged upper parts (b. 4). Such exchanges between the flutes thereafter become the dominant aspect of the texture throughout. The second movement, Allegro ma non presto, a fugue in ABA1 reprise form, finds Bach at his most captivatingly melodious. The bariolage-like episode that intervenes at several points (bb. 10, 43, and 86) serves as a reminder that the work might originally have been scored for violins rather than flutes. The tonic full-close that ends the A-section (b. 32) promises a fresh start; but, unlike in the obbligato-harpsichord sonatas, no real contrast follows, only the inverted fugue subject. The arioso-style Adagio e piano, no. 3, with its repeated quavers and imitative figures, recalls its equivalent in the fifth obbligatoharpsichord sonata. The Presto-finale is, like no. 2, a fugue in ABA1 reprise form. Its charming character-subject recalls several fugue subjects from the obbligatoharpsichord sonatas (BWV 1014 no. 2, 1015 no. 4, and 1016 no. 2). The Sonata in G for violin and continuo, BWV 1021, whose sole surviving score was copied out by Anna Magdalena with her husband’s assistance around 1730/4, exhibits the same traditional movement structure as the flute sonatas—Adagio, Vivace, Largo, Presto—but the dimensions are unusually small for Bach, and the rather conventional bass line forms a striking contrast with the characteristically elaborate and eloquent violin part. A likely explanation has been put forward:62 namely, that the bass of the composition was adapted by Bach from an anonymous figured-bass exercise. Not long after Anna’s preparation of the score, this bass was reused by a member of Bach’s household, perhaps C. P. E. Bach, as the basis of the Sonata in G, BWV 1038, for flute, violin, and continuo. At a still later date, that work was in turn adapted by someone from Bach’s circle to form the Sonata in F, BWV 1022, for violin and harpsichord. All this strongly suggests that the authentic Violin Sonata in G (BWV 1021) was composed

61 It is thought to go back to an original in the same key for two violins and continuo; see H. Eppstein, ‘J. S. Bachs Triosonate G-dur (BWV 1039) und ihre Beziehungen zur Sonate fu¨r Gamba und Cembalo G-dur (BWV 1027)’, Die Musikforschung, 18 (1965), pp. 126–37. 62 By Ulrich Siegele, Kompositionsweise und Bearbeitungstechnik in der Instrumentalmusik J. S. Bachs (Neuhausen and Stuttgart, 1975), pp. 24–31.

sonata s w ith obbligato harpsichord

105

by Bach as a model to assist his pupils—especially Carl Philipp Emanuel, then only about 18 years old—in learning the art of composing upon a given bass. The violin part of the two slow movements (nos. 1 and 3) is florid and improvisatory in style—full of ‘Corellian graces’. In the binary-form Adagio (no. 1) each strain opens with a decorated suspension figure in the continuo, later freely echoed by the violin (bb. 5 and 10–11). A virtually identical figure occurs repeatedly in the bass of the corrente-like but through-composed Vivace, no. 2 (bb. 4–5, 8–9, 11, 20–1, 35–6, 40–1, 44–5). On several occasions Bach anticipates it in the violin part (bb. 3–5, 34–6), giving rise to imitation at the lower 4th. The violin part of the Largo, no. 3, can afford to be free, since it is built over the firm structure of a six-note basso quasi ostinato (bb. 1 and 13) and its inversion (b. 5). The only place where violin and continuo achieve true parity is the Presto-finale, an alla breve fugue, built on a neutral soggetto of three long notes—a plain version of the aforementioned suspension figure from the preceding movements. It is accompanied by two regular countersubjects, the first in crotchets (b. 2, bass), the second in quavers (b. 3, bass).

I.4 Sacred and secular: the vocal works

Secular cantatas Title, occasion

Earliest source/s

Scribe, date

Die Zeit, die Tag und Jahre macht, BWV 134a (New Year’s Day) Durchlauchtster Leopold, BWV 173a (birthday) Entﬂiehet, verschwindet, entweichet, ihr Sorgen, BWV 249a (birthday) Schwingt freudig euch empor, BWV 36c (birthday) Der zufriedengestellte Aeolus, BWV 205 (name-day) Vereinigte Zwietracht, BWV 207 (homage) Trauer Music, BWV 198 (memorial) Von der Vergnu¨gsamkeit, BWV 204 O! angenehme Melodei, BWV 210a (homage)

Paris, Ms. 2

Autograph, for 1 Jan. 1719

Berlin, P 42/1

Autograph, for 10 Dec. 1722?

[lost]

—, for 23 Feb. 1725

Berlin, P 43/1

Autograph, 1725

Berlin, P 172

Autograph, for 3 Aug. 1725

Berlin, P 174, St 93

Autograph, part-autograph, for 11 Dec. 1726 Autograph, 15 Oct. 1727 Autograph, 1726/7 Part-autograph, for 12 Jan. 1729

Berlin, P 41/1 Berlin, P 107 [fragment]

Since Prince Leopold of Anhalt-Co¨then adhered to the Calvinist faith of his dynasty, the composition of sacred cantatas there did not belong to Bach’s regular duties.1 As in other central German courts, however, the Capellmeister was expected to compose and perform festive music in the form of secular congratulatory cantatas for New Year’s Day and for the birthday of the reigning prince, which in Leopold’s case took place on 10 December. Of the twelve cantatas that Bach presumably composed for these occasions between December 1717 and January 1723, only two survive complete: Die Zeit, die Tag und Jahre macht, BWV 134a, for New Year’s Day, 1719; and Durchlauchtster Leopold, BWV 173a, for the prince’s birthday, probably in 1722. Of Der Himmel dacht auf Anhalts Ruhm und Glu¨ck, BWV 66a, written for the same occasion in 1718, only the printed libretto survives, but the music can be largely reconstructed from later sacred

1 However, Bach is known to have performed one sacred cantata for Leopold’s birthday, on 10 Dec. 1718, namely Lobet den Herrn, alle seine Heerscharen, BWV Anh. I 5, which has not survived.

se c ular ca ntat as

107

parodies (BWV 66 and 42).2 Sacred parodies (BWV 184 and 194), alongside a few performing parts, also provide musical evidence for two Co¨then secular cantatas whose texts are no longer extant, BWV 184a and 194a.3 With the possible exception of BWV 194a, all these works take the form of serenatas; that is, evening serenades or Abendmusiken sung in praise of the reigning prince. They are essentially dialogues between two singers, who in some cases represent allegorical characters: ‘Fama’ and ‘Glu¨ckseligkeit Anhalts’ (Fame and Anhalt’s Felicity) in BWV 66a; ‘Zeit’ and ‘Go¨ttliche Vorsehung’ (Time and Divine Providence) in BWV 134a. The concluding chorus of rejoicing may be sung by the same two singers, either exclusively or augmented by two more singers to create a four-part (SATB) ensemble. The dialogue structure of the librettos is responsible for the duet writing that abounds in arias, recitatives, and choruses alike. In the early Co¨then years (1718–20), Bach’s librettos were devised by Christian Friedrich Hunold, who wrote under the pseudonym Menantes; but he died in 1721, and it is not known who took over his role. Der Himmel dacht (BWV 66a) and Die Zeit, die Tag (BWV 134a), first performed within a month of each other in 1718–19, form a sister-pair of serenatas that display a judicious mixture of French dance and Italian concerto and operatic styles. The intimate, closely intertwined nature of the duet writing is clearly drawn from the world of opera, as is the exclusive use of da capo form for all the arias, duets, and choruses. The idiomatic obbligato parts for solo violin in the duets introduce a certain concertante element, as does the sinfonia of BWV 66a, whose ritornello-withinda-capo form recalls so many of Bach’s concerto movements. The concertino of this sinfonia, however, which returns in the aria no. 7, is made up of the French trio of two oboes and bassoon, which goes back to Lully. Moreover, the da capo-form finales of the two serenatas are written in a light 3/8 dance rhythm of French gigue-passepied type,4 with largely homophonic textures and regular phrase structure. It is a style that Bach had employed in several da capo-form choruses during the Weimar period, both sacred and secular (BWV 208 no. 15, 172 no. 1, and 63 no. 1); and not only would it become a staple of his Leipzig secular cantatas, but it would continue to be transferred fruitfully to the realm of sacred music. Durchlauchtster Leopold (BWV 173a) and Erwu¨nschtes Freudenlicht (BWV 184a) form another sister-pair of serenatas, probably performed within a month in late 1722 and early 1723. They are considerably more diverse and adventurous than the earlier pair in forms, styles, and dance types. In BWV 173a the ritornello of the first aria (no. 2) is somewhat modish in style, with its triplet semiquavers against dotted rhythms. Accordingly, the vocal motto is heavily stylized in the galant manner of the 2 See Friedrich Smend, Bach in Ko¨then (Berlin, 1951); Eng. trans. by J. Page, ed. and rev. by S. Daw (St Louis, Mo., 1985), pp. 50–6; and also Joshua Rifkin, ‘Verlorene Quellen, verlorene Werke: Miszellen zu Bachs Instrumentalkomposition’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke (Witten, 1997), pp. 59–75 (esp. 65–7). 3 Details are given by Alfred Du¨rr, Krit. Bericht, NBA I/35 (1964), pp. 138–51. 4 See Doris Finke-Hecklinger, Tanzcharaktere in J. S. Bachs Vokalmusik (Trossingen, 1970), pp. 89–103.

108

sacred and secular: the vocal works

time. In the second aria (no. 3) a brief theme made up of vocal antecedent and instrumental consequent takes the place of an opening ritornello. In the duet no. 4, subtitled ‘Al tempo di minuetto’, a complete minuet is stated three times in different keys a 5th apart (G, D, A), the second and third statements acting as doubles to the first. The soprano aria no. 6 is a reprise-form (ABA1) movement whose 16-bar ritornello is a binary dance in the rhythm of a bourreé. The bass aria no. 7, also somewhat bourreé-like but now in Bar form (AA1B), has an obbligato part for cello and bassoon in unison. The concluding chorus (so-called though sung by soprano and bass only) is a binary dance in minuet rhythm with the vocal parts built into the repeats. BWV 184a, no less varied in its forms and dance rhythms, opens with a motivic accompagnato, whose florid motive for two flutes in 3rds illustrates the ‘desired light of joy’.5 The second movement is a da capo-form duet in the style of a pastorale. The soprano aria no. 4, cast in ABA1 reprise form, is in the rhythm of a polonaise. In the gavotte-finale, a binary dance with the voices built into the repeats, as in the equivalent movement of BWV 173a, becomes the A-section of an overall ABA da capo scheme. It is clear, then, that by 1722/3 Bach employed a greater range of forms in his secular cantatas, no longer being confined to da capo form, and drew from a greater range of dance types: not just the gigue-passepied type but also menuet, bourreé, pastorale, polonaise, and gavotte. This process reaches its logical conclusion in Ho¨chsterwu¨nschtes Freudenfest, BWV 194a, of which only a few instrumental parts survive.6 Its sacred parody BWV 194 shows that it was conceived as a vocal-instrumental version of a French ouverture-suite. The work was on a grander scale than the serenatas, comprising eleven movements, including an expansive opening chorus in the form of an ouverture—with a central imitative section in ritornello form—and a finale in the familiar gigue-passepied style. These movements form a frame around four arias in different dance rhythms: pastorale, gavotte, gigue (in a variant with dotted rhythms against triplets), and menuet. Since the work differs so radically in conception from the serenatas, it might perhaps have been written for a special occasion, possibly during the first three months of 1723 (before Bach’s move to Leipzig that April), in which case the sacred version would have been performed later in the same year (2 November).7 During Bach’s first five or six years in Leipzig (1723–9), many different occasions arose for the composition and performance of secular cantatas. He continued to write occasional music for the courts of Weißenfels (BWV 249a) and Co¨then (BWV 36a), as well as secular wedding cantatas (BWV 216 and Anh. I 196) and solo cantatas (BWV 5 It seems likely that the sacred text here corresponded with the secular original; see Alfred Du¨rr, The Cantatas of J. S. Bach, rev. Eng. trans. by R. D. P. Jones (Oxford, 2005), p. 367. 6 Since the text is lost, the title of the sacred parody is used here for identification purposes. 7 For an organ consecration at Sto¨rmthal, near Leipzig; it might also have been performed in Leipzig on the previous Sunday, 31 Oct., which doubled as the 23rd Sunday after Trinity and the Reformation Festival; see Peter Wollny, ‘Neue Bach-Funde’, BJ 83 (1997), pp. 7–50 (esp. 21–6).

secular ca nta tas

109

204). In addition, he now wrote music of homage for nobles and prominent citizens (BWV 36c, 249b, and 210a) and festive music for university and student events (BWV 205, 207, 198, Anh. I 195, and Anh. I 20). Although most of his festive music for the Electoral House of Saxony dates from the 1730s, as early as 1727 he composed congratulatory cantatas for the name-day and birthday of Augustus II (BWV 193a and Anh. I 9). Of the fifteen or so secular cantatas from this period that are known to have existed, however, only five survive complete, BWV 36c, 205, 207, 198, and 204, though two others can be largely reconstructed from later derivatives (BWV 249a and 210a). At this stage Bach had no ready-made vocal and instrumental ensemble of his own for the performance of secular cantatas, and it seems more than likely that soon after his arrival in Leipzig he came to an agreement with Georg Balthasar Schott, director of the Collegium musicum founded by Telemann, enabling him to make use of its resources (in 1729 Bach would take over its directorship from Schott).8 For his librettos Bach increasingly relied on Christian Friedrich Henrici (pseudonym: Picander), who collaborated with him, perhaps for the first time, in the Weißenfels pastoral cantata BWV 249a of 1725. On occasion (BWV 198 and Anh. I 196) Bach also set to music librettos by Johann Christoph Gottsched, a key figure in the early German Enlightenment and author of the influential Versuch einer critischen Dichtkunst of 1730. In these secular cantatas we often encounter Bach recycling his own music for subsequent occasions. The later occasion might also be secular, as when music written for an academic dignitary (BWV 205, 207) was re-texted as a homage to the Elector of Saxony (BWV 205a, 207a). Or a secular work might be given a sacred text at a later date, as in the case of BWV 193a, written for the name-day of Augustus II on 3 August 1727, then only about three weeks later (for 25 August) partially adapted as BWV 193 for the local council election service. In several cases a particularly fine composition might become a repertoire piece that could be recycled over and over again, regardless of whether the occasion was sacred or secular. The pastoral cantata BWV 249a, for example, written for the birthday of Duke Christian of Saxe-Weißenfels on 23 February 1725, became an Easter cantata six weeks later (1 April), then a birthday cantata for Count J. F. von Flemming in the following year (BWV 249b, 25 August 1726), and finally it was transformed into the Easter Oratorio (BWV 249, 1735?). Schwingt freudig euch empor was originally written for the birthday of a local university teacher (BWV 36c, April–July 1725) but later celebrated the birthday of the Princess of Anhalt-Co¨then (BWV 36a, December 1725 or 1726). Two Advent Sunday versions followed (BWV 36, 1726–30 and 2 December 1731), after which it finally served as a congratulatory piece for a scholar from the Rivinus family (BWV 36b, summer 1735). One last example is

8 See Andreas Glo¨ckner, ‘Bachs Leipziger Collegium Musicum und seine Vorgeschichte’, in C. Wolff (ed.), Die Welt der Bach-Kantaten, vol. ii (Stuttgart, 1997), pp. 105–17 (esp. 112); and his ‘Zur Vorgeschichte des “Bachischen” Collegium musicum’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke, pp. 293–303 (esp. 299).

110 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s O holder Tag, BWV 210, which, having begun life in some unknown form before 1729, then served as a homage cantata for Duke Christian of Saxe-Weißenfels on 12 January 1729 (O! angenehme Melodei, BWV 210a) and subsequently for Count J. F. von Flemming and various unknown patrons. Finally, it took its definitive form as a wedding cantata (O holder Tag, BWV 210) around 1738/41. Thus BWV 249, 36, and 210 were each performed in four or five different versions, some secular and others sacred, over a period of roughly ten to twelve years. One of the most prominent types of secular vocal composition cultivated by Bach during his early Leipzig years is the dramma per musica. It is represented by five substantial works—BWV 249a, 249b, 205, 207, and 193a—of which all but one (BWV 207) are settings of texts by Picander. Dramma per musica was at that time a standard term for Italian opera; but in Leipzig, whose opera house had closed down in 1720, it came to represent a small theatrical piece with plot and dramatis personae, lasting about as long as one act of a grand opera. Such dramas were seldom staged in the theatre; they were more often sung in assembly rooms without acting or costumes.9 The sung roles are predominantly mythological characters, as often in contemporary opera seria, or allegorical figures, as in Bach’s Co¨then serenatas. In the earliest dramma per musica of the Leipzig period—the pastoral cantata Entfliehet, verschwindet, entweichet, ihr Sorgen, BWV 249a, written for the birthday of Duke Christian of Saxe-Weißenfels on 23 February 172510—the four singers take the parts of the shepherds Menalcas and Damoetas and the shepherdesses Doris and Sylvia, all drawn from Virgil’s Eclogues. The work is designated ‘Tafel-Music’ in Picander’s libretto, but this refers merely to its function as musique de table—a function it shares with the Hunt Cantata, BWV 208, composed for the same occasion twelve years earlier (1713). In musico-dramatic form it undoubtedly belongs to the dramma per musica type. The standard three-movement (fast–slow–fast) sinfonia of contemporary Italian opera is subtly elided with the first vocal movement, no. 3, which not only introduces the four characters in duet pairs but also acts as finale to the sinfonia, returning to the gigue-passepied rhythm and trumpets-and-drums scoring of no. 1. The text banishes the minor-mode pathos of the Adagio, no. 2, in the words ‘Flee, vanish, fade, you cares; do not unsettle our merry feelings’ (‘Entfliehet, verschwindet, entweichet, ihr Sorgen; verwirret die lustigen Regungen nicht’). The three arias, nos. 5, 7, and 9, all have woodwind obbligatos in accordance with the pastoral theme. The lullaby no. 7 comes nearest to an evocation of shepherds’ music. The finale, in which all four characters sing together for the first time, sets up expectations of ABA da capo form, which are then deliberately thwarted. A is 9 This description, drawn from Johann Christoph Gottsched’s Versuch einer critischen Dichtkunst (1730), is quoted by Hans-Joachim Schulze, ‘Kantatenformen und Kantatentypen’, in Wolff (ed.), Die Welt der BachKantaten, vol. ii, pp. 157–65 (esp. 158). 10 The music is lost but can be reconstructed from the sacred parody BWV 249—see the reconstruction by Paul Brainard in NBA II/7 (1977), p. 99. Picander’s text survives and is reproduced in German-English parallel text in Du¨rr, The Cantatas of J. S. Bach, pp. 805–7.

secular cantatas

111

constructed in binary dance form, with the voices built into the repeats, as in the finales of some of the Co¨then serenatas, though no longer in a recognizable dance rhythm. B then brings the standard da capo-aria modulation from vi to iii, but the da capo is unexpectedly replaced by new music in the 3/8 gigue-passepied rhythm of the first movement in order to round off the whole composition. Bach’s next major music drama, Der zufriedengestellte Aeolus (Aeolus Placated), BWV 205, was performed under six months later, on 3 August 1725. The plot is loosely based on Virgil’s account (from the Aeneid, Book I) of how Aeolus (bass) keeps the winds imprisoned in a cave. He threatens to free them, unleashing the autumn storms, but Zephyrus (tenor), god of mild summer breezes, Pomona (alto), goddess of fruit, and Pallas Athene (soprano), patroness of the arts, all beg deferment in order that Professor A. F. Mu¨ller’s name-day may be celebrated. The opening chorus, a da capo ritornello structure involving an exceptionally large instrumental ensemble, presents a truly remarkable tone-picture of the winds in all their elemental fury. In the end, the bluster of Aeolus and his winds (nos. 2, 3, and 11) cannot prevail over the cool, refreshing shade of Zephyrus, represented by a gentle obbligato duet of viola d’amore and viola da gamba (no. 5), a florid solo violin part (no. 9), or whispering obbligato flutes (no. 13). The fully scored finale, in which all four protagonists come together, is not dissimilar to that of the pastoral cantata. Binary, rondeau, and da capo forms are all united. In the framing A-section, a binary dance—again, not in any standard dance rhythm—is given twice, first for instruments only, then with inbuilt vocal parts. In the central B-section, two modulatory episodes frame a tonic reprise of the binary dance, creating an overall rondeau design. A third music drama, Vereinigte Zwietracht der wechselnden Saiten, BWV 207— again, written in honour of a university teacher, here Dr Gottlieb Kortte—was performed about sixteen months later, on 11 December 1726. The excessively abstract, moralizing libretto appears to be the work of an academic. The four voices represent allegorical figures; soprano: Das Glu¨ck (Fortune); alto: Die Dankbarkeit (Gratitude); tenor: Der Fleiß (Diligence); and bass: Die Ehre (Honour). The opening chorus is a magnificent vocal version of the Allegro third movement from Brandenburg Concerto No. 1. As already noted, the vocal version might be the original one, in which case this would represent a later manifestation of it. Bach’s intention was clearly that the ‘united discord of changing strings’ of the title and the ‘rolling drums’ penetrating boom’ (‘Der rollenden Pauken durchdringender Knall’) should be reflected in the instrumental parts. Immediately after the duet no. 5, Trio II (here entitled ‘Ritornello’) from the finale of the same concerto enters, perhaps as an illustration of Honour’s words ‘My laurel shall be his tutelary adornment’ (‘Den soll mein Lorbeer schu¨tzend decken’). The two arias are both clearly illustrative. The modish syncopation of the B minor tenor aria no. 3 reflects the words of Diligence: ‘Do not draw back your foot, you who choose my path’ (‘Zieht euren Fuß nur nicht zuru¨cke, ihr die ihr meinen Weg erwa¨hlt’). And in the G major alto aria no. 7, the ostinato dotted-rhythm figure for unison upper strings, in sharp contrast with the fugal flute duet that it accompanies,

112 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s clearly reflects the words of Gratitude: ‘Etch this commemoration in the hardest ¨ tzet dieses Angedenken in den ha¨rtsten Marmor ein’). The finale for the marble’ (‘A four characters combined plus instrumental ensemble is similar to that of Aeolus Placated: a binary dance without definite dance rhythm within an overall da capo form. In the A-section, each strain is first performed tutti, then played by the instruments only; in B two episodes, in keys vi and iii, are linked by a central relative-minor ritornello. Bach cultivated not only the dramatic forms of secular vocal music, the serenata and dramma per musica, but also the true ‘cantata’, a lyrical, contemplative genre for one or more solo voices that remained relatively close to its seventeenth-century Italian roots. The absence of the dramatic element meant that for Bach there was in principle little to distinguish this cantata type from its sacred equivalent beyond the text, which could easily be altered, and the Lutheran chorale, which could be added in a sacred parody. Four secular cantatas of this lyrical type survive from the early Leipzig years: BWV 36c, 204, 198, and 210a.11 Schwingt freudig euch empor, BWV 36c, which is expressly designated ‘cantata’, was performed in the spring or summer of 1725 to celebrate the birthday of an elderly teacher. In the opening chorus, an alternating form, ABA1B1, is united with an overall ritornello structure. The ritornello is notable for its concertante oboe d’amore part and for its triplet-semiquaver figures—a representation of ‘soaring up joyfully’— which impart a certain galant quality to the musical material. The strongly contrasted arias, nos. 3, 5, and 7, culminate in the inspired A major trio no. 7, whose florid obbligato for viola d’amore illustrates the idea that ‘Even with subdued, weak voices our teacher’s praise is proclaimed’ (‘Auch mit geda¨mpften, schwachen Stimmen verku¨ndigt man der Lehrer Preis’). The finale is a gavotte in binary dance form within a slightly modified da capo structure. The da capo is written out in full in order that the relation between voices and instruments may be reversed; thus, section A: each strain for instruments first, then voices built into the repeats; A1: each strain with voices first, then repeats for instruments only. The true innovation is reserved for section B: two further gavotte strains of eight bars each alternate with recitative for each of the three soloists in turn. Since the seventeenth-century Italian cantata was usually for solo voice with continuo, it is no wonder that this was viewed in eighteenth-century Germany as the authentic type. As late as 1732 Johann Gottfried Walther in his Musicalisches Lexicon defined ‘cantata’ as a musical piece with Italian (later French or German) words, consisting of arias and recitatives, for solo voice and basso continuo, and sometimes with two or more additional instruments. It is not surprising, then, that this intimate species of chamber cantata occupied a special place in Bach’s output. He had already composed several such works in Weimar, both sacred and secular (BWV 11 BWV 198 is a special case, as we shall see; BWV 210a survives as a fragment only but can be reconstructed from its later adaptation as BWV 210.

se c ular ca ntat as

113

199, 54, and probably 202); and during the early Leipzig years, for an unknown occasion in 1726 or 1727, he composed the solo-soprano cantata Von der Vergnu¨gsamkeit (On Contentedness), BWV 204, to a medley of texts by Hunold and others. Despite the tiresomely moralistic text (as it seems to our ears), Bach’s setting achieves remarkable variety, ranging from the dance rhythms of the G minor aria with oboe duet (no. 2), via the modish half-bar repeats in the F major aria with obbligato violin (no. 4), to the intricate dialogue of soprano and flute in the D minor aria (no. 6). The finale, no. 8, a B♭ aria with tutti instrumental ensemble, is in a dance-like 2/4 metre, often associated with homophonic textures. This type is not found in Bach before 1726 but occurs quite frequently in his vocal works thereafter.12 A keyboard equivalent may be found in certain Italianate dance movements of 1727–8: Capriccio (Partita 2 no. 6), Scherzo (Partita 3 no. 6), and Aria (Partita 4 no. 4). Laß, Fu¨rstin, laß noch einen Strahl, BWV 198, one of the most remarkable secular compositions of the early Leipzig years, was performed on 17 October 1727 in memory of the recently deceased Christiane Eberhardine, Electress of Saxony and Queen of Poland. The work is nowhere described as a cantata in the original sources. The original printed text is designated ‘ode’ in accordance with the form of Gottsched’s poem. The title page of the autograph score describes the composition as ‘Trauer Music’ (Music of Mourning), whereas the first page of music is headed ‘Tombeau de S. M. la Reine de Pologne’ (Lament for Her Majesty the Queen of Poland) in accordance with an old French tradition. Turning to the music itself, we discover that Bach has ignored Gottsched’s strophic-ode form and set the words in a cantatastyle alternation of recitatives and arias, framed by choruses. Indeed, since the composition falls into two parts, with the mourning oration delivered during the interval, the closest parallel lies in the two-part cantatas of Bach’s first months in Leipzig (BWV 75, 76, etc.), in which the sermon intervened between part I and part II. Apart from the conclusion of each part with a chorus that is freely composed rather than chorale-based, there is little to differentiate the Trauer Music from Bach’s sacred cantata style. The great solemnity of the occasion, the ecclesiastical venue (the university church), and no doubt Bach’s personal veneration for a woman who resisted her husband’s apostasy, remaining faithful to the Lutheran Church—all these circumstances help to explain the lofty, refined tone of Bach’s setting and its close affinity with his sacred vocal music of the same period. The great opening lament, in particular, would hardly have been out of place in the St Matthew Passion, first performed only six months earlier, on 11 April 1727. The motivic accompagnati of the Trauer Music might have found a place there too, especially no. 6, with its elegiac oboes d’amore. The masterly evocation of funeral bells in no. 4 has numerous precedents in the sacred cantatas of the 1720s; and in the alto aria no. 5 Bach exploited a rich vein of lullaby, as in the roughly contemporaneous ‘Ruhet hie, matte Sinnen’

Instances of it are enumerated in Finke-Hecklinger, Tanzcharaktere in J. S. Bachs Vokalmusik, pp. 142–3.

114 sa cred and s ecular: the vocal works from O! angenehme Melodei, BWV 210a, no. 4 (discussed at the end of this section) and ‘Schlummert ein’ from Ich habe genung, BWV 82, no. 3 (Ex. 1). The finale of Part I, an alla breve chorus in motet style, might have graced any mature sacred work of Bach’s. Only the finale of Part II is clearly rooted in secular style—a solemn giga in binary dance form with repeats, framed by an instrumental ritornello. It is as moving and appropriate as any of the preceding movements, however, excelling itself at the moment in the text when poets are called upon to write about the queen, whereupon what they write is sung in a great choral unison: ‘She has been virtue’s property, j Her subjects’ delight and glory, j The prize of queens’ (‘Sie ist der Tugend Eigentum, j Der Untertanen Lust und Ruhm, j Der Ko¨niginnen Preis gewesen’). It is hardly surprising, then, that the outer movements of the Trauer Music were reused alongside extracts from the St Matthew Passion in the lost funeral music for Prince Leopold of AnhaltCo¨then about eighteen months later, in March 1729. By the same token, one imagines that the same two choruses from the Trauer Music, together with all the arias, would have been perfectly suited to the lost St Mark Passion, where they found their final home in March 1731. The last cantata to be considered here, BWV 210a, is (like Von der Vergnu¨gsamkeit) a ‘Cantata a voce sola’. It is thought to have existed in some form before 1729—perhaps dedicated to the Co¨then prince August Ludwig, immediate successor to Prince Leopold. But the earliest evidence of it that we possess is O! angenehme Melodei, BWV 210a, performed on 12 January 1729 as a homage cantata for Duke Christian of Saxe-Weißenfels on the occasion of a visit to Leipzig. Only the solo soprano part survives, but the instrumental parts can be reconstructed from the later version, O holder Tag, BWV 210, of 1738/41. Though identical in type to Von der Vergnu¨gsamkeit, it outstrips the earlier work in virtually every respect. The quite exceptional quality of the music seems to be connected with the subject matter. Bach was always susceptible

Ex. 1

Wie

starb

die Hel

din so

ver - gnügt,

a) 5th movement of Trauer Music, BWV 198, bb. 9–10 (voice part only)

het

hie,

mat

Sin

nen

b) 4th movement of O! angenehme Melodei, BWV 210a, bb. 9–10 (voice part only)

Schlum -

mert

ein,

ihr

mat

ten

(gen)

c) 3rd movement of Ich habe genung, BWV 82, bb. 10–11 (voice part only)

sacred cantatas: leipzig cycle i

115

to aural imagery, but normally only isolated movements are affected. Here, however, the entire text is devoted to the power of music. The opening accompanied recitative begins with the invocation ‘O! angenehme Melodei!’ (‘O pleasant melody’). The words ‘Spielet, ihr beseelten Lieder’ (‘Play, you inspired songs’) then call forth an exquisite A major aria in minuet rhythm (no. 2). After a recitative observation that ‘A song being sung makes bitter griefs sweet’ (‘Ein singend Lied macht herbes Gra¨men su¨ße’), an exceptionally beautiful lullaby in the rhythm of a pastorale (no. 4), remarkably similar to BWV 198 no. 5 (see Ex. 1), praises ‘tender harmony’ (‘Eine zarte Harmonie’) as affording ‘rest for tired senses’ (‘Ruhet hie, matte Sinnen’). According to the recitative no. 5, ‘Beloved Music’ (‘beliebte Musica’) is distressed by those who do not value her—no doubt a veiled reference to Bach’s Leipzig enemies— hence her twice silencing the lovely ﬂute ritornellos of the following aria no. 6. Yet at least the dedicatees of this cantata13 value music (no. 7): above all, they love ‘an agreeable melody’ (‘ein angenehme Melodei’), hence the oboe d’amore obbligato in the rhythm of a polonaise in the C♯ minor aria no. 8. Finally, in a motivic accompagnato followed by a Vivace-ﬁnale that requires the complete instrumental ensemble, Bach calls upon his patrons to continue their support for ‘noble harmony’ (‘der edlen Harmonie . . . geneigt’). There can be no doubt that Bach’s word-setting in this composition, carried out with such supreme artistry, was heartfelt in view of its subject matter, ‘Beloved Musica’ herself. Nor is it likely to be mere chance that, not long after the composition of this exquisite work, Bach, newly appointed as director of the Leipzig Collegium musicum in the spring of 1729, took up the cause of Musica again in his artistic credo, Der Streit zwischen Phoebus und Pan (The Dispute between Phoebus and Pan), BWV 201.

Sacred cantatas: Leipzig Cycle I 14 Title, occasion

Earliest source/s15

Du wahrer Gott und Davids Sohn, BWV 23, Berlin, P 69, St 16 Quinquagesima Jesus nahm zu sich die Zwo¨lfe, BWV 22, Berlin, P 119 Quinquagesima Die Elenden sollen essen, BWV 75, Trinity 1 Berlin, P 66

Scribe, date16 Autograph, part-autograph, for 7 Feb. 1723 Autograph, for 7 Feb. 1723 Autograph, for 30 May 1723 (cont.)

13 At various times Duke Christian of Saxe-Weißenfels, Count J. F. von Flemming, and various unknown patrons. 14 The list includes not only Cycle I but other cantatas performed during Bach’s first year in Leipzig: BWV 23, 22, 119, and 194. 15 P (Partitur) refers to a score and St (Stimmen) to performing parts. 16 Date refers to performance, not composition. Dates are drawn primarily from Du¨rr Chr 2. Where several scribes are given, separated by a comma, the first refers to the first source in the previous column, the second to the second source.

116 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s (cont.) Title, occasion

Earliest source/s15

Die Himmel erza¨hlen die Ehre Gottes, BWV 76, Trinity 2 Ein ungefa¨rbt Gemu¨te, BWV 24, Trinity 4 Ihr Menschen, ru¨hmet Gottes Liebe, BWV 167, St John’s Day Herz und Mund und Tat und Leben, BWV 147, Visitation ¨ rgre dich, o Seele, nicht, BWV 186, Trinity 7 A Erforsche mich, Gott, BWV 136, Trinity 8 Herr, gehe nicht ins Gericht, BWV 105, Trinity 9 Schauet doch und sehet, BWV 46, Trinity 10 Siehe zu, daß deine Gottesfurcht, BWV 179, Trinity 11 Lobe den Herrn, meine Seele, BWV 69a, Trinity 12 Du sollt Gott, deinen Herren, lieben, BWV 77, Trinity 13 Es ist nichts Gesundes, BWV 25, Trinity 14 Preise, Jerusalem, den Herrn, BWV 119, council election Warum betru¨bst du dich, mein Herz, BWV 138, Trinity 15 Christus, der ist mein Leben, BWV 95, Trinity 16 Ich elender Mensch, BWV 48, Trinity 19

Berlin, P 67, St 13b

Scribe, date16

Autograph, J. A. Kuhnau et al., for 6 June 1723 Berlin, P 44/3 , St 19 Autograph, J. A. Kuhnau et al., for 20 June 1723 Berlin, St 61 J. A. Kuhnau et al., for 24 June 1723 Berlin, P 102, St 46 Autograph, J. A. Kuhnau et al., for 2 July 1723 Berlin, P 53 B. C. Kayser, for 11 July 1723 Berlin, St 20 J. A. Kuhnau et al., for 18 July 1723 Berlin, P 99 Autograph, for 25 July 1723 Berlin, St 78 J. A. Kuhnau et al., for 1 Aug. 1723 Berlin, P 146, St 348 Autograph, J. A. Kuhnau et al., for 8 Aug. 1723 Berlin, St 68 Part-autograph, for 15 Aug. 1723 Berlin, P 68

Autograph, for 22 Aug. 1723

Berlin, St 376 Berlin, P 878

J. A. Kuhnau et al., for 29 Aug. 1723 Autograph, for 30 Aug. 1723

Berlin, P 158

Autograph, for 5 Sept. 1723

Berlin, St 10

J. A. Kuhnau et al., for 12 Sept. 1723

Berlin, P 109, St 53

Autograph, J. A. Kuhnau et al., for 3 Oct. 1723 Ich glaube, lieber Herr, BWV 109, Trinity 21 Berlin, P 112, St 56 Autograph, J. A. Kuhnau et al., for 17 Oct. 1723 Was soll ich aus dir machen, BWV 89, Berlin, St 99 C. G. Meißner et al., Trinity 22 for 24 Oct. 1723 Ho¨chsterwu¨nschtes Freuden fest, BWV 194, Berlin, P 43/2 , St 48 Autograph, J. A. Kuhnau et al., organ consecration for 2 Nov. 1723 O Ewigkeit, du Donnerwort, BWV 60, Berlin, St 74 J. A. Kuhnau et al., for 7 Nov. 1723 Trinity 24 Es reißet euch ein schrecklich Ende, BWV 90, Berlin, P 83 Autograph, for 14 Nov. 1723 Trinity 25 Wachet, betet, betet, wachet, BWV 70, Berlin, St 95 J. A. Kuhnau et al., for 21 Nov. 1723 Trinity 26 Berlin, P 63, St 11 Autograph, J. A. Kuhnau et al., Darzu ist erschienen der Sohn Gottes, for 26 Dec. 1723 BWV 40, 2nd Day of Christmas Sehet, welch eine Liebe, BWV 64, Berlin, St 84 J. A. Kuhnau et al., for 27 Dec. 1723 3rd Day of Christmas Singet dem Herrn, BWV 190, Berlin, P 127, St 88 Autograph, J. A. Kuhnau et al., New Year’s Day for 1 Jan. 1724 Schau, lieber Gott, BWV 153, Berlin, St 79 J. A. Kuhnau et al., for 2 Jan. 1724 Sunday after New Year Sie werden aus Saba alle kommen, BWV 65, Berlin, P 147 Autograph, for 6 Jan. 1724 Epiphany Mein liebster Jesus ist verloren, BWV 154, Berlin, P 130, St 70 B. C. Kayser, J. A. Kuhnau et al., Epiphany 1 for 9 Jan. 1724

s a cr e d c a n t a t a s : l e i p z i g c y c l e i Herr, wie du willt, so schicks mit mir, BWV 73, Epiphany 3 Jesus schla¨ft, was soll ich hoffen, BWV 81, Epiphany 4 Erfreute Zeit im neuen Bunde, BWV 83, Puriﬁcation Nimm, was dein ist, und gehe hin, BWV 144, Septuagesima Leichtgesinnte Flattergeister, BWV 181, Sexagesima Erfreut euch, ihr Herzen, BWV 66, Easter Monday Ein Herz, das seinen Jesum, BWV 134, Easter Tuesday Halt im Geda¨chtnis Jesum Christ, BWV 67, Easter 1 Du Hirte Israel, ho¨re, BWV 104, Easter 2 Wo gehest du hin, BWV 166, Easter 4 Wahrlich, wahrlich, ich sage euch, BWV 86, Easter 5 Wer da gla¨ubet und getauft wird, BWV 37, Ascension Sie werden euch in den Bann tun, BWV 44, Ascension 1 Wer mich liebet, BWV 59, Whit Sunday Erho¨htes Fleisch und Blut, BWV 173, Whit Monday Erwu¨nschtes Freudenlicht, BWV 184, Whit Tuesday

117

Berlin, St 45

J. A. Kuhnau et al., for 23 Jan. 1724

Berlin, P 120, St 59 Berlin, St 21

Autograph, J. A. Kuhnau et al., for 30 Jan. 1724 J. A. Kuhnau et al., for 2 Feb. 1724

Berlin, P 134

Autograph, for 6 Feb. 1724

Berlin, St 66

J. A. Kuhnau et al., for 13 Feb. 1724

Berlin, P 73

Autograph, for 1731 revival, 1st perf. 10 Apr. 1724 Berlin, P 44/2 , St 18 Autograph, part-autograph, 1st perf. 11 Apr. 1724 Berlin, P 95, St 40 Autograph, J. A. Kuhnau et al., for 16 Apr. 1724 Berlin, St 17 J. A. Kuhnau et al., for 23 Apr. 1724 Berlin, St 108 J. A. Kuhnau et al., for 7 May 1724 Berlin, P 157 Autograph, for 14 May 1724 Berlin, St 100

J. L. Krebs et al., for 18 May 1724

Berlin, P 148, St 86

Autograph, J. A. Kuhnau et al., for 21 May 1724 Berlin, P 161, St 102 Autograph, C. G. Meißner et al., for 28 May 1724 Berlin, P 74 C. G. Meißner, c. 1728, 1st perf. 29 May 1724? Berlin, St 24 Part-autograph, for 30 May 1724

Before he took up his Leipzig post on 30 May 1723, Bach had already performed at the Thomaskirche two audition cantatas, BWV 22 and 23 (the latter brought with him from Co¨then), one before the sermon and the other afterwards, on Quinquagesima (Estomihi) Sunday, 7 February 1723. He may also have performed the Whit Sunday cantata BWV 59 on 16 May 1723.17 On the very day that he embarked on his new duties as Thomascantor, Sunday 30 May, Bach performed the ﬁrst cantata of his new churchyear cycle (Cycle I), BWV 75, ‘mit guten applausu’ (with good applause).18 Since 30 May was the First Sunday after Trinity, Bach’s cantata cycles henceforth began on that occasion, rather than—as one might otherwise expect—at the beginning of the church year, the First Sunday in Advent. During his ﬁrst year as Leipzig music director, Bach performed at least sixty-three church cantatas, of which forty were new compositions and twenty-three revivals of existing works from the Weimar or Co¨then periods.

17 Perhaps in the university church; the autograph score seems to have been written for that date, but the performing parts for the following year. 18 According to the Acta Lipsiensium Academica, 1723; BD II, No. 139; NBR, No. 103.

118

sacred and secular: the vocal works

It was Bach’s own decision to perform at the main Leipzig service exclusively cantatas of his own composition, a self-imposed task of enormous magnitude, which nonetheless satisﬁed his deep-seated ambition to create ‘a well-regulated church music to the glory of God’.19 Quite frequently he even performed two cantatas or parts I and II of a two-part work during the same service, before and after the sermon. Many of these double compositions are at least in part of Weimar or (less often) Co¨then origin, as shown in the following list (W = Weimar; C = Co¨then):20 BWV

Occasion

Date

22 + 23 (C) 75, Parts I and II 76, Parts I and II 21 (W), Parts I and II 24 + 185 (W) 147 (W), Parts I and II 186 (W), Parts I and II 179 + 199 (W) 70 (W), Parts I and II 181 + 18 (W) Anh. I 199 + 182 (W) 4 (pre-W) + 31 (W) 59 + 172 (W) 194 (C) + 165 (W)

Quinquagesima Trinity 1 Trinity 2 Trinity 3 Trinity 4 Visitation Trinity 7 Trinity 11 Trinity 26 Sexagesima Annunciation Easter Whit Trinity

7 Feb. 1723 , 20 Feb. 1724 30 May 1723 6 June 1723 13 June 1723 20 June 1723 2 July 1723 11 July 1723 8 Aug. 1723 21 Nov. 1723 13 Feb. 1724 25 Mar. 1724 9 Apr. 1724 28 May 1724 4 June 1724

By reviving or adapting pre-Leipzig cantatas, Bach must have sought to relieve the huge burden he had placed on himself, not to mention prolonging the life of ﬁne compositions from the not-too-distant past. Many of the Weimar works could be revived without major changes. In certain cases, however, Bach found it expedient to enlarge the vocal and/or instrumental ensemble in order to suit the larger resources available in Leipzig. In Ich hatte viel Beku¨mmernis (BWV 21), for example, choirs of trombones and vocal ripienists were added; in Gleichwie der Regen und Schnee vom Himmel fa¨llt (BWV 18) two recorders were added to the ensemble of four violas and continuo, doubling the ﬁrst two violas at the upper octave; and additional violins, oboe, and violone were added to the instrumental ensemble of Himmelsko¨nig, sei willkommen (BWV 182). The last three cantatas composed in Weimar, BWV 70a, 186 a, and 147 a, could not be revived without change, since they were written for the penitential period between Advent and Christmas when no concerted music was permitted in the Leipzig churches. Consequently textual alterations were made that allowed them to be performed on other occasions (Trinity 26, Trinity 7, and the Feast of the Visitation respectively) and they were expanded to the two-part structure of the

19 20

BD I, No. 1; NBR, No. 32. Other Weimar cantatas revived during Cycle I are BWV 162, 163, 61, 63, 155, and 12.

s a cr e d c a nt a t a s : l e ip z i g c y c l e i

119

first Cycle I cantatas (BWV 75, 76, and 21) by inserting recitatives and adding new chorales as the finale of each part. The earliest known sacred parody of Bach’s first year in Leipzig (parody being the re-texting of existing compositions) was performed on 2 November 1723 on the occasion of an organ consecration at Sto¨rmthal, near Leipzig. The work concerned is Ho¨chsterwu¨nschtes Freudenfest, BWV 194, which Bach adapted from a Co¨then secular cantata, BWV 194a (see the section on ‘Secular Cantatas’). The Co¨then work was in effect a vocal French ouverture-suite in which every aria was cast in a different dance rhythm. Bach clearly saw no incongruity in applying this secular style to the setting of sacred texts. Indeed, dance rhythms would proliferate in his sacred vocal works during the early Leipzig years. He makes an important concession to sacred style, however, by concluding each part of the two-part composition with a chorale, as in the other two-part cantatas of Cycle I (except BWV 21). Bach considered this imposing composition, with its mixed sacred–secular style, suitable for revival on one of the major feasts of the church year, Trinity Sunday (4 June 1724), a performance that brought Cycle I to a close. Bach also undertook sacred parodies of four Co¨then serenatas towards the end of the cycle, in the spring of 1724. They were performed on the Second and Third Days of Easter (BWV 66 and 134; 10–11 April) and of Whit (BWV 173 and 184; 29–30 May). The three Lutheran High Feasts, Christmas, Easter and Whit, were each celebrated over three successive days, and Bach had to provide cantatas for each of them. The use of parodied compositions must have lightened his burden; and it is also possible that he wished to lighten the tone on the second and third days after the relatively serious, demanding music of the first day (Easter: BWV 4 and 31; Whit: BWV 172 and 59). When the sacred parodies were first made, in spring 1724, relatively little change appears to have been made to the secular originals beyond what was necessary to accommodate the new sacred text. Yet despite the preponderance of secular dance style and the virtual absence of traditional sacred elements, Bach must have relished the opportunity to render this exceptionally attractive music, otherwise consigned to ephemera, a permanent part of his Leipzig repertoire, for he revised it on several occasions in later years and made numerous significant improvements to it. It is clear that Bach did not have the advantage of a single librettist who was consistently available during his first year in Leipzig. This would explain why Cycle I lacks unity in terms of the overall structure of its constituent cantatas. The only texts by known librettists are those of Cantata No. 24 (Neumeister, 1714) and Nos. 69a, 77, and 64, which are condensed versions of texts from Johann Knauer’s Gott-geheiligtes Singen und Spielen (Gotha, 1720), a cycle that was set to music by Gottfried Heinrich Sto¨lzel.21 It has been observed that the first four Leipzig cantatas, BWV 22, 23, 75, and 76, all appear to have texts by the same librettist, possibly the Leipzig Burgomaster

See Helmut K. Krausse, ‘Eine neue Quelle zu drei Kantatentexten J. S. Bachs’, BJ 67 (1981), pp. 7–22.

120

sacred and secular: the vocal works

Gottfried Lange.22 In addition, three groups of cantatas have been identified on the basis of similar text structure, which may point to common authorship of their librettos.23 Altogether it is thought that for Cycle I Bach might have collaborated with at least four librettists, whose identity is perhaps to be sought among the local clergy. The type of libretto introduced by Neumeister in his Geistliche Cantaten statt einer Kirchen-Music (Spiritual Cantatas in place of Church Music) of 1700, with its madrigalian verse, to be set in alternating recitative and arias, was first taken up by certain older, well-established composers, such as Johann Philipp Krieger and Johann Kuhnau. During the period 1715–25, however, composers of the younger generation, such as Telemann and Bach, applied to the Neumeister cantata a musical style that was in some ways radically new and involved a huge increase in the technical demands made on the performers.24 This operatic style of church music was a major innovation and accorded with Neumeister’s view of the new cantata as ‘nothing other than a piece from an opera, assembled from recitative style and arias’ (‘nicht anders . . . als ein Stu¨ck aus einer Opera, von Stylo Recitativo und Arien zusammengesetzet’).25 According to Gottfried Ephraim Scheibel, a former Leipzig theology student writing in 1721, ‘If secular music can give us pleasure and often take away our cares, why should we not also have similar pleasure when we hear music being made in church?’, for ‘religious and secular music have no distinctions as far as the movement of the affections is concerned’.26 Such thinking explains why Bach considered it appropriate to create sacred parodies out of secular compositions, as in the Co¨then serenatas discussed earlier. The admission of secular styles, whether operatic, concertante, or dance-based, into the sacred cantatas of Bach, Telemann, and others would have been unthinkable in the church music of the previous generation. Bach had already adopted this strongly secular-influenced style in the sacred cantatas he composed in Weimar from 1713 to 1717. Indeed, the Weimar cantatas might well be viewed as more radical in this respect than those of the first Leipzig cycle. Such is the dominant role of madrigalian verse in the Weimar texts, mostly by Salomo Franck, that sixteen of the twenty-two surviving sacred cantatas of that period lack biblical-text movements altogether. Where biblical words are present they are set not as a chorus (except in the early and in some ways decidedly retrospective Cantata No. 21) but as a recitative. Since these recitatives usually quote the words of the Lord, they are sung by the bass, the traditional vox Christi of Passion settings (BWV 182, 172, 61, and 18). Where 22

According to Hans-Joachim Schulze, ‘Texte und Textdichter’, in C. Wolff (ed.), Die Welt der BachKantaten, vol. iii (Stuttgart, 1999), pp. 109–25 (esp. 110–11). 23 For a summary of research on this issue, see Du¨rr, The Cantatas of J. S. Bach, pp. 26–8. 24 See Peter Wollny and Christoph Wolff, ‘Allgemeine Strategien in Bachs I. Leipziger Kantatenjahrgang’, in M. Geck (ed.), Bachs 1. Leipziger Kantatenjahrgang (Dortmund, 2002), pp. 23–40. 25 Preface to Erdmann Neumeister’s Geistliche Cantaten statt einer Kirchen-Music (1700). 26 Gottfried Ephraim Scheibel, Zufa¨llige Gedancken von der Kirchen-Music, wie sie heutiges Tages beschaffen ist (Frankfurt and Leipzig, 1721); Eng. trans. as ‘Random Thoughts about Church Music in Our Day’, introd. and trans. by Joyce L. Irwin, in Carol K. Baron, Bach’s Changing World: Voices in the Community (Rochester, NY and Woodbridge, 2006), pp. 227–49 (see 236 and 238).

sacred cantatas: leipzig cycle i

121

choruses are present in the Weimar cantatas—in BWV 182, 12, 172, 63, 70a, 186a, and 147a, all of which were revived during Leipzig Cycle I—they are settings of freely invented, madrigalian texts and thus far indistinguishable from the arias. Indeed, Franck in his printed texts often designated these movements ‘aria’. Chorales play a more prominent part in the Weimar cantatas than biblical-text movements, though elaborate chorale-choruses are present only in the early Cantata No. 21 and in the Advent cantata No. 61 (1714), and three cantatas lack a chorale altogether (Nos. 63, 152, and 54). The earlier Weimar cantatas tend to have an intermediate chorale (BWV 21, 199, 182, 172, 18); only occasionally is it placed at the end of the work (BWV 12, 61, and 18). Later on, however, in Bach’s settings of texts from Franck’s 1715 and 1717 collections, the chorale-finale becomes the norm. During the Weimar years Bach often enhanced a plain chorale setting by furnishing it with an instrumental descant (BWV 12, 172, 31, 161, and 70a). Furthermore, he often used his own initiative by introducing a complete instrumental chorale quotation in an aria or duet (BWV 12, 172, 80a, 31, 185, 161, 163), in some cases anticipating the melody of the chorale-finale.27 Only against this background is it possible to understand Bach’s achievement in the first Leipzig cycle. When he arrived in Leipzig for his audition in February 1723, evidently he brought with him the original version of the cantata Du wahrer Gott und Davids Sohn, BWV 23. Not surprisingly, it fuses elements of the Weimar sacred cantatas with aspects of the Co¨then secular cantatas. From Weimar are derived the instrumental chorale quotation (here in a recitative, no. 2, however, rather than in an aria), the madrigalian-verse chorus no. 3, and the absence of biblical text; from Co¨then, the duet treatment of the opening aria and of the episodes in the chorus, and also the rondeau form, dance rhythm, and homophonic texture of that chorus. In Leipzig Bach added a sublime, deeply moving chorale-finale, a setting of Christe, du Lamm Gottes— the same chorale quoted instrumentally in the second movement, in keeping with Bach’s Weimar procedure noted earlier. The structure of the finale as a set of three chorale variations accords with the origin of the chorale in the ancient threefold prayer of the Agnus Dei. Each of the three prayers is embedded in a different, elaborate instrumental setting, of which the middle one gives the chorale in strict three-part canon for soprano, unison oboes, and first violin. This movement decisively shifts the stylistic balance of the cantata from secular to sacred. What led Bach to take this step? One wonders whether he sensed, or was informed of, a climate of opinion in Leipzig resistant to secular style unless it was adequately counterbalanced by traditional sacred elements. Relevant here, perhaps, is the well-known stipulation, at the joint assembly of the three Leipzig councils in which Bach was elected, that ‘he should make compositions that were not theatrical’.28 Ecclesiastical and operatic elements are well balanced in Jesus nahm zu sich die Zwo¨lfe, BWV 22, presumably the later of the two audition cantatas, since it no longer 27 28

For further details see the coverage of Bach’s Weimar cantatas in Vol. I, pp. 243–96. BD II, No. 129; NBR, No. 98.

122 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s looks back to Weimar or Co¨then. On the contrary, in certain important respects it seems to establish a blueprint for the Leipzig cantatas of Cycle I that followed. The overall form is as follows: biblical-text movement—aria—recitative—aria—chorale. Thus settings of authoritative ecclesiastical texts form a frame around the madrigalian-text movements. Such a frame occurs more often than not in the Cycle I cantatas. No doubt it would have satisfied the Leipzig clergy, but there is good reason to suppose that it would also satisfy Bach himself, for he consistently poured some of his finest music into these ecclesiastical-text settings. In the opening movement of Cantata No. 22, the biblical narrative is set as a dramatic scena worthy of the Bach Passions—a highly relevant comparison in view of the subject matter. The tenor Evangelist introduces the scene, whereupon the bass, the traditional vox Christi of Passion music, sings Jesus’s words announcing the journey to Jerusalem and the forthcoming Passion. Finally, the twelve disciples show their lack of comprehension of Jesus’s words in a four-part permutation fugue. The vivid drama of this movement has no real counterpart in Bach’s Cycle I cantatas—BWV 81 (nos. 4–5), 67 (no. 6), and 44 (nos. 1–2) come closest. But two aspects of it recur repeatedly in Cycle I: the setting of biblical words as a choral fugue, and the singing of Jesus’s words by the bass voice in a heightened form that lies midway between arioso and aria. Bach’s intention was presumably to raise the tone of such Jesus settings above the level of recitative, employed at Weimar, while avoiding the secular connotations of aria. In the cantata under consideration, the florid arioso of Christ is built into repeated, varied ritornello returns for the full instrumental ensemble. This Vokaleinbau (vocal insertion) technique is later used in several vox Christi movements from Cycle I (BWV 89 no. 1, 153 no. 3, and 166 no. 1). Elsewhere, however (BWV 154 no. 5 and 81 no. 4), bass voice and continuo collaborate in a two-part imitative texture; in the first case, the imitative texture presumably represents the Son’s imitation of the Father, for Jesus asks, ‘Do you not know that I must be about my Father’s business?’ (‘Wisset ihr nicht, daß ich sein muß in dem, das meines Vaters ist?’). In the vox Christi movement that opens Cantata No. 86, Wahrlich, wahrlich ich sage euch, imitation turns into full-blown fugue, with two regular countersubjects, for bass voice and instrumental ensemble. The authority of Christ’s words is here reflected in the strictness of the fugal writing and in the traditional, pseudo-vocal polyphonic style in alla breve metre. In two other vox Christi movements, Jesus’s voice features in a dramatic scene that may be compared with the opening movement of Cantata No. 22. In Jesus schla¨ft, was soll ich hoffen, BWV 81 no. 4, in which bass and continuo engage in two-part imitation, Jesus sings to his disciples, fearful of shipwreck, ‘You of little faith, why are you so fearful?’ (‘Ihr Kleinglaübigen, warum seid ihr so furchtsam?’). Jesus then calms the storm in the madrigalian words of the following da capo aria. In the remarkable penultimate movement of Halt im Geda¨chtnis Jesum Christ, BWV 67, the bass sings Christ’s first words to the disciples after the Resurrection, ‘Peace be with you!’ (‘Friede sei mit euch!’), in arioso, accompanied by a ‘peace’ motive for flute and two oboes d’amore. This arioso alternates with a strophic chorus—built into varied returns of the opening

s a c r e d can t a t a s : l e i p z ig c y c l e i

123

string ritornello—which represents earthly commotion and strife, the very antithesis of the peace offered by Christ. The great biblical-text choruses that open many of the cantatas from Cycle I are one of its defining features. There was no precedent among the Weimar cantatas, except in the early Cantata No. 21 (1713), whose great psalm choruses are nonetheless highly retrospective, inhabiting the somewhat antiquated world of the pre-Weimar cantatas. The heading of each cantata with biblical words made the first movement analogous to the text of a sermon, giving it divine authority—that of the Word of God itself. Bach clearly considered fugue, with its palpable order, strictness and relative antiquity, to be the most appropriate method of setting these biblical texts. Part I of Cantata No. 21 concludes with what might be described as a choral prelude and fugue, a setting of Psalm 42: 11. This cantata was revived on 13 June 1723 (Trinity 3), immediately after the first two cantatas of Cycle I, BWV 75 and 76. Here the same choral prelude and fugue form, again in association with a psalm text, is used for the opening dictum. The ‘prelude’ is updated, however: it no longer takes the sectional form of the motet, as in Cantata No. 21, but is instead a unified, concertante piece in ritornello form. The fugue, which on occasion employs permutation technique (Cantatas Nos. 76 and 105), tends to build up from solo choir with continuo to tutti choir with doubling instruments. In every case, using fugal and concertante means, Bach conjures up a powerfully affective musical correlative to the text, setting the scene for the entire cantata. In Die Elenden sollen essen, BWV 75, the halting dotted rhythms and sharp dissonances of the ‘prelude’ portray ‘the poor’ (‘die Elenden’), whereas the repeated quavers and running semiquavers of the fugue stand for ‘eternal life’ (‘ewiglich Leben’). A week later, in Cantata No. 76, Bach gave a magnificent, celebratory setting of the psalm words (Psalm 19: 1) later set so memorably by Haydn in The Creation: ‘The heavens are telling the glory of God’ (‘Die Himmel erza¨hlen die Ehre Gottes’). In two successive cantatas, Nos. 105 and 46, a few weeks later, Bach turned to darker aspects of the Christian life, in accordance with the Gospel readings. The Adagio ‘prelude’ of Herr, gehe nicht ins Gericht, BWV 105, illustrates fear of divine judgement in a close canon at the upper 5th, involving a sequence of rising semitones over a slow tremolo bass. On the following Sunday, in Cantata No. 46, Bach set the words ‘Behold and see if there be any sorrow like my sorrow’ (‘Schauet doch und sehet, ob irgend ein Schmerz sei wie mein Schmerz’; Lamentations 1: 12)—familiar from Handel’s setting in the Messiah—as a deeply moving lament. It gives way to a fugue, whose tonally wayward, seemingly indeterminate subject depicts the words ‘For the Lord has made me full of misery’ (‘Denn der Herr hat mich voll Jammers gemacht’), while the contrasting countersubject, rhythmic with short note-values, portrays ‘the day of his fierce anger’ (‘am Tage seines grimmigen Zorns’). Bach often attempted to integrate prelude and fugue into a single, compound entity. A step in this direction is taken in Cantata No. 24 no. 3, where the material of the prelude returns in a coda at the end of the double fugue. Full integration of concertante and fugal elements occurs in Cantata No. 69a no. 1, a festive song of praise

124

sacred and secular: the vocal works

to a psalm text (Psalm 103: 2). Here a double fugue is surrounded by a choral prelude and postlude, which are in turn framed by instrumental ritornellos. The prelude and postlude are in effect ritornello reprises with built-in vocal parts, and even the fugue subjects are derived from ritornello material, so that the entire structure hangs together. The opening movement of the Christmas cantata No. 40 is not dissimilar, except that the text is drawn from the New Testament (1 John 3: 8), the concluding instrumental ritornello is lacking, and the two subjects of the double fugue contrast sharply, one representing ‘the Son of God’ (‘der Sohn Gottes’), and the other ‘the works of the devil’ (‘die Werke des Teufels’). A related form is found in the ﬁrst movement of the Epiphany cantata Sie werden aus Saba alle kommen, BWV 65: ritornello—prelude—fugue—postlude. Again, the choral prelude and postlude that surround the fugue are built out of the ritornello in a manner that involves Choreinbau (the insertion of choral parts into ritornello material). The text, from Isaiah 60: 6—‘They will all come out of Sheba, bearing gold and incense, and proclaiming the Lord’s praise’ (‘Sie werden aus Saba alle kommen, Gold und Weihrauch bringen und des Herren Lob verku¨ndigen’)—inspires the composer to present a graphic, colourful tone-picture of a multitude of Gentiles travelling from the east towards Bethlehem. A rather more complex amalgamation of ritornello and fugue occurs in the opening movement of Cantatas Nos. 136, 67, and 104: ritornello—prelude—fugal exposition I—ritornello—fugal exposition II—ritornello. In Erforsche mich, Gott, und erfahre mein Herz, BWV 136, the bold horn theme that opens the ritornello is not only sung by soprano in the prelude but forms the subject of the two fugal expositions, so that the whole movement is thematically united. Much the same applies to the similarly bold, memorable, triadic horn theme in the same key (A) in Halt im Geda¨chtnis Jesum Christ, BWV 67. Here, however, Bach writes a double fugue, employing the horn theme for the opening words, ‘Keep in remembrance Jesus Christ’, and a contrasting quaver theme for the following words, ‘who is arisen from the dead’ (‘der auferstanden ist von den Toten’). Also, in this case both ritornello returns are furnished with inbuilt vocal parts. The opening chorus of Du Hirte Israel, ho¨re, BWV 104, which has strong pastoral overtones in keeping with its psalm text (Psalm 80: 1), takes the form: Sinfonia—chor. a—fugue b—chor. a1—fugue b1—chor. a2. Since the instrumental introduction does not recur, it is not a ritornello but rather an integral sinfonia. From it the three homophonic choruses that surround the fugal expositions are derived. Although the fugue subject is new, it is accompanied by sinfonia material in the instrumental parts. In three cases, BWV 179, 64, and 144, the opening dictum is set simply as a motetstyle fugue, without prelude, ritornello, or any other concertante element. These fugues are written for four-part choir with colla parte instruments and continuo. In keeping with their alla breve metre, the polyphonic style employed is decidedly retrospective. It is occasionally diversiﬁed by individual features, however, such as the chromatic word-painting in Cantata No. 179 on ‘with false heart’ (‘mit falschem Herzen’), or the rhythmic countersubject on ‘go your way’ (‘gehe hin’) in No. 144.

s a c r e d can t a t a s : l e i p z ig c y c l e i

125

In a few exceptional cases Bach sets the opening dictum without recourse to fugue. The opening chorus of the Ascension Day cantata Wer da glaübet und getauft wird, BWV 37, takes the form: sinfonia (key A)—chor. a (A–E)—chor. a1 (E–A). Both choral sections are built out of the very complex, thematically rich ritornello by means of expansion, transposition, and Choreinbau. The equivalent movement of Cantata No. 109 is constructed in true ritornello form thus: rit. (key d)–chor. a (d–a)–rit. (a–C)– chor. a1 (a–g)–rit. (g)–chor. a2 (g–d)–rit. (d). The choral sections possess their own theme, which is, however, ultimately derived from the initial ritornello motive. Each chorus is prefaced by an extended vocal solo, which creates an intimate atmosphere appropriate to the first person of the text: ‘I believe, dear Lord, help my unbelief ’ (‘Ich glaube, lieber Herr, hilf meinem Unglauben’). In Preise, Jerusalem, den Herrn, BWV 119, Bach employs French-overture form for a festive occasion—the annual Leipzig council-election service—as he did again only two months later in Cantata No. 194, written for an organ and church consecration. In both cases, the majestic frame is purely instrumental (save for a brief vocal coda in No. 194), so that the voices participate in the fast middle section only. Bach’s setting of the psalm text (Psalm 147: 12–14) in Cantata No. 119 is essentially monothematic as in a fugue, but the giga-like theme is invariably answered at the octave rather than the 5th. Finally, in Sie werden euch in den Bann tun, BWV 44, the prophecy of the persecution of Jesus’s disciples from John 16: 2 is sung in a duet–chorus sequence. A tenor–bass duet is preceded by an oboe duet in which the fugal entries are prolonged in strict canon for 10 bars. The same canon is then taken over by the voices, perhaps as a symbol of the imitatio Christi. ‘As I have been persecuted,’ Jesus seems to be saying, ‘so too will you be.’ A dramatic chorus follows, in which each phrase of the text is set to different material in the style of a motet. For all Bach’s espousal of the ‘modern’, operatic type of cantata during the Weimar years, the Lutheran chorale continued to play a significant part, and in the later Weimar cantatas the chorale-finale became the norm. This was still the case in Leipzig, where dictum chorus and chorale provided an ‘objective’ frame within which ‘subjective’ arias and recitatives could alternate. Already in Weimar the chorale-finales took the form of plain, four-part chorale settings, albeit amply decorated with passing notes in Bach’s customary manner. The instrumental descants with which he sometimes adorned them (BWV 12, 172, 31, 161, 70a) still occur in two Leipzig cantatas from Cycle I, Nos. 136 and 95. More often in the first cantatas of this cycle, the plain chorale is embedded in an elaborate instrumental texture, which has the effect of highlighting the hymn and placing it in a particular affective light (BWV 75, 76, 24, 167, 147, 186, 105, 46, 138, 109). Bach had already adopted this procedure in the finale of the audition cantata No. 22; and, going further back, a clear analogy lies in some of the larger organ chorales of the ‘Eighteen’. Particularly striking is the string tremolo in the chorale-finale of Cantata No. 105, illustrating God’s gradual stilling of the conscience that torments the sinner; and the obbligato recorder duets of the equivalent movement in No. 46, representing the calming of God’s wrath through the suffering and death of Jesus Christ. The chorale-finales of Nos. 138 and 109 are particularly extensive and broadly

126 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s conceived. Both feature a substantial, intricately wrought instrumental ritornello and a significant degree of polyphonic movement in the accompanying vocal parts. The later cantatas of Cycle I often contain two or three different chorales (Cantata No. 95 even contains four); but the earlier ones are more often unified by including several statements of one and the same chorale. In the big two-part cantatas at the start of the cycle (BWV 75, 76, 147, 186), the same chorale arrangement acts as the finale of both parts. Other cantatas include two (Nos. 23, 48, and 190) or even three (No. 138) different arrangements of the same chorale. The earlier of the two chorale statements in Cantatas Nos. 23 and 48 is wordless and instrumental—a revival of a Weimar procedure of Bach’s that allowed him, independently of the librettist, to rely on the listener’s associations with the chorale melody. In Leipzig, however, the wordless chorale is no longer built into arias or duets, as it was in Weimar, but rather into recitative (Nos. 23 and 70), a sinfonia (No. 75), and choruses (Nos. 77, 25, and 48). In the last-mentioned cantatas—the first two performed on successive Sundays (22 and 29 August) and the third a few weeks later (3 October)—Bach introduced a complete instrumental chorale into a biblical-text chorus. For the congregation, the chorale would presumably act as an unspoken commentary on the biblical words. The celebrated opening chorus of Du sollt Gott, deinen Herren, lieben, BWV 77, quotes Jesus’s words about loving God and one’s neighbour (Luke 10: 27). Bach simultaneously employs the chorale melody Dies sind die heilgen zehn Gebot (‘These are the ten holy commandments’) to indicate that the entire Law of the Old Testament is contained in Jesus’s love commandment.29 The chorale melody is given in canon by augmentation—its strictness representing the Law—at the lower 5th between trumpet and continuo. The voices and strings have a freely fugal texture based on a variant of the first chorale line in shorter note-values. In the equivalent movement of Ich elender Mensch, wer wird mich erlo¨sen, BWV 48, the wordless chorale melody, which returns (with words) in the finale, is again treated in strict canon, here at the lower 4th between trumpet and unison oboes. The intensely expressive ritornello and vocal melodies illustrate the agonizing words of Paul from Romans 7: 24: ‘Wretched man that I am, who shall redeem me from the body of this death?’ Redemption is to be sought in Jesus Christ, to whom the chorale appeals in the words ‘Lord Jesus Christ, I cry to you’. He is the ultimate figure of divine authority, hence the canonic treatment of the chorale. In the opening chorus of Es ist nichts Gesundes an meinem Leibe, BWV 25, the words of Psalm 38: 3, ‘There is nothing sound in my body . . . nor any peace in my bones’, are sung to a double fugue, whose two subjects are first treated independently in canonic stretto and then combined. At the same time a wind choir, made up of three unison recorders, cornett, and three trombones, delivers the chorale Ach Herr, 29 Other layers of symbolism have also been unearthed by the numerous commentators on this move¨ ber Bach und anderes: ment. See, in particular, Georg von Dadelsen, ‘Bachs Kantate 77’, repr. in Dadelsen, U Aufsa¨tze und Vortra¨ge 1957–1982 (Laaber, 1983), pp. 185–93; and Gerhard Herz, ‘Thoughts on the First Movement of J. S. Bach’s Cantata No. 77’, programme notes at annual meeting of American Musicological Society, 1974; repr. in Herz, Essays on J. S. Bach (Ann Arbor, 1985), pp. 205–17.

s a c r e d c a n t a t a s: l e i pzi g c y c l e i

127

mich armen Su¨nder, straf nicht in deinem Zorn (‘Ah Lord, do not punish me, a poor sinner, in your anger’)—the bodily illness described by the Psalmist is the result of sin, which brings down God’s wrath, hence the plea for mercy. Among the first movements of the Cycle I cantatas, there are two cases in which a biblical text is combined with a sung chorale rather than a wordless, instrumental one, namely Nos. 60 and 190. In O Ewigkeit, du Donnerwort, BWV 60, the movement concerned is not a chorus but a dialogue, in which the alto and tenor are cast as the allegorical figures ‘Furcht’ (Fear) and ‘Hoffnung’ (Hope) respectively. The two characters represent opposite reactions to the prospect of death. Fear (doubled by horn) sings a plain chorale cantus firmus—the first verse of Johann Rist’s hymn O Ewigkeit, du Donnerwort—while simultaneously Hope sings Jacob’s words from Genesis 49:18, ‘Lord, I await your salvation’ (‘Herr, ich warte dein Heil’) in a wide-ranging arioso. Each protagonist is supported by one of the two main instrumental themes from the framing ritornello: Fear by the opening string tremolo figure and Hope by the gentler legato figures of the oboe d’amore duet. The result is one of Bach’s most imaginative conceptions, vivid in its portrayal of conflicting states of the soul. If this movement is essentially a chorale arrangement to which a biblical text is added, the opening movement of the New Year cantata Singet dem Herrn ein neues Lied, BWV 190, is primarily a biblical-text chorus within which chorale quotations are inserted. A central fugue to words from Psalm 150 is framed by the first two lines of Luther’s German Te Deum, sung as a chorale cantus firmus in a very powerful choral unison. These Te Deum extracts are in turn framed by concertante psalm sections. As a whole, the movement constitutes one of Bach’s most thrilling and effervescent songs of praise, marred only by the incomplete state in which it is preserved.30 In the second movement of this New Year cantata, the same two chorale lines recur twice, now in a plain four-part vocal setting and troped by recitative, which explains the grounds for the praise and thanksgiving to which the Te Deum lines refer. Such a combination of chorale and recitative occurs in three first movements whose primary text is drawn from a chorale rather than the Bible, namely those of Cantatas Nos. 138, 95, and 73. Of all the Cycle I cantatas, No. 138 approaches nearest to the chorale cantatas of Cycle II, for three of its six movements, nos. 1, 2, and 6, are based on the first three verses of one and the same chorale, namely Warum betru¨bst du dich, mein Herz. The first two movements share the same structure: the chorale asks, ‘Why are you distressed?’, whereupon troping recitatives give the outpourings of the anguished soul, whereas the chorale encourages trust in God, who ‘stands by you in distress’. A week later, Bach performed Christus, der ist mein Leben, BWV 95, whose opening chorus contains two different chorales plus a troping recitative. Here, all constituents share the same outlook, namely positive acceptance of death, as affirmed in the Nunc

In the ﬁrst two movements only the voice and violin parts survive.

128 sacred and s ecular: the vocal works Dimittis paraphrase of the second chorale, ‘With peace and joy I go to that place’ (‘Mit Fried und Freud ich fahr dahin’). Like Nos. 138 and 95, Cantata No. 73, Herr, wie du willt, so schicks mit mir, incorporates troping recitatives in the chorale arrangement of its first movement, providing a madrigalian commentary on the chorale text. Of the three chorale-choruses under discussion, it perhaps comes closest in design (leaving aside the tropes) to the first movements of the Cycle II chorale cantatas. The four-part vocal chorale texture is embedded in an independent instrumental accompaniment, derived from the framing ritornello, which, however, includes quotations of the first two chorale lines in the horn and first violin parts. Indeed, the first four notes of the chorale, as stated here—diminished and staccato—act as a musical motto throughout, constantly reminding the listener of the key words with which they are associated, ‘Lord, as you will’ (‘Herr, wie du willt’). Due to the ubiquity of this and other ritornello themes, the whole movement is tightly motivic. The final example of the conjunction of chorale and recitative in Cycle I is found not in a chorus but in a bass solo, with unison strings and continuo—the second movement of the Purification cantata Erfreute Zeit im neuen Bunde, BWV 83. Here the chorale is not a Lutheran hymn but the German Nunc Dimittis, sung to its traditional melody, the eighth psalm tone. It is sung by the bass as a long-note cantus firmus. A canonic ritornello for unison strings and continuo provides accompaniment and interludes throughout. Between the verses of the canticle a troping recitative intervenes, meditating upon Simeon’s words (Luke 2: 29–31). A solo chorale arrangement of a more orthodox kind—based on a Lutheran hymn and without troping recitative—forms the third movement of Cantata No. 95. The plain cantus firmus is sung in long notes by the soprano against an ostinato motive in the continuo and a florid obbligato melody in unison oboes d’amore. Similar solo chorale arrangements form the third movement of four successive cantatas performed in May 1724, Nos. 166, 86, 37, and 44. In each case the essential principle is: plain vocal cantus firmus, prefaced by an instrumental ritornello, which furnishes the material of the obbligato part that accompanies the chorale. Only Cantata 37 no. 3 is exceptional: a soprano–alto duet with continuo, in which the cantus firmus is delivered by one voice and freely imitated by the other. Whereas the outer portions of the Cycle I cantatas are often settings of ecclesiastical texts, whether biblical words or chorale verses, the inner portions often tend to be secular-influenced madrigalian texts, set as a succession of arias and recitatives. This allows librettist and composer to present a subjective commentary on the authoritative biblical and chorale texts, a commentary that can potentially elicit a response from the inner world of the individual member of the congregation. The aria forms that Bach employs for setting the meditative portions of these texts are derived from secular music, the most common being the ABA da capo form of contemporary opera and an ABA1 reprise form of Bach’s own invention, developed out of the concerto.31

31 Also called ‘recapitulation form’ or ‘concerto-aria form’; see Miriam K. Whaples, ‘Bach’s Recapitulation Forms’, Journal of Musicology, 14 (1996), pp. 475–513, and Daniel E. Freeman, ‘J. S. Bach’s “Concerto” Arias: A Study in the Amalgamation of Eighteenth-Century Genres’, Studi musicali, 27 (1998), 123–62.

s a c r e d c a n t a t a s : l e i pzi g c y c l e i

129

The latter arose in Weimar by assimilating features of the ritornello forms of aria and concerto-Allegro. Typically, paragraph A ends in the dominant; then, after a modulatory middle paragraph (B), A returns, now adapted so that dominant-key material is transposed to the tonic. This structure, more subtle than da capo form, is employed in all the arias of the audition cantatas, Nos. 22 and 23, and quite often thereafter, though it does not equal da capo form in frequency. Aspects of concerto style are fairly often encountered in the arias of Cycle I. A particularly notable example is the alto aria that opens Cantata No. 83, Erfreute Zeit im neuen Bunde, whose thematic material, solo violin part, and relatively large instrumental ensemble (two horns, two oboes, strings, and continuo) give it something of the character of a concerto-Allegro. This and the following movement, already mentioned, with its biblical text, liturgical chant, and canonic instrumental parts, represent one of the most extreme juxtapositions of sacred and secular styles in the Leipzig cantatas. Another prominent secular feature of the arias from Cycle I is the high incidence of dance rhythms. This would have met with the disapproval of G. E. Scheibel, who opined that ‘minuets, jigs, gavottes, passepieds, and so on are not appropriate in church because they induce idle thoughts in the listeners’.32 Both arias of the audition cantata No. 22 are cast in dance rhythms, perhaps because the Co¨then serenatas were then still fresh in Bach’s mind. The C minor alto aria with oboe obbligato (no. 2) is a giga-pastorale in compound time, a type that recurs in the penultimate movement of Cantatas Nos. 40 and 64.33 The 12/8 pastorale proper occurs for the first time in the early months of 1724, in Cantatas 154 no. 4 and 104 no. 5. In both cases, Bach invents a melody of sublime simplicity, in keeping with the pastoral atmosphere. The second dance aria in the audition cantata, a tenor aria in B♭ (no. 4), has been described as a passepied-menuet, a type that recurs in Cantatas 48 no. 4, 89 no. 5, and 190 no. 5. In the first of these examples, the ritornello forms a complete binary dance of 16 (8 + 8) bars. The minuet proper occurs in two different forms: an older, presumably slower version in 3/4 (BWV 77 no. 5 and 138 no. 4) and a lighter, more modern, perhaps quicker version in 3/8 (BWV 25 no. 5 and 65 no. 6). All these movements have simple, memorable themes and a regular phrase structure in accordance with their dance rhythms. The minuet-aria from Cantata No. 138 even falls into a rondeau structure such as occurs frequently in French dance music. Several minuet-arias suggest a tempo slow enough to resemble sarabandes (BWV 109 no. 5 and 44 no. 3). The sarabande proper occurs in the very first aria that Bach performed after taking up his Leipzig post, Cantata 75 no. 3. A dotted-rhythm variant in ‘tempo di sarabanda’ is found in Cantatas 69a no. 5 and 154 no. 1. Other dance rhythms occur occasionally among the Cycle I arias: the gavotte (BWV 64 no. 5), the polonaise (BWV 190 no. 3), and the forlane or loure (BWV 40 no. 4 and 153 no. 3). In Cantata No. 64 no. 5, the

Scheibel, Zufa¨llige Gedancken, trans. by Joyce Irwin, p. 241. The identiﬁcation of dance-types here is throughout indebted to Finke-Hecklinger, Tanzcharaktere in J. S. Bachs Vokalmusik. 33

130

sacred and secular: the vocal works

secular connotations of the dance rhythm are viewed negatively: the gavotte clearly represents ‘the world’ (‘die Welt’), which is rejected in the text. One of the most remarkable features of Bach’s arias is the manner in which he creates an arresting musical image of whatever is described in the text, whether it be a pictorial scene or a particular state of being. Contemporary opera composers attempted this, of course; but Bach, using operatic means, carried it out with consummate skill and— when dealing with the inner life—with exceptional penetration and insight. One of the finest tone-pictures among the arias of Cycle I is Bach’s musical representation of the calming of the storm (Matthew 8: 23–7) in Jesus schla¨ft, was soll ich hoffen, BWV 81. All three arias from this work are thoroughly operatic in character: the opening portrayal of the sleeping Jesus, with its legato ‘sleep’ motive in the violins, doubled by recorders at the upper octave; the storm aria, no. 3, with its ‘foaming, raging waves’, depicted in the rapid demisemiquaver motion of the first violin part; and the calming of the storm itself, no. 5, in which the bass voice represents the vox Christi, despite the non-biblical, madrigalian text. In this movement, the unisono strings represent the fury of the sea; and the oboe d’amore duet, Jesus’s calming influence over it. The graphic quality of Bach’s tone-picture is in no way diminished by the allegorical interpretation of these events that is clearly spelt out in the libretto.34 Bach’s depiction of states of the soul and the transformation from one state to another is illustrated with exceptional clarity in the two arias from Herr, gehe nicht ins Gericht, BWV 105. The soprano aria with obbligato oboe and strings, no. 3, is to be played senza basso continuo, which has a symbolic significance: the absence of a firm musical basis to the texture signifies the sinful soul’s lack of a firm basis in the divinity. Violin tremolo depicts the ‘trembling and wavering thoughts of sinners’; and eloquent oboe and soprano lines, the frightened conscience of the soul separated from God. The transformation takes place in the accompanied recitative no. 4, after which the tenor, in an aria with horn, strings, and continuo, no. 5, sings of the soul’s determination to make a friend of Jesus, hence the lively continuo part and the firm, confident tones of the horn. The violin tremolo of no. 3 is here replaced by legato demisemiquavers, which embroider the horn part and perhaps represent the exuberance of the soul that has found new meaning in life. The recitatives of Cycle I, like those of the Weimar cantatas, often juxtapose or alternate secco and arioso. Such an alternation is eminently suited to the task of clarifying in musical terms the opposition between different states of being presented in the text. In the dialogue recitative, no. 4, from O Ewigkeit, du Donnerwort, BWV 60, for example, the fearful reflections on death of the allegorical figure ‘Furcht’ (Fear), sung by the alto in secco, are three times interrupted by the bass, the vox Christi, who, in the most eloquent arioso, announces, ‘Blessed are the dead who die in the Lord’ (‘Selig sind die Toten, die in dem Herren sterben’). In several cases—Cantatas 147 no. 8, 105 no. 4, 46 no. 2, and 40 no. 5—a recitative takes the form of a motivic accompagnato, anticipating Bach’s highly fruitful use of this type in the Passions.

It is outlined by Du¨rr, The Cantatas of J. S. Bach, pp. 215–17.

sacred cantatas: leipzig cycle i

131

Among the finest of the Cycle I cantatas are those in which a specific sacred theme, established in the text, is associated with an appropriate musical correlative, which is then sustained, in one form or another, throughout the whole length of the composition. Those that immediately come to mind are, in chronological order, Nos. 23, 105, 46, 60, 73, and 67. The audition cantata No. 23 gives, from beginning to end, a profound musical portrayal of the afflicted, yet penitent and suppliant soul of man. The opening duet already presents a deeply moving tone-picture of human affliction and need for the divine. The blind man’s cry for mercy in the recitative no. 2, drawn from the Sunday Gospel (Luke 18: 31–43), is echoed by an instrumental rendition of the German Agnus Dei, ‘Christ, you Lamb of God, who bears the sin of the world, have mercy upon us!’ The secular-style chorus no. 3 continues the theme of sight: since ‘All eyes wait upon You, Lord’, ‘Do not leave them forever in darkness’— darkness being not merely lack of sight but wrongdoing and tribulation. Finally, the German Agnus Dei returns in the incomparable concluding chorale, which assigns a universal application to the individual’s prayer for mercy. Herr, gehe nicht ins Gericht, BWV 105, presents two opposing inner states of being: fear of divine judgement and reconciliation with God. In the opening psalm chorus (Psalm 143: 2), the mortal prays that he or she shall not be judged, since no living person can be justified. The chromatic, canonic writing no doubt represents the consciousness of sin; and the throbbing bass, the fear of judgement. This bass is not only recalled in the viola bassett of the soprano aria no. 3 but also intensified in the violins’ tremolo—a reference to the ‘trembling thoughts of sinners’ in the text. The bass accompagnato and tenor aria, nos. 4–5, point clearly to the solution, namely Jesus’s redemption of sinners, which allows the tormented conscience to be stilled in the chorale-finale, a gradual process heard in the progressive slowing-down of the string tremolo. In the sister-cantata Schauet doch und sehet, BWV 46, performed only a week later, divine wrath vents its fury on unrepentant sinners, hence the destruction of Jerusalem—prophesied by Jesus in Luke 19: 41–8—which provokes the great lament of the opening chorus. The same theme is explored further in the tenor’s accompagnato, no. 2, hence its opening words ‘Lament, then, you ruined city of God’. The ostinato motive here for two recorders, however, presumably represents ‘Jesus’s tears’—the first hint of a counter to the divine wrath. Nevertheless, the bass solo with obbligato trumpet and strings, no. 3, is a storm aria in which God’s wrath breaks out on unrepentant sinners. The sharpest possible contrast is then heard in the alto aria no. 5, which represents Jesus’s gentleness and protection of the devout, whose innocence is symbolized by the bassett for two unison oboes da caccia. The florid, legato semiquaver figures of the two recorders in this movement refer back to those of the accompagnato no. 2 and forward to those of the concluding chorale (all three movements are in the key of G minor), where they are associated with Jesus’s calming of God’s wrath through the redeeming efficacy of his Passion. O Ewigkeit, du Donnerwort, BWV 60, is concerned with two conflicting attitudes towards death, hence its conception as a dialogue between the allegorical characters

132 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s Fear (alto) and Hope (tenor). The opening chorale-arioso has already been described. In the dialogue-aria no. 3 the two characters are still at loggerheads, a situation reflected in the contrast between the dotted rhythms of the oboe d’amore (Fear) and the decorative legato semiquavers of the solo violin (Hope). In the dialoguerecitative no. 4 Hope is vanquished and Fear has a new dialogue partner, Christ himself (bass), who, in a deeply moving arioso, announces, ‘Blessed are the dead who die in the Lord’. Thereupon Fear is able to call: ‘Then appear again, O Hope!’ In the concluding chorale, the rising 4th that opened the first movement becomes an augmented 4th, symbolizing the mystical translation of the soul into new life. To extraordinary chromatic harmony, the soul is able to say, ‘My great woe remains down below’ (‘Mein großer Jammer bleibt danieden’). Herr, wie du willt, so schicks mit mir, BWV 73, deals with the theme of God’s will, as opposed to the will of humankind. Acceptance of God’s will, in words drawn from the Sunday Gospel, ‘Lord, if you will’ or ‘as you will’ (Matthew 8: 2), is treated as a verbal and musical motto, both in the opening chorale-chorus and in the sublime bass aria no. 4. In each case, chorale or biblical words are troped by madrigalian texts outlining the obstacles to acceptance of the divine will, then gradually removing them. The bass recitative no. 3 is still concerned with human will, which ‘remains perverse’. By the chorale-finale no. 5, however, all that matters is ‘the will of the Father’. Finally, we turn to Halt im Geda¨chtnis Jesum Christ, BWV 67, perhaps the greatest of Bach’s Easter cantatas (though written for the First Sunday after Easter, not for Easter Sunday itself). The magnificent opening chorus with biblical text (2 Timothy 2: 8) announces the Resurrection, but the tenor aria no. 2 nonetheless shows the soul in a divided state: on the one hand, ‘My Jesus is risen’, but on the other, the soul feels frightened and at war (compare the motives of bb. 1 and 2). The inner turmoil of the soul is expressed in recitative (Nos. 3 and 5), though it is also possible to celebrate Christ’s triumph in the chorale no. 4. A direct confrontation between the two themes takes place in the bass aria with chorus no. 6: the bass as vox Christi, accompanied by a ‘peace’ motive for flute and two oboes d’amore, announces: ‘Peace be with you!’ (John 20: 19), to which Christians respond in the madrigalian words of the upper voices, accompanied by the ‘strife’ motive of the opening string ritornello. The choir in this movement gradually comes to accept Jesus’s help in dealing with the war within, which enables Christ to be addressed in the chorale-finale as the Prince of Peace.

Magniﬁcat and Passion Title, occasion

Earliest source/s

Scribe, date

Magniﬁcat in E♭, BWV 243a, Visitation, Christmas St John Passion, BWV 245, Version I, Good Friday

Berlin, P 38

Autograph, for 2 July and 25 Dec. 1723

Berlin, P 28, St 111 Part-autograph, for 7 Apr. 1724

magnificat and passion

133

The original E♭ version of Bach’s Magnificat may well have received its first performance at the Feast of the Visitation on 2 July 1723, only five weeks after he took up his Leipzig post.35 It was revived later that year—at Vespers on Christmas Day. In Leipzig, the Magnificat—the song of Mary from Luke 1: 46–55—was sung in German on ordinary Sundays, but a concerted setting in Latin was performed on festive occasions,36 including the three High Feasts, Christmas, Easter, and Whit, and the three Marian Feasts, the Purification, Annunciation, and Visitation. Accordingly, in setting the text Bach employed his full festive orchestra of three trumpets and drums, two oboes, strings, and continuo. In addition, he calls for a five-part choir (SSATB), otherwise virtually unknown in his Leipzig church music—the only other examples are the motet Jesu, meine Freude, BWV 227, and the Dresden Missa of 1733 plus associated works (BWV 191 and 232). Without exception these works lie outside the normal routine of Bach’s sacred vocal works. In accordance with an old German custom (not peculiar to Leipzig) the text of the Magnificat is, in Bach’s Christmas version, troped with laudes in the form of German or Latin songs of praise. These four movements, A, B, C, and D, were composed after the entire Latin text of the Magnificat had been set and are appended at the end of the autograph manuscript.37 It is clear, then, that they are not an integral part of Bach’s conception—they would be performed at Christmas, but could be omitted if and when the Magnificat was performed on other occasions. The laudes tell the story of the angels and shepherds as recorded in Luke 2: 8–20. The angel comes down to the shepherds ‘from heaven on high’, hence A, the four-part (SATB) cantus firmus arrangement of the chorale Vom Himmel hoch, verse 1 (Martin Luther, 1535). The angel’s message is recorded in B, the motet-style ‘Freut euch und jubilieret’ for four voices (SSAT) and continuo, with its different imitative point for each portion of text (bb. 1, 20, and 30) and its motivic link with the Magnificat itself (‘Et exsultavit’). The multitude of the heavenly host appear and sing C, ‘Gloria in excelsis . . .’, the angelic hymn from Luke 2: 14, here set for five voices (SSATB) with doubling instruments, violin I descant, and continuo. Again, each portion of text receives a different setting in motet style: ‘Gloria in excelsis’ (b. 1), ‘Et in terra pax’ (b. 7), and ‘Bona voluntas’ (b. 11). Finally D, ‘Virga Jesse floruit’ is conceived as a lullaby, reflecting the shepherds’ arrival in Bethlehem, where they find the child lying in a manger. This is the only one of the four laudes set in the form and style of an aria: it is a florid, imitative duet for soprano, bass, and continuo.38

35 See Andreas Glo¨ckner, ‘Bachs Es-Dur Magnificat BWV 243a—eine genuine Weihnachtsmusik?’, BJ 89 (2003), pp. 37–45. 36 According to the chronicler Christoph Ernst Sicul; see Glo¨ckner, ‘Bachs Es-Dur Magnificat’, p. 37 n. 1. 37 See Alfred Du¨rr, Krit. Bericht, NBA II/3 (1955), p. 37. 38 It survives in a fragmentary state but can be reconstructed from a later version, the fifth movement of the Christmas cantata BWV 110 (1725).

134

sacred and secular: the vocal works

As in contemporary Italian concerted settings of the Magniﬁcat, the verses of the canticle are mostly set as separate movements, roughly alternating between solos and choruses. Bach’s overall structure is as follows: 1 2 3 4 5 6 7 8 9 10 11 12

Chor. Solo Solo Chor. Solo Duet Chor. Solo Solo Chor. Chor. Chor.

Magniﬁcat Et exsultavit Quia respexit Omnes generationes Quia fecit Et misericordia Fecit potentiam Deposuit Esurientes Suscepit Israel Sicut locutus est Gloria

E♭ E♭ c–g g B♭ f A♭–E♭ g F c E♭ B♭–E♭

tutti S II, strgs., bc S I, ob., bc tutti B, bc AT, strgs., bc tutti T, unis. strgs., bc A, 2 recs., bc SSA, tr. I, bassett SSATB, bc tutti

Thus the great tutti choruses form pillars that surround pairs of solo movements. Only at the end is this conception modified by three consecutive choruses, nos. 10–12, which, however, show a progressive build-up in required forces. The work opens and closes with pairs of movements in the tonic E♭; and the music of the first chorus returns in the last, as a pointed illustration of the words ‘sicut erat in principio’ (‘as it was in the beginning’), a musical witticism that goes back at least as far as Monteverdi.39 In the interests of compactness the solo movements are in general shorter than Bach’s cantata arias, and da capo form is nowhere to be found. The first pair of solos, nos. 2 and 3, are both sung by soprano voice, reflecting the nature of the canticle as Mary’s song of praise. The key words in ‘Et exsultavit’ are ‘My spirit hath rejoiced’, hence the major mode and the joy motives in the continuo (bb. 2, 6, 8, etc.) and violin I (b. 12 etc.). ‘Quia respexit’, on the other hand, is a slow, expressive movement (later marked ‘Adagio’), whose minor mode reflects ‘the lowliness of His handmaiden’. An even greater contrast occurs in the second pair of solos, nos. 5 and 6. The key words of the two movements represent opposite attributes of the divinity, ‘might’ (no. 5) and ‘mercy’ (no. 6). ‘He that is mighty’ is portrayed in ‘Quia fecit’ by the bass voice, accompanied by a continuo part that makes constant use of its opening ritornello, and in a ‘strong’ key, the overall dominant B♭. In ‘Et misericordia’, on the other hand, Mary’s words ‘And His mercy is on them that fear Him’ elicit a setting in F minor, full of pathos and built over a partly chromatic, quasi-ostinato bass. The divine quality of mercy is expressed in the beatific parallel 3rds of the violins in the ritornello, later turned to 6ths in the voices. The last pair of solos, nos. 8 and 9, exhibit an analogous contrast—between God’s ‘putting down the mighty from their seat’ and His ‘filling the hungry with good things’. In ‘Deposuit potentes’, a tenor solo in G minor with 39 As noted by Robert L. Marshall, ‘On the Origin of the Magnificat’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 3–17.

magnificat and passion

135

obbligato for unison upper strings, the main figure of both ritornello and vocal solo gives a vivid representation of the toppling of the mighty. Significantly, ‘Esurientes’ follows in an unrelated key, F major. This movement possesses its own internal contrast, based on that of the canticle: God not only ‘filled the hungry with good things’, but He also ‘sent the rich empty away’, hence the two contrasting themes of the ritornello, both taken up by the voice—a charming theme for two recorders in 3rds and 6ths for ‘the hungry’, and an imitative theme for the same pair of instruments for ‘the rich’ (b. 5). The magnificent opening chorus, ‘My soul doth magnify the Lord’, consists of a single vocal section framed by an instrumental ritornello. This substantial binary ritornello (30 bars, 12 + 18), which modulates to the dominant and back, is exceedingly rich in themes. These are subsequently taken over by the vocal ensemble, for the three vocal periods (bb. 31, 45, and 61) are each built into portions of the returning ritornello music. The second great ‘pillar’ chorus, no. 4, follows without a break after ‘Quia respexit’, no. 3, whose last sentence, ‘beatam me dicent . . .’, is completed by the words of the chorus, ‘omnes generationes’ (‘all generations shall call me blessed’). It is a motet-style, imitative chorus with largely colla parte instruments. Though not fugal, it is designed throughout on the stretto principle in relation to its repeated-note subject. The stretto descends through all five voices at the beginning (b. 2) and ascends through them at the end (b. 21). The middle portion of the movement contains two eight-entry strettos by rising tones (bb. 5 and 15) and a ten-entry stretto through the circle of 5ths (b. 10). The third ‘pillar’ chorus, ‘Fecit potentiam’, no. 7—‘He hath shewed strength with His arm’—is a powerful fugue whose subject entries, however, are characteristically masked by homophonic material in the form of an incisively rhythmic setting of ‘fecit potentiam’ in block chords, imitated by inversion between antiphonal groups of voices and instruments. After a climactic subject entry in trumpet I (b. 21), the fugue is dispersed (‘dispersit’) in favour of a highly graphic, dramatic portrayal of the words ‘He hath scattered the proud in the imagination of their hearts’. The three concluding choruses, nos. 10–12, lack intervening solos. The first of them, ‘Suscepit Israel’, no. 10, is in effect a chorale arrangement with instrumental cantus firmus, a type that Bach had employed in Weimar and revived during his first year in Leipzig. The ninth psalm tone, the chant traditionally associated with the Magnificat, is played in long notes by first trumpet. Meanwhile, a trio of upper voices (SSA) sing the words in an imitative texture, accompanied by a bassett of unison upper strings. It seems likely that the bassett refers to the divine quality of mercy (‘misericordia’) alluded to in the text. A different imitative point is used for each of the two sections: a scale theme, answered by inversion, for ‘He hath holpen his servant Israel’ and a contrasting, disjunct theme for ‘In remembrance of his mercy’. The following movement, ‘Sicut locutus est’, no. 11, completes the text of the Magnificat in an alla breve fugue for five-part choir and continuo. It is

136 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s constructed on the permutation principle, which Bach had employed since his earliest years as a composer.40 Only two themes stand out, however: the subject itself, ‘Sicut locutus est’ (‘As he promised to our forefathers’), and the countersubject, ‘Abraham et semini eius’ (‘Abraham and his seed for ever’). These two subjects are combined at every entry. In the 16-bar conclusion (b. 37), it is the bold, rhythmic ‘Abraham’ theme that is treated in sequential stretto in the outer parts, whereas the subject itself receives only one more entry (bass, b. 45). The setting of the doxology, ‘Gloria Patri’ (no. 12), is bipartite: a majestic common-time opening, in which florid triplet rhythms for imitative voices alternate with tutti block chords; and a lighter, quicker conclusion in triple time. Bach would return to this bipartite, AB conception several times later on: at the following Christmas (25 December 1724) in the six-part Sanctus (BWV 232111) that would eventually be incorporated in the B minor Mass, and a few months after that in the finale of the pastoral cantata (BWV 249a, 23 February 1725) and its sacred parody (BWV 249, 1 April 1725), which later became the Easter Oratorio. The first performance of the St John Passion took place on Good Friday, 7 April 1724—that is, during Bach’s first year in Leipzig, when he was busy composing and performing the cantatas of Cycle I. The six-week Lenten period that preceded Good Friday gave him a break from cantata production (concerted music in the Leipzig churches was forbidden during Lent), which enabled him to devote his time and energy to the huge task of composing the new Passion. Bach adhered to the genre of oratorio-Passion developed by Lutheran composers ever since the St John Passion of 1641 by the Hamburg cantor Thomas Selle. Accordingly, the Gospel account of the Passion, John 18–19, is delivered in three forms: the narration of the Evangelist, sung in tenor recitative; dialogue between Jesus (bass) and other characters (Peter, Pilate, etc.), also sung in recitative; and turba (crowd) choruses, representing the disciples, priests, soldiers, onlookers, etc. A meditative commentary is provided by Lutheran chorales and by non-biblical reflective, devotional texts, set in the form of ariosos and arias. The Passion as a whole is framed by an exordium and conclusio (introduction and conclusion), set as choruses, which, like the arias, are based on freely composed texts. It is not known who was responsible for the freely composed texts, but whoever it was made substantial use of material by other poets, B. H. Brockes, C. H. Postel, Christian Weise, and Salomo Franck. Typically, a portion of John’s Gospel that represents a particular stage in the narrative is sung by the Evangelist, in dialogue, and in turbae; it is then followed by a reflective commentary on the events described in the form of ariosos, arias, and chorales. Sequences of this kind may be regarded as scenes or sub-scenes by analogy with opera, as shown in the following table (the commentary is shown in italics; sub-scenes are marked a–d):

See Vol. I of the present study, pp. 97–117.

magnifica t and passion

137

Part I No. 1: Exordium: chor. Act I: Arrest and Interrogation Sc. 1: the arrest of Jesus a) Armed men come in search of Jesus Nos. 2–3: recit.—turba a—recit.—turba a1—recit.—chorale A b) Peter draws his sword and strikes Malchus Nos. 4–5: recit.—chorale B c) Jesus is arrested Nos. 6–7: recit.—aria d) Peter follows Jesus Nos. 8–9: recit.—aria Sc. 2: Peter’s denial and Jesus’s interrogation a) First denial and interrogation Nos. 10–11: recit.—chorale C b) Second and third denials; Peter’s remorse Nos. 12–14: recit.—turba b—recit.—aria—chorale D Part II No. 15: Exordium: chorale E Act II: Run-up to the Crucifixion Sc. 1: Jesus brought before Pilate a) First part of Pilate’s interrogation Nos. 16–17: recit.—turba c—recit.—turba c1—recit.—chorale A b) Second part of Pilate’s interrogation Nos. 18–20: recit.—turba a2—recit.—arioso—aria Sc. 2: Call for Jesus’s Crucifixion a) Jesus mocked; first call for crucifixion Nos. 21–2: recit.—turba d—recit.—turba e—recit.—turba f—recit.—chorale F b) second call for crucifixion; Jesus carries cross to Golgotha Nos. 23–4: recit.—turba f—recit.—turba e—recit.—turba a3—recit.—aria + chor. Act III: The Crucifixion Sc. 1: Jesus on the Cross a) Jesus crucified; Pilate’s inscription Nos. 25–6: recit.—turba d—recit.—chorale G b) Soldiers share Jesus’s clothes; Jesus attends to his Mother and a disciple Nos. 27–8: recit.—turba g—recit.—chorale D c) Jesus is thirsty; his last words from the Cross Nos. 29–30: recit.—aria d) Jesus’s death on the Cross Nos. 31–2: recit.—aria + chorale D

138 sacred and s ecular: the vocal works Sc. 2: Attendant circumstances a) Supernatural events Nos. 33–5: recit.—arioso—aria b) Jesus pierced by a spear; Jesus buried by Joseph and Nicodemus Nos. 36–8: recit.—chorale E—recit. Nos. 39–40: Conclusio: chor.—chorale H The letters attached to the chorales and turbae indicate the recurrence of their music—in varied form and with different words—which acts as a powerful means of integration. The most prominent chorale is Jesu Leiden, Pein und Tod (Paul Stockmann, 1633), which occurs three times in the course of the Passion (see chorale D). Its prominence results not just from repeated returns but from strategic placing—at the end of Part I and at two key points in the later stages of Part II. In all three cases it is concerned with Jesus’s loving relationship with an individual—with Peter (No. 14), with his mother and a disciple (No. 28), and with the Christian (No. 32). Two verses from the chorale Christus, der uns selig macht (Michael Weiße, 1531) are placed at the beginning and end of Part II (see chorale E), providing a firm theological context for the events that are recorded between them: the evil perpetrated against Jesus, who has not only ‘committed no evil’ but who ‘makes us blessed’; and also the need to consider Jesus’s death and the explanation for it, namely human sinfulness. Recurrent turbae play a significant role in the scenes that immediately precede the Crucifixion. The first call for Jesus’s crucifixion (Nos. 21–2) forms the context for three powerful turbae, which then recur in reverse order (in variant versions and with largely different words) during the second call for crucifixion (Nos. 23–4) and immediately after the Crucifixion itself (No. 25). In every case there is a clear link between the two occurrences of the same turba and a clear explanation for Bach’s use of the same music. The chorus in which Jesus is mocked by the Roman soldiers as ‘King of the Jews’ (turba d, No. 21) recurs later when the Jews object to Pilate’s use of this title for Jesus in the inscription attached to the cross (No. 25). The first call for Jesus’s crucifixion (No. 21) is sung in a frenzied chorus (turba e) with violently dissonant suspensions against obsessive dactyl figures to the words ‘Kreuzige, kreuzige!’. Later, when Jesus’s crucifixion is demanded once more (No. 23), we hear a variant of the same chorus, transposed down a semitone and with a new three-bar introduction to the words ‘Weg, weg mit dem’ (‘Away, away with him!). Finally, a rational case is presented to justify Jesus’s death, adducing Jewish law (No. 21, turba f) and, a little later, portraying Jesus as a traitor to the Roman Empire (No. 23). Essentially the same music is employed in both cases, a regular fugue being used to signify law and reason. The framing choruses of the St John Passion, the exordium and conclusio (Nos. 1 and 39), differ fundamentally from the vast majority of Cycle I cantata-choruses, being settings of freely invented rather than biblical texts. In the first Leipzig year non-biblical choruses occur only in revivals of Weimar cantatas, in parodies of Co¨then secular cantatas, and in two newly-composed cantatas, No. 23 (third movement), which Bach

magnificat and passion

139

evidently composed at Co¨then and brought to Leipzig for his audition, and No. 119 (seventh movement), composed for a festive occasion outside the church year, the annual council election service. Freely composed, madrigalian texts in choruses are indistinguishable from those of arias; hence their musical settings are often similar in form, frequently employing da capo (ABA) or reprise (ABA1) structures. Half of the non-biblical choruses mentioned earlier—almost all of Weimar or Co¨then origin—are constructed in da capo form, so it was natural that Bach should employ it in the great exordium that introduces the St John Passion. It was also natural that Bach should avoid fugue, since he tended to associate it with biblical words. The da capo form of the exordium is fused with a ritornello structure of great power. The main choral formulation is derived from the ritornello, whereas two recurring, imitative vocal themes are independent. An illuminating parallel may be drawn between the opening ritornello of the exordium and that of the great psalm chorus that opens Herr, gehe nicht ins Gericht, BWV 105, performed only about nine months earlier (25 July 1723). Not only are the two ritornellos in the same key of G minor, but both are led by a woodwind duet (doubled by violins in BWV 105) of great solemnity and pathos, involving imitation and chromaticism in long, sustained notes over repeated continuo quavers (Ex. 2a). Both ritornellos, moreover, lead to repeated calls of ‘Herr’ (‘Lord’). A more immediately relevant parallel may be drawn with the double-counterpoint phrase, with its long notes and suspensions, that forms the framework of the ‘Kreuzige’ choruses (Nos. 21d and 23d), of which the first is in the same key as the exordium (Ex. 2b). Turbae and exordium alike open with a piercing semitonal dissonance, clearly associated with the pain and horror of the Crucifixion. This element plays a dominant role in the exordium alongside the Johannine view of Christ glorified, reflected in the string figuration which is taken up by the voices on the word ‘Herrscher’ (‘Ruler [whose praise is glorious in all the lands]’). The conclusio is made up of two movements—the chorus ‘Ruht wohl’, in the key of C minor, and the chorale Herzlich lieb hab ich dich, verse 3 (Martin Schalling, 1571), in the relative major E♭. The minor–major sequence reflects the progression from thoughts of Jesus’s grave to thoughts of the afterlife. The link is established in the chorus: ‘The grave . . . opens to me the [gates of] heaven’. This chorus (No. 39) is in some respects not dissimilar to the finales of Bach’s secular cantatas. It is a simple homophonic movement in the dance rhythm of a sarabande, cast in rondeau-ritornello form. Occurring directly after the Evangelist’s account of Jesus’s burial, its ritornello features a descending arpeggio ‘burial’ figure (bb. 5 and 7), which comes into its own during the rondeau episodes (bb. 64, 68, 116, and 120). The eight arias are designed as meditations on specific events in the course of the Evangelist’s narrative. The alto aria ‘Von den Stricken’, No. 7, follows the Gospel account of Jesus’s arrest: ‘The band . . . took Jesus and bound Him’. The text, drawn from the Brockes Passion, attaches theological significance to the image of binding: ‘My Saviour is bound to unbind me from the bonds of sin’. The four-bar canon at the upper 2nd between the two obbligato oboes is presumably a symbol of binding, in which case the freer writing of the consequent phrase, with its parallel 3rds, might

140 sacred a nd s ecula r: the vocal works

Ex. 2

a) 1. St. John Passion, exordium, bb. 9b–15 (woodwind only) 2. 1st movement of Herr, gehe nicht ins Gericht, BWV 105 (viola and continuo omitted)

b) 1. St. John Passion, exordium, bb. 1–3 (woodwind only) 2. St. John Passion, ‘Kreuzige’ chor. (No. 21d), SA + oboes only (words omitted) stand for unbinding. Immediately afterwards, the Gospel account of Peter’s following Jesus is applied to the individual Christian in the soprano aria ‘Ich folge dir gleichfalls’ (No. 9): ‘I will follow you likewise with joyful steps’. The paired continuo quavers no doubt signify footsteps, while the continuous semiquavers of the unison ﬂutes clearly represent eager hurrying after the Lord. In certain cases the biblical words that prompt an aria convey such a powerful image that they are set in arioso prior to the aria that meditates upon them. Peter’s bitter weeping, for example, on recalling Jesus’s prediction of his denial is sung as an Adagio arioso (No. 12c) with chromatic lines in both tenor and continuo. The ‘remorse’ aria that follows (No. 13), a dotted-rhythm sarabande, turns the chromaticism of the arioso into the time-honoured lamento bass. The anguished tenor part, supported by dense string harmonies, portrays Peter’s impassioned soul as he tells of his extreme mental and spiritual agony over the denial of his Lord. Much later on, at the end of the Evangelist’s recitative No. 29, Jesus’s last words from the Cross, ‘Es ist vollbracht’ (‘It is accomplished’), are sung in a brief but memorable phrase of arioso. This recurs in the following ‘Molto Adagio’ aria with solo alto and obbligato viola da gamba. The arioso phrase is embellished in its ﬂorid, deeply expressive opening paragraph, but in the much abridged reprise it recurs twice in its original form, framing the gamba ritornello. The central paragraph is a Vivace song of victory—‘The hero from Judah triumphs with power’—in the relative major D, with a fanfare-like vocal theme and a concitato string accompaniment. A more extreme contrast than between this and the framing paragraphs could hardly be imagined. The Vivace, of

m a g n i f i c a t a n d p a s s io n

141

course, accords with the Johannine conception of Christ as victor, even in His Crucifixion. But the words ‘Es ist vollbracht’ are also drawn from John’s Gospel and may be viewed in a positive light, as if Jesus were saying, ‘This is the completion of my earthly ministry’. On the other hand, Bach associates Jesus’s words with a mood of the deepest sorrow, partly no doubt because they immediately precede his death, but also surely prompted by the phrases ‘Die Trauernacht’ (‘the night of mourning’) and ‘die letzte Stunde’ (‘the last hour’) from the aria text. In two cases a biblical-text arioso or accompagnato provides a cue for an exceptionally protracted meditation, comprising arioso as well as aria, both to freely invented texts borrowed from the Brockes Passion (Nos. 19–20 and 34–5). Both arioso-aria sequences occur at key points in the drama. ‘Betrachte, meine Seel’ and ‘Erwa¨ge’ (Nos. 19–20) mark the end of Pilate’s interrogation of Jesus, when Barabbas is released in his place and Jesus is scourged in preparation for his Crucifixion. ‘Mein Herz’ and ‘Zerfließe’ (Nos. 34–5) are a response to the supernatural events recorded upon Jesus’s death. The specific stimulus for the first arioso-aria sequence is the Evangelist’s graphic musical description of Jesus’s scourging (No. 18c), with its huge melisma against ‘whipping’ dotted rhythms in the continuo. In the double meditation that follows, Christians are invited to see ‘in Jesus’s agonies [their] highest good’. The bass arioso (No. 19), scored for the delicate combination of two violas d’amore, lute, and continuo, gives perfect expression to the bitter-sweet aspect of the Passion. The second arioso-aria sequence is prompted by the Evangelist’s graphic description of the Temple veil’s being ‘rent in two pieces’, a passage borrowed from Mark 15: 38. This is expanded in the Brockes-text arioso that follows, No. 34, by an account of other supernatural events, paraphrased from Matthew 27: 51–2 and illustrated by tremolo and other figures in short note-values. The arioso begins, however, with a change from G major to its subdominant minor C as the background for an incomparable musical account of the entire world suffering with Jesus, while ‘the sun clothes itself in mourning’. In both sequences, the arioso and the following aria differ in key and voice type, but the obbligato instruments of the aria are selected from those used to accompany the arioso: two violas d’amore in ‘Erwa¨ge’ (No. 20) and transverse flute plus oboe da caccia in ‘Zerfließe’ (No. 35). The two arias have so much in common that they might have been conceived as close relatives. In both cases a pair of obbligato instruments moves frequently in parallel 3rds or 6ths, while also engaging in a trio with the continuo. In both cases water imagery is treated in florid melismas: ‘Wasserwogen’ (‘flood waves’) in No. 20 (bb. 22 and 32); and ‘Fluten der Za¨hren’ (‘floods of tears’) in No. 35 (e.g. bb. 28–9 and 31–2). There are also clear links between the vocal and instrumental headmotives of the two arias (Ex. 3). The parallel writing between the two obbligato instruments and the florid themes of both movements create an elegiac tone of intense pathos. There is no real precedent for these arias in the cantatas of Cycle I. Rather, they represent a new and profound

142

sacred and secular: the vocal works

Ex.3

Er - wä - ge,

er - wä - ge

a) St. John Passion, tenor aria ‘Erwa¨ge’, instrumental incipit (viola d’amore I, II; continuo omitted) and vocal entry

Zer - flie

ße,

mein Her - ze,

b) St. John Passion, soprano aria ‘Zerfließe’, instrumental incipit (flute and oboe da caccia; continuo omitted) and vocal entry inspiration for the Passion of Good Friday 1724, not to be exceeded thereafter (even in the St Matthew Passion). Nor is there any precedent in Cycle I for the two arias with chorus, Nos. 24 and 32. In both cases the texts are modelled on arias from the Brockes Passion that take the form of dialogues between the Daughter Zion (‘Tochter Zion’) and the Faithful Soul/s (‘Glaübige Seele’). Bach treats them in responsorial fashion, as bass solos with periodic contributions from the rest of the choir (SAT, No. 24) or from the choir as a whole (SATB, No. 32). The first of the two arias with chorus, ‘Eilt’ (No. 24), was prompted by the Evangelist’s naming of the place where Jesus was to be crucified, Golgotha. The dialogue that ensues takes the dramatic form of command, question, and response: ‘Hurry! Where? To Golgotha’. The rapid scale figures of the theme convey the image of haste. The second aria with chorus, ‘Mein teurer Heiland’ (No. 32), is prompted by the brief Evangelist’s recitative that reports the death of Jesus. The aria forms the culmination of the death scene (Nos. 29–32). It is fitting, therefore, that it should be combined with a four-part chorale—the last rendering of the most prominent chorale in the whole Passion, Jesu Leiden, Pein und Tod.

Leipzig Cycle II: chorale cantatas Title, occasion

Earliest source/s

Scribe, date

O Ewigkeit, du Donnerwort, BWV 20, Trinity 1 Ach Gott, vom Himmel sieh darein, BWV 2, Trinity 2 Christ unser Herr zum Jordan kam, BWV 7, St John’s Day Ach Herr, mich armen Su¨nder, BWV 135, Trinity 3

Basle, Leipzig TS

Autograph, J. A. Kuhnau et al., for 11 June 1724 Autograph, J. A. Kuhnau et al., for 18 June 1724 J. A. Kuhnau et al., for 24 June 1724 Autograph, for 25 June 1724

Berlin, N.Mus.ms. 681 , Leipzig TS Leipzig TS Leipzig BA

leipzig cycle ii: chorale cantatas Meine Seel erhebt den Herren, BWV 10, Visitation, Trinity 4 Wer nur den lieben Gott la¨ßt walten, BWV 93, Trinity 5 Was willst du dich betru¨ben, BWV 107, Trinity 7 Wo Gott der Herr nicht bei uns ha¨lt, BWV 178, Trinity 8 Was frag ich nach der Welt, BWV 94, Trinity 9 Nimm von uns, Herr, du treuer Gott, BWV 101, Trinity 10 Herr Jesu Christ, du ho¨chstes Gut, BWV 113, Trinity 11 Allein zu dir, Herr Jesu Christ, BWV 33, Trinity 13 Jesu, der du meine Seele, BWV 78, Trinity 14 Was Gott tut, das ist wohlgetan, BWV 99, Trinity 15 Liebster Gott, wenn werd ich sterben, BWV 8, Trinity 16 Herr Gott, dich loben alle wir, BWV 130, Michaelmas Ach, lieben Christen, seid getrost, BWV 114, Trinity 17 Herr Christ, der einge Gottessohn, BWV 96, Trinity 18 Wo soll ich ﬂiehen hin, BWV 5, Trinity 19 Schmu¨cke dich, o liebe Seele, BWV 180, Trinity 20 Aus tiefer Not schrei ich zu dir, BWV 38, Trinity 21 Mache dich, mein Geist, bereit, BWV 115, Trinity 22 Wohl dem, der sich auf seinen Gott, BWV 139, Trinity 23 Ach wie ﬂu¨chtig, ach wie nichtig, BWV 26, Trinity 24 Du Friedefu¨rst, Herr Jesu Christ, BWV 116, Trinity 25 Nun komm, der Heiden Heiland, BWV 62, Advent 1 Gelobet seist du, Jesu Christ, BWV 91, Christmas Day Christum wir sollen loben schon, BWV 121, 2nd Day of Christmas Ich freue mich in dir, BWV 133, 3rd Day of Christmas Das neugeborne Kindelein, BWV 122, Sun. after Christmas Jesu, nun sei gepreiset, BWV 41, New Year

143

Washington, Leipzig TS Leipzig TS

Autograph, J. A. Kuhnau et al., for 2 July 1724 Part-autograph, for 9 July 1724

Leipzig TS

J. A. Kuhnau et al., for 23 July 1724 J. A. Kuhnau et al., for 30 July 1724 Autograph, J. A. Kuhnau et al., for 6 Aug. 1724 J. A. Kuhnau et al., for 13 Aug. 1724 Autograph, for 20 Aug. 1724

Leipzig TS, MB Berlin, P 47/1 , Leipzig TS Leipzig TS Stuttgart Priv. poss., Leipzig TS Leipzig TS Berlin, P 647, Leipzig TS Brussels Priv. poss. Priv. poss., Leipzig TS Berlin, P 179, Leipzig TS London BL, Leipzig TS Stuttgart Leipzig TS Cambridge FM Leipzig TS Berlin, P 47, Leipzig TS Paris BN, Leipzig TS Berlin, P 877, Leipzig TS Berlin, P 869, St 392 , Leipzig TS Berlin, P 867, St 390 , Leipzig TS Berlin, P 1215, St 387 , Leipzig TS Berlin, P 868, St 391 , Leipzig TS Berlin, P 874, St 394 , Leipzig TS

Autograph, J. A. Kuhnau et al., for 3 Sept. 1724 J. A. Kuhnau et al., for 10 Sept. 1724 Autograph, J. A. Kuhnau et al., for 17 Sept. 1724 C. G. Meißner et al., for 24 Sept. 1724 Autograph, for 29 Sept. 1724 Autograph, J. A. Kuhnau et al., for 1 Oct. 1724 Autograph, J. A. Kuhnau et al., for 8 Oct. 1724 Autograph, J. A. Kuhnau et al., for 15 Oct. 1724 Autograph, for 22 Oct. 1724 J. A. Kuhnau et al., for 29 Oct. 1724 Autograph, for 5 Nov. 1724 J. A. Kuhnau et al., for 12 Nov. 1724 Autograph, J. A. Kuhnau et al., for 19 Nov. 1724 Autograph, J. A. Kuhnau, for 26 Nov. 1724 Autograph, J. A. Kuhnau et al., for 3 Dec. 1724 Autograph, J. A. Kuhnau et al., for 25 Dec. 1724 Autograph, part-autograph, for 26 Dec. 1724 Autograph, part-autograph, for 27 Dec. 1724 Autograph, J. A. Kuhnau et al., for 31 Dec. 1724 Autograph, part-autograph, for 1 Jan. 1725 (cont.)

144

sacred and secular: the vocal works

(cont.) Title, occasion

Earliest source/s

Scribe, date

Liebster Immanuel, Herzog der Frommen, BWV 123, Epiphany Meinen Jesum laß ich nicht, BWV 124, Epiphany 1 Ach Gott, wie manches Herzeleid, BWV 3, Epiphany 2 Was mein Gott will, das gscheh allzeit, BWV 111, Epiphany 3 Ich hab in Gottes Herz und Sinn, BWV 92, Septuagesima Mit Fried und Freud ich fahr dahin, BWV 125, Puriﬁcation Erhalt uns, Herr, bei deinem Wort, BWV 126, Sexagesima Herr Jesu Christ, wahr’ Mensch und Gott, BWV 127, Quinquagesima Wie scho¨n leuchtet der Morgenstern, BWV 1, Annunciation

Berlin, P 875, St 395 , Leipzig TS Berlin, P 876, St 396 , Leipzig TS Priv. poss., Leipzig TS, St 157 Berlin, P 880, St 399

Autograph, J. A. Kuhnau et al., for 6 Jan. 1725 Autograph, J. A. Kuhnau et al., for 7 Jan. 1725 Autograph, J. A. Kuhnau et al., for 14 Jan. 1725 Autograph, part-autograph for 21 Jan. 1725 Autograph, J. A. Kuhnau et al., for 28 Jan. 1725 J. A. Kuhnau et al., for 2 Feb. 1725 J. A. Kuhnau et al., for 4 Feb. 1725 Autograph, J. A. Kuhnau et al., for 11 Feb. 1725 J. A. Kuhnau et al., for 25 Mar. 1725

Berlin, P 873, Leipzig TS Berlin, St 384 , Leipzig TS Leipzig TS Berlin, P 872, St 393 , Leipzig TS Leipzig TS

In Cycle I the main emphasis was placed on the musical setting of the biblical text, to which the chorale was subordinate. In Cycle II, on the other hand, the focus of attention is on the chorale and its musical setting, and biblical words are absent unless brieﬂy quoted or paraphrased within the chorale. The ‘chorale cantata’ that resulted was entirely based upon the text of a single chorale, chosen on account of its clear association with the Sunday Gospel. The outer verses were preserved unchanged in the framing movements, whereas the inner verses were mostly paraphrased so that they could be set in the ‘modern’ fashion as arias and recitatives. Certain inner verses or parts of them were, however, kept in their original form, giving the composer further scope for the use of the associated chorale melody. Some of Bach’s most eminent predecessors had worked within the tradition of the so-called ‘chorale concerto’, the forerunner of the chorale cantata, notably Buxtehude, J. P. Krieger, Zachow, and Pachelbel. Moreover, Leipzig had been an important centre of its cultivation in the seventeenth and early eighteenth centuries, as demonstrated by the compositions of previous Leipzig music directors, Johann Hermann Schein, Sebastian Knu¨pfer, Johann Schelle, and Johann Kuhnau. In Schelle’s day, around 1690, a single chorale served multiple purposes: it was sung by the congregation, expounded by the minister in the sermon, and elaborated by the music director to form a chorale cantata.41 It is not unlikely that the same happened during Bach’s tenure, in 1724–5. His cycle of chorale cantatas might well have resulted from his 41 See Friedhelm Krummacher, Bachs Zyklus der Choralkantaten: Aufgaben und Lo¨sungen (Go¨ttingen, 1995), pp. 28–9.

leipzig cycle ii: chorale cantatas

145

collaboration with a particular preacher at the Thomaskirche or the Nicolaikirche. The forty cantatas of his cycle do not fill the entire church year: they extend only from the First Sunday after Trinity, 11 June 1724, to the Feast of the Annunciation, 25 March 1725. The most likely explanation for Bach’s breaking off at this point is that his librettist was no longer available. It has been suggested that this librettist might have been Andreas Stu¨bel, former conrector of the Thomasschule, who was not only theologically trained but experienced in the art of poetry.42 Stu¨bel died on 31 January 1725 after an illness of only three days, which might explain the cycle’s coming to an abrupt end. By Bach’s time, the chorale was no longer current among Lutheran composers to the extent that it had been in the seventeenth century. What, then, might have induced him and his librettist to revive the old Leipzig tradition of the chorale concerto/ cantata as late as 1724? One factor was no doubt Bach’s deep personal interest in the chorale, amply demonstrated in the Mu¨hlhausen and Weimar cantatas, as well as in those of the first Leipzig cycle. At an early date, around 1707/9, he had composed Christ lag in Todes Banden, BWV 4, a magnificent chorale cantata per omnes versus— one that retains all verses in their original form—albeit in a relatively antiquated style.43 It is a testimony to Bach’s partiality for this powerful work and to his willingness to intermingle ‘ancient’ and ‘modern’ styles that he appears to have revived it on his first two Easter Sundays in Leipzig (9 April 1724 and 1 April 1725). Indeed, it is conceivable that the earlier revival, within Cycle I, might have prompted the idea of the chorale cantatas of Cycle II or at least acted as a spur to their composition. Another factor in the conception of the chorale-cantata cycle might have been connected with the year in which they were conceived, 1724. Two hundred years earlier, in 1524, publication had begun of the great series of Wittenberg chorales by Martin Luther, Johann Walther, and others. Among Bach’s chorale cantatas there is a strong emphasis on these Reformation hymns and their associated melodies. It is possible, then, that the chorale cantatas were conceived to mark the bicentenary of the Lutheran chorale, thereby celebrating two hundred years of Lutheran hymnody.44 In keeping with such an exalted purpose, the chorale cantatas represent the greatest cyclical project that Bach ever undertook and the only one based on a thoroughly unified textual and musical concept.45 For Bach it involved two major challenges: first, immense pressure of work. In Cycle I he had found some relief in reviving older compositions from the Weimar or Co¨then years, but in Cycle II, due to the choralecantata format, there could be no revivals (except Cantata No. 4, which essentially adhered to that format)—every cantata had to be composed afresh. Secondly, Bach

See Schulze, ‘Texte und Textdichter’, p. 116. See Vol. I, pp. 113–17. 44 See Robin A. Leaver, ‘Cantata’, in M. Boyd (ed.), Oxford Composer Companions: J. S. Bach (Oxford, 1999), p. 86. 45 As pointed out by Krummacher, Bachs Zyklus der Choralkantaten, p. 7. 43

146

sacred and secular: the vocal works

was faced with the daunting task of reinventing an old Leipzig tradition in order to suit his own time. He had to find new ways of reconciling old cantus firmus techniques and other modes of chorale treatment with relatively modern forms and styles derived from the opera and concerto. The most common overall structure that emerged contained six movements thus: chorale chor.—recit.—aria—recit.—aria—chorale; or else: chorale chor.—aria—recit.—aria—recit.—chorale. There are, however, numerous examples of an additional aria or recitative, or an intermediate chorale arrangement (preserving the wording of an inner verse). Moreover, the chorale melody is often combined in some way with an aria or recitative. Thus, although the inner verses in principle adhere to the operatic style that Neumeister introduced in his reforms of 1700, in practice they are often saturated with the words and/or music of the chorale. For the finale, Bach adheres to the custom he had established in the first Leipzig cycle (and, before that, in the late Weimar cantatas) of composing a plain four-part chorale setting. This allowed the Gospel-based message of the cantata, summed up in the last verse of the hymn, to be presented in a suitably congregational form, whether or not the congregation actually took part. Bach’s attempt to enhance the plain chorale by adding elaborate, obbligato instrumental parts, encountered in the finales of the two audition cantatas, Nos. 22 and 23, and in twelve of the earlier Cycle I compositions, is revived only once among the chorale cantatas—in the finale of Was willst du dich betru¨ben, BWV 107, where the plain chorale is sung against the background of a siciliana. This cantata is exceptional in other respects too. It is the only chorale cantata of the cycle in which the original hymn text is preserved throughout—perhaps on this occasion Bach’s librettist was for some reason unable to supply the usual hymn paraphrase. It may be for this reason that the cantata contains four arias in succession, without intervening recitatives—a phenomenon not encountered since the last three Weimar cantatas, Nos. 70 a, 186 a, and 147 a. The opening movement of the chorale cantatas takes the form of a great choralechorus—the largest and most elaborate movement in the whole work. Typically, the chorale melody is presented as a plain cantus firmus in the soprano part, often in augmented note-values. The lower voices (ATB) accompany, sometimes in plain homophony, but more often with some degree of polyphonic elaboration. The fourpart vocal texture of chorale plus accompaniment is embedded in an independent, concertante instrumental texture of great sophistication. Its themes, which often have some motivic connection with the chorale melody, are first stated in the opening ritornello, whose material furnishes interludes and postlude as well as an instrumental backing to the vocally delivered chorale-lines. There is no evidence that Bach had arrived at this form in its entirety before the beginning of Cycle II. The concept of an independent instrumental texture as a setting for the chorale, however, had been realized not only in the audition and Cycle I finales mentioned earlier but also in three opening chorale-choruses from Cycle I, those of Cantatas 138, 95, and 73. In all three cases, however, the chorale is troped by recitative.

l e ip z i g c y c l e i i : c h o r a l e c a nt a t a s

147

In the first month of Cycle II Bach sought the maximum diversity of form and style in his opening chorale-choruses; only later did he settle, to some extent, into more regular patterns. Thus, in the first four cantatas, Nos. 20, 2, 7, and 135, the chorale cantus firmus is situated in turn in the soprano, alto, tenor, and bass part; and in the fifth cantata, No. 10, it migrates from the soprano to the alto part. Thereafter it is situated in the soprano part, with only two exceptions, Nos. 96 (alto) and 3 (bass). In the first cantata of the series, O Ewigkeit, du Donnerwort, BWV 20, the opening chorale-chorus is cast in the form of a French overture. During Cycle I Bach had twice employed that form to celebrate a particular festive occasion (BWV 119 and 194). A closer parallel with Cantata No. 20, however, lies in the opening movement of the Weimar cantata No. 61 of 1714. In both cases the French overture marks a new beginning—the start of a new church year (No. 61) or of a new cantata cycle (No. 20). The inaugural function of Cantata No. 20 perhaps explains why it is conceived on a grander scale than any of its successors: it is the only Cycle II cantata constructed in two parts, a format found among the earlier cantatas of Cycle I. The second chorale-chorus of the cycle, that of Ach Gott, vom Himmel sieh darein, BWV 2, could hardly be more different. It is based on Martin Luther’s chorale of 1524, a German paraphrase of Psalm 12. In accordance with the authority and antiquity of the text, Bach sets its first verse as a cantus firmus motet in alla breve metre. The four voices are doubled by trombones and strings, creating an effect of great solemnity. A similar motet style was later employed in two other choralechoruses from Cycle II: the opening movement of Cantatas Nos. 38 and 121. It is interesting to note that Bach clearly had no compunction in juxtaposing the timehonoured motet style with the concertante and operatic styles employed in the following movements. The first aria of Cantata No. 2, for example, with its florid obbligato part for solo violin, comes across as strikingly ‘modern’ after the relatively archaic introductory movement. Indeed, its frequent mixing of different rhythms gives it a certain galant flavour. In the third cantata of the cycle, Christ unser Herr zum Jordan kam, BWV 7, the ‘modern’ concertante style is applied to the opening chorus. The instrumental ensemble, which includes two concertato violins, conjures up a vivid tone-picture of the waters of baptism and of Christ’s stately approach to the River Jordan. Bach then returns to motet style in the opening chorale-chorus of Ach Herr, mich armen Su¨nder, BWV 135. Its style is very different from that of the motet movements already mentioned, however. It might be described as a chorale motet with instrumental introduction and interludes. All parts, whether vocal or instrumental, are saturated with the chorale melody. Each line of the chorale cantus firmus is first played by unison upper strings (without continuo), then sung by the bass, doubled by trombone and continuo. The penitential text, a paraphrase of Psalm 6, suggests that the bassett cantus (unison strings) might represent the lost soul of the sinner; the bass cantus (with continuo), the solid foundation of prayer and the assurance of forgiveness. Not

148 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s long afterwards, in Nimm von uns, Herr, du treuer Gott, BWV 101, Bach went a step further, combining the traditional chorale-motet style in the vocal texture with independent instrumental music in his ‘modern’, descriptive vein. The instrumental music has two components: formal counterpoint of a deeply reverential character, and a sharply contrasting appoggiatura figure with diminished 3rd (bb. 25ff.), which creates violent dissonances and presumably stands for the many dangers and evils listed in the text. After the first four cantatas of Cycle II, Bach thenceforth drew from his extremely rich and varied stylistic palette on an ad hoc basis. A number of chorale-choruses exhibit special formal features that are worthy of note. In Wer nur den lieben Gott la¨ßt walten, BWV 93, each chorale line is prefaced by a duet or quartet in which a florid variant of that line is sung in imitation, presumably by the concertists (solo singers) only. The chorale line is then sung in plain four-part harmony. The instrumental music is to a large extent led by oboe duet, the strings often playing a subordinate role (this applies also to the first movement of the Cycle II cantatas Nos. 113, 8, 3, and 92). Cantata No. 62, based on Martin Luther’s Advent hymn Nun komm, der Heiden Heiland, opens with an exceptionally rich instrumental ritornello, containing no fewer than four significant motives, which are combined not only with each other but with the first and last chorale lines. The first chorale line enters at the outset as a long-note cantus firmus in the continuo—perhaps a conscious reminiscence of the Weimar cantata based on the same chorale, No. 61 of 1714. Since the first and last chorale lines are identical, Bach is able to construct the movement as a whole in reprise form (ABA1), the outer lines being distinguished by motet-style fore-imitation in the accompanying vocal parts. The amalgamation of motet and concertante modes of writing is still more evident in the New Year cantata Jesu, nun sei gepreiset, BWV 41. The two Stollen are embedded in a concertante instrumental setting, derived from the opening ritornello, which is led by a duet for two trumpets. The Abgesang, on the other hand, is motet-like, not only in its sectional form but in the style of its two main sections: a quiet Adagio in 3/4 (‘In goodly stillness we have completed the old year’) and an alla breve Presto in chorale-motet style, concerned with the congregation’s wishes and prayers for the coming year. Bach then returns to the concertante style of the Stollen in order to bind the whole movement together. If the chorale-chorus of Cantata No. 41 is notable for its heterogeneous character, demanding special measures to integrate it, that of the Epiphany cantata Liebster Immanuel, Herzog der Frommen, BWV 123, performed within a week later, is unified to an exceptional degree. The opening 20-bar ritornello is built entirely out of the opening two-bar phrase of the chorale, which then proceeds to permeate the instrumental music (and to some extent the vocal parts too) between the chorale lines. The effect is as if all participants in turn are calling out ‘Beloved Emmanuel’. In a number of cases the entire character of a chorale-chorus is clearly dictated by Bach’s response to a particular image in the text. The opening ritornello of Wo Gott der

leipzig cycle ii: chorale cantatas

149

Herr nicht bei uns ha¨lt, BWV 178—and as a result the instrumental background to the chorale, with its furioso dotted rhythms and trilling semiquavers—is clearly inspired by the textual image of raging enemies (‘Wenn unsre Feinde toben’). Only a week later, the lovely, florid instrumental music, led by concertante flute, that opens Was frag ich nach der Welt, BWV 94, no doubt represents ‘the world and all its treasures’ (‘der Welt und allen ihren Scha¨tzen’), which in the text is rejected in favour of Jesus. In Liebster Gott, wenn werd ich sterben, BWV 8, the florid melody for two oboes d’amore, the evocation of funeral bells in broken chords for muted strings, pizzicato continuo, and high repeated notes for the flute, in addition to the decorated chorale cantus firmus (also found in Cantatas Nos. 113, 33, and 78)—all these features evoke what has been described46 as ‘a sublime vision of the hour of death’. In Ach wie flu¨chtig, ach wie nichtig, BWV 26, the transitory nature of earthly life is conveyed by evanescent rising scale figures, by homophonic accompanying vocal parts in short note-values, and by the unisono singing of the diminished first chorale line by the lower voices at the end of each line of the cantus, reiterating time and again the words ‘Ah, how fleeting, ah, how empty [is the life of man].’ Certain chorale-choruses are no less remarkable for vivid text illustration than for special features of design. The two aspects are indelibly interlinked in the celebrated opening movement of Jesu, der du meine Seele, BWV 78, one of those ‘movements particularly strict in formal structure’ that ‘achieve the maximum expressive power’.47 Three themes are combined throughout: the chorale melody, presented as a lightly decorated cantus firmus in the soprano part, and two instrumental themes presented in the opening ritornello. One is the descending chromatic 4th, which acts as a ground bass throughout. This theme, the traditional lamento bass of Italian opera, had already been used by Bach in the Weimar cantata Weinen, Klagen, BWV 12 (1714) to express weeping and lamenting. In Cantata No. 78, this same chromatic bass is associated with Jesus’s ‘bitter death’ and ‘heavy affliction of the soul’. Combined with it in the ritornello and often thereafter is a theme for strings and oboes in the rhythm of a French chaconne. Since the chaconne is a courtly dance in Lullian opera, the theme might be associated with the positive aspect of the text, the rescue of the soul through Jesus’s self-sacrifice. Whenever the chorale cantus firmus enters, all three themes are combined—a remarkable tour de force on the part of the composer (Ex. 4). The chromatic 4th also plays an essential role in the first movement of Ach Gott, wie manches Herzeleid, BWV 3—no longer as a ground bass, however, but as the framework of the main instrumental theme, a highly expressive, melodious idea presented by two oboes d’amore, accompanied by strings and continuo (Ex. 5). The chromatic 4th, the Adagio tempo, and the ‘sighing’ appoggiatura figures of the string accompaniment are all no doubt intended to convey the ‘heartbreak’ and ‘tribulation’

46 47

By Du¨rr, The Cantatas of J. S. Bach, p. 552. According to Du¨rr, The Cantatas of J. S. Bach, p. 526.

150

sacred and secular: the vocal works

Ex. 4

su,

der

mei

See

1st movement of Jesu, der du meine Seele, BWV 78, bb. 21–4 (thematic parts only)

Ex. 5

1st movement of Ach Gott, wie manches Herzeleid, BWV 3, incipit (oboe d’amore II; strings and continuo omitted) plus editorial chromatic-4th scaffolding of the text. This impression is strengthened by the omnipresence of the chromatic theme: it is by no means restricted to the oboe duet but forms the basis of foreimitation before each chorale line, and to some extent it also accompanies the cantus firmus. The plain cantus is sung by the bass, doubled by trombone and continuo. This creates a deep layer of great power, in sharp contrast to the upper layer of the oboe duet and the higher voices (SAT), which are pervaded by the chromatic-4th theme. The lower level perhaps represents the biblical authority of the chorale text, for lines 3–4 clearly paraphrase a well-known passage from Acts 14: 22: ‘Through much tribulation must we enter into the Kingdom of God.’ Where a chorale verse is retained in its original form in an inner movement, Bach sets it in a modernized adaptation of the seventeenth-century chorale concerto for few voices. In Cantatas 178, 114, 92 (no. 4 in each case), and 113 (no. 2), the chorale is sung by a single voice as a plain cantus firmus in standard note-values (long notes in Cantata 113) within a texture of two, three or four parts. The instrumental parts, which flow directly out of the opening ritornello, are often imitative and tightly derived from a single motive. Bach had already cultivated this type of chorale arrangement in a series of cantatas from Cycle I (Nos. 95, 166, 86, 37, and 44). Another type, however, employed only twice in Cycle II, in Cantatas 10 (no. 5) and 93 (no. 4), represents a significant innovation. Here the role of voices and instruments is reversed: the plain chorale cantus firmus is delivered in the instrumental part (No. 10:

l e i p z i g c y c l e i i: c h o r a l e ca n t a t a s

151

trumpet or unison oboes; No. 93: unison strings), while the unaltered chorale verse is sung to a freely invented theme in an imitative vocal duet. In Cantata No. 93 this theme is clearly derived from the chorale melody, whereas that of No. 10 is independent but descriptive: its chromaticism identifies it as a ‘mercy’ motive, illustrating the first line of text, ‘He remembers his mercy’ (‘Er denket der Barmherzigkeit’). Since these two chorale arrangements resemble organ chorales, it is hardly surprising that Bach later arranged both for the organ (BWV 647 and 648) and published them among the six Schu¨bler Chorales of 1748/9. Among the chorale cantatas, chorale and recitative are often combined in various ways. Most of the methods in which this is carried out are already foreshadowed in the cantatas of Cycle I. The simplest method is for secco recitative to preface a chorale arrangement, sung by the same voice, as in Cantata 95 nos. 2–3, from Cycle I—a disposition to which Bach returns in Cantata 180 no. 3 from Cycle II. Elsewhere the verbal chorale paraphrase is declaimed as recitative during an instrumental rendering of the plain chorale cantus firmus (Cantatas 5 no. 4, 38 no. 4, 122 no. 3). There are clear precedents for this procedure among the cantatas of Bach’s first year in Leipzig (Cantatas 23 no. 2 and 70 no. 9). A more common mode of combining chorale and recitative in Cycle II is the chorale-trope. This procedure had already been employed in Cantata 18 no. 3 (Weimar) and in Cantatas 138 no. 2 and 190 no. 2 (Cycle I), but only in Cycle II did it become a frequent resource (Cantatas 93 nos. 2 and 5, 178 no. 2, 94 no. 5, 101 no. 5, 113 no. 4, 91 no. 2, and 92 no. 2). An intermediate chorale verse in its original form is ‘troped’ by the insertion of freely invented text between its lines, amplifying it and shedding further light on its contents. The chorale verse is sung to its associated melody, plain or lightly decorated, normally by a single voice accompanied by continuo. Between the chorale lines, the same voice sings the interpolated lines in secco recitative. On a few occasions, however, in accordance with the aforementioned Weimar and Cycle I precedents, the chorale is sung in a plain four-part (SATB) setting in alternation with secco for different voices in turn (Cantatas 178 no. 5, 3 no. 2, and 92 no. 7). More elaborate variants of the chorale-trope principle, involving ritornello, ostinato bass, decorated chorale melody (sung as arioso), and accompagnato in place of secco, are found in the third movement of Cantatas Nos. 101, 94, 125, and 126. Rarely in Cycle II does Bach unite aria and chorale arrangement (Cantatas 101 nos. 4 and 6, 113 no. 7, and 122 no. 4). Only in Cantata No. 122 is the chorale verse preserved in full in its original wording; the other movements preserve only one or two lines, the remainder being paraphrased. Cantata 113 no. 7, despite its heading ‘Aria duetto’, is in reality a chorale arrangement for soprano and alto with continuo. Alternate lines of the chorale melody are retained, albeit in an embellished form. In Cantata 101 no. 4, Bach alternates a Vivace bass aria with Andante presentations of the chorale cantus firmus. There could hardly be a greater contrast than between these two elements. In effect the bass sings a furioso rage aria—the Vivace ritornello, for three oboes and

152 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s continuo, on which it is based clearly represents God’s anger (‘Why would you be so angry?’). The first two Andante chorale interpolations question God’s anger (first line of chorale in the bass voice); then in the third, the bass sings in freely composed arioso the plea for leniency of lines 4–6, while at the same time the oboe choir with continuo play a four-part setting of the entire chorale melody. It is not often that Bach juxtaposed operatic and ecclesiastical styles as dramatically as he did here, but the procedure was clearly warranted by the text. Aria and chorale are united again in the soprano–alto duet that forms the sixth movement of the same cantata—no longer in alternation, however, but rather in simultaneous combination. Chorale lines 1, 3 and 4 are preserved in their original form both in text and in music. In the opening ritornello, the obbligato flute and oboe da caccia twice play the first chorale line as a plain, long-note cantus firmus in counterpoint with a florid, elegiac theme in the rhythm of a siciliana. This theme is no doubt associated with the first line of the chorale text, ‘Remember Jesus’s bitter death’ (‘Gedenk an Jesu bittern Tod’). The combination of this theme with a vocal counterpoint and with the first chorale line forms the chief material of the outer paragraphs, whereas the middle one combines decorated versions of the third and fourth chorale lines. The secular sphere of dance is here fused intimately with the sacred sphere of the Lutheran chorale, resulting in a movement of quite exceptional beauty and sophistication. Certain features of this movement recur in Cantata 122 no. 4: both are in the key of D minor, both employ dance rhythm, and both combine aria and chorale simultaneously. Cantata 122 no. 4, however, unlike the earlier movement, is a chorale-trope: the hymn text is preserved in its original form but troped by freely invented lines. Moreover, it is a terzetto (for SAT) with continuo and unison upper strings, which permits a clear division into three distinct strands of texture: the plain chorale cantus firmus, sung by the alto and reinforced by unison upper strings; a soprano–tenor duet, in which the freely invented lines are sung to a different imitative point for each line in the manner of a motet; and the continuo, which adheres to the dotted rhythms of the canarie or loure and their associated motives, as presented in the framing ritornello. The chorale is thus accompanied by a simultaneous commentary on each line—a feat that Bach nowhere else attempted—while the dancing continuo rhythms perhaps reflect a world in which (to quote from the first chorale line) ‘God is reconciled and our friend’. In Cycle II, aria themes are occasionally derived from the chorale melody (as in Cantatas 93 no. 3 or 107 no. 5), or more frequently the chorale melody is quoted during an aria, particularly where the original hymn words are present.48 Melodic chorale quotations also occur quite frequently during recitatives, again normally in conjunction with the original words. Usually the chorale lines are sung in arioso at an

This takes place in BWV 2 no. 3, 135 nos. 3 and 5, 93 no. 6, 107 no. 5, 113 no. 5, and 111 no. 2.

l e i p z i g cy cl e i i : c h o r a l e c a n t a t a s

153

Adagio or Andante tempo, thereby standing out from the surrounding secco. The two most striking examples occur in Cantatas 2 no. 2 and 78 no. 5. In the latter case a vividly illustrative bass accompagnato flows directly into an Andante arioso in which the entire Abgesang of the chorale (lines 5–8) in its original textual form is sung in a very free melodic paraphrase. The intensely expressive decoration of the chorale melody and the rich, dense string accompaniment were clearly inspired by the words, which speak of ‘my heart, mingled with grief ’ and ‘your precious blood, which was shed on the Cross’.49 As in Cycle I, the arias are predominantly cast in da capo (ABA) or reprise form (ABA1 or a variant thereof). In two cases, however (Cantatas 20 no. 6 and 180 no. 5), aria form is assimilated to binary dance form, which draws attention to the significant role of dance in the arias of Cycle II. Dance rhythms abound, including the menuet (Cantatas 93 no. 3 and 1 no. 5), passepied-menuet (62 no. 2 and 92 no. 8), dottedrhythm sarabande (114 no. 2, 96 no. 5, and 125 no. 2), siciliana (101 no. 6, 115 no. 2), gavotte (130 no. 5), bourreé (26 no. 4), giga (94 no. 6, 8 no. 4), giga-pastorale (113 no. 3), and gigue-passepied (124 no. 5).50 Clearly Bach did not baulk at the idea of introducing the strong secular associations of the dance within the sacred context of the chorale cantata. On the other hand, where the context demanded it, he was equally capable of devising arias in a style traditionally associated with sacred music. Thus aria ritornellos and the vocal paragraphs that follow are sometimes constructed fugally (for example, in Cantatas 38 no. 5 and 116 no. 4), and arias are occasionally structured in a motet-like series of quite different sections in accordance with the text (for example, Cantatas 139 no. 4 and 127 no. 4). One of the most prominent movement types of Cycle I, the vox Christi solo, is in its strict form not found in Cycle II due to the absence of biblical words. On some occasions, however, an inner chorale verse includes a paraphrase of the words of the Father or the Son, and these quotations call forth special treatment (Cantatas 2 no. 4, 7 no. 5, 113 no. 6). The most remarkable of these paraphrased vox Christi solos occurs in the fourth movement of the Quinquagesima cantata Herr Jesu Christ, wahr’ Mensch und Gott, BWV 127. A bass accompagnato that deals with the Last Judgement introduces an aria in which the bass voice represents the vox Christi, since the words purport to be those of Jesus. Two contrasting paragraphs alternate: a chorale arioso in common time for bass and continuo, announcing that the Christian is absolved from eternal punishment; and a 6/8 furioso with trumpet and strings that portrays the end of the world. Accompagnato, or accompanied recitative, on some occasions anticipates the motivic form it will take in the St Matthew Passion (Cantatas 10 no. 6, 107 no. 2, 125 no. 3,

49 Other examples of melodic chorale references within recitatives occur in BWV 135 no. 4, 116 no. 3, 133 nos. 3 and 5, and 41 no. 5. 50 All references to dance rhythm are indebted to Finke-Hecklinger, Tanzcharaktere in J. S. Bachs Vokalmusik.

154

sacred and secular: the vocal works

and 127 no. 4). The two most elaborate examples, from Cantatas 125 and 127, both date from towards the end of the chorale-cantata cycle (2 and 11 February 1725). In No. 125 the bass accompagnato and the chorale with which it alternates are united by the same florid string motive throughout—a motive presumably suggested by the line ‘[Christ] delights the spirit with heavenly sweetness’. In the accompagnato for bass, accompanied by trumpet, strings, and continuo, that introduces the vox Christi aria from Cantata 127 described earlier, we hear the ‘last trump’ and a tremolo ‘terror’ motive, evoking the feelings of extreme fear aroused by the Last Judgement. Two relatively young instruments, the transverse flute and the violoncello piccolo, were given demanding obbligato parts in the second half of the year 1724. Bach liked to combine the transverse flute with the oboe d’amore, as in the opening movements of Cantatas Nos. 99, 115, and 125. In the last-named case, the two instruments (plus continuo) are also combined in the second movement, a florid sarabande whose intimate sound is associated with the paraphrased words of the aged Simeon from the Nunc Dimittis (Luke 2: 29–32): ‘I will even with weak eyes look to You, my faithful Saviour’. Another choice chamber trio, this time made up of transverse flute, violoncello piccolo, and continuo, conjures up the intimacy of prayer in the Molto Adagio soprano aria from Cantata No. 115 (fourth movement). In some of the opening chorale-choruses, rich and diverse instrumentation helps to create a tone-picture relevant to the text. The ‘sublime vision of the hour of death’ in Cantata No. 8 has already been described. Other prominent examples are the first movements of Cantatas Nos. 1 and 180. At the opening of Schmu¨cke dich, o liebe Seele, BWV 180, the words ‘Adorn yourself, O dear soul’ are illustrated by rich instrumental adornment of the texture: unison upper strings pitted against a four-part woodwind choir, made up of two recorders, oboe, and oboe da caccia, followed by antiphonal exchanges between recorders and oboes. In Wie scho¨n leuchtet der Morgenstern, BWV 1, the words ‘How lovely shines the Morning Star’, referring to the coming of the Saviour, give rise to a lavish tone-picture involving antiphonal exchanges between pairs of concertato violins, oboes da caccia, and horns, accompanied by ripieno strings and continuo. In the arias of Cycle II, musical illustration of the text is often rendered exceptionally vivid by extreme contrasts between the A and B paragraphs of the da capo or reprise structure. In the second movement of Cantata No. 114, for example, a tenor aria with flute obbligato, ‘this vale of tears’ is illustrated by a slow, florid sarabande in D minor and in 3/4 with dotted rhythms, then the comfort of ‘Jesus’s fatherly hands’ by a Vivace in F major and in 12/8 gigue rhythm. In the alto aria with oboe d’amore and strings that forms the second movement of Cantata 115, an Adagio in siciliana rhythm represents the lullaby of the ‘slumbering soul’, then an Allegro in gigue or passepied rhythm brings the sudden awakening caused by ‘punishment’, after which the Adagio returns to depict the ‘sleep of death’. Finally, in the third movement of Cantata No. 123, a tenor aria with two oboes d’amore and continuo, the ‘hard journey of the cross’ and the ‘bitter food of

l e i p z i g cy cl e i i : c h o r a l e c a n t a t a s

155

tears’ are illustrated by a highly expressive Lento in F♯ minor with intricate, chromatic lines. At the words ‘when storms rage’ this gives way to a diatonic Allegro in A major with a lengthy operatic melisma on the word ‘toben’ (‘rage’). As in Cycle I, among the most remarkable cantatas are those in which musical imagery, employed in accordance with the text to represent a state of the soul, is sustained throughout the entire composition. In Nimm von uns, Herr, du treuer Gott, BWV 101, the solemn polyphony of the oboe and string choirs in the opening ritornello and cantus firmus passages no doubt represents the prayer for mercy and protection of the text. But the dissonant appoggiatura figure that emerges towards the end of the ritornello (b. 25) and then accompanies the chorale lines presumably stands for the feared ‘severe punishment’. The tenor aria with obbligato violin (originally flute), no. 2, is suffused with the tone of beseeching (see line 4), but the middle aria, no. 4, for bass, three oboes, and continuo, returns to the sharp contrast of the opening movement: divine wrath in the Vivace furioso; a prayer for lenient treatment in the Andante chorale quotations. In the last aria, no. 6, a soprano–alto duet with obbligato flute and oboe da caccia, the Christian’s grounds for hoping to receive ‘compassion’ from a ‘merciful God’ are expressed in the opening chorale line, repeated at the end, ‘Remember Jesus’s bitter death’ (‘Gedenk an Jesu bittern Tod’). Both in the opening instrumental duet and in the following vocal duet, the chorale melody associated with this line is kept plain but combined in invertible counterpoint with a florid siciliana theme of exceptional refinement and beauty. In this movement final release from conflicting states of the soul comes with remembrance of the Atonement. The exceptional integration of this cantata is greatly aided by the retention of the associated chorale melody—Vater unser im Himmelreich, Luther’s German version of the Paternoster—in all but one of its seven movements. Mit Fried und Freud ich fahr dahin, BWV 125, is based on Luther’s German translation of the Nunc Dimittis, Simeon’s canticle from Luke 2:29–32. This in itself creates a strong sense of unity, which is enhanced by the occurrence of the associated chorale melody in three of the six movements. The exceptionally high quality of the music may reflect Bach’s response to the divine authority of the Nunc Dimittis, as mediated by the revered founder of the Lutheran Church. The opening choralechorus exhibits the character of a pastorale, appropriate to the quiet resignation of the aged Simeon as he confronts death, saying: ‘In peace and joy I depart, according to God’s Will.’ The movement shares its key of E minor, its 12/8 metre, and much else with the great chorale-chorus that opens the St Matthew Passion, first performed only two years later. The B minor alto aria no. 2, with obbligato flute and oboe d’amore, is written in the slow triple time of a dotted-rhythm sarabande. Profusely decorated with appoggiaturas, it is a deeply expressive movement that tells of Simeon’s weak sight as he beholds the Saviour. The florid string ostinato figure of the motivic accompagnatocum-chorale trope no. 3 was, as already suggested, probably called forth by ‘the faithful Saviour, who on one’s deathbed already delights the spirit with heavenly sweetness’. The tenor–bass duet no. 4 is concerned with Jesus as the ‘Light’ who

156 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s comes ‘to lighten the Gentiles’. The pure three-part counterpoint in trio-sonata texture (two violins and continuo), in its vigour and fluency, perhaps represents the powerful, continuous sound to which the text refers, the sound of the divine promise that ‘whoever believes shall be saved’. Just over a week after the performance of this cantata upon the Feast of the Purification (2 February 1725), the Quinquagesima cantata Herr Jesu Christ, wahr’ Mensch und Gott, BWV 127, was performed (11 February). It is not only the penultimate chorale cantata in the cycle but one of the most remarkable of them all. In the opening chorale-chorus, the Christian prays to Jesus Christ for mercy ‘through your bitter Passion’. Bach reinforces this double theme, mercy and Passion, by introducing a second chorale, the German Agnus Dei Christe, du Lamm Gottes, which had already played a major role in a previous Quinquagesima cantata, No. 23 (audition and Cycle I). Cantata No. 127 includes both a vocal rendition of the chorale Herr Jesu Christ, with cantus firmus in standard note-values in the soprano part, and an instrumental rendition of Christe, du Lamm Gottes, with the cantus in long notes. During the opening ritornello, the first line of the Agnus Dei theme is combined in counterpoint with two others themes: the diminished first line of Herr Jesu Christ and a freely invented theme in dotted rhythms, doubled in 3rds or 6ths—perhaps a musical image of the merciful Christ. The second and third lines of the Agnus Dei theme are stated between or during the lines of the sung chorale. Moreover, while the soprano sings Herr Jesu Christ, the lower voice parts invariably sing its diminished first line in imitation. Thus the movement is remarkable not only for its unique combination of two chorale melodies but for its thematic cogency and concentration, which aid the listener’s focus on the dual themes of the text, the prayer for mercy and the forthcoming Passion. Associated thoughts on death in the tenor recitative no. 2 lead to the sublime soprano aria no. 3, ‘My soul rests in Jesus’s hands’ (‘Die Seele ruht in Jesu Ha¨nden’), with its florid, intricate duet for soprano and oboe, accompanied very lightly by staccato recorders and pizzicato bass, and its graphic illustration in the central paragraph of the words ‘Ah, call me soon, you death-bells’ (‘Ach ruft mich bald, ihr Sterbeglocken’). Just as reflection on Christ’s Passion and death led to thoughts of the individual Christian’s death, so now the latter leads to consideration of the end of the world and the Last Judgement. These matters form the theme of the fourth movement, a bass solo which, despite its heading ‘Recit.’, defies categorization. It is perhaps best described as a motivic accompagnato followed by an alternation of arioso and chorale arrangement. Both key and metre are unsettled and remain so throughout. The illustrative motives in the accompagnato have already been described. The bass is here the individual Christian facing the Last Judgement. In the chorale and arioso passages that follow, however, he takes over the role of vox Christi, for henceforth all the words are supposed to be those of Jesus Christ, despite being chorale-based rather than biblical. The hymn quotations are sung by the bass (with continuo only) to a variant of the first chorale line, one of the main themes of the first movement. Gentle assurance is here offered to the Christian, who ‘shall not come under judgement, nor taste death for ever’. These chorale passages alternate with quick

s t j o h n p a s s i o n , v e r s i o n ii

157

Ex. 6

Ich bre - che mit

star - ker und helf - en - der

Hand

a) 4th movement from Herr Jesu Christ, wahr’ Mensch und Gott, BWV 127, bb. 44–5, bass only (instrumental parts and continuo omitted)

Sind Blit - ze, sind Don - ner in Wol - ken ver - schwun - den?

b) St. Matthew Passion, No. 27b, initial subject entry (B I + II + continuo) 6/8 episodes in which the furioso of the trumpet and strings depicts the end of the world. The central 6/8 episode has a different musical theme from the others— recognizable as that of the similarly furioso chorus ‘Sind Blitze, sind Donner’ from the St Matthew Passion, ﬁrst performed two years later (Ex. 6). The concluding chorale is a prayer for forgiveness and for faith in the face of death.

St John Passion, Version II Title, occasion

Earliest source/s

Scribe, date

St John Passion, BWV 245, Version II, Good Friday

Berlin, St 111

Part-autograph, for 30 Mar. 1725

On Good Friday 1725 (30 March), less than a week after the chorale-cantata cycle had come to a premature end with the performance of Cantata No. 1 on 25 March,51 Bach revived the St John Passion—only one year after its ﬁrst performance. Exceptional circumstances must have led to his decision to perform the same setting of the Passion in two successive years. He did, however, take pains to ensure that the revived Passion sounded rather different from the original one by incorporating six new movements, of which ﬁve were replacements of existing movements and the sixth was an interpolation. The new movements for Version II are: 1. ‘O Mensch, bewein dein Su¨nde groß’, chorale-chorus in E♭; replacement for original exordium. 2. ‘Himmel reiße, Welt erbebe’, chorale-aria in F♯ minor; interpolation after chorale no. 11.

51 Though the revival of BWV 4 on Easter Sunday (1 Apr.) could be regarded as the true termination of the cycle.

158 sacred and s ecular: the vocal works 3. ‘Zerschmettert mich, ihr Felsen und ihr Hu¨gel’, tenor aria in A; replacement for tenor aria no. 13. 4. ‘Ach windet euch nicht so, geplagte Seelen’, tenor aria in C minor; replacement for arioso-aria sequence nos. 19–20. 5. ‘Und siehe da, der Vorhang im Tempel’, Evangelist’s recitative no. 33; replacement for original version. 6. ‘Christe, du Lamm Gottes’, chorale-chorus in G minor; replacement for original conclusio (chorale no. 40). The chorale-chorus ‘O Mensch, bewein’, later (1736) incorporated in the St Matthew Passion, belongs to the type established in the opening movements of the chorale cantatas: the chorale cantus firmus, sung by the soprano and accompanied polyphonically by the lower voice parts, is embedded line by line in a concertante instrumental texture derived from the opening ritornello. A four-note motive, prominent in the chorale itself, is used in direct and inverted, plain and decorated, standard and augmented forms throughout the ritornello and hence in all the instrumental music. Moreover, during the cantus firmus the lower voice parts are based tightly on this same motive, employing it as an imitative point. This integrated chorale structure is clearly intended to reveal the central theological significance of the Passion, namely the Atonement. At the same time, however, mankind is called upon to ‘bewail [its] great sin’ that made the Atonement a necessity. The movement is thus not only an authoritative statement of church doctrine but also a great lament, hence the ‘weeping’ semiquaver couplets that clothe the four-note motive (already associated with the Passion in the Orgelbu¨chlein chorale O Lamm Gottes, unschuldig, BWV 618); hence also the minor-mode pathos of the rhyming-close to the two periods of the binary ritornello (bb. 5–8 and 13–16), into which grief-stricken melismas for the accompanying voices are subsequently built (bb. 21–3 and 64–7). ‘Himmel reiße, Welt erbebe’ is a bass aria with chorale, just like ‘Mein teurer Heiland’, no. 32, and both movements are based on the same chorale, Paul Stockmann’s Jesu, Leiden, Pein und Tod of 1633. By introducing a fourth arrangement of this chorale into the Passion, Bach creates a symmetrical arrangement of two verses from it towards the end of Part I and two more near the end of Part II; and in each case the first of the pair is sung in a plain four-part setting and the second united with an aria. The striking of Jesus (Nos. 10–11) seems an inappropriate context for the words of ‘Himmel reiße’, which deal with the Passion itself. However, it is possible to view the movement as a concluding meditation upon the ‘arrest scene’. The relationship between aria and chorale texts is uniquely close: the aria falls in with the rhyme scheme of the chorale, and rhyming words in the two texts often form opposite pairs, such as ‘Freude’/‘leide’ or ‘Dornen’/‘Rosen’. This antithesis is reflected in opposite modes—the aria is in F♯ minor and the chorale in A major—and in contrasting voices and instruments: the aria is performed by bass voice and continuo, the chorale by soprano and two obbligato flutes. The low pitch of the bass and continuo clearly

st john pa ssion, version ii

159

represents the earthly plane, the ‘vile earthly vault’ with its grief, sorrow, fear, suffering, agonies, thorns, and stormy tempest. The high pitch of the soprano and flutes, on the other hand, stands for the heavenly plane of ‘pure joy’ and ‘my heart’s pasture’ where ‘my soul walks on roses’. Before Good Friday 1725 extreme simultaneous contrast of this kind occurs only in the opening movement of O Ewigkeit, du Donnerwort, BWV 60, from Cycle I, the only other chorale-aria of this kind before Cycle III. ‘Zerschmettert mich, ihr Felsen und ihr Hu¨gel’ is scored for exactly the same voice and instruments as the movement it replaces, ‘Ach, mein Sinn’ (no. 13), namely tenor, strings, and continuo. Moreover, both are designed as meditations on Peter’s denial. ‘Ach, mein Sinn’ expressed Peter’s feelings of grief and torment over his denial, but it also included the image ‘Or do I wish mountains and hills to fall upon my back?’, an allusion to Luke 23: 26–31.52 This image, clothed in different words, opens ‘Zerschmettert mich’: ‘Crush me, you rocks and hills; heaven, cast your thunderbolt upon me!’ These words chime in with those that opened the choralearia ‘Himmel reiße’ immediately before the denial scene: ‘Heaven, tear apart; world, quake!’ In both cases the image injects a powerfully dramatic element into the commentary, similar to that which inhabits parts of the Gospel narrative itself. In ‘Zerschmettert mich’ the concitato style is used to depict the crushing power of rocks and hills. Peter’s remorse is heard, however, in the Adagio accompagnato setting of the words ‘How outrageously . . . have I forgotten You, O Jesus!’ and in the poignant melisma on ‘bitter tears’. Nevertheless, the bright, lively A major of the ritornello sounds strangely incongruous after the Evangelist’s deeply felt, chromatic arioso describing how Peter ‘wept bitterly’. The aria ‘Ach, windet euch nicht so’ is placed immediately after the Evangelist’s arioso that vividly describes the scourging of Jesus (no. 18c). The madrigalian-text arioso ‘Betrachte, meine Seel’ (no. 19) is omitted at this point in Version II, but the new aria is clearly intended to replace not this movement but ‘Erwa¨ge’ (No. 20): both are da capo arias in C minor for tenor, instrumental duet (violas d’amore in ‘Erwa¨ge’, oboes in its replacement), and continuo. The two chief motives of the new aria are both illustrative: a ‘writhing’ motive in the first bar, later linked to the words ‘Oh, writhe not so, tormented souls’, and a slow, chromatic ascent in bar 3, later associated with the word ‘Kreuzesangst’ (‘fear of the cross’). In the central paragraph, this chromatic motive is treated in four-part stretto (bb. 61–3), then in imitation by inversion in the two oboe parts. Despite these felicities, it is hard to see any justification for the substitution beyond that of introducing new music into Version II. The original version of the Evangelist’s recitative that tells of supernatural events following Jesus’s death (no. 33) was only three bars long. Only the continuo part survives, but the text was probably an interpolation from Mark 15: 38, recording no

52 See Michael Marissen, Bach’s Oratorios: The Parallel German-English Texts with Annotations (Oxford, 2008), p. 109 n. 26.

160

sacred and secular: the vocal works

more than the rent in the temple veil.53 For Version II Bach replaced it with a sevenbar version—that with which we are familiar today—based on Matthew 27: 51–2, which tells not only of the rent veil but of the earth quaking, the rocks being split, and the graves opened. All these events are described in the following arioso, ‘Mein Herz’ (no. 34), which almost certainly explains why Bach made the alteration: it would surely be considered inappropriate for the meditative commentary to deal with events that were absent from the preceding biblical narrative. The expanded, seven-bar version of the Evangelist’s recitative no. 33 is the only one of the six new movements introduced in Version II that was retained in the final version of the Passion. Finally, the plain four-part chorale that originally concluded the Passion was replaced in Version II by Bach’s sublime setting of the German Agnus Dei, ‘Christe, du Lamm Gottes’, borrowed from the Leipzig audition cantata No. 23. It must have been Bach’s intention that this movement and ‘O Mensch, bewein’ should form a powerful chorale-based frame around the Passion narrative. ‘Christe, du Lamm Gottes’ was well equipped to counterbalance ‘O Mensch, bewein’ on account of its elaborate instrumental parts and its large dimensions—a product of its threevariation structure that reflects the threefold prayer of the Agnus Dei. Moreover, the texts of the two chorale movements are closely interrelated: both speak of Christ’s ‘bearing the sins of the world’. However, whereas the exordium is a chorus of lamentation, the conclusio is a prayer for mercy. Until recently it was widely believed that ‘O Mensch, bewein’, together with at least one of the arias peculiar to Version II, ‘Himmel reiße’, originated in Weimar.54 The obvious objection to this theory is that ‘O Mensch, bewein’ exhibits at its highest point of development a form of chorale-chorus that Bach devised specifically for the chorale cantatas of Cycle II. It seems surprising that scholars have been so slow to notice that there is no counterpart to ‘O Mensch, bewein’ among the Weimar cantatas, whereas its type is reflected in almost all the opening movements of the chorale cantatas.55 It cannot be mere chance, then, that ‘O Mensch, bewein’ was performed— within Version II of the St John Passion—only five days after the last chorale cantata (BWV 1, 25 March 1725). The tenuous source-critical arguments for a Weimar origin of ‘O Mensch, bewein’ have recently been refuted.56 It has been shown that for the 53 These conclusions were reached by Arthur Mendel, Krit. Bericht, NBA II/4 (1974), p. 270; see also Alfred Du¨rr, Johann Sebastian Bach: St. John Passion—Genesis, Transmission, and Meaning, Eng. trans. by A. Clayton (Oxford, 2000), pp. 119–20. 54 See Mendel, Krit. Bericht, NBA II/4, p. 80; the same author’s ‘Traces of the Pre-history of Bach’s St. John and St. Matthew Passions’, in Festschrift Otto Erich Deutsch zum 80. Geburtstag (Kassel, 1963), pp. 31–48; and his ‘More on the Weimar Origin of Bach’s “O Mensch, bewein” (BWV 244/35)’, Journal of the American Musicological Society, 17 (1964), pp. 203–6. 55 This has recently been pointed out independently by Joshua Rifkin, ‘ “O Mensch, bewein dein Su¨nde groß” and the History of Bach’s Passions’ (unpublished article, c. 1990), by Krummacher, Bachs Zyklus der Choralkantaten, pp. 88–90, and by Ulrich Leisinger, ‘Die zweite Fassung der Johannes-Passion von 1725: Nur ein Notbehelf ?’, in Leisinger (ed.), Bach in Leipzig—Bach und Leipzig (Hildesheim, 2002), pp. 29–44 (esp. 39–40). 56 By Rifkin, ‘ “O Mensch, bewein” and the History of Bach’s Passions’.

s t j o h n p a s s i on, ve rs i on i i

161

Passion Bach copied it from an original in the key of D that included parts for transverse flutes and oboes d’amore. This would place the origin of the movement in Leipzig, which is confirmed by the nature of the music itself. It has also been shown that arguments for the Weimar origin of the chorale-aria ‘Himmel reiße’, on the basis of different versions of the chorale melody, do not hold water.57 Nor does there appear to be a strong reason for dating any of the other movements earlier than the Leipzig period.58 If, as seems highly likely, ‘O Mensch, bewein’ was composed in the key of D59 in 1725, it cannot have been originally intended for Version II of the St John Passion, for Bach had to transpose it up a semitone to E♭ in order to accommodate it to the key scheme of that work (the first recitative, no. 2a, centres around the key of C minor). His original intention, then, must have been to perform a new Passion on Good Friday 1725—one that began with the original D major version of ‘O Mensch, bewein’. The close link between that movement and the opening choruses of the chorale cantatas suggests that Bach might have been planning a chorale-Passion as the culmination of Cycle II.60 If so, it is likely that the whole of Sebald Heyden’s 23verse chorale O Mensch, bewein dein Su¨nde groß (1525) would have been used in the course of the work, the last verse (like the first, and no doubt several intermediate ones) being retained in its original form and other verses being paraphrased for use in arias and ariosos. The most likely reason why Bach apparently abandoned work on the new Passion after only the exordium had been composed is also the explanation for his leaving the chorale-cantata cycle incomplete: due to illness, death, or some other reason, his librettist could no longer supply texts. Version II of the St John Passion was thus purely a makeshift.61 After the performance of Version I in 1724 he seems to have lent out the performing parts, which proves that he did not plan a repeat performance in the following year. The revival seems to have become necessary only when his librettist dropped out some time in the early months of 1725, preventing Bach from undertaking further work on the projected new Passion. He did his best to avoid the impression of an exact revival by incorporating six new movements, of which three were chorale-based in accordance with his preoccupations at that time. But it seems unlikely that he regarded the new version as an improvement. Otherwise, among the six new Version II movements, why did he retain only the brief recitative ‘Und siehe da’, no. 33, in the final version of the Passion? In every other case the original version was restored. 57 See Markus Rathey, ‘Weimar, Gotha oder Leipzig: Zur Chronologie der Arie “Himmel, reiße” in der zweiten Fassung der Johannes-Passion (BWV 245/11+)’, BJ 91 (2005), pp. 291–300. 58 According to Leisinger, ‘Die zweite Fassung der Johannes-Passion’, pp. 41–4, and Rathey, ‘Weimar, Gotha oder Leipzig’. 59 This theory of Mendel’s (Krit. Bericht, NBA II/4, p. 80) has been accepted by Du¨rr (J. S. Bach: St. John Passion) and by Rifkin (‘ “O Mensch, bewein” and the History of Bach’s Passions’). 60 This theory was put forward by Leisinger, ‘Die zweite Fassung der Johannes-Passion’, pp. 40–1. 61 Both Rifkin (‘ “O Mensch, bewein” and the History of Bach’s Passions’) and Leisinger (‘Die zweite Fassung der Johannes-Passion’, p. 44) take this view.

162 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s

Leipzig Cycle II: non-chorale cantatas Title, occasion

Earliest source/s

Scribe, date

Kommt, ﬂiehet und eilet, BWV 249, Easter Sunday Bleib bei uns, BWV 6, 2nd Day of Easter

Berlin, St 355

J. A. Kuhnau et al., for 1 Apr. 1725

Berlin, P 44/1 , St 7

Berlin, St 103

Autograph, J. A. Kuhnau et al., for 2 Apr. 1725 Autograph, J. A. Kauhnau et al., for 8 Apr. 1725 Autograph, J. A. Kuhnau et al., for 15 Apr. 1725 Autograph, J. A. Kuhnau et al., for 22 Apr. 1725 Autograph, J. A. Kuhnau et al., for 29 Apr. 1725 Autograph, J. A. Kuhnau et al., for 6 May 1725 Autograph, J. A. Kuhnau et al., for 10 May 1725 Autograph, J. A. Kuhnau et al., for 13 May 1725 J. A. Kuhnau et al., for 20 May 1725

Leipzig TS

J. A. Kuhnau et al., for 21 May 1725

Berlin, P 75, St 22

Autograph, J. A. Kuhnau et al., for 22 May 1725 Autograph, J. A. Kuhnau et al., for 27 May 1725

Am Abend aber desselbigen Sabbats, BWV 42, Easter 1 Ich bin ein guter Hirt, BWV 85, Easter 2 Ihr werdet weinen und heulen, BWV 103, Easter 3 Es ist euch gut, daß ich hingehe, BWV 108, Easter 4 Bisher habt ihr nichts gebeten, BWV 87, Easter 5 Auf Christi Himmelfahrt allein, BWV 128, Ascension Sie werden euch in den Bann tun, BWV 183, Ascension 1 Wer mich liebet, der wird mein Wort halten, BWV 74, Whit Sunday Also hat Gott die Welt geliebt, BWV 68, 2nd Day of Whit Er rufet seinen Schafen mit Namen, BWV 175, 3rd Day of Whit Es ist ein trotzig und verzagt Ding, BWV 176, Trinity

Berlin, P 55, St 3 Berlin, P 106, St 51 Berlin, P 122, St 63 Berlin, P 82, St 28 Berlin, P 61, St 6 Priv. poss., Berlin, St 158 Berlin, P 149, St 87

Berlin, P 81, priv. poss.

The last chorale cantata of Cycle II was a revival of the early Christ lag in Todes Banden, BWV 4, on Easter Sunday (1 April) 1725. On the same day, though perhaps in a different church,62 Bach performed the Easter cantata Kommt, fliehet und eilet, BWV 249, which later became the Easter Oratorio (1735?). A greater contrast than between these two Easter compositions could hardly be imagined. Christ lag in Todes Banden is a complete setting by Bach of the great Easter hymn by Martin Luther, which in turn derives from the Latin sequence Victimae paschali laudes and from the old German hymn Christ ist erstanden. Accordingly, it is composed in a conservative church style that Bach had largely abandoned by 1713. Kommt, fliehet und eilet, on the other hand, is a straightforward parody—involving relatively little change beyond the new text and minor musical accommodations to it—of the pastoral cantata Entfliehet, verschwindet, entweichet, ihr Sorgen, performed at Weißenfels on 23 February 1725, only about five weeks before Easter Sunday. It is written in an up-to-date, operatic style

See Du¨rr, The Cantatas of J. S. Bach, p. 32.

l e i p z i g cy cl e i i co n t .

163

which is fully maintained in the sacred parody. No chorale or biblical text is added, nor is there any Gospel narrative sung by an Evangelist, as in the Passions and the other oratorios. In effect, the Easter cantata is a sacred dramma per musica with sung plot and dramatis personae. It thus differs fundamentally from all Bach’s other sacred works. As a piece of genuinely theatrical music, it belongs to a type that Bach was expressly told not to write at the 1723 council meeting in which he was elected to his Leipzig post.63 That he did so is perhaps a sign of his increasing boldness and independence. It is also possible that the inherently dramatic nature of the work could be justified by linking it with the old tradition of the Easter play, a dramatic representation of the events surrounding the Resurrection. Sacred parody had been excluded from the chorale cantatas by the nature of their texts, but Bach returned to it afterwards, not only in the Easter cantata Kommt, fliehet und eilet but in some of the Cycle II cantatas that followed, though only in isolated movements rather than entire compositions. In Am Abend aber desselbigen Sabbats, BWV 42, Bach returned to Cantata No. 66a, a Co¨then serenata he had already parodied during the previous Eastertide (as Cantata No. 66 of Cycle I). For Cantata No. 42 he now selected two movements from No. 66a not so far used, the sinfonia and the aria ‘Beglu¨cktes Land von su¨ßer Ruh und Stille’ (‘Fortunate land of sweet calm and quiet’).64 Both movements share the same concertino, the Lullian trio of two oboes and bassoon. As the first line indicates, the original aria describes a scene of peace—the woodwind trio is accompanied by a warm, placid background of held string chords. In the parodied version, both this movement and the preceding recitative refer to Jesus’s first appearance before the disciples after the Resurrection (John 20: 19–25), when ‘[he] came and stood among them and said, “Peace be with you”’. The music seems ideally suited to the new text. In the Whitsuntide cantata Er rufet seinen Schafen mit Namen, BWV 175, Bach again returned to a Co¨then serenata (BWV 173a) that he had parodied during the same season in the previous year (BWV 173).65 And again he selected an aria omitted from the previous sacred parody, namely ‘Dein Name gleich der Sonnen geh’. Here a movement written in praise of the ‘most illustrious Leopold’, prince of AnhaltCo¨then, is adapted without difficulty to refer to the ‘Good Shepherd’ of the Gospel (John 10: 1–11). In 1725 Bach took further measures to relieve his Whitsuntide burden. In Cantata No. 68, Also hat Gott die Welt geliebt, both arias were parodied from his earliest known secular vocal work, the Hunt Cantata, BWV 208, of 1713. This substantial secular input is offset by the strong ecclesiastical frame that surrounds it: a chorale-chorus at the beginning and a biblical-text chorus at the end. Particularly interesting are the changes Bach made to the first of the two arias, ‘Mein glaübiges

BD II, No. 129; NBR, No. 98. See Rifkin, ‘Verlorene Quellen, verlorene Werke’, pp. 65–7. 65 According to Tatiana Shabalina, however, BWV 173 is more likely to have originated in 1727 than in 1724; see her ‘ “Texte zur Music” in Sankt Petersburg: Neue Quellen zur Leipziger Musikgeschichte sowie zur Kompositions- und Auffu¨hrungsta¨tigkeit J. S. Bachs’, BJ 94 (2008), pp. 33–98 (esp. 68–71). 64

164

sacred and secular: the vocal works

Herze’. The original, BWV 208 no. 13, is a brief, simple aria for soprano and continuo—38 bars with a single vocal paragraph—that uses pastoral imagery to praise Duke Christian of Weißenfels. In its sacred context, it turns into a far more substantial piece—52 bars in ABA1 form—that celebrates the continuing presence of Christ. Above all, the former hymn-like soprano part is replaced by a highly attractive melody in the style of a popular song, sung in counterpoint with the instrumental theme (Ex. 7). Even Bach rarely achieved transformations of such quality in his parodied compositions. Finally, the first two movements of the Whit Sunday cantata Wer mich liebet, der wird mein Wort halten, BWV 74, were adapted from the outer movements of an earlier cantata composed for the same occasion, No. 59, which apparently originated in 1723 (for 16 May) but may not have been performed till the following year (on 28 May; see the section ‘Sacred cantatas: Leipzig Cycle I’). The chorale from this cantata was also plundered in 1725, serving as the finale of Cantata No. 175 on Whit Tuesday. Thus a little work of great quality, though with severe limitations, no doubt due to the circumstances of its original performance, was finally brought in from the cold. For the librettos of the non-chorale cantatas of Cycle II, from Easter to Trinity 1725, Bach first made use of some texts left over from Cycle I (BWV 6, 42, 85), then set nine texts by the local Leipzig poet Mariane von Ziegler (BWV 103, 108, 87, 128, 183, 74, 68, 175, 176). As in Cycle I, biblical words and chorale verses form the main content of the librettos, diversified by freely invented, madrigalian verse suitable for setting as recitatives and arias. First, one might ask what effect Bach’s recent experience of composing a long, unbroken series of chorale cantatas had on the treatment of chorales in these later Cycle II cantatas. The great chorale-choruses that open Cantatas Nos. 128 and 68 are

Ex. 7

Mein gläu - bi

- lok - ke,

ges Her - ze, froh - lok - ke,

sing, scher - ze

dein

sing, scher - ze, froh -

sus

ist

da;

Soprano aria from Also hat Gott die Welt geliebt, BWV 68 No. 2, bb. 9–12 (soprano and violoncello piccolo; continuo omitted)

l e i p z i g cy cl e i i co n t .

165

unthinkable without the first movements of the chorale cantatas behind them. Both essentially exhibit the pattern Bach had established, yet both also show unusual features of their own. In the Ascension Day cantata Auf Christi Himmelfahrt allein, BWV 128, the ritornello is constructed as a concertante fugue. In Also hat Gott die Welt geliebt, BWV 68, the instrumental music is written in the 12/8 dotted rhythms of the siciliana. This is also true of the chorale-finale of the Cycle II cantata No. 107, but in No. 68 it appears to have special significance. The chorale opens with a paraphrase of John 3.16, ‘For God so loved the world . . .’, which suggests that ‘the world’ might be represented by the secular element of dance. In addition, however, the chorale cantus firmus is expressively decorated, often partaking of the instrumental material with its siciliana rhythms. It is varied to such an extent that at times it bears little relation to the original chorale melody. We encounter here, then, an exceptionally subjective response to the chorale—unheard of among the chorale cantatas—which perhaps relates to the Son: ‘For God so loved the world that He gave His only Son . . .’. Some cantatas of this period incorporate several different chorales (BWV 6, 42, 85, and 128), which was also true of certain Cycle I cantatas but was precluded by the structure of the chorale cantatas. In two cases (Nos. 6 and 85) the intermediate chorale is a cantus firmus arrangement for solo voice, obbligato instrument/s, and continuo. The ritornello, and hence the instrumental music as a whole, is considerably more extensive and highly organized than in earlier examples of this type. In Cantata No. 6, Bleib bei uns, third movement, later transcribed as one of the Schu¨bler Chorales (BWV 649), the ritornello—music of great charm for obbligato violoncello piccolo—is a substantial period incorporating three ideas: a derivative of the chorale’s headmotive (Vordersatz), a sequential broken-chordal figure (Fortspinnung), and a cross-string figuration (Epilog). The third movement of Cantata No. 85, Ich bin ein guter Hirt, performed only a fortnight later, is overtly pastoral in character due to its text, a German version of Psalm 23. The two oboes, the soprano, and even the continuo have a paraphrased version of the first chorale line, and all but one of the following lines are similarly decorated. As a result, a well-integrated, homogeneous quartet texture emerges. The most unusual of these intermediate chorale arrangements is the fourth movement of Cantata No. 42, Am Abend aber desselbigen Sabbats, which, despite its chorale text, is largely free of references to the associated chorale melody. This freedom recalls the opening chorale-chorus of Cantata No. 68 and raises the question whether Bach was in these cases reacting against the almost excessive dominance of the chorale melody in the chorale cantatas. In the later Cycle II cantatas, biblical-text settings—in the form of choruses, recitatives, and vox Christi solos—are no less important than in Cycle I. The first biblical-text chorus of the series, that which opens Bleib bei uns (BWV 6), is among the finest of them all. As in Cantata No. 69a (Cycle I), ritornello and fugal structures are united within an overall ABA1 reprise form. The sarabande rhythm of the ritornello and paragraph A, far from lowering the tone through its secular

166 s acred and secular: the voca l w orks associations, imparts a certain gravitas and ‘imposing grandeur’66 to the setting. The same text is set with greater urgency in the fugal B-section, where the permutation of three subjects creates the effect of the clamouring disciples as they beg their risen Lord to ‘remain with us’ (‘Bleib bei uns’). Afterwards, the ritornello returns with inbuilt vocal parts (A1). The Ziegler cantatas include a series of choruses that quote the words of Jesus from St John’s Gospel: Cantatas 103 (no. 1), 108 (no. 4), 74 (no. 1), and 68 (no. 5). It is a matter of interest in itself that Bach chose to set Jesus’s words for multiple voices, for this is virtually unknown in Cycle I (BWV 44 nos. 1–2 is an isolated example). It may well be that he wished to impart extra weight and power to the divine utterances. The fourth movement of Es ist euch gut, daß ich hingehe, BWV 108, and the fifth of Also hat Gott die Welt geliebt, BWV 68, are both motet-style fugues with doubling instruments. The former in fact comprises three fugues on different portions of text, of which the third is a varied reprise of the first, giving rise to an overall ABA1 scheme. In the finale of Cantata No. 68, the voices are doubled not only by woodwind and strings but by cornett and three trombones, which lends a special solemnity to Christ’s severe words (from John 3: 18) about belief and condemnation. These words are set in the form of a double fugue, which helps to point up the antithesis ‘Whoever believes . . .’/‘Whoever does not believe . . .’. The opening movement of the Whit Sunday cantata Wer mich liebet, der wird mein Wort halten, BWV 74, is an adaptation for four-part choir and enlarged instrumental ensemble of the soprano–bass duet that opened the smaller cantata of the same name, BWV 59 (16 May 1723 and/or 28 May 1724). The most complex and imposing of these choral settings of Jesus’s words is the first movement of Cantata No. 103, Ihr werdet weinen und heulen, a remarkable synthesis of ritornello form, fugue, and recitative. The musical sequence of events turns on the sorrow–joy antithesis of the text. The ritornello, with its concertante flauto piccolo (sopranino recorder), signifies ‘rejoicing’, whereas the main fugue subject, with its semitonal steps, augmented intervals, and chromatic counterpoints, sets the ‘weeping and lamenting’ text. Ritornello and fugue, in varied and developed forms, alternate throughout. The ritornello, after its initial statement, is furnished with inbuilt voice parts that set the ‘joy’ text. The fugue, after its first exposition, is enhanced to double fugue by the addition of a second subject—none other than the ‘joy’ theme of the ritornello. Before the final fugue–ritornello sequence, the bass—presumably Christ himself—sings the words of sorrow in a deeply expressive accompagnato marked ‘Adagio e piano’. In setting Jesus’s words in these late Cycle II cantatas, Bach frequently casts the bass voice as vox Christi, as in Cycle I. The arioso that opens Cantata No. 87, Bisher habt ihr nichts gebeten in meinem Namen, is clearly modelled on that of Cantata 86, written for the same Sunday (Fifth after Easter) in the previous year. In both cases, the bass voice

According to Du¨rr, The Cantatas of J. S. Bach, p. 280.

le ip zi g c yc l e i i c o n t.

167

participates as an equal member—without thematic differentiation—in a strictly polyphonic, five-part texture alongside strings with doubling oboes. This rich, homogeneous texture lends great weight and solemnity to Christ’s words. By contrast, the vox Christi solos that open Cantatas Nos. 85 and 108 are aria-like in their vocal mottos, their articulation by internal ritornellos, and their concertante oboe parts. The long, florid melismas in the voice part of Cantata 108 no. 1 recall the first movement of the audition cantata No. 22 in their decidedly operatic character. Biblical words are sometimes set to recitative—a procedure unknown in Cycle I but common in the Weimar cantatas. A clear reference to the tradition of Passion settings is encountered in Cantata No. 42: after the sinfonia, the tenor takes the role of Evangelist, narrating the Gospel events in recitative, accompanied by throbbing semiquavers in the continuo to describe the ‘fear’ of the text. At the opening of the pastoral cantata No. 175, Er rufet seinen Schafen mit Namen, the Gospel words, set as tenor accompagnato, are not a narration of events but a quotation from Jesus’s parable about the Good Shepherd and his sheep, hence the pastoral accompaniment for three recorders. The usual bass voice for Jesus’s words is not used, presumably because the parable uses the third person: ‘He calls his sheep by name and leads them out.’ Outstanding achievements among the later cantatas of Cycle II are many, but for a perfect wedding between text and music it would be hard to exceed Cantata No. 87, Bisher habt ihr nichts gebeten in meinem Namen. The parallel between the two halves of Ziegler’s text, clearly reflected in Bach’s musical setting, is eminently satisfying. Each half is introduced by a quotation of Jesus’s words from John 16, and in each case the response of the soul is given in an aria of quite exceptional beauty and immense profundity. Ziegler’s text opens with Jesus’s words, ‘Until now you have not asked for anything in my name’ (John 16: 24). These words are taken as a reproach in the recitative no. 2, since mortal sinfulness means that one should ‘pray in penitence and devotion’, which is exactly what happens in the following G minor alto aria with two oboes da caccia, ‘Vergib, o Vater, unsre Schuld’ (‘Forgive, O Father, our guilt’), one of the most deeply moving of all Bach’s penitential arias. A figured arpeggio in the continuo forms an ostinato with the character of a beseeching gesture. Above it, the ritornello theme, later taken over by the voice, is made up of eloquent, prayerful gestures, played in 3rds and 6ths by the two obbligato oboes da caccia. In the following tenor accompagnato, the penitent soul seeks comfort from Jesus, who then provides it in the words of John 16: 33: ‘In the world you will have tribulation, but be of good cheer: I have overcome the world.’ The soul’s response is then given in a tenor aria ‘of overwhelming beauty’.67 The dance rhythms of the siciliana, in conjunction with the major mode, convey a lightness of heart—the soul is willing to suffer in the knowledge that Jesus’s comfort is at hand—at the opposite pole to the heavy-hearted soul-searching of the previous aria, with its minor-mode pathos and ‘sighing’ appoggiatura figures.

Du¨rr’s description, The Cantatas of J. S. Bach, p. 324.

168 sacred a nd s ecula r: the vocal works

Leipzig Cycle III Title, occasion

Earliest source/s

Scribe, date

Tue Rechnung! Donnerwort, BWV 168, Trinity 9 Lobe den Herren, BWV 137, Trinity 12

Berlin, P 152, St 457, etc. Leipzig TS

Ihr, die ihr euch von Christo nennet, BWV 164, Trinity 13 Bringet dem Herrn, BWV 148, Trinity 17

Berlin, P 121, St 60

Gott der Herr ist Sonn und Schild, BWV 79, Reformation Festival Unser Mund sei voll Lachens, BWV 110, Christmas Day Selig ist der Mann, BWV 57, 2nd Day of Christmas Su¨ßer Trost, mein Jesus ko¨mmt, BWV 151, 3rd Day of Christmas Gottlob! Nun geht das Jahr zu Ende, BWV 28, Christmas 1 Herr Gott, dich loben wir, BWV 16, New Year Liebster Jesu, mein Verlangen, BWV 32, Epiphany 1 Meine Seufzer, meine Tra¨nen, BWV 13, Epiphany 2 Alles nur nach Gottes Willen, BWV 72, Epiphany 3 Der Friede sei mit dir, BWV 158, Puriﬁcation Wir mu¨ssen durch viel Tru¨bsal, BWV 146, Easter 3 Gott fa¨hret auf mit Jauchzen!, BWV 43, Ascension Brich dem Hungrigen dein Brot, BWV 39, Trinity 1 Siehe, ich will viel Fischer aussenden, BWV 88, Trinity 5 Vergnu¨gte Ruh, beliebte Seelenlust, BWV 170, Trinity 6 Es wartet alles auf dich, BWV 187, Trinity 7

Berlin, P 89, St 35

Berlin, P 1047, St 634

Autograph, J. A. Kuhnau et al., for 29 July 1725 J. A. Kuhnau et al., for 19 Aug. 1725? Autograph, J. A. Kuhnau et al., for 26 Aug. 1725 J. C. Altnickol; 1st perf. 23 Sept. 1725 Autograph, J. A. Kuhnau et al., for 31 Oct. 1725 Autograph, J. A. Kuhnau et al., for 25 Dec. 1725 Autograph, J. A. Kuhnau et al., for 26 Dec. 1725 Autograph, J. A. Kuhnau et al., for 27 Dec. 1725 Autograph, J. A. Kuhnau et al., for 30 Dec. 1725 Autograph, C. G. Meißner et al., for 1 Jan. 1726 Autograph, part-autograph, for 13 Jan. 1726 Autograph, part-autograph, for 20 Jan. 1726 Autograph, part-autograph, for 27 Jan. 1726 C. F. Penzel; perf. 2 Feb. 1726?

Berlin, Am.B.538

J. F. Agricola; perf. 12 May 1726?

Berlin, P 44/5, St 36

Autograph, C. G. Meißner et al., for 30 May 1726 Autograph, C. G. Meißner et al., for 23 June 1726 Autograph, C. G. Meißner et al., for 21 July 1726 Autograph, C. G. Meißner et al., for 28 July 1726 Autograph, C. G. Meißner et al., for 4 Aug. 1726 Autograph, C. G. Meißner et al., for 11 Aug. 1726 Autograph, part-autograph, for 25 Aug. 1726 Autograph, part-autograph, for 8 Sept. 1726 Autograph, C. G. Meißner et al., for 22 Sept. 1726

Es ist dir gesagt, Mensch, BWV 45, Trinity 8 Herr, deine Augen sehen, BWV 102, Trinity 10 Geist und Seele wird verwirret, BWV 35, Trinity 12 Wer Dank opfert, der preiset mich, BWV 17, Trinity 14

Berlin, P 46/4

Berlin, P 153, St 92 Berlin, P 144, St 83 Coburg, Berlin St 89 Berlin, P 92, St 37 Berlin, P 45/6 , St 44 Berlin, P 126, St 67 Berlin, P 45/3 , St 69 Berlin, P 54, St 2

Berlin, P 62, St 8 Berlin, P 145, St 85 Berlin, P 154, St 94 Berlin, P 84, St 29, priv. poss. Berlin, P 80, St 26 Berlin, P 97, St 41 Berlin, P 86, St 32 Berlin, P 45/4 , St 101

l e ip z i g c y c l e i i i Es erhub sich ein Streit, BWV 19, Michaelmas, Trinity 15 Wer weiß, wie nahe mir mein Ende, BWV 27, Trinity 16 Wer sich selbst erho¨het, BWV 47, Trinity 17 Gott soll allein mein Herze haben, BWV 169, Trinity 18 Ich will den Kreuzstab gerne tragen, BWV 56, Trinity 19 Ich geh und suche mit Verlangen, BWV 49, Trinity 20 Was Gott tut, das ist wohlgetan, BWV 98, Trinity 21 Ich armer Mensch, BWV 55, Trinity 22 Falsche Welt, dir trau ich nicht, BWV 52, Trinity 23 O ewiges Feuer, o Ursprung der Liebe, BWV 34a, wedding Ach Gott, wie manches Herzeleid, BWV 58, Sun. after New Year Ich habe genung, BWV 82, Puriﬁcation Ich bin vergnu¨gt mit meinem Glu¨cke, BWV 84, Septuagesima

Berlin, P 45/7, St 25 Berlin, P 164, St 105 Berlin, P 163, St 104 Berlin, P 93, St 38 Berlin, P 118, St 58 Berlin, P 111, St 55 Berlin, P 160, St 98 Berlin, P 105, St 50 Berlin, P 85, St 30 Berlin, St 73 Berlin, P 866, St 389, Leipzig TS Berlin, P 114, St 54 Berlin, P 108, St 52

169

Autograph, C. G. Meißner et al., for 29 Sept. 1726 Autograph, part-autograph, for 6 Oct. 1726 Autograph, part-autograph, for 13 Oct. 1726 Autograph, part-autograph, for 20 Oct. 1726 Autograph, C. G. Meißner et al., for 27 Oct. 1726 Autograph, C. G Meißner et al., for 3 Nov. 1726 Autograph, C. G. Meißner et al., for 10 Nov. 1726 Autograph, C. G. Meißner et al., for 17 Nov. 1726 Autograph, C. G. Meißner et al., for 24 Nov. 1726 C. G. Meißner et al., 1726/7 Autograph, part-autograph, for 5 Jan. 1727 Autograph, part-autograph, for 2 Feb. 1727 Autograph, part-autograph, for 9 Feb. 1727?

Bach’s third Leipzig cycle of cantatas is not confined to a single church year but extends over the period 1725–7. Nor is his composition of sacred music as regular as before, with the result that he sometimes performed cantatas by other composers instead of his own. During the long Trinity period of 1725, June to November, new compositions are sparse and for a while Bach seems to have performed cantatas by Telemann—or more likely, perhaps, left a deputy to perform them in his absence.68 His productivity became more regular during the Christmas and Epiphany seasons of 1725–6, but from the Feast of the Purification onwards (2 February 1726) he performed no fewer than eighteen cantatas by his Meiningen cousin Johann Ludwig Bach.69 At first these cantatas were performed exclusively, but from Ascension Day onwards (30 May) Bach interspersed them with newly composed cantatas of his own (BWV 43, 39, 88, 187, 45, 102, and 17)

68 See Wolf Hobohm, ‘Neue “Texte zur Leipziger Kirchen-Music” ’, BJ 59 (1973), pp. 5–32, and Andreas Glo¨ckner, ‘Bemerkungen zu den Leipziger Kantatenauffu¨hrungen vom 3. bis 6. Sonntag nach Trinitatis 1725’, BJ 78 (1992), pp. 73–6. 69 See William H. Scheide, ‘J. S. Bachs Sammlung von Kantaten seines Vetters Johann Ludwig Bach’, BJ 46 (1959), pp. 52–94; BJ 48 (1961), pp. 5–24; and BJ 49 (1962), pp. 5–32; Walter Blankenburg, ‘Eine neue Textquelle zu sieben Kantaten Johann Sebastian Bachs und achtzehn Kantaten Johann Ludwig Bachs’, BJ 63 (1977), pp. 7–25; Konrad Ku¨ster, ‘Meininger Kantatentexte um Johann Ludwig Bach’, BJ 73 (1987), pp. 159–64, and the ¨ berlieferung der Kantaten Johann Ludwig Bachs’, BJ 75 same author’s ‘Die Frankfurter und Leipziger U (1989), pp. 65–106.

170 sacred a nd s ecula r: the vocal works based on librettos drawn from the same cycle, Sonntags- und Fest-Andachten (Meiningen, 1704), attributed to Duke Ernst Ludwig of Saxe-Meiningen. The cantatas of Cycle III are not unified in textual and musical form, unlike the chorale cantatas of Cycle II. Yet certain types may be clearly differentiated—types that are hardly, if at all, represented among Bach’s earlier cantatas—namely the solo cantata, the dialogue cantata, and what might be termed the ‘concertante’ cantata, which is characterized by a significant input of instrumental music. The predominant type of Italian secular cantata, as represented by those of Alessandro Scarlatti, for example, is written for a single solo voice with continuo (or instrumental ensemble). Bach contributed to this secular type periodically (BWV 202, 203, 204), but he also transferred it to the sacred domain in two Weimar cantatas, Nos. 199 and 54, both settings of texts from Georg Christian Lehms’s Gottgefa¨lliges KirchenOpffer of 1711. When Bach revived the solo cantata within the last phase of Cycle III, July 1726 to February 1727 (there are no such cantatas in Cycles I and II), he first reverted to this Lehms publication (Cantatas Nos. 170 and 35) before turning to the work of other librettists (Nos. 169, 56, 55, 52, 82, and 84). In Bach’s hands the solo cantata reflects its secular origin more clearly than that written for four-part choir. One obvious sign of this is the complete absence of biblical texts—the librettos consist almost entirely of madrigalian verse, designed to be set as recitatives and arias. Even the contribution of the Lutheran chorale, so substantial in Bach’s other cantatas, is here minimal. Cantatas Nos. 170, 35, and 82, like the Weimar cantata No. 54, contain no chorale at all, and the other five solo cantatas contain only a plain four-part chorale at the end. In the Weimar cantata No. 199, the chorale is sung by solo soprano, which preserves the integrity of the solo form. The four-part chorale that concludes Cantatas Nos. 169, 56, 55, 52, and 84, on the other hand, necessitates the introduction of three additional voices, so that their form might be described as a ‘modified’ solo cantata. If the chorales may be viewed as a concession to sacred style, two other aspects of these compositions tend rather to reinforce the secular associations of the solo-cantata genre. One is the demanding writing for solo voice, which no doubt exploited the vocal technique and interpretative skills of particular singers; the other is the introduction of concerto movements (Cantatas Nos. 35, 169, and 52) and the use of obbligato organ (Nos. 170, 35, and 169), to which we shall return in due course. Some of these solo cantatas deservedly belong among Bach’s best-loved sacred compositions, particularly Nos. 170, 56, and 82. This is surely to be explained by his profoundly moving portrayal of the various states of the soul that are represented in the text. Vergnu¨gte Ruh, beliebte Seelenlust, BWV 170, presents two opposite states: the opening aria, a pastorale of heart-warming beauty, depicts the ‘contented rest’ and ‘beloved pleasure of the soul’; in the next aria, no. 3, an intricate and subtle piece of chromatic counterpoint (comparable in this respect with the Sinfonia in F minor, BWV 795), the tortured chromaticism of the obbligato organ part portrays ‘the perverted hearts that are so very contrary to God’, while the bassett for unison

l e i p z i g cy cl e i i i

171

upper strings symbolizes the loss of God’s support. A similarly fugal texture in the opening aria of Ich will den Kreuzstab gerne tragen, BWV 56, together with the rising augmented 2nd of the subject and the deeply affecting quaver couplets that follow, conjures up the heavy burden of the cross that the follower of Christ has to carry in imitation of his or her master. In the opening aria of Ich habe genung, BWV 82, whose text is loosely paraphrased from the Nunc Dimittis (Luke 2: 29–32), the musical/verbal motto ‘I have enough’, the siciliana rhythm, and the florid oboe obbligato with its rich string accompaniment—all these things help to convey the mingled sadness and joy that attend the elderly Simeon’s wish to depart from this life now that he has seen the Saviour. In the third movement Simeon looks forward to death in the words ‘Slumber, you tired eyes; close peacefully and blessedly’ (‘Schlummert ein, ihr matten Augen, fallet sanft und selig zu’). These words are set as a lullaby that gives them perfect expression, one in which death is viewed as an entry into ‘sweet peace, quiet repose’. Another significant genre represented in Cycle III is the dialogue cantata. All four of Bach’s dialogues between Jesus and the Soul belong to this cycle: two with texts by Lehms, Cantatas Nos. 57 and 32, date from winter 1725–6; and two with anonymous texts, Nos. 49 and 58, from the following winter, 1726–7. Isolated movements or twomovement sequences within the early and Weimar cantatas possess dialogue character, notably Cantatas 106 (nos. 6–7), 21 (nos. 7–8), 172 (no. 5), and 152 (no. 6).70 But not until the first Leipzig cycle did Bach devote a whole cantata to the dialogue format, namely No. 60, O Ewigkeit, du Donnerwort, which Bach himself entitled ‘Dialogus zwischen Furcht u[nd] Hoffnung’. At the following Easter (April 1724) Bach returned to the dialogue partners Fear and Hope in two movements (nos. 4 and 5) from Cantata No. 66, a sacred parody of a Co¨then serenata. The dialogue form was very popular in seventeenth-century Germany but must have sounded distinctly old-fashioned by the 1720s,71 had not Bach taken special measures to update it using operatic and concertante methods. The type he employed in Cycle III, the Jesus–Soul dialogue, derives from the Song of Solomon where, according to the Christian interpretation, the bridegroom is identified with Jesus (or God or the Holy Spirit) and the bride with the Faithful Soul (or the Church or the congregation). Andreas Hammerschmidt, J. R. Ahle, and W. C. Briegel all published collections of such dialogues from 1645 onwards. In addition, six compositions of this kind have come down to us from Schu¨tz and a further six from Buxtehude. It was customary for the Faithful Soul to be sung by soprano and Jesus by bass, and Bach adheres to this custom in all four of his Jesus–Soul dialogues.72

See Vol. I of the present study, pp. 106, 250, 253, 263–4, and 268–70. Johann Mattheson (Der vollkommene Capellmeister, Hamburg, 1739) says only that ‘Ihr Styl ist etwas madrigalisch’. 72 Regarding Bach’s compositions in dialogue form and their background, see Friedhelm Krummacher, ¨ ber Bachs geistliche Dialoge’, and Michael Ma¨rker, ‘Die Tradition der Jesus-Seele‘Gespra¨ch und Struktur: U Dialoges und ihr Einﬂuß auf das Werk Bachs’, both in K. Lehmann (ed.), Schaffenskonzeption, Werkidee, Textbezug; pub. as Beitra¨ge zur Bach-Forschung, 9/10 (Leipzig, 1991), pp. 45–59 and 235–41 respectively. 71

172

sacred and secular: the vocal works

In Selig ist der Mann, BWV 57, written for St Stephen’s Day (26 December) 1725, the two characters are designated ‘Jesus’ and ‘Anima’ by Bach himself. The opening movement is a deeply felt setting of an authoritative text from the New Testament, ‘Blessed is the man who endures temptation, for after he is tested he will receive the crown of life’ (James 1: 12), sung by the bass as vox Christi. Like many movements of this type, it is not strictly speaking an aria (though so-called), for within the framing ritornellos the bass sings without intermission. The densely polyphonic five-part texture for bass voice and strings, doubled by oboes, anticipates the opening movement of the ‘Kreuzstab’ Cantata, No. 56, performed in the following autumn. The first aria of Anima (the Soul) is not dissimilar, but her second, no. 7, introduces something radically different. Jesus having offered her eternal life, she expresses her longing for the afterlife in the words ‘Ich ende behende mein irdisches Leben. Mit Freuden zu scheiden verlang ich itzt eben’ (‘Swiftly I end my earthly life. Joyfully I now long just to depart’). Lehms here employs the poetic metre of popular song, to which Bach responds by setting it in gigue-passepied rhythm. Thus, while the solo violin obbligato expresses passionate longing for the bliss of the afterlife, both Lehms and Bach employ secular means to deliver the words. This is no doubt connected with the analogy between earthly and heavenly love, central to the Christian interpretation of the Song of Solomon: as bride and bridegroom are united in marriage, so too are Christ and the Soul in heaven. The second dialogue cantata, Liebster Jesu, mein Verlangen, BWV 32, is based on the Gospel account of Jesus’s parents’ losing him in Jerusalem when he was 12 years old (Luke 2: 41–52). Lehms employs the dialogue format to give an allegorical interpretation of these events. Thus, in the opening soprano aria, the Soul has lost Jesus in a spiritual sense. It is no mere coincidence that music of similar character to that which here expresses intense longing for Jesus—a florid Adagio for interacting solo soprano and obbligato oboe, accompanied by broken-chordal string figures—was employed by Bach in the context of a secular wedding cantata, No. 202. For, as the mystical theology of the time might put it, as the bride yearns for the bridegroom so the Soul yearns for Jesus. The recitative–aria sequence for bass, nos. 2–3, may be attributed to the vox Christi, for although only the recitative quotes the Gospel text (Luke 2: 49), it is given a spiritual interpretation in the freely composed words of the aria, ‘Here in my Father’s abode a distressed spirit finds me’ (‘Hier in meines Vaters Sta¨tte, findt mich ein betru¨bter Geist’). The cantata culminates in a Vivace duet, no. 5, in which the newfound unity of Jesus and the Soul is represented by canonic imitation between the two voices. In the two dialogue cantatas from winter 1726–7, Nos. 49 and 58, a key role is played by the chorale-aria. Bach introduces overtly secular elements in the earlier stages of Ich geh und suche mit Verlangen, BWV 49: a long concertante sinfonia and, as culmination of the recitative no. 3, an operatic love duet between bride (the Soul) and bridegroom (Jesus) with split-text dialogue as in the Co¨then serenatas, after which the voices unite in 3rds and 6ths as a sign of their perfect union. Ecclesiastical tradition returns,

l ei p z i g c y c l e i ii

173

however, in the chorale-finale. Bach here unites chorale and aria in an innovative fashion, though there are partial precedents in ‘Bleibt, ihr Engel’, BWV 19 no. 5, performed only five weeks before, and in ‘O Ewigkeit, du Donnerwort’, BWV 60 no. 1, from Cycle I. The latter is essentially a chorale arrangement, despite the brief biblical quotation in one of the voices. In ‘Bleibt, ihr Engel’ a wordless, instrumental rendering of the chorale cantus firmus accompanies the vocal solos, as in many of Bach’s Weimar cantatas. The ritornello form of the aria and the Bar form of the chorale are superimposed. This is also true of the chorale-finale of Cantata No. 49, which may thus be described as a chorale-aria. The crucial difference is that the cantus firmus is no longer instrumental but sung to the words of the hymn—verse 7 of Philipp Nicolai’s Wie scho¨n leuchtet der Morgenstern—by one of the dialogue partners, the soprano (the Soul, the bride of Christ). The other dialogue partner, the bass (Christ, the bridegroom), simultaneously sings madrigalian words that express his love for the Soul in the form of an aria. Chorale, aria, and ritornello themes are all interrelated. Bach returned to this apparently new form of chorale-aria two months later in Ach Gott, wie manches Herzeleid, BWV 58. Here, both outer movements follow the design mapped out in the finale of Cantata No. 49, being conceived as soprano–bass dialogues in which the bass (Jesus) sings the aria theme and its derivatives, while the soprano (Soul) sings the chorale cantus firmus. The theme of the cantata is the familiar Lutheran antithesis between temporal suffering and heavenly joy. This seems to be reflected in the two ritornello themes of the opening chorale-aria, both subsequently taken up by the bass: the dancing dotted rhythms of the initial theme and the chromatic descent of its counterpoint. For the concluding chorale-aria Bach employs the same chorale melody as in the first movement, despite their texts being drawn from different hymns. This congruence of the outer movements has a strongly unifying effect and no doubt explains why the work was later included among the chorale cantatas of Cycle II. The triadic headmotive of the finale’s ritornello recalls the opening of the E major Violin Concerto (BWV 1042) and pervades the entire movement. Its import is clear: it is sung by the bass (Jesus) to the key words ‘nur getrost’ (‘just be of good cheer’) in response to the soprano’s (the Soul’s) singing of the ‘hard journey’ she has before her. Another new departure, largely restricted to Cycle III and later, is Bach’s inclusion of instrumental music borrowed from his concertos and other instrumental works, or else in some cases, perhaps, newly composed.73 An isolated case occurs in the earlier months of the cycle: the opening chorus of the Christmas cantata Unser Mund sei voll lachens, BWV 110. In order to mark the festive occasion this movement is cast in the form of a French overture, like its equivalents in two Cycle I cantatas, Nos. 119 and 194. As in No. 119 again, the text is drawn from the Psalms, and the dotted-rhythm prelude

73 As suggested by Konrad Ku¨ster, ‘Konzertvorlage oder Originalkomposition? Zu den obligaten Orgelanteilen in Bachs Kantaten aus dem Jahr 1726’, in Leisinger (ed.), Bach in Leipzig—Bach und Leipzig, pp. 45–58.

174

sacred and secular: the vocal works

and postlude are purely instrumental, for the voices participate only in the quick, imitative middle section. In this case, however, the vocal (and trumpet) choirs have been added to a purely instrumental movement—that which opens the Ouverture in D, BWV 1069.74 This must count as one of Bach’s most remarkable adaptations, for it would be hard to imagine a more successful musical portrayal of the words from Psalm 126: 2, ‘May our mouth be full of laughter and our tongue full of praise’ (‘Unser Mund sei voll Lachens und unsre Zunge voll Ru¨hmens’). The other cases of borrowed (or newly composed) instrumental music in the cantatas of Cycle III, Nos. 35, 169, 49, and 52, all occurred during the Trinity season of 1726. In each case what is believed to have been a quondam concerto movement (now lost) is adapted to form a sinfonia, and in all but one case (No. 52) the solo part is played on obbligato organ. Independent sinfonias are common among the early and Weimar cantatas, but Cycles I and II almost invariably include integral sinfonias (in other words, the ritornello of the opening chorus doubles as sinfonia). The nonchorale cantatas of Cycle II, however, include two Easter cantatas in successive weeks, Nos. 249 and 42, that open with independent sinfonias on the scale of concerto movements. That of No. 249 started life as such, since it was drawn from a secular vocal work (BWV 249a), and in all probability the same is true of No. 42.75 But the adaptation of actual concerto movements to form cantata sinfonias that took place within Cycle III appears to have been a new venture—one that G. E. Scheibel would have seriously questioned, since he argued that instrumental music sometimes exceeds the listeners’ attention span and understanding, so that it is better to combine voices and instruments around an edifying text.76 Numerous questions arise. How did the church authorities react to the importation of instrumental music into the Sunday morning service on this massive, unprecedented scale? No doubt by late 1726 Bach had built up a powerful position as city music director in Leipzig. Yet he was not impregnable, as the many disputes with his colleagues and superiors demonstrate, though no criticism is recorded over this particular issue. We might also ask what was Bach’s purpose in introducing all this instrumental music into the cantata. The use of obbligato organ might have served as a platform for his own keyboard virtuosity or perhaps for exhibiting the skills of his eldest son Wilhelm Friedemann, then 16 years old. The use of concerto movements might signify a renewed interest in instrumental ensemble music in late 1726, following several years of almost exclusive preoccupation with vocal music—a new interest that would lead to his assumption of the directorship of the Leipzig Collegium musicum a few years later. Finally, it seems unlikely to be mere chance that

74 Regarding the addition of trumpets and drums for the cantata version, see Joshua Rifkin, ‘Klangpracht und Stilauffassung: zu den Trompeten der Ouvertu¨re BWV 1069’, in M. Geck and K. Hofmann (eds.), Bach und die Stile (Dortmund, 1999), pp. 255–89. 75 According to Rifkin, ‘Verlorene Quellen, verlorene Werke’, pp. 65–7. 76 Scheibel, Zufa¨llige Gedancken, trans. by Joyce Irwin (see n. 26), p. 241.

l ei p z i g c y c l e i ii

175

all the compositions with imported concerto-sinfonias are solo or dialogue cantatas. In this context the sinfonia might be considered to compensate for the absence of a choir. Moreover, due to the presence of obbligato organ, the vocal and instrumental spheres are each represented by a soloist of whom great demands are made. Another question that might be asked is whether any justification in the text can be found for the lavish role played by instrumental music. In Geist und Seele wird verwirret, BWV 35, in which both of the two parts are prefaced by a sinfonia, adapted from the first movement and possibly the finale of a lost oboe concerto (a fragment of an unfinished harpsichord arrangement survives in BWV 1059), it is not at all easy to see a clear connection with the text. But the sinfonia of Gott soll allein mein Herze haben, BWV 169, based on the first movement of a lost oboe or oboe d’amore concerto (later adapted as Harpsichord Concerto No. 2 in E, BWV 1053), surely represents ‘the world’ which ‘alone would gladly be the beloved of my soul’ but is rejected, since ‘God alone shall have my heart’ (arioso no. 2). Similarly, the sinfonia of Falsche Welt, dir trau ich nicht, BWV 52, borrowed from the first movement of the Sinfonia in F, BWV 1046a (an early version of the first Brandenburg Concerto), presumably represents the world, for the following recitative, no. 2, opens with the words ‘False world, I do not trust you’. Ich geh und suche mit Verlangen, BWV 49, performed only a fortnight after No. 169, borrows from the same concerto (on this occasion, the finale) for its sinfonia. As we have seen, the text concerns the wedding of Christ as bridegroom with the Soul as bride. The work opens with the bridegroom awaiting the arrival of his bride, so it seems reasonable to interpret the sumptuous concerto-sinfonia as wedding music. In the cantatas with concerto-based sinfonias, Nos. 35, 169, 49, and 52, the obbligato organ (not used in No. 52) is not restricted to the sinfonia but is employed in most of the arias too. During the Trinity season of 1726, when these cantatas originated, obbligato organ was also employed in isolated arias from cantatas that lack a concerto-based sinfonia: Cantatas 170 (nos. 3 and 5), 27 (no. 3; originally harpsichord obbligato), and 47 (no. 2; later violin obbligato). With respect to its origin and genesis, by far the most interesting of the arias with obbligato organ is ‘Stirb in mir’, BWV 169 no. 5. Apart from the French ouverture BWV 110 no. 1, this is the only case among the Cycle III cantatas in which an instrumental piece—the slow movement of the lost concerto (preserved as BWV 1053) whose outer movements formed the sinfonias of Cantatas Nos. 169 and 49—was transformed into a vocal movement. It is a siciliana (so titled in the harpsichord version BWV 1053) whose great beauty and dance connotations represent ‘the world’, which is nonetheless rejected in the words ‘Die in me, world and all your love’. The solo alto and obbligato organ do not collaborate in a duet but rather present different forms of one and the same melodic line in a kind of heterophony. The richly decorative effect of this mode of writing no doubt illustrates the glittering external beauty that the world loves, as opposed to the inner beauty of ‘the love of God’. Cycle III includes a single chorale cantata, Lobe den Herren, BWV 137, which belongs to the per omnes versus type in which the hymn is preserved in its original

176

sacred and secular: the vocal works

wording throughout. Bach had rarely cultivated this type before—only the early Cantata No. 4 and the Cycle II cantata No. 107 are known—but it was to become standard in his late chorale cantatas. The concertante opening chorus of No. 137, a festive movement of great splendour, hardly differs in form or style from the great chorale-choruses of Cycle II. The three inner movements, though designated ‘aria’, are closer to chorale arrangements than arias, since not only the hymn text but its associated melody is retained. The second movement, a chorale trio, resembles Cantata 6 no. 3 from earlier in the same year (1725). In both cases, an elaborate concertante part for string solo (Cantata No. 6: violoncello piccolo; No. 137: violin) furnishes both ritornellos and an accompaniment to the plain or lightly decorated chorale cantus firmus, delivered by solo voice. Both were later adapted for organ and included among the Schu¨bler Chorales (BWV 649 and 650, 1748/9). The third movement of Cantata No. 137, a soprano–bass duet with two obbligato oboes, is as far removed as one could imagine from the operatic duet type that Bach often cultivated elsewhere. The ritornellos and vocal paragraphs all open with fugal entries in stretto, and the theme they share is a minor-mode version of the first two bars of the chorale melody. The fourth movement is a chorale-aria of the Weimar type: a tenor aria with continuo is combined with a wordless, instrumental rendering of the chorale cantus firmus, played here by solo trumpet. Not only the opening movement of Cantata No. 137 but the other chorale-choruses from Cycle III, those of Nos. 28, 16, 27, and 98, on the whole adhere to the patterns established in Cycle II. The second movement of Cantata No. 28 is a motet-style arrangement of the chorale Nun lob, mein Seel, den Herren, marked ‘Alla breve’, with soprano cantus firmus and with the voices doubled by cornett and trombones (as well as oboes and strings), which lends great weight and solemnity to the texture. Similar chorale-choruses in motet style feature in several cantatas from Cycle II (Nos. 2, 38, and 121). Two days later, on New Year’s Day 1726, Bach presented a modernized version of the chorale motet (as in the Cycle II cantatas Nos. 135 and 101) as the opening movement of Herr Gott, dich loben wir, BWV 16. Whereas the inner instrumental parts largely double the voices, the outer ones, first violin (with doubling oboe) and continuo, have independent parts, but in no way do they differ in style from the vocal counterpoints. Neither ritornellos nor instrumental interludes are included, but merely a four-bar continuo introduction whose figures furnish much of the highly motivic counterpoint to the chorale throughout. The opening chorale-chorus of Cantata No. 27, Wer weiß, wie nahe mir mein Ende, differs from its equivalents in Cycle II in that the chorale is troped by recitative insertions, a clear reference back to Cycle I (Cantatas Nos. 138, 95, 190, and 73). An extremely expressive, imitative duet for two oboes over a motivic string accompaniment in the opening ritornello establishes a mood of great depth and seriousness, highly appropriate for a text that considers the end of our earthly existence. Only about six months after the first performance of this cantata (on 6 October 1726), similar expressive means would be used to describe Christ’s death on the Cross in the St Matthew Passion (11 April 1727).

l ei pz ig c yc le ii i

177

As in Cycle I and the non-chorale cantatas of Cycle II, the chief formats for biblicaltext settings are the opening chorus and the vox Christi bass solo. Three cantatas from the early months of Cycle III, Nos. 148, 79, and 110, open with great psalm choruses. And for Ascension Day and the Trinity season of 1726 Bach embarked on a new venture that involved setting two biblical texts in each cantata, namely the composition of seven cantatas—Nos. 43, 39, 88, 187, 45, 102, and 17—to texts drawn from the same cycle as the eighteen cantatas by his relative Johann Ludwig Bach that he performed during that year, a cycle whose texts were perhaps written by J. L. Bach’s employer Duke Ernst Ludwig of Saxe-Meiningen.77 These cantatas are in two parts, the first headed by an Old Testament text and the second by one from the New Testament. In all but one case (No. 88) Bach set the opening Old Testament words as a chorus of large dimensions. After this series, only one further Cycle III cantata opens with a biblical-text chorus, namely No. 47. The festive song of praise that opens the Ascension cantata Gott fa¨hret auf mit Jauchzen, BWV 43, may be construed as a massive prelude and fugue, a form that Bach had employed in some of his earliest Cycle I cantatas (Nos. 75, 76, 24, 105, and 46). In every other case he finds new ways of uniting ritornello form and fugue, as in some of the later Cycle I cantatas (Nos. 136, 69a, 40, 65, 67, and 104). Several of those Cycle I choruses exhibit an ABA1 reprise structure: ritornello plus derived vocal music— central fugue—ritornello plus derived vocal music (Nos. 69a, 40, and 65). Bach returns to this structure in some of the equivalent movements from Cycle III (Nos. 148, 79, 187, 45, and 47). At the end of both framing paragraphs or at the end of the whole movement only, the derived vocal music usually culminates in a ritornello return with inbuilt choral parts (that is, employing the technique of Choreinbau), one of Bach’s most effective and powerful resources. In Cantata No. 187, the subject of the central fugue is new in relation to the surrounding ritornello material. No. 148, however, includes two central fugues with different subjects, drawn from the antecedent and consequent phrases respectively of the opening ritornello, which is itself fugal in structure. The opening movement of No. 45 is still more obviously monothematic and the reprise structure is here expanded, so that each of the three vocal paragraphs takes the form: episode—fugue—ritornello with Choreinbau. The most remarkable of these reprise structures is that which opens the Reformation Festival cantata No. 79, Gott der Herr ist Sonn und Schild. Both the massive integral sinfonia and the following chorus are constructed in ABA1 reprise form with central fugue. The ritornello-sinfonia is exceptionally long, perhaps reflecting the interest Bach showed elsewhere in incorporating instrumental music into the cantatas of Cycle III. Integration is achieved here not by monothematicism but by thematic combination. In the sinfonia, the main theme for two horns and timpani is followed by the instrumental fugue with its own subject, after which the two themes are

See Konrad Ku¨ster, ‘Meininger Kantatentexte um Johann Ludwig Bach’, BJ 73 (1987), pp. 159–64.

178 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s combined. The following chorus introduces its own vocal theme in combination with the fugue subject, and after the vocal fugue all three themes are combined. Cantata No. 47, Wer sich selbst erho¨het, opens with an almost equally long and complex ritornello-sinfonia, which recurs at the end with inbuilt vocal parts, forming an outer frame (AA). The massive central fugue, comprising three expositions, is built on a subject that rises and falls through an octave to illustrate Christ’s words ‘Whoever exalts himself shall be abased’ (Luke 14: 11). It is accompanied by a regular countersubject that falls and rises through an 11th or 12th, illustrating the continuation of the New Testament text, ‘and whoever humbles himself shall be exalted’. Cantatas Nos. 17, 102, and 39 involve ritornello, fugue, and Choreinbau in the structure of their opening biblical-text chorus, but they lack the ABA1 reprise scheme. In No. 17, Wer Dank opfert, der preiset mich, a substantial sinfonia-ritornello introduces a choral fugue that culminates in ritornello-cum-Choreinbau (keys I–V). After a brief modulatory episode, the same choral sequence returns in subdominant recapitulation (IV–I), so that the overall form is bipartite, AA1. The opening chorus of Cantatas Nos. 102 and 39 in each case takes the form ABC and is thus related to the motet principle of juxtaposing a series of different, often contrasting sections in accordance with the import of the text. In Herr, deine Augen sehen nach dem Glauben, BWV 102, however, ritornello modes of structuring are employed to integrate the disparate elements. In Brich dem Hungrigen dein Brot, BWV 39, on the other hand, ritornello structuring applies only to the A-section of the ABC structure, which is based on Isaiah 58: 7–8: ‘Break your bread with the hungry, and bring those who are in distress into your house.’ The opening ritornello, whose antiphonal exchanges between recorders, oboes, and strings presumably represent the sharing out of bread, is afterwards furnished with vocal parts in order to frame a fugue whose intricately wrought subject sets the same words once more. All this constitutes the A-section, which itself falls into the ABA1 structure characteristic of entire biblical-text movements. Here, however, it is followed by two further sections in different metres (i and 3/8), B and C. Section B is a mere interlude, a responsorial treatment of the next sentence, ‘If you see someone naked, then clothe him, and do not avoid your own kin’, in the concertato style of the seventeenth century. Section C falls into its own reprise structure, but unlike A is no longer ritornello-based. The two framing paragraphs, fugues on slightly different versions of the same subject, are linked by a central homophonic passage. In general, ritornello form is in this movement marginalized in favour of motet-like modes of structuring, though reprise form gives the outer sections internal coherence. Only in one case among the seven Meiningen-text cantatas is the opening biblical text set as an aria rather than a chorus, namely the ﬁrst movement of Siehe, ich will viel Fischer aussenden, BWV 88. The explanation is simple: Jeremiah (16: 16) here quotes the words of the Lord in the ﬁrst person singular, so it is appropriate to have them sung by the bass as vox Dei. Once again, motet tradition lies behind the movement’s division into two halves that contrast to the strongest degree imaginable, in key,

l ei pz ig c yc le ii i

179

metre, tempo, theme, texture, and instrumentation. In each part Bach presents a vivid tone-picture: first a pastorale, similar to that which would open Vergnu¨gte Ruh (BWV 170) on the following Sunday, but here sketching a lakeside scene, hence the wave figures in strings and oboes; and then a hunt, hence the addition of a pair of hunting horns to the instrumental ensemble. It has already been noted that the Meiningen-text cantatas fall into two parts, of which the first opens with an Old Testament text, the second with one from the New Testament. The New Testament text, the fourth movement in each case, is in the first and last cantatas of the series, Nos. 43 and 17, a passage of Gospel narrative set as secco recitative. In No. 17 Luke’s account of the healing of the ten lepers (17: 11–19) is sung by the tenor, the traditional voice of the Evangelist. In No. 43, on the other hand, the account of the Ascension from Mark 16: 19 is unexpectedly sung by soprano, perhaps because a high voice was deemed appropriate for the image of the Lord being ‘lifted up into heaven’. In the other five cantatas in the series, the solo bass represents the vox Christi. This is so even where the text is drawn from the Epistles (Nos. 39 and 102), since the words, being so clearly inspired by Jesus Christ, are consequently deemed to be spoken by him. In Cantatas Nos. 39 and 88, the vox Christi movements are bass ariosos with continuo accompaniment. Those of Nos. 187, 45, and 102, on the other hand, though designated ‘Arioso’ (except in No. 187), are aria-like in dimensions and accompanied by obbligato strings, which introduce the movement with a substantial ritornello and harbour the chief thematic content. In No. 187, as in the earlier Leipzig cantatas Nos. 86 and 87, the ritornello is constructed as a fugal exposition in which the bass subsequently joins, imparting an air of authority and solemnity to the delivery of Christ’s words. Bach’s setting of Jesus’s words ‘Fu¨rchte dich nicht’ (‘Fear not’) in Siehe, ich will viel Fischer, BWV 88 no. 4, recalls Schu¨tz’s style of word-setting and, like so many biblical solos from Cycles I–III, illustrates the abiding importance for Bach of the monodic style developed in the seventeenth century by Schu¨tz and his contemporaries. One noteworthy feature of Cycle III is the use of verbal mottos to link several movements together: Cantatas 79 (nos. 1–2), 43 (nos. 8–10), 169 (nos. 2–3), 56 (nos. 1 and 4), 49 (nos. 2–3), 55 (nos. 3–4), and 82 (nos. 1–2). In two cases Bach establishes a musical link between the movements concerned. In the ‘Kreuzstab’ Cantata, No. 56, the last two lines of the opening aria, ‘Da leg ich den Kummer auf einmal ins Grab, da wischt mir die Tra¨nen mein Heiland selbst ab’ (‘There I lay my sorrows all at once in the grave; there my Saviour Himself wipes away my tears’), are quoted at the end of the recitative no. 4, immediately before the concluding chorale. The associated music from the opening aria (bb. 143–51) returns, somewhat varied, as arioso in the recitative. This has the effect of emphasizing the Christian view of death, since the deeply moving music linked to it is heard twice, knitting the whole cantata together. A parallel case occurs in the dialogue cantata Ich geh und suche mit Verlangen, BWV 49, performed on the following Sunday. The first two lines of the bass aria no. 2, ‘Ich geh und suche mit Verlangen dich, mein Taube, scho¨nste Braut’ (‘I go and seek you

180

sacred and secular: the vocal works

with longing, my dove, my fairest bride’), recur in the following dialogue-recitative, set to a variant of their original melody. On occasion Bach creates purely musical interconnections between movements, links not specifically motivated by the text, as in Cantatas 164 (nos. 1 and 5), 79 (nos. 1 and 3), and 110 (nos. 1 and 4). In the great Reformation Festival cantata No. 79, Gott der Herr ist Sonn und Schild, for example, Bach links the authoritative biblical and chorale statements of the first and third movements respectively by employing the ritornello theme of the one—festive music for two horns and timpani—as chorale accompaniment in the other. Finally, mention must be made of a special case. A cantata of the highest quality that nonetheless might easily be overlooked due to its small dimensions is Der Friede sei mit dir, BWV 158, which is known to have been performed on Easter Tuesday and at the Feast of the Purification.78 It cannot be dated on the basis of the existing sources, but on internal grounds there are strong reasons for supposing that it might have originated within Cycle III, either in 1726 or in early 1727. The opening bass recitative presents an alternation of arioso (a) and secco (b and c) in the form a b a1 c a2, just like the equivalent movement in Cantata 169 (no. 2: a b a1 c a2 d). And in both cases the arioso highlights a motto text, sung at the beginning, middle, and end to variants of the same theme. In Cantata No. 158 the motto text is ‘Der Friede sei mit dir’ (‘Peace be with you’), Christ’s first words on His appearance to the disciples after the Resurrection (Luke 24: 36). The bass might be viewed as the vox Christi, were it not that Jesus’s arioso is troped by a commentary, sung as secco by the same singer. The second movement is a chorale-aria of a type not found in Bach before Cycle III, where it occurs in Cantatas 49 (no. 6) and 58 (nos. 1 and 5). Like those movements, it is a soprano–bass duet in which the aria is sung by the bass and the chorale by the soprano. In this case, however, there is no Jesus/Soul identification, nor are the two singers conceived as separate characters of some other kind. The first line of the aria and chorale is identical, ‘Welt, ade, ich bin dein mu¨de’ (World, farewell, I am weary of you’), and soprano and bass share their sentiments throughout, echoing the thoughts of Simeon as he approaches death. The bass recitative no. 3 ends by quoting in full the last two lines of the aria text from the second movement, ‘Da bleib ich, da hab ich Vergnu¨gen zu wohnen, da prang ich gezieret mit himmlischen Kronen’ (‘There I remain, there I delight to dwell, there I shine forth, adorned with a heavenly crown’). In the third movement, these words are sung in bass arioso to music closely related to the setting of the same words in the chorale-aria. This joint textual and musical link between movements clearly parallels those that have already been noted in Cantatas 56 (nos. 1 and 4) and 49 (nos. 2–3). In sum, then, Der Friede sei mit dir exhibits enough of the characteristics of the late Cycle III cantatas to render an origin in 1726/7 highly likely.

78 The few conclusions that can be tentatively drawn about the history of the work are summarized in Du¨rr, The Cantatas of J. S. Bach, pp. 289–90.

st matthew passion

181

St Matthew Passion Title, occasion

Earliest source/s

Scribe, date

Passio unseres Herrn Jesu Christi nach dem Evangelisten Mattha¨us, BWV 244, Good Friday

Berlin, P 25, St 110

Autograph, part-autograph, 1st perf. 11 Apr. 1727

The first performance of the St Matthew Passion is most likely to have taken place on Good Friday (11 April) 1727,79 and it was probably revived two years later, on 15 April 1729. We do not know exactly what form the original version took, since the surviving score and parts transmit the later, definitive version of 1736. However, a copy of the score survives that must be derived from the lost autograph of the pre-1736 version.80 This shows that the conclusio of Part I was originally a four-part chorale, ‘Jesum laß ich nicht von mir’—the great chorale-chorus ‘O Mensch, bewein dein Su¨nde groß’ was not transferred from Version II of the St John Passion until 1736. Another major difference concerns the continuo: whereas Bach specifies for the 1736 version two spatially separated continuo groups, one for each ‘choir’ (that is, vocal-instrumental ensemble), the early version has a single continuo group, common to both choirs. Other differences between the 1727/9 and 1736 versions are concerned with matters of vocal or instrumental scoring and with increased elaboration of the kind that Bach was accustomed to add in the course of revision. The text is made up of three interwoven components: the biblical account of the Passion in Matthew 26–7; numerous Lutheran chorales; and a poetic commentary by Bach’s usual Leipzig librettist Picander (Christian Friedrich Henrici), which was published in Vol. II of his collected verse, Ernst-Scherzhaffte und Satyrische Gedichte (Leipzig, 1729). As in the St John Passion, each event in the biblical narrative is first set in recitative, with choral interpolations for crowd utterances where necessary, then commented upon in chorales and in Picander’s verse, set as accompagnato, arias, and choruses. This format allows the work to be notionally divided into acts and scenes as illustrated in the following table (the commentary is shown in italics): Part I No. 1: Exordium: chorale-chor. Act I: Prehistory of the Passion Sc. 1: Jesus’s announcement of his Passion Nos. 2–3: recit.—chorale A 79 See Joshua Rifkin, ‘The Chronology of Bach’s St. Matthew Passion’, Musical Quarterly, 61 (1975), pp. 360–87. In the preface to the libretto for the Berlin centenary performance in 1829, Carl Friedrich Zelter mentioned an old church text (‘alte Kirchentext’) that referred to a performance at Good Friday Vespers in the Thomaskirche, Leipzig, in 1729. This was probably a revival. 80 Berlin, Am.B.6/7 in the hand of Johann Christoph Farlau, c. 1756. Of course, we cannot be sure that the 1727 and 1729 versions were identical, nor which of the two Farlau reproduces.

182 sacred and s ecular: the vocal works Sc. 2: Assembly of the chief priests Nos. 4a–b: recit.—turba 1 Sc. 3: Anointing of Jesus at Bethany Nos. 4c–6: recit.—turba 2—recit.—accompagnato—aria Sc. 4: Judas’s betrayal Nos. 7–8: recit.—aria Act II: The Last Supper Sc. 1: Preparations for the Passover Nos. 9–11 (b. 16): recit.—turba 3—recit.—turba 4—chorale B—recit. Sc. 2: The Sacrament of Holy Communion Nos. 11 (b. 16) –13: recit.—accompagnato—aria Sc. 3: At the Mount of Olives Nos. 14–15: recit.—chorale C Sc. 4: Jesus foretells Peter’s denial Nos. 16–17: recit.—chorale C Act III: The Garden of Gethsemane Sc. 1: The agony in the garden I Nos. 18–20: recit.—accompagnato + chorale A—aria + chor. Sc. 2: The agony in the garden II Nos. 21–3: recit.—accompagnato—aria Sc. 3: The agony in the garden III Nos. 24–5: recit.—chorale D Sc. 4: Jesus’s betrayal and arrest Nos. 26–8: recit.—duet + chor.—chor.—recit. No. 29: Conclusio (chorale-chor.) Part II No. 30: Exordium (aria + chor.) Act I: Jesus before the Chief Priests Sc. 1: False witness against Jesus Nos. 31–5: recit.—chorale F—recit.—accompagnato—aria Sc. 2: Trial, condemnation, and mockery Nos. 36–7: recit.—turba 5—recit.—turba 6—chorale Sc. 3: Peter’s denial Nos. 38–40: recit.—turba 7—recit.—aria—chorale G Sc. 4: Judas’s repentance Nos. 41–43 (b. 16): recit.—turba 8—recit.—aria—recit. Act II: Jesus before Pontius Pilate Sc. 1: The charge against Jesus Nos. 43 (b. 16) -44: recit.—chorale C Sc. 2: First call for Cruciﬁxion Nos. 45–6: recit.—turba 9a—chorale A

s t m a t t h ew pa s s io n

183

Sc. 3: The defence of Jesus Nos. 47–9: recit.—accompagnato—aria Sc. 4: Second call for Cruciﬁxion Nos. 50–2: recit.—turba 9b—recit.—turba 10—recit.—accompagnato—aria Sc. 5: The crown of thorns Nos. 53–4: recit.—turba 11—recit.—chorale C Sc. 6: Via crucis Nos. 55–7: recit.—accompagnato—aria Act III: Golgotha Sc. 1: The Cruciﬁxion Nos. 58–60: recit.—turba 12—recit.—turba 13—recit.—accomp.—aria + chor. Sc. 2: Jesus dies on the Cross Nos. 61–2: recit.—turba 14a—recit.—turba 14b—recit.—chorale C Sc. 3: The earthquake Nos. 63a–b: recit.—turba 15 Sc. 4: Joseph of Arimathea Nos. 63c–65: recit.—accompagnato—aria Sc. 5: Burial of Jesus Nos. 66–7: recit.—turba 16—recit.—accompagnato + chor. No. 68: Conclusio (chor.) In the course of devising his poetic commentary Picander drew upon a variety of sources. The exordium to Part II, no. 30, quotes from the Song of Solomon 6: 1. About half of the aria texts are indebted to a series of eight Passion sermons by the seventeenthcentury Rostock theologian Heinrich Mu¨ller, published in his posthumous collection of sermons, of which Bach possessed a copy.81 Two poems by Bach’s Weimar librettist Salomo Franck were used in the accompagnati nos. 5 and 64. The Brockes Passion of 1712, already heavily plundered in the St John Passion, provided the text for the ‘thunder and lightning’ chorus no. 27b. Finally, Picander drew from two earlier poetic works of his own: an epic poem of 1726 for the aria ‘Blute nur’, no. 8; and his Passion-oratorio of 1725 for three arias (nos. 13, 39, and 49), an accompagnato (no. 19), and the conclusio to Part II (no. 68). In setting this heterogeneous text, Bach perhaps drew as near as possible to the concept of the Passion as Gesamtkunstwerk (complete work of art) and to the Passionoratorio genre while retaining the biblical text in full and unaltered. In his libretto for the St Matthew Passion, Picander subscribed to a prominent feature of the Hamburg Passion-oratorio, as represented by Hunold’s Der blutige und sterbende Jesus and Brockes’s Der fu¨r die Su¨nde der Welt gemartete und sterbende Jesus, namely the dialogue format, already imitated by Picander in his own Passion-oratorio

81 See Elke Axmacher, ‘Ein Quellenfund zum Text der Mattha¨us-Passion’, BJ 64 (1978), pp. 181–91, and her ‘Aus Liebe will mein Heyland sterben’: Untersuchungen zum Wandel des Passionsversta¨ndnisses im fru¨hen 18. Jahrhundert (Neuhausen and Stuttgart, 1984), pp. 28–52 and 166–203.

184 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s libretto Erbauliche Gedancken auf den Gru¨nen Donnerstag und Charfreytag u¨ber den Leidenden Jesum, in einem Oratorio of 1725. Typically, the dialogue takes place between an individual, the ‘Tochter Zion’ (Daughter Zion), and a group, ‘Die Glaübigen’ (the Faithful). The ‘Daughter Zion’ is an allegorical figure who in the Old Testament personifies the city of Jerusalem. In Christian tradition she came to symbolize the Church, the bride of Christ (Ephesians 5: 22–33).82 In the St Matthew Passion, Picander introduces ‘Tochter Zion’ and ‘Die Glaübigen’ at corresponding points in Parts I and II: in the exordium and at the beginning and end of the last ‘act’ (Part I, nos. 1, 19–20, and 27; Part II, nos. 30, 59–60, and 67–8). In setting Picander’s text Bach apparently ignored the names his librettist had attached to the individual and collective dialogue partners. This is possibly due to their origin in the theatrical Passion-oratorio genre—Bach might have considered them inappropriate in a liturgical context. Accordingly, the part that Picander gave to the Daughter Zion is not restricted by Bach to a single singer, nor even to voices in the female range. It thus ceases to exist as a clearly defined part in its own right. On the other hand, the dialogue format—fundamental to certain Cycle III cantatas in the period immediately preceding the Passion—was a crucial part of Bach’s conception of the work. It led to his decision to employ two vocal-instrumental ensembles, Chorus I and II, both flexible enough to tackle arias as well as choruses but differing clearly in technical expectations. The singers of Chorus I were concertists, capable of performing Bach’s most demanding arias; those of Chorus II, on the other hand, were ripienists—primarily choral singers, who could nonetheless tackle relatively simple arias.83 It seems likely that Bach’s conception gradually evolved towards the independence of the two choirs.84 In the early version, with its single continuo group, all the performers were probably situated in one location. Bach’s allocation of a separate continuo to each of the two Choruses in 1736 permitted their spatial separation. In Bach’s setting, every one of the dialogue movements prescribed by Picander is to be performed with Chorus I in a leading role and Chorus II in a subordinate role. A prominent form of dialogue in certain movements (nos. 1 and 60) is the sequence ‘imperative—question—response’. Picander clearly found his model for this sequence in an aria from the Brockes Passion, ‘Eilt, ihr angefochtne Seelen’, whose text had been borrowed almost word-for-word in the St John Passion, no. 24.85 Here we encounter the sequence ‘Kommt!—Wohin?—nach Golgotha’. Such forms of dialogue are very effective in injecting a dramatic element into the lyrical meditations. Furthermore, they arouse the visual imagination, for we are invited to visualize whatever is pointed out in the exchange. In the recitative–aria sequence nos. 59–60 from the St Matthew 82

For further details, see Marissen, Bach’s Oratorios, p. 29 n. 1. See Daniel R. Melamed, ‘The Double Chorus in J. S. Bach’s St. Matthew Passion, BWV 244’, Journal of the American Musicological Society, 57 (2004), pp. 3–50, and the same author’s Hearing Bach’s Passions (Oxford, 2005), pp. 49–65 (esp. 52). 84 Melamed, Hearing Bach’s Passions, p. 65. 85 The text of Brockes’s aria is given in Du¨rr, J. S. Bach: St. John Passion, p. 45. 83

s t m a t t h ew pa s s io n

185

Passion, the Daughter Zion first points out Golgotha, the scene of the Cross, in the framing words of her recitative, then asks us to visualize Jesus’s loving gesture from the Cross. The exchange between ‘Tochter Zion’ and ‘Die Glaübigen’ that follows is very similar to that of ‘Eilt’ (St John Passion, no. 24) and set by Bach in a similar fashion: ‘Kommt!—Wohin?—In Jesu Armen’. In both Passions the questions are sung in brief choral interjections in the course of an ongoing meditative aria. Fundamental to the contemplative atmosphere in ‘Sehet’ (St Matthew Passion, no. 60) are the semiquaver couplets that adorn the second phrase of the ritornello, bb. 5–8, played in 3rds or 6ths by the two oboes da caccia. This anticipatory-note figure is present in the instrumental accompaniment to ‘O Mensch, bewein dein Su¨nde groß’ (see the section ‘St. John Passion, Version II’). Indeed, their ritornellos are at times so similar as to border on the identical (Ex. 8). Of course, they were probably composed no more than two years apart; and it is natural that Bach should have had the earlier Passion in mind when he wrote the later one. But the resemblance also encourages the belief that when ‘O Mensch, bewein’ was transferred to the St Matthew Passion in 1736 it found its natural home. Dialogue between the Daughter Zion and the Faithful also plays a prominent part in the opening chorus of the Passion, where a similar question–answer formula is employed: ‘Sehet!—Wen?—Den Braütigam’ (‘Look!—At whom?—At the Bridegroom’), a reference to Paul’s image of the Church as Christ’s bride in Ephesians 5: 22–33. As in the arias ‘Eilt’ and ‘Sehet’, the questions are set as brief choral interjections, but the imperative and response, though attributed by Picander to ‘die Tochter Zion’, are sung by the full vocal-instrumental ensemble of Chorus I. In keeping with Picander’s attribution and the poetic text he devised, he entitled the movement ‘Aria’. This is highly

Ex. 8

a) St. Matthew Passion, ‘Sehet’ (No. 60), bb. 48b–51, oboe da caccia I, II and continuo (ﬁguring omitted)

b) St. Matthew Passion, ‘O Mensch, bewein’ (No. 29), bb. 11b–14, ﬂute I, II and continuo (oboes and strings omitted)

186 sacred a nd s ecula r: the vocal works relevant to Bach’s setting, for the closest analogue to the movement’s form lies in a certain type of chorale-aria cultivated in the St John Passion (nos. 32, ‘Mein teurer Heiland’, and 11+, ‘Himmel reiße’) and in certain dialogue cantatas from Cycle III, performed not long before the St Matthew Passion, Cantata 49 no. 6 (3 November 1726) and 58 nos. 1 and 5 (5 January 1727). In every case, chorale and aria texts are sung simultaneously by different protagonists, each to its own type of music—the chorale verse to its associated melody and the aria text to elaborate, freely invented music. In the Exordium from the Passion, the aria text is mostly sung by Chorus I, though with brief interjections from Chorus II as described. Chorus I takes up the great double fugue from the opening ritornello, turning it into a massive, solemn chorus of lamentation. But we are also invited to witness the procession to Golgotha—‘Sehet ihn aus Lieb und Huld Holz zum Kreuze selber tragen’ (‘Look at Him, out of love and favour, bearing the wood of the Cross Himself ’)—hence the underlying trochaic rhythm. If the ‘aria’ represents our subjective response to the Passion in the form of a great lament, the chorale represents the objective element of doctrinal statement and theological truth. Accordingly, it is clearly set off from the rest of the texture: it lies outside Chorus I and II, being sung as a plain cantus firmus by ripieno soprano doubled by organ. Furthermore, it is in the major key of G, as distinct from the E minor of the surrounding aria-lament. The chorale, ‘O Lamm Gottes, unschuldig’, a German paraphrase of the Agnus Dei (Nicolaus Decius, 1531), incorporates some of the main themes of the Passion: Christ as the innocent Lamb, his forbearance in suffering, his taking upon himself the burden of human iniquity, and the human response in the prayer for mercy ‘Erbarm dich unser, o Jesu’. The image of Christ as the innocent Lamb explains other features of the instrumental music, uniting ‘aria’ and chorale: the slow 12/8 rhythms of the pastorale and the leading role taken by the ‘pastoral’ woodwind instruments, flutes and oboes. Prompted by Picander’s dialogue format, Bach uses a related technique of choral interjection in the meditative aria that immediately follows Jesus’s arrest, ‘So ist mein Jesus nun gefangen’, no. 27a. This is a lament on Jesus’s capture: the soprano and alto of Chorus I, who together represent ‘die Tochter Zion’, express intense grief in a sublime duet, accompanied by unison flutes and oboes with a bassett for unison strings. Since the bassett plays a ‘binding’ figure, the absence of continuo perhaps symbolizes the lack of divine support for Christians due to the capture of their Lord. Upon this meditative surface Chorus II, representing ‘Die Glaübigen’, bursts out in furious, sharply rhythmic interjections calling for Jesus to be freed. Nowhere else does Bach so completely contravene the contemporary doctrine of unity of affect. For this is more than a dramatic interruption of quiet contemplation: it is a sharp clash between opposing responses to Jesus’s arrest, namely extreme sorrow and fierce anger. The Daughter Zion and the Faithful (Chorus I and II) then unite to call down thunder and lightning upon Jesus’s betrayer in the fugal chorus ‘Sind Blitze, sind Donner’ (No. 27b) that follows without a break. After the opening tutti exposition, Chorus I and II split up into antiphonal exchanges to represent storm and fury breaking out on all sides.

st matthew passion

187

Bach enriches the other dialogue movements from the Passion by incorporating elements of motet (no. 30), chorale (no. 19), and refrain (nos. 20, 67, and 68). The exordium to Part II, the aria ‘Ach, nun ist mein Jesus hin’, no. 30, comes nearest to a ready identiﬁcation of the singers with the Daughter Zion and the Faithful, but only in the revised version of 1736. In the earlier version, the Daughter Zion was sung by solo bass, despite being addressed by the Faithful as ‘o du Scho¨nste unter den Weibern’ (‘O you most beautiful among women’). It might have been the incongruity of having the person so addressed sung by the bass voice that induced Bach to transfer the part to solo alto in the course of the 1736 revision. The soloist and instrumental ensemble, drawn from Chorus I, perform a deeply affecting aria, expressing anguish at the loss of Jesus, who has now been taken captive. The four-part choir of Chorus II (the Faithful), doubled by strings, respond by singing a biblical-text motet that shows their deep sympathy in the words of the Song of Solomon 6: 1. Aria and motet alternate in a rondeau-like structure (a b a1 c a2 d a3). The motet episodes are constructed in stretto fugue, each on the basis of a different subject in accordance with the series principle of the motet. Bach would have considered this style appropriate to the biblical authority of the text. The aria-motet form as a whole seems to represent a new departure—there is no obvious parallel in the cantatas of Cycles I–III. The words of Daughter Zion and the Faithful also receive a responsorial setting in ‘O Schmerz’, no. 19, but here Zion is sung as a motivic accompagnato by the tenor, recorders, and oboes da caccia of Chorus I, while Chorus II (the Faithful) respond by singing verse 3 of the chorale Herzliebster Jesu in a four-part setting with doubling strings. Chorus I and II alternate throughout, as in the Exordium to Part II. While ‘O Schmerz’ is in certain respects unique, there are numerous precedents for its basic form in the chorale-tropes of Cycle II. The tenor represents the subjective element of personal feeling—the individual soul expressing great anguish over Jesus’s distress. The chorale, on the other hand, represents the impersonal, objective element of Church doctrine, and as such looks for the cause of Jesus’s distress—the sinfulness of humankind. In the following movement, ‘Ich will bei meinem Jesu wachen’, no. 20, we encounter a new form of dialogue in which solo aria is answered by choral refrain. Daughter Zion is again sung by the tenor of Chorus I (with blithe disregard for dramatic realism) but now in an aria, accompanied by obbligato oboe and continuo. Chorus II respond with refrains, sung by four-part choir doubled by ﬂutes and strings. Since the refrains are later expanded to form lengthy paragraphs in motet style, the closest parallel lies in the aria-motet no. 30. The contrasting music for Chorus I and II is explained by the text: Zion (tenor, Chorus I) sings, ‘I will stay awake with Jesus’ to an alert, determined theme, to which the Faithful (Chorus II) respond, ‘Then our sins will fall asleep’, hence the lullaby quality of the refrains, whose quaver-couplet motive is nonetheless derived from the aria-ritornello (b. 3), so that the antithetical elements are integrated. Picander attributes two more movements to the dialogue partners, namely the last two movements of the whole Passion, nos. 67 and 68. The penultimate movement,

188

sacred and secular: the vocal works

‘Nun ist der Herr zur Ruh gebracht’, no. 67, unites the motivic accompagnato of ‘O Schmerz’ (no. 19) with the choral refrains of ‘Ich will bei meinem Jesu wachen’ (no. 20). As in the last-named movement, the refrains, sung by Chorus II (the Faithful), are associated with rest and sleep, albeit at this stage the sleep of death—‘Mein Jesu, gute Nacht’. Whereas the accompagnato of ‘O Schmerz’ was sung by one solo voice, that of ‘Nun ist der Herr’ is sung by four different solo voices in turn (in the order BTAS), each of whom is answered by the choral refrain. This accompagnato is clearly intended to introduce the choral finale, ‘Wir setzen uns mit Tra¨nen nieder’, no. 68, since this too is attributed by Picander to ‘die Tochter Zion’ and ‘die Glaübigen’, though they no longer engage in dialogue. Instead, Bach introduces his own moments of antiphony and refrain. The words called softly to Jesus in his grave, ‘Ruhe sanfte, sanfte ruh!’ (‘Rest in peace, in peace rest’) are sung as a brief antiphonal exchange between the two choirs. Then in the middle paragraph, the exchange is sung entirely by Chorus II as a choral refrain that three times responds to the contemplation of Chorus I. The refrain technique employed here is very similar to that of nos. 20 and 67. In all three cases, the subject is rest or sleep, which thus forms a key theme in the Passion for symbolic reasons: ‘sleep’ indicates a view of death as no mere cessation but rather a stage that has to be passed through. In many respects this great finale is clearly modelled on its equivalent in the St John Passion (‘Ruht wohl’, no. 39). Both are in the key of C minor; both contain falling and rising arpeggio figures in the continuo; and both are cast in the rhythm of a sarabande—indeed, the headmotive of the St Matthew Passion’s finale turns up in several of Bach’s sarabandes (Ex. 9). The overall form of this movement may be viewed as binary dance form with repeats within an overall da capo structure, or else as a sarabande en rondeau with a single episode.

Ex. 9

a) St. Matthew Passion, conclusio (No. 68), headmotive (bb. 1–2, 13–14)

b) Sarabande from English Suite No. 1 in A, BWV 806, headmotive

c) Sarabande en rondeau from Suite in F minor, BWV 823, bb. 21–2, treble

d) Sarabande from Lute Suite in C minor, BWV 997, headmotive

st matthew passion

189

The aria introduced by accompagnato or arioso, exceptional in the St John Passion (nos. 19–20 and 34–5 only), became standard in the St Matthew Passion. Here, arias without such introduction are the exception. As the exordium to Part II, the aria-motet ‘Ach, nun ist mein Jesus hin’, no. 30, requires no introduction. The arias ‘Blute nur’ (no. 8), ‘Erbarme dich’ (no. 39), and ‘Gebt mir meinen Jesum wieder’ (no. 42) belong in the context of sub-plots—Judas’s betrayal and Peter’s denial—and therefore do not merit the same expansiveness of treatment as that which is accorded to Jesus Himself. ‘Erbarme dich’, on the other hand, one of the most memorable arias in the entire Passion, is prepared by the immediately preceding biblical-text recitative, with its deeply moving arioso setting of ‘und weinete bitterlich’ (‘and [Peter] wept bitterly’). The headmotive of the aria is drawn from the same recitative, specifically from Peter’s third and most emphatic denial, ‘Ich kenne des Menschen nicht’ (‘I do not know of the man’), which is immediately echoed by the Evangelist a 5th higher to the words ‘Und alsbald kra¨hete der Hahn’ (‘and immediately the cock crowed’)—an ingenious musical means of representing Peter’s sudden recollection of Jesus’s prediction that before the cock crew he would deny Him three times. The aria, with its florid solo violin part and rich string accompaniment, expresses the feelings of intense grief occasioned by Peter’s denial in one of the most beautiful and affecting of all Bach’s sicilianas. The remaining ten arias are each preceded by a meditative introduction in the form of an accompanied recitative. No doubt Bach had in mind the two arioso-aria sequences from the St John Passion (nos. 19–20 and 34–5). But the changed name of the introductory movement from ‘Arioso’ to ‘Recitat[ivo]’ betokens a definite change of type. Moreover, the two arias from the earlier Passion, ‘Erwa¨ge’ and ‘Zerfließe’, are longer and more elaborate than anything in the later one. Perhaps, in view of the massive scale of the later work, Bach was reluctant to hold up the Gospel narrative for too long. There are other differences too. The textual link between arioso and aria in the earlier work is somewhat tenuous, whereas in the St Matthew Passion it is invariably clear. Furthermore, in the St John Passion the aria is in both cases sung by a different voice from the arioso, whereas in the later work recitative and aria are invariably sung by the same voice. In musical terms the accompanied recitative acts as a mediator between the free declamation of the Evangelist’s secco recitative and the structured lyricism of the aria. In terms of textual content, the accompanied recitative meditates poetically on the preceding biblical passage. The aria that follows then draws the conclusion in the form of a definite response—a commitment to a certain spiritual attitude, such as penitence (no. 6) or patience (no. 35), or else a resolution to act in a certain way, often accompanied by a prayer that the Lord will strengthen this determination (nos. 13, 20, 23, 57, 65). The recitative type employed is the motivic accompagnato, for which precedents may be found throughout Leipzig Cycles I–III of 1723–7.86 Both there and in the Passion an ostinato-like, constant recurrence of a single brief motive 86 Cycle I: BWV 147 (no. 8), 105 (no. 4), 46 (no. 2), and 40 (no. 5). Cycle II: BWV 10 (no. 6), 107 (no. 2), 125 (no. 3), and 127 (no. 4). Cycle III: BWV 110 (no. 3), 43 (no. 6), 102 (no. 6), and 56 (no. 2).

190 sacred a nd s ecula r: the vocal works throughout creates a quiet, contemplative atmosphere as a background to the voice’s declamation. The motives of the Passion’s accompagnati are often illustrative, depicting the pouring of water (no. 5), tears (no. 12), Jesus’s agony (no. 19), his falling down in prayer before his Father (no. 22), his silence at the testimony of the false witnesses (no. 34), and Pilate’s having him scourged (no. 51). Only in nos. 48, 56, 59, and 64 does the prevailing motive apparently lack a reference of this kind. Both in text and music the link between biblical narrative, accompagnato, and aria is often strong and clear. After the canonic duet of the two false witnesses, no. 33, representing the ‘copycat’ nature of their testimony, the Evangelist simply reports, ‘Aber Jesus schwieg stille’ (‘But Jesus remained silent’). This brief line prompts the lyrical commentary that follows. The accompagnato, no. 34, begins: ‘My Jesus remains silent at false lies’, and draws the moral that ‘we should be like Him and remain silent in persecution’. Silence is represented by short quaver chords for two oboes and continuo (later also multiple-stops on viola da gamba), separated by rests. In the following tenor aria ‘Geduld’, no. 35, the Christian hopes for ‘patience when false tongues stab me’. The simple directness of the headmotive, with its tonic-dominant oscillation, recalls other arias, such as nos. 6 and 8. This gentle motive in slurred quaver couplets clearly stands for ‘patience’, whereas the angular dotted rhythms of the rest of the theme no doubt allude to the ‘false tongues’ that ‘stab me’. Both of these motives are confined to the continuo part, which leaves the tenor voice enormous scope for freedom of declamation. Much the same applies to the relation between voice and accompaniment in the soprano aria ‘Aus Liebe will mein Heiland sterben’, no. 49. The secco–accompagnato–aria sequence to which this movement belongs, nos. 47–9, forms a quiet oasis of contemplation amidst the ferocity of the crowd’s calls for Jesus’s crucifixion. The cue for the lyrical commentary is Pilate’s question ‘What evil has he done?’ The soprano immediately replies, ‘He has done good to us all’ in an accompagnato with two oboes da caccia, which goes on to list all the good things done by Jesus. The following soprano aria ‘Aus Liebe’ expands upon the evil–good antithesis: He has done good ‘out of love’ and is even ‘willing to die’ for the same reason; as far as evil is concerned, ‘He knows nothing of any sin’. Jesus’s innocence is here symbolized by the absence of basso continuo: the two oboes da caccia from the accompagnato provide bassett and harmonic filling, while an added transverse flute plays the florid obbligato part. This movement, with its simple texture of melody and accompaniment—slow-moving and purely harmonic—represents one of Bach’s closest approaches to the progressive style of the younger generation. The revision of the St Matthew Passion over the years from 1727 onwards, which cannot be followed in detail due to the loss of the relevant sources, seems to have been concerned to a large extent with the role of Bach’s vocal and instrumental resources.87 It is possible that the differentiation between Chorus I and II was at first restricted to

See Melamed, Hearing Bach’s Passions, p. 65.

st m att he w pass io n

191

dialogue movements as prescribed by Picander. Thus many of the turbae (crowd choruses) have no antiphonal element, and those that do in some cases give the impression that they might derive from a single-choir original version. However this may be, in the version we possess Bach shows great freedom and flexibility over the marshalling of his vocal and instrumental resources. Small groups of people are sung by a single choir: the disciples by Chorus I; the bystanders who interrogate Peter (no. 38b) by Chorus II; ‘some’ bystanders at the Cross by Chorus I and ‘others’ by Chorus II (nos. 61b and d). Large groups of people are represented by both choirs together in various combinations. For the Jewish religious authorities the two choirs sing either in unison or in antiphony. The antiphonal writing gives drama to their utterances, but also perhaps reflects their mixed composition—chief priests, scribes, and elders of the people. Unison doubling evidently reflects the unanimity of their sentiments. The great choral shout of ‘Barrabam!’ and the crowd choruses that follow, ‘Laß ihn kreuzigen’ etc. (nos. 45–50), are sung by the two choirs in unison throughout, as if to emphasize the united determination of the crowd that Jesus should be crucified. Similarly, after the earthquake the onlookers at the Crucifixion are at one in their recognition of Jesus’s divinity, hence the doubling of the two choirs at the words ‘Wahrlich, dieser ist Gottes Sohn gewesen’ (‘Truly this was the Son of God’, no. 63b). Despite occasional thematic or motivic correspondences between the turbae, they do not possess the same integrating function as in the St John Passion. In the later Passion this role is taken primarily by the chorales. The Lutheran chorale represents the traditional voice of the Church and as such is second in authority to the Gospel itself. The chorales would have been thoroughly familiar to Bach’s Leipzig congregation and would therefore act as a valuable aid to comprehension and accessibility. Bach not only reinforces this function of the chorales but gives them a significant unifying role by incorporating several verses of the same hymn at various points in the narrative. Thus O Welt, sieh hier dein Leben (Paul Gerhardt, 1647) is heard twice, once in each Part (nos. 10 and 37); three verses of Herzliebster Jesu (Johann Heermann, 1630) are included (nos. 3, 19, and 46), and no fewer than five of O Haupt voll Blut und Wunden (Paul Gerhardt, 1656)—two in each Part (nos. 15, 17, 54, and 62) plus the same melody to a different hymn (Befiehl du deine Wege, no. 44). Finally, in the definitive version of 1736, the chorale-based exordium and conclusio to Part I (nos. 1 and 29) complement each other in an ideal fashion. The one is in E minor, the other in E major. Each is based on one of the earliest of the great Passion chorales, dating back to the time of Martin Luther himself: O Lamm Gottes, unschuldig (no. 1), a German paraphrase of the Agnus Dei (Nicolaus Decius, 1531); and O Mensch, bewein dein Su¨nde groß (Sebald Heyden, 1525). The two chorale melodies have much in common (Ex. 10). The one opens with an appeal to the innocent Son of God, the other to guiltladen mankind: ‘O Lamb of God, guiltless’/‘O man, bewail your great sin’. And a reference to Christ’s bearing the burden of our sins is common to the conclusion of both chorales: ‘All Su¨nd’ hast du getragen’ (‘All sin have You borne’)/‘Tru¨g’ unser Su¨nden schwere Bu¨rd’ (‘carry the heavy burden of our sins’). In both cases, the

192

sacred and secular: the vocal works

Ex. 10

Lamm

tes

Mensch, be - wein

Stamm

Äu

Got

des

ßert

un - schul

dein

Kreu - zes

und kam

auf

dig

Sün - de

groß

ge - schlach

den

tet

a) St. Matthew Passion, No. 1, chorale O Lamm Gottes, lines 1–2 b) St. Matthew Passion, No. 29, chorale O Mensch, bewein, lines 1 and 3 chorale represents an authoritative statement of Church doctrine. In the exordium, however, the objective element of the Agnus Dei, sung by ripieno soprano, was clearly separated from our emotional response to the Passion as articulated in the dialogue between the two choirs. In the conclusio, on the other hand, all forces unite: Chorus I and II are combined, and all the sopranos join together to sing the chorale melody. There is no division in this movement between subjective and objective, personal and impersonal: the chorale acts as the vehicle of strong feelings as well as of great truths. Not only does it state the central meaning of the Passion, namely the Atonement, the expiation by Christ of the sins of humankind; it also calls upon us to bewail the great sin that brought this about. Both elements are equally present in Bach’s setting: the theological statement in the venerable chorale melody, sung by unison sopranos (doubled by organ), and our grief-stricken response in the accompanying voice parts and in the deeply expressive instrumental framework.

Cantatas of 1727–1728 and the Picander Cycle (1728–1729) Title, occasion

Earliest source/s

Scribe, date

Schwingt freudig euch empor, BWV 36, Advent 1 Ich lasse dich nicht, BWV 157, memorial, Puriﬁcation O ewiges Feuer, BWV 34, Whit Sunday

Berlin, Am.B.106

C. Nichelmann; perf. 1726/30

Berlin, P 1046, St 386

C. F. Penzel; 1st perf. 6 Feb. 1727

Berlin, Am.B.39 , St 73

Autograph, part-autograph, 1 June 1727

c a n t a t a s o f 1 7 2 7–17 2 8 e tc . Gelobet sei der Herr, BWV 129, Trinity Sunday Ihr Tore zu Zion, BWV 193, council election Wir mu¨ssen durch viel Tru¨bsal, BWV 146, Easter 3 Man singet mit Freuden vom Sieg, BWV 149, Michaelmas Ich habe meine Zuversicht, BWV 188, Trinity 21 Ehre sei Gott in der Ho¨he, BWV 197a, Christmas Day Gott, wie dein Name, BWV 171, New Year Ich steh mit einem Fuß im Grabe, BWV 156, Epiphany 3 Sehet, wir gehn hinauf gen Jerusalem, BWV 159, Quinquagesima Ich lebe, mein Herze, BWV 145, Easter Tuesday Ich liebe den Ho¨chsten, BWV 174, Whit Monday

Leipzig TS

193

Berlin, St 62

C. G. Meißner et al., 8 June 1727 J. H. Bach et al., 25 Aug. 1727

Berlin, Am.B.538

J. F. Agricola; 1st perf. 1728?

Berlin, P 1043, St 632

C. F. Penzel; 1st perf. 29 Sept. 1728? Autograph fragment, for 17 Oct. 1728? Autograph fragment, for 25 Dec. 1728? Autograph, for 1 Jan. 1729? Anon.; 1st perf. 23 Jan. 1729?

Various owners New York PML New York PL Leipzig TS Berlin, P 1048, St 633 Berlin, P 151

C. F. Penzel; 1st perf. 27 Feb. 1729? Anon.; 1st perf. 19 Apr. 1729?

Berlin, P 115, various owners

Autograph, C. P. E. Bach et al., for 6 June 1729

There is little evidence of the regular composition of sacred cantatas by Bach after Cycle III. However, the recent discovery of the printed texts of four cantatas performed in succession in June 1727—BWV 34, 173, 184, and 129, for the three Whit feast days and Trinity Sunday88—sheds a new light on this period. For it raises the possibility of the regular composition or revival of cantatas before and/or after the Whit-Trinity period, for which there is at present no evidence. If that were the case, the cantatas of 1727 and the ﬁrst half of 1728 might have constituted Bach’s fourth Leipzig cycle. A period of about four months would have to be left out of account, however, for no cantatas could be performed from 7 September 1727 to 6 January 1728—the period of public mourning for the queen, Christiane Eberhardine, whose death was mourned and life celebrated in the Bach–Gottsched Trauer Music, BWV 198 of 1727. Of the three Whit cantatas whose printed texts have been discovered, O ewiges Feuer, BWV 34, related by parody to the wedding cantata of the same name (BWV 34a, 1726/7),89 was formerly dated 1746/7 on the basis of the autograph score, but the performance at that date must have been a revival, for the recently discovered printed text dates from Whit Sunday 1727. Cantatas 173 and 184, both

88 See Tatiana Shabalina, ‘ “Texte zur Music” in Sankt Petersburg: Neue Quellen zur Leipziger Musikgeschichte sowie zur Kompositions- und Auffu¨hrungsta¨tigkeit J. S. Bachs (BJ 94 (2008), pp. 33–98, esp. 65–77) and ‘ “Texte zur Music” in Sankt Petersburg—Weitere Funde’ (BJ 95 (2009), pp. 11–48, esp. 28–9). Her conjecture concerning Cycle IVon the basis of her ﬁndings in St Petersburg ties in with questions previously raised by Georg von Dadelsen, Beitra¨ge zur Chronologie der Werke Johann Sebastian Bachs (Trossingen, 1958), pp. 139–42, and Alfred Du¨rr, ‘Noch einmal: Wo blieb Bachs fu¨nfter Kantatenjahrgang?’, BJ 72 (1986), pp. 121–2. 89 See T. Shabalina, ‘Neue Erkenntnisse zur Entstehungsgeschichte der Kantaten BWV 34 und 34a’, BJ 96 (2010), pp. 95–109.

194

sacred and secular: the vocal works

parodies of Co¨then serenatas, were probably first performed in 1724 in the context of Cycle I (this can be verified only in the case of BWV 184), but are now known to have been revived on Whit Monday and Tuesday 1727. The Trinity cantata Gelobet sei der Herr, BWV 129, was formerly thought to have received its first performance in 1726 (within Cycle III), but the printed text found recently proves that it was written for Trinity Sunday 1727. Apart from these new findings, our knowledge of Bach’s composition and performance of church cantatas in 1727 and early 1728 is sparse. An early version of the Advent cantata Schwingt freudig euch empor, BWV 36, might have originated around this time as a straightforward parody of the secular cantata of that name (BWV 36c, 1725). Bach simply took the opening chorus and all the arias of the secular model, had the text adapted, and replaced the original gavotte-finale with a four-part chorale. At a later stage, for Advent 1731, he would undertake a root-and-branch revision of the work, introducing much new material. The original version of Cantata No. 157, Ich lasse dich nicht, du segnest mich denn, written for a memorial service on 6 February 1727, is lost, but the work survives in a later version for the Feast of the Purification (2 February, year unknown). The council election cantata Ihr Tore zu Zion, BWV 193, which survives in an incomplete state, was performed on 25 August 1727, only three weeks after its secular model Ihr Haüser des Himmels (BWV 193a, 3 August 1727). Finally, the Eastertide cantata Wir mu¨ssen durch viel Tru¨bsal, BWV 146, cannot be dated with any precision due to the loss of the original sources, but its strong connection with BWV 188 (discussed later) points to 1728 as the most likely year of its original performance. Picander’s church-year cycle of librettos Cantaten auf die Sonn- und Fest-Tage durch das gantze Jahr (Leipzig, 1728) was to have been set to music by Bach—in the preface, dated 24 June, the author remarks that Ich habe solches Vorhaben desto lieber unternommen, weil ich mir schmeicheln darf, daß vielleicht der Mangel der poetischen Anmuth durch die Lieblichkeit des unvergleichlichen Herrn Capell-Meisters, Bachs, du¨rfte ersetzet, und diese Lieder in den Haupt-Kirchen des anda¨chtigen Leipzigs angestimmet werden. (I have undertaken such a project all the more willingly because I flatter myself that perhaps the lack of poetic charm might be compensated by the loveliness of the incomparable Capellmeister Mr Bach’s music, and that these songs will be sung in the principal churches of devout Leipzig.)90

Whether Bach did set the entire cycle in the event or only a certain proportion of it has long been a matter of dispute. The original autograph scores and performing parts are largely lost, so that the few cantatas that survive—seven complete (BWV 149, 188, 171, 156, 159, 145, and 174), half of an eighth (BWV 197a), and a few bars of a ninth (BWV

90 The title and preface are known only from Spitta II, pp. 172–5. Many issues connected with the cycle have recently been clariﬁed by the discovery of an exemplar of Picander’s printed text of 1728; see Shabalina, BJ 95 (2009), pp. 20–30.

c a n t a t a s o f 1 7 2 7–17 2 8 e tc .

195

Anh. I 190)—do so purely by chance. The date of Picander’s preface (24 June 1728) suggests that Bach might have started the cycle on the First Sunday after Trinity, according to his custom. In addition, one of the cantatas, No. 174, is dated 1729 in the original performing parts. It is possible, then, that the cantatas were composed and performed between June 1728 and June 1729. This would assume that Bach set the texts within a single year, as in Cycles I and II, but it is perfectly possible that he set them over several years, as in Cycle III. It has also been suggested recently that Bach might have delegated some of the settings to his eldest sons and most able pupils. In particular, there is evidence that the young C. P. E. Bach was involved in the project. A newly discovered cantata of his, Ich bin vergnu¨gt mit meinem Stande, is based on a text from the Picander Cycle. The fragment Ich bin ein Pilgrim, Anh. I 190, whose text is drawn from the same cycle, is in C. P. E. Bach’s hand and might well have been composed by him. And the Easter Tuesday cantata from the cycle, Ich lebe, mein Herze, zu deinem Ergo¨tzen (BWV 145) is untypical of J. S. Bach and might also have been composed by his young son. Since C. P. E. Bach left Leipzig in September 1734, the joint father–son project of setting the cycle might have extended over several years— from the late 1720s to the mid-1730s—and it was possibly still incomplete when the son left his father’s home.91 In the surviving cantatas of the Picander Cycle and in the half-dozen or so cantatas that immediately preceded them, from the period 1727–8, Bach in many respects pursued further the trends that had been established in Cycle III. This applies, in particular, to instrumental music, chorale treatment, and the setting of biblical texts. In four cantatas, Nos. 146, 188, 156, and 174, Bach adapted concerto movements as sinfonias, as he had already done in some of the later Cycle III compositions (Nos. 35, 169, 49, and 52). And in two cases, Nos. 146 and 188, the instrumental ensemble is led by obbligato organ—not only in the sinfonia but occasionally in other movements too. There is a clear parallel between Cantata 52 from Cycle III and No. 174 from the Picander Cycle. In both cases, the first movement of a Brandenburg Concerto forms a sinfonia of enormous dimensions, out of all proportion to the modestly scored solo vocal music that follows. In No. 174 the first movement of the third Brandenburg Concerto is even expanded in instrumentation: the original nine solo strings now become a concertino, set against a new ripieno of two horns, three oboes, and additional strings. There appears to be no justification in the text for this vast sinfonia, but there may be some link with Bach’s assuming the directorship of the Collegium musicum in 1729, the year of the cantata’s first performance. An entire concerto is divided between Cantatas Nos. 146 and 188—the lost original version of BWV 1052— just as the Cycle III cantatas Nos. 169 and 49 had shared the lost original of BWV 1053. In all four cases, the outer movements of the concertos form sinfonias. But whereas the slow movement of the one concerto became an aria in No. 169, the slow movement 91 See Peter Wollny, ‘Zwei Bach-Funde in Mu¨geln: C. P. E. Bach, Picander und die Leipziger Kirchenmusik in den 1730er Jahren’, BJ 96 (2010), pp. 111–51.

196

sacred and secular: the vocal works

of the other was transformed into a chorus in No. 146 (no. 2). The biblical text, ‘Wir mu¨ssen durch viel Tru¨bsal in das Reich Gottes eingehen’ (‘We must through much tribulation enter into the Kingdom of God’), is reflected in the pathos-filled voice parts which, using long-established Choreinbau techniques, Bach builds into the framing ritornellos and into the episodes, led by obbligato organ. Elsewhere, in Cantata No. 156, a concerto slow movement92 is uniquely employed as a sinfonia. The lovely Adagio melody for concertante oboe with string accompaniment creates a contemplative atmosphere far better suited to the first words of the libretto—‘Ich steh mit einem Fuß im Grabe’ (‘I stand with one foot in the grave’)—than a concertoAllegro would have been. Two of Bach’s later modes of chorale treatment are represented not only in Cycle III (1725–7) but in the cantatas of 1727–9, namely the chorale-aria with vocal cantus firmus (as opposed to the earlier type with instrumental chorale) and the chorale cantata per omnes versus (that is, with the original text preserved throughout). The last-named type is represented in Cycle III by Cantata No. 137, Lobe den Herren, den ma¨chtigen Ko¨nig der Ehren, and in the post-Cycle III group by No. 129, Gelobet sei der Herr, mein Gott, composed for Trinity Sunday 1727. The two works are identical in number and order of movements: there are no recitatives, only a succession of three arias framed by chorale-choruses. Whereas the arias of No. 137 are more like chorale arrangements, however, preserving the associated chorale melody, those of No. 129 are entirely independent of it. Since the first two lines of each verse are virtually identical, the three arias are a strong testimony to Bach’s ability to set any text, however seemingly intractable, to beautiful music. Whereas the opening chorale-chorus of each cantata belongs to the type established in Cycle II, the finales differ: both are plain four-part chorales, but that of No. 137 is enhanced by a four-part trumpet choir; in No. 129, on the other hand, the chorale is built into a concertante instrumental texture with its own theme, treated as ritornello as well as accompaniment. Such elaborate instrumental music would eventually adorn the chorale-finales of the Christmas (Part IV and VI) and Ascension Oratorios. The chorale-aria is represented in the Picander Cycle by the second movement of Cantatas Nos. 156 and 159. In both cases the chorale cantus firmus is sung by soprano, as in Cycle III (Nos. 49, 58, and 158); but whereas the aria voice was there invariably bass, it is now tenor (No. 156) or alto (No. 159). Soprano and bass in Cycle III (Nos. 49 and 58) were cast as Jesus and the Soul, but after Cycle III no such identification applies. In Cantatas Nos. 156 and 159, as also in No. 158, the free, madrigalian aria text tropes that of the chorale, amplifying and embroidering it, so that in terms of the sentiments expressed the two singers are at one throughout. In ‘Ich steh mit einem Fuß im Grabe’, BWV 156 no. 2, one of Bach’s many profound

92 Which also survives as the Largo of Harpsichord Concerto No. 5 in F minor, BWV 1056; see Joshua Rifkin, ‘Ein langsamer Konzertsatz J. S. Bachs’, BJ 64 (1978), pp. 140–7.

c a n t a t a s o f 1 7 2 7–17 2 8 et c.

197

treatments of departure from this world, there is a certain irony in the tenor’s entry: he sings ‘I stand’ (‘Ich steh’) to a long-held note, while the unison strings and continuo contradict him, portraying his sinking into the grave. In ‘Ich folge dir nach’, BWV 159 no. 2, the soprano sings verse 6 of O Haupt voll Blut und Wunden (Paul Gerhardt, 1656), nowadays associated above all with the St Matthew Passion. Both this and the aria text, set to music in giga-pastorale rhythm, express the soul’s determination to stand by Jesus throughout His suffering and death on the Cross. Only the opening movement of this cantata, ‘Sehet, wir gehn hinauf gen Jerusalem’, and its equivalent in No. 145 (also from the Picander Cycle) are conceived as dialogues between Jesus and the Soul, whereas in the dialogues of Cycle III (Nos. 57, 32, 49, and 58) the entire cantata is laid out as such. Nor is the identiﬁcation of the bass with Jesus and the soprano with the Soul adhered to so strictly: in Cantata No. 159, the Soul is sung by alto; and in No. 145, Jesus by tenor. The dialogue movement (no. 1) from the last-named cantata possesses many of the characteristics of a secular love duet, and both it and the bass aria no. 3, which is virtually identical in structure, are generally thought to have been parodied from a lost Co¨then serenata,93 though this theory is contradicted by the more recent hypothesis of C. P. E. Bach’s authorship mentioned earlier. At the opposite pole from this secular style is the dialogue movement that opens Cantata No. 159, ‘Sehet, wir gehn hinauf gen Jerusalem’, where we encounter Bach’s sacred style at its most profound. Jesus’s words from the Gospel, ‘See, we are going up to Jerusalem’ (Luke 18: 31), sung by the bass as vox Christi in arioso with continuo accompaniment, are troped by a free commentary, sung by the alto as Soul in accompagnato with held string chords. The deeply felt bass aria no. 4 has a certain afﬁnity with the bass recitative that opens Der Friede sei mit dir, BWV 158. In both cases, Jesus’s words—here, the last words from the Cross as recorded in John 19:30, ‘Es ist vollbracht’ (‘It is accomplished’)—are sung by the bass as vox Christi in the form of a motto text at the beginning, middle, and end of the movement. On each occasion it is sung to a musical motto, namely the headmotive of the ritornello and its exact inversion, which follows immediately. It is a tribute to the instinctive, second-nature quality of Bach’s contrapuntal technique that he is able to make use of the mechanical device of inversion amidst such depth and intensity. As in Cantata No. 158, Jesus’s words are troped by a commentary that gives the human response to them, sung by the same singer—the conception is contemplative rather than dramatic. As a whole, this composition is outstanding among the Picander series; and, as a cantata for Quinquagesima, when the thoughts of the faithful turn towards Passiontide, it is fully equal to the remarkable audition and Cycle II cantatas that preceded it (Nos. 22, 23, and 127).

93 This widely accepted theory of Friedrich Smend’s—Bach in Ko¨then (Berlin, 1951), pp. 45–7—is rejected by Wollny, BJ 96 (2010), p. 139.

198 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s

The motets Title, occasion

Earliest source/s

Scribe, date

Singet dem Herrn ein neues Lied, BWV 225, funeral? Der Geist hilft unser Schwachheit auf, BWV 226, burial Jesu, meine Freude, BWV 227, funeral? Fu¨rchte dich nicht, BWV 228, funeral? Komm, Jesu, komm, BWV 229, funeral?

Berlin, P 36, St 122 Berlin, P 36/1 , St 121 Berlin, P 48/6 Berlin, P 569 Berlin, P 609

Autograph, J. A. Kuhnau et al., c. 1727 Autograph, part-autograph, for 20 Oct. 1729 Anon.; orig. late 1720s? Anon.; orig. early 1720s? C. Nichelmann; orig. before 1732/5

Bach’s motets belong to a conservative genre much cultivated in his own locality of Thuringia and Saxony, notably by his older relatives, the brothers Johann Christoph and Johann Michael Bach. ‘Modern’ operatic and concertante forms play no part here. Instead polyphony, rhetorical homophony, and antiphonal writing are employed for the setting of biblical and chorale texts, as they had been throughout the seventeenth century. Also included are strophic sacred songs or ‘arias’, which differ from chorales only in their late origin and their absence from the hymnbooks. All but one of the five fully authenticated motets considered here,94 BWV 225–9, are scored for double choir (SATB, SATB), a Venetian speciality taken up in Germany by Michael Praetorius, Heinrich Schu¨tz, and others in the early seventeenth century. Bach employed it elsewhere only in rare circumstances (the St Matthew Passion; Preise dein Glu¨cke, BWV 215, and related pieces, such as the ‘Osanna’ from the B minor Mass). The exception, Jesu, meine Freude, BWV 227, is scored for five voices (SSATB), which Bach hardly used elsewhere other than in his Latin church music (Magnificat, 1733 Missa, etc.). This vocal scoring already distinguishes the motets from the choruses of Bach’s sacred cantatas, but a still more fundamental distinction lies in the use of instruments. In the cantatas they have a leading role as the carriers of important themes and as vehicles of a concertante texture. In the motets, on the other hand, their role is restricted at most to the colla parte doubling of the vocal parts. Performing parts for instruments and continuo survive only in the case of Der Geist hilft, BWV 226, but such instrumental support is to be assumed elsewhere too, since it was standard practice at the time. Bach’s motets, unlike his sacred cantatas, originated as occasional music—they were not tied to regular events in the church year as the cantatas were. There is evidence that two of them, Der Geist hilft and Komm, Jesu, komm, were designed to be sung as Trauermusik (funeral music), and the texts of the other motets suggest that they too were probably written for funerals or memorial services. Only in one case is the identity of the deceased person known: according to the original sources, Der Geist hilft was written for the burial of J. H. Ernesti, Rector of the Thomasschule, on 20 October 1729. The original score and parts of Singet dem Herrn date from only 94

Doubtful motets, BWV 230, Anh. III 159, and Anh. III 160, are left out of account.

the mo te ts

199

two or three years earlier (1726/7); and both Jesu, meine Freude and Komm, Jesu, komm originated some time during Bach’s ﬁrst decade or so in Leipzig (1723–32/5). Only Fu¨rchte dich nicht cannot be dated even approximately, though its mature style is similar enough to that of the other motets to suggests a common period of origin.95 Fu¨rchte dich nicht is a setting of two verses from Isaiah and two strophes from the chorale Warum sollt ich mich denn gra¨men (Paul Gerhardt, 1653) as follows: A

Double-choir chor. Is. 41: 10

Prelude, double fugue + chorale, postlude Is. 43: 1 + chorale, vv. 11–12

‘Movement’ A96 follows the long-established motet principle of allotting a different, often contrasted setting to each portion of text. Consequently, it falls into three sections: ‘Fu¨rchte dich nicht’ (b. 1), ‘Weiche nicht’ (b. 10), and ‘ich sta¨rke dich’ (b. 29). Movement B is framed by a brief double-choral prelude and postlude (bb. 73 and 151), but its main content is the central chorale-cum-fugue, in which the eight voices double up to sing in four parts. The chorale cantus ﬁrmus in the soprano part is accompanied by a double fugue for alto, tenor, and bass, sung to words from the second Isaiah verse. The fugue subjects, S I and II, are a traditional combination of seventeenth-century origin: a chromatic sequence in S I combined in contrary motion with a diatonic sequence in S II. The chromatic theme tells of the Redemption and, by implication, of its cost (‘Denn ich habe dich erlo¨set’), while the soprano (chorale) hovers above, singing radiantly of the mystical union with the Saviour. The effect is not dissimilar to that of ‘Es ist der alte Bund’ from the Actus Tragicus, BWV 106, though a closer formal parallel would be ‘Sei nun wieder zufrieden’ from Ich hatte viel Beku¨mmernis, BWV 21. For all the motet-style composing along to the words, Bach takes steps to enmesh the two movements together, taking his cue from the text. Both of the Isaiah verses begin with the same words, ‘Fu¨rchte dich nicht’ (‘fear not’), with the result that the prelude of B recalls the opening section of A. Moreover, the same words, sung to a variant of the same music, recur once more in the postlude at the end of B. Similarly, the words ‘Du bist mein’ are common to both chorale strophes and to the Isaiah verse that accompanies them. Bach draws special attention to them by setting them as an extra countersubject in a rhythmic sequence divided by rests. This concern for formal unity is just one of several ripe features in the work, others being the major-9th chords in bb. 11 and 16, and the impressive extended phrase-lengths in the third section of A (‘Ich sta¨rke dich’), a feature that recurs repeatedly in the other motets.

95 For the dating of BWV 227, see Daniel R. Melamed, J. S. Bach and the German Motet (Cambridge, 1995), ¨ berlieferung im 18. Jahrhundert p. 101; for that of BWV 229, Hans-Joachim Schulze, Studien zur Bach-U (Leipzig and Dresden, 1984), pp. 130 ff. Melamed (Bach and the German Motet, p. 101) dates BWV 228 c. 1715. 96 Using the word ‘movement’ loosely, since there is no tonal closure.

200

sacred and secular: the vocal works

In Der Geist hilft unser Schwachheit auf, Bach sets the same textual elements, biblical words and chorale, using similar musical means to those of Fu¨rchte dich nicht, doublechoir chorus, double fugue, and chorale, but they are differently disposed: A

Double-choir chor. Rom. 8: 26

Double fugue Rom. 8: 27

Chorale Komm, heiliger Geist, v. 3

Again, the double-choir chorus falls into three contrasting sections according to the words: a ‘Der Geist hilft’ (b. 1); b ‘Denn wir wissen nicht’ (b. 41); c ‘Sondern der Geist selbst’ (b. 124). On this occasion, however, the first two sections are both repeated in a varied form—a new means of counteracting the motet’s traditional series form— giving rise to the alternating structure a b a1 b1 c. In section a the activity of the Holy Spirit is conveyed in a long semiquaver melisma against the dance rhythm of a passepied, hence the regular eight-bar phrases, exceptional in a motet. Section b, being concerned with prayer, is gentler and more lyrical, like its equivalent in Fu¨rchte dich nicht. The last section, c, is one of the most remarkable passages in all Bach’s motets: a fugato on a syncopated, sequential subject, whose surrounding counterpoint depicts the ‘inexpressible sighs’ with which ‘the Spirit intercedes for us’ in angular lines, melismas broken by rests, and suspiratio or sigh figures. After the extremely unsettled minor-mode harmony of this fugato, a complete contrast ensues in movement B, with the clear major mode of its alla breve stretto fugue on the second verse of the text from Romans 8. It has two features in common with the fugue from Fu¨rchte dich nicht: the two choirs combine to create a strengthened four-part texture; and the two subjects are combined in double fugue, though here they are worked in separate expositions first (bb. 146 and 178; they are first combined at b. 199). The concluding four-part chorale arrangement was not newly composed but taken over from an existing composition, in which it was probably in the key of G major, not B♭.97 Komm, Jesu, komm is exceptional among Bach’s motets in that it lacks both fugue and biblical text. Nor does it include a chorale: the text is drawn from a strophic sacred song, written by Paul Thymich for a funeral in 1684 and set to music by Johann Schelle, a predecessor of Bach’s as Leipzig Thomascantor. Bach sets the first and last verses of the Lied as follows:

Double-choir chor. Verse 1

Aria Verse 11

See Klaus Hofmann, Johann Sebastian Bach: Die Motetten (Kassel, 2003), pp. 92–5.

the mo te ts

201

In movement A each line of the text is set in a different manner in accordance with the traditional motet principle that Bach elsewhere applies to biblical texts. As in Der Geist hilft, however, varied repeats are used to create coherence. In this case they are dictated by the Bar form (AAB) of the text: Stollen I: a ‘Komm, Jesu, komm’ (b. 1); b ‘Die Kraft verschwind’t’ (b. 16); Stollen II: a1 ‘Ich sehne mich’; c + b1 ‘Der saure Weg’ (b. 44). The text here is an appeal to Jesus as death approaches, expressing a longing for his peace that lies beyond ‘the bitter path’ of life, hence the deeply expressive minor-mode music, with its dissonant minor-9th appoggiaturas (bb. 9, 28, and 61) and its falling diminished 7th in the imitative point on ‘der saure Weg’. A complete contrast ensues at the Abgesang (b. 64), with its eager cries of ‘Komm, komm, ich will mich dir ergeben’ in close imitation. Another sharp change of tempo and metre—to a dance-like 6/8— occurs for the last line of the Abgesang, to which Bach gives enormous emphasis due to its special status as a paraphrase of John 14: 6: ‘You are the right Way, the Truth, and the Life’ (‘Du bist der rechte Weg, die Wahrheit und das Leben’). These words call forth a seemingly endless ﬂow of lyricism—shared between the two choirs—which, at 88 bars, exceeds in length everything that has preceded it (78 bars). Movement B— entitled ‘Aria’ by Bach, meaning an essentially homophonic composition on a metrical text—is written in the style of a four-part chorale, except that the concluding echo of John 14: 6 is sung melismatically and calls forth great fervour. Singet dem Herrn is perhaps (alongside Jesu, meine Freude) the most inspired and certainly the most imposing in dimensions of all Bach’s motets. Magniﬁcent songs of praise, in the form of psalm-choruses, frame a central chorale-aria, creating a symmetrical structure thus: A

Antiphonal chor.—fugue Ps. 149: 1–3

chorale + aria

Antiphonal chor.—fugue Ps. 150: 2, 6

Movements A and A1 are related only in structure (though their openings are motivically linked—cf. bb. 1 and 221): in both cases, as often in Bach’s vocal music, there is a clear analogy with the instrumental form of prelude and fugue. A, however, falls into three distinct sections according to the text: a ‘Singet dem Herrn’ (b. 1); b ‘die Gemeine der Heiligen’ (b. 28); c ‘Israel freue sich’ (b. 59); though all three are motivically linked (compare, for example, bb. 1, 42, and 60). A1, on the other hand, is a single uniﬁed chorus of praise, culminating in fugue. The antiphonal choruses of A and A1, however, have in common their subtly varied answers in the exchanges between the two choirs and their ample, generous phrase-lengths, a recurring feature of Bach’s double-choir motets. The two fugues could hardly be more different. For that of A, ‘Die Kinder Zion’ (b. 75), the double-choir texture is maintained: the fugal exposition of Choir I is accompanied by a return of the ‘Singet’ music (or a variant thereof) in Choir II, so that a full texture is preserved from the outset. The fugue

202 sacred and s ecular: the vocal works subject is playful and joyous in accordance with the text, ‘Let the children of Zion be joyful in their King’. In the fugue that concludes A1, ‘Alles, was Odem hat’ (b. 225), on the other hand, the two choirs double up to form a reinforced four-part texture (as in Fu¨rchte dich nicht and Der Geist hilft). Moreover, the antiphonal music gives way to a quick, dance-like 3/8 time. The central movement B takes the form of a chorale-trope, a frequent mode of chorale treatment in Bach’s cantatas. Here, however, double-choir texture is maintained and the chorale is troped not by recitative but by an ‘aria’, a metrical sacred song (with anonymous text) similar in kind to the text of Komm, Jesu, komm. Verse 3 of the chorale Nun lob, mein Seel, den Herren (J. Gramann, 1530) is sung by Choir II in a four-part harmonization, each phrase being answered by words from the aria, sung in a homophonic or lightly polyphonic texture by Choir I. In essence, the comforting words of the chorale, ‘As a father takes pity on his young children, so does the Lord on us all’, are supported by the prayer of the aria, ‘O God, continue to look after us’. Bach sets lines 4, 7, and 8 of the aria in invertible counterpoint (bb. 195 and 210), whose crotchet theme appears to be quoted from the last line of Johann Hermann Schein’s chorale Machs mit mir, Gott, nach deiner Gu¨t of 1628 (Ex. 11), where it is sung to the words ‘Ist alles gut, wenn gut das End’ (‘All is well that ends well’). This supports the view that Bach’s motet was written for a funeral or memorial service, for the chorale by Schein, a predecessor of Bach’s as Leipzig Thomascantor, was in Bach’s day sung primarily as a funeral hymn.

Ex. 11

Drum sei

un - ser Schirm und

Licht

a) Singet dem Herrn, BWV 225, bb. 195–6, T only (SA omitted)

b) J. H. Schein, Machs mit mir, Gott, last line, treble only (words omitted) Jesu, meine Freude is singular not only in its five-voice texture but in structure: the six verses (A 1–6 in the table) of the chorale Jesu, meine Freude by Johann Franck (1653), sung to the well-known melody by Johann Cru¨ger, alternate with five verses (B 1–5) from Romans 8 (vv. 1–2 and 9–11). While the chorale verses are saturated with Jesusliebe (love of Jesus), the biblical words are concerned with life in the Spirit as opposed to the flesh. The eleven movements form a symmetrically organized structure,98 as in Singet dem Herrn, though of a very different kind: 98 The first to draw attention to the symmetry of the motet was Friedrich Smend, ‘Bachs MatthaüsPassion’, BJ 25 (1928), pp. 1–95.

the m o tets

A1 4vv

203

A6 4vv B1 5vv

B5 5vv A2 5vv

A5 4vv B2 3vv

B4 3vv A3 5vv

A4 4vv B3 5vv

A1 and A6 are identical four-part chorales, creating an outer frame. B5 is a varied reprise of B1 (to different words), establishing a frame for the biblical verses. Within these movements, chorale settings for five (A2, A3) or four voices (A4, A5) surround biblical-text trios (B2, B4). The centre-piece is a five-part biblical setting, B3, the only movement constructed as a fugue. It has been pointed out99 that the style of chorale setting in this motet is too advanced for 1723, Bach’s first year in Leipzig. Thus, if the work was performed then, as was formerly assumed, it might have been in a version that lacked most of the chorale movements. It is perhaps more likely, however, that the whole motet—with the possible exception of one or two movements—originated in the late 1720s, along with Singet dem Herrn and Der Geist hilft. The musically identical outer movements, A1 and A6, are plain four-part chorales, albeit of great beauty and with an exceptionally expressive tenor part. The five-part A2 and the four-part A4 represent a further development of the same type: the soprano continues to deliver the plain chorale melody, but the lower parts are more elaborate than usual, often in the interests of text illustration. A3 and A5 move beyond the realm of chorale harmonization altogether. A3 borrows the rhetorical, homophonic style of the first biblical setting, B1, to give a vivid musical illustration of the defiance of the ‘old dragon’. The first soprano part incorporates a paraphrase of the entire chorale melody, whose Bar form is preserved. Extremely effective is the unison writing that depicts united defiance of the devil, death, fear, and the world (bb. 4, 8, 20, and 24). The same figure, derived from the second line of the chorale melody, returns in the Abgesang (line 7, b. 37), but now in mellifluous 3rds; and the resemblance between this passage and the 3/8 dolce treatment of the same line in Bach’s organ trio on the same chorale, BWV 713, is most striking (Ex. 12).100 A5 is a full-blown cantus firmus chorale, with the cantus in the alto part of an SSAT quartet. Since the text says ‘good night’ to the world, the absence of bass 99 By Werner Breig, ‘Grundzu¨ge einer Geschichte von Bachs vierstimmigem Choralsatz’, Archiv fu¨r Musikwissenschaft, 45 (1988), pp. 165–85, 300–19 (esp. 183–5). 100 The two passages are compared in Peter Williams, The Organ Music of J. S. Bach (Cambridge, 1980–4), vol. ii, p. 252.

204 s a c r e d a n d s e c u l a r : t h e v o c a l w o r k s might signify the world’s lack of a foundation in Christ, though other possible interpretations have been put forward. This movement gives a different version of line 2 (and line 5) of the chorale melody—one used by Bach in Weimar—from that of the other chorale movements; consequently it has been suggested101 that A5 might be drawn from an older composition (Ex. 13). However, Bach uses both versions in A3, and he might have selected the older one in A5 due to its character as an inversion of line 1 (see the accompanying parts in bb. 19–24). In the Abgesang (line 8) he uses the later version. Moreover, the bewitchingly lyrical setting of ‘Gute Nacht’—short phrases with appoggiatura endings, cut off by rests—is foreign to Bach’s Weimar style, but can be paralleled in Leipzig in the mid-1720s (Ex. 14). The ﬁrst biblical-text movement, B1, whose music recurs in varied form in B5, is intensely serious and fully commensurate with its life-and-death theme—the opposing demands of the ﬂesh and the Spirit. As in Der Geist hilft, several distinct sections—

Ex. 12

a) Jesu, meine Freude, BWV 227, 5th movement, bb. 37–9 (alto omitted)

b) Jesu, meine Freude, BWV 713, bb. 53–5

Ex. 13

a) Jesu, meine Freude, BWV 227, chorale, line 2, version of 9th movement (bb. 19–22)

b) Jesu, meine Freude, chorale, line 2, version of 1st movement (bb. 3–4) etc.

101 By Alfred Du¨rr, ‘Melodievarianten in J. S. Bachs Kirchenliedbearbeitungen’, in A. Du¨rr and W. Killy (eds.), Das protestantische Kirchenlied im 16. und 17. Jahrhundert (Wiesbaden, 1986), pp. 149–63 (see 153).

the m o tets

205

Ex. 14

Gu - te

Nacht,

gu - te

Nacht,

gu - te

Nacht,

gu - te

Nacht,

a) Jesu, meine Freude, 9th movement, bb. 1–8, S I, II (T bassett omitted)

b) Sarabande from Partita No. 3, BWV 827, bb. 2–4, upper parts only (bass omitted) here in Bar form, a a b (bb. 20, 28, and 36)—later recur in somewhat altered versions (bb. 56, 64, and 84) and with a new interpolation (bb. 72–83). The reiteration of the word ‘nichts’ separated by rests in the Stollen (bb. 20 and 28) recalls that of ‘Komm’ at the start of Komm, Jesu, komm. But there the keyword was exchanged antiphonally between the two choirs; here Bach employs echo dynamics instead. Rhetorical gestures of this kind are, of course, rooted in the seventeenth-century German motet tradition. In particular, Bach would have encountered them in the work of his older relatives, the brothers Johann Christoph and Johann Michael Bach. The Abgesang (b. 36) begins in the homophonic style of the Stollen, but the full harmony characteristically conceals a fugal subject entry in the tenor part, which then inaugurates a fugato of immense power and depth. B2 and B4 are both trios, though for different combinations of voices. B2 is a gentle SSA trio in which the notion of ‘living in Christ Jesus’ conjures up beatiﬁc 3rds in the two soprano parts, accompanied by an alto bassett. B4 is an ATB trio in a pastoral 12/8 metre. The opening is remarkably similar to that of B2, both in melodic shape and in the parallel 3rds between the upper voices, presumably because both are concerned with life in Christ or the indwelling of Christ. B3 receives full, thorough fugal treatment, partly no doubt due to its central place in Bach’s symmetrical scheme, but also perhaps due to the centrality of its text: God’s indwelling Spirit, it tells us, renders one spiritual rather than carnal. Like the fugue in Der Geist hilft, it takes the form of a double fugue, in which each subject (setting a different clause of the text) is ﬁrst treated independently, after which the two subjects are combined.

I.5 Conclusion

One of the most noticeable features of Bach’s instrumental and keyboard music of the Co¨then and early Leipzig periods, as opposed to that of earlier years, is that compositions are often grouped together in sets of six (or multiples thereof), as shown in the following list: Original title1

Modern title

Date

Six Suittes avec leurs Pre´ludes Six Suites a Violoncello solo senza Basso Sei Solo a Violino senza Basso Six Concerts avec plusieurs Instruments Das Wohltemperirte Clavier Auffrichtige Anleitung Sei Sonate a Cembalo certato e Violino solo Six Suites de pie`ces pour le Clavecin ¨ bung, Op. 1 Clavier U

Six English Suites, BWV 806–11 Six Cello Suites, BWV 1007–12

c. 1720 c. 1720

Six Violin Solos, BWV 1001–6 Six Brandenburg Concertos, BWV 1046–51

1720 1721

Twenty-Four Preludes and Fugues, BWV 846–69 Thirty Inventions and Sinfonias, BWV 772–801 Six Sonatas for violin and obbligato harpsichord

1722 1723 c. 1722–5

Six French Suites, BWV 812–17 Six Keyboard Partitas, BWV 825–30

c. 1722–5 1725–30

Such groupings were, of course, standard in printed editions of the time, but none of Bach’s great collections of this period was published, with the single exception of Clavieru¨bung I. However, the ordering of compositions in sets of this kind must have had a special signiﬁcance for Bach at the time. He might have been aware that they marked the arrival of full maturity. Of his earlier keyboard and instrumental works, written during the Weimar period, only the Orgelbu¨chlein was ordered as a set, but it was left unﬁnished and did not receive a title till the Co¨then period. To a large extent Bach seems to have composed the pieces listed in the table in groups rather than subsequently arranging his music into them. This implies that the set might have acted as a self-imposed framework within which to order his musical ideas, for it has been remarked more than once that Bach was often at his best when writing within

1 In the case of the English Suites, the Cello Suites, and the French Suites, the titles are drawn from a pupil’s copy rather than an autograph, but it is highly likely that they reproduce the original wording.

c on c lusi o n

207

tight formal constraints.2 The pieces that made up a set would have a common format and would share certain underlying formal and stylistic principles. On the other hand, the maximum variety of invention would be sought. Hence stylistic diversity, ever an important consideration, is eventually elevated to the status of a governing principle. As a whole, each set is designed to provide a fully comprehensive and exhaustive survey of the instrumental, formal, and stylistic properties that Bach considered appropriate to the genre concerned. In keeping with this encyclopaedic tendency, certain sets have a clear didactic purpose: the keyboard works listed are intended not only for the delectation of accomplished players but as composition models and playing studies for students. Although these are Bach’s earliest sets of keyboard and instrumental works (only the fragmentary Orgelbu¨chlein is older), he was at the height of his powers when he composed them and in many cases they exhibit radical, innovative tendencies. In the field of instrumental chamber music, Bach must have had long experience of the standard basso continuo texture by 1720, certainly as a performer and possibly also as a composer—only one such work survives from earlier years, namely the Fuga in G minor for violin and continuo, BWV 1026 (Weimar, c. 1708/17), but others may be lost. His own major instrumental sets of the Co¨then years, however, represent an implied criticism of the standard continuo texture of his day. The continuo is either omitted altogether, as in the Cello Suites and Violin Solos, or else the harpsichord is elevated from its continuo role to that of obbligato instrument, on a par with the violin, as in the Six Sonatas, BWV 1014–19. Neither procedure is without precedent, though Bach seems to have been the first composer to employ it systematically throughout an entire set, with the single exception of J. P. von Westhoff, whose Six Suites of 1696 are the only known set for violin senza basso before Bach’s Six Violin Solos of 1720. Just as Bach had adapted a large instrumental ensemble to solo keyboard in his Weimar concerto transcriptions, so in his unaccompanied music for violin or cello he sought to create the impression of a complete chambermusic texture on a single string instrument, using multiple stops to engender a species of pseudo-polyphony. In the case of the violin, he may well have been inspired by the playing of the German virtuosos J. P. von Westhoff and J. G. Pisendel, as well as by the compositions of other violinists of the Austro-German tradition, such as J. H. Schmelzer, H. I. F. von Biber, and J. J. Walther. In the sonatas with obbligato harpsichord, the keyboard instrument takes upon itself one of the two melodic parts (the other being allotted to the violin) as well as the bass and, where necessary, the harmonic filling-in parts. In other words, it occupies solo and continuo roles simultaneously. Bach’s wish to release it from its purely accompanying role may well be connected with his own keyboard virtuosity. It can be no mere coincidence that around the same time the harpsichord rose to the 2 See, for example, Alfred Du¨rr, The Cantatas of J. S. Bach, rev. Eng. trans. by Richard D. P. Jones (Oxford, 2005), p. 204.

208

par t i

rank of soloist in Brandenburg Concerto No. 5, for in both cases the instrument has a concertato role alongside the violin (plus flute in the concerto). These pioneering works were to shape many future compositions of the same type. The solo harpsichord of the Brandenburg Concerto would act as a template for its use in the Six Concertos for solo harpsichord, BWV 1052–7, and in the Concerto in A minor for harpsichord, violin, and flute, BWV 1044; and the coupling of obbligato harpsichord with violin in the Six Sonatas, BWV 1014–19 would provide a precedent for its coupling with flute (BWV 1030 and 1032) and viola da gamba (BWV 1027–9). The use of harpsichord as solo instrument is by no means the only pioneering aspect of the Brandenburg Concertos. Equally significant is the juxtaposition of different types of concerto according to instrumental grouping and function. While Concerto No. 1 may be described as a concerto ripieno (since it lacks concertino), albeit with a small solo part for violin, Nos. 3 and 6 are concerti senza ripieno, consisting of concertino (plus continuo) only (though a certain unofficial concertino–ripieno differentiation is noticeable in No. 6). Concertos Nos. 2, 4, and 5, on the other hand, are concerti grossi, exhibiting a clear distinction between concertino and ripieno, though within the concertino in each case one instrument acts as primus inter pares—trumpet, violin, and harpsichord respectively—so that the concerto grosso type is modified by a strong element of the solo concerto. All the basic concerto types represented in the set are to be found among Vivaldi’s works. Bach’s distinction lies in realizing and amalgamating them in subtle new ways and in juxtaposing them within a single set of Concerts avec plusieurs instruments. Bach was at the forefront of new developments around 1720 not only in the instrumental disposition of his sonatas and concertos but also in the sphere of temperament and tonality. Although the use of all twenty-four major and minor keys of the tonal system was in the air at the time, Bach seems to have been the first to write a substantial composition in each key within a collection that included all keys, Das Wohltemperirte Clavier of 1722. This classed him with progressive thinkers such as Johann Mattheson, who championed the ‘modern’ tonal system alongside the fashionable galant style. Bach’s music, however, shows him to be an advocate of tradition as well as innovation, hence his employment of the conservative prelude and fugue genre for his comprehensive demonstration of the tonal system. We are informed by Forkel that he also used all twenty-four keys in his improvised fantasies, no doubt in conjunction with enharmonic change, chromatic harmony, and modulation to remote tonal regions. His Fantasias in D minor (BWV 903 no. 1) and G minor (BWV 542 no. 1) exhibit these tendencies and, despite the likelihood of subsequent refinements at some distance from the original conception, no doubt give an approximate idea of the improvisations that Forkel had in mind. This pseudo-improvisatory style is not confined to the fantasia, however. In English Suite No. 3 in G minor, Bach even applies it to the sarabande. Such transgressing of the normal boundaries of style and genre would become increasingly characteristic of the composer.

c on c lusi o n

209

The freedom of style, form, and genre that we encounter in Bach’s keyboard and instrumental works from 1717 onwards owes much to his cultivation of the vermischte Geschmack (‘mixed taste’), a judicious blend of native German characteristics with elements imported from other nations, particularly the Italian sonata and concerto and French dance music and ornamentation. The concept of the ‘mixed taste’ was not fully developed in print till J. J. Quantz’s Versuch einer Anweisung die Flo¨te traversiere zu spielen (Berlin, 1752), but from 1713 onwards Johann Mattheson had talked of the unification of French and Italian styles by German composers, and this became a reality at the Dresden court where Quantz was employed from 1729 to 1741. The court orchestra was led by the Concertmaster Jean Baptiste Volumier and the first violinist Johann Georg Pisendel (eventually Volumier’s successor), who together set about blending the two foremost national styles. By 1717, the year of Bach’s famous contest with Louis Marchand in Dresden, the ‘mixed taste’ was current in the Saxon capital, and it is likely that his first major encounter with it took place at that time.3 With regard to the French and Italian styles and their interaction, a clear stylistic evolution can be traced in Bach’s suites and partitas for harpsichord, violin, and cello. In the English and Cello Suites, the stylistic constituents are regularized and kept within certain predetermined boundaries. The French Suites, by comparison, exhibit a greater range of style and a wider choice of dance types. The solo violin Partitas belong not so much to the French dance suite as to the Italian sonata da camera tradition, hence the Italian movement titles of Nos. 1 and 2. In No. 3, on the other hand, a ‘Preludio’ in Italian violin style introduces a suite of French ballet dances, a Lullian element that in the keyboard and cello suites is restricted to the intermezzi. The keyboard Partitas of Clavieru¨bung I elevate stylistic diversity, already prominent in the French Suites, to the status of a governing principle. Like the solo violin Partitas, they draw on the Italian sonata da camera tradition, but in this case Italianate movements are intermingled with French dances. Linguistic differentiation is used to clarify the intended stylistic reference. As in the French Suites, Bach’s cantabile style is brought to bear on the traditional dances, but now so too are features of the progressive galant style—the two modes of writing become inseparable. In addition, partly as a result of the da camera influence, the link with dance style is often tenuous, hence the formula ‘Tempo di Minuetto/Gavotta’, borrowed from Corelli, and the non-dance titles Capriccio, Burlesca, and Scherzo. In such cases the individuality of Bach’s invention takes precedence over conventional aspects of genre, with the result that the movements concerned are best described as character pieces. They are not without precedent. Already in the Menuet from the Suite in A minor, BWV 818a, and in the Courante and Sarabande from English Suite No. 5 in E minor, both from around 1720, characterization of the dance type recedes in favour of unique, unrepeatable strokes of invention. Not only in the Partitas but in many of Bach’s dance movements from the 3 See Ulrich Siegele, ‘Bachs vermischter Geschmack’, in M. Geck and K. Hofmann (eds.), Bach und die Stile (Dortmund, 1999), pp. 9–17.

210

par t i

period 1717–29 the French and Italian styles are indissolubly merged. In some cases, as in the Burlesca from Partita No. 3 in A minor (BWV 827), for example, he presents an opening theme in French dance style, ornamented and variegated in rhythm, but later moves into the even flow in quick note-values of the Italian sonata da chiesa. The sharp contrast between the initial idea of the Burlesca, with its rhythmic theme exchanged between the hands, and the highly fluent, largely treble-dominated concluding paragraph (from b. 25 onwards) perhaps partly explains Bach’s decision to alter the movement title from ‘Menuet’ (Clavierbu¨chlein for Anna Magdalena Bach, 1725) to ‘Burlesca’ (original edition, Leipzig, 1727).4 The history of Bach’s keyboard and instrumental suites from 1717 to 1729 may be described as a process of individuation. A parallel process takes place in the preludes of The Well-Tempered Clavier (WTC I) which, upon revision, shook off their original character as brief, impromptu-style introductions to the fugues and became substantial, highly developed entities in their own right. Some of the enlarged preludes were now furnished with impressive toccata-style conclusions (c, C♯, e); the three-voice preludes included a pastorale (E) and a Corellian trio-sonata movement (b); and Bach’s most expressive cantabile style inhabits the minor-mode preludes, particularly those in the more remote keys (c♯, e♭, f, g, b♭). In the fugues, the character of subject and countersubjects is all-important—fugue is viewed primarily as a mode of thematic treatment rather than a study in counterpoint. This has two important consequences: contrapuntal devices are employed not so much for their own sake as for expressive purposes; and fugues are no longer seamless but divided by prominent cadences into melodic periods and paragraphs, often falling into overall bipartite or tripartite schemes. The majority of the large-format fugues—those in C♯, c♯, d♯, A, and a—fall into an overall three-phase form according to their fugal structure. This is also true of the large-format organ fugues from the earlier Leipzig years, those in b, e, F, and c (BWV 544, 548, 540, and 537). Two of them (b and c) are cast in ABA1 reprise form, like the C♯ major Fugue from the WTC I; one (F) in a tripartite form dictated by its triple-fugue structure, as in the C♯ minor Fugue from the WTC I; and one (e) in da capo form, which is not represented among the fugues of the WTC I. The Inventions and Sinfonias build upon the achievements of The Well-Tempered Clavier. In evolving the concept of the inventio, Bach created a new genre to show how a character theme with potential for development was capable of generating an entire composition. For this purpose canonic and fugal means were used, involving stretto, subject inversion, and above all interchange of parts. Bach’s character themes are often in unspecified dance rhythm or in cantabile style. Among the most individual of the

4 This and other ways in which Bach fuses the French and Italian styles are described by David Ledbetter in ‘Les Gouˆts re´unis and the Music of J. S. Bach’, Basler Jahrbuch fu¨r historische Musikpraxis, 28 (2004), pp. 63–80, and in his ‘A Question of Genre: J. S. Bach and the “Mixed Style” ’, in T. Donahue (ed.), Music and Its Questions: Essays in Honour of Peter Williams (Richmond, Va., 2007), pp. 205–24.

c o nc l usi on

211

last-named type is the Invention in E, with its continuous syncopation and diminution ﬁgures. In the course of the second strain, the customary two-part counterpoint with Stimmtausch (exchanged parts) gives way to a substantial mediant episode (bb. 29–42) for decorative treble and plain supporting bass. Bach stated as one of the main aims of the Inventions and Sinfonias: ‘am allermeisten aber eine cantable Art im Spielen zu erlangen’ (‘above all, to arrive at a cantabile style of playing’). To play in this manner implies a corresponding style of music, and Bach does in fact build on certain cantabile preludes from The Well-Tempered Clavier to achieve a similar style throughout much of the later collection. This is also true of the roughly contemporaneous French Suites, particularly their Allemandes and Sarabandes. Bach was here writing in accordance with contemporary taste, for the chief rule of composition for the ‘galant homme’ of Johann Mattheson’s Das Neu-Ero¨ffnete Orchestre (Hamburg, 1713) was ‘daß man Cantable setze’ (‘to compose in a singing style’).5 And Bach’s cantabile compositions of this period culminate in the keyboard Partitas of Clavieru¨bung I, in which the singing style is allied to all manner of galant features, such as the triplets and sigh ﬁgures of the Sarabande from Partita No. 3 or the syncopated rhythms and cut-off cadences of the Aria from No. 4. An earlier collection, the Sei Sonate a cembalo certato e violino solo, BWV 1014–19, lacks such galant gestures, but the cantabile writing it contains counts as one of its greatest virtues, as noted by C. P. E. Bach: ‘Es sind einige Adagii darin, die man heut zu Tage nicht sangbarer setzen kann’ (‘There are several Adagios among them which even today could not be composed in a more singing style’).6 Sangbarkeit of this kind was a fundamental component of the galant style, hence the late approbation of the Bach son, himself a noted exponent of that style. The ritornello principle plays a fundamental role in Bach’s music of this period, not only in the Brandenburg Concertos, where it would be expected, but in the pre´ludes to the English Suites and in the fast second and fourth movements of the Six Sonatas for violin and obbligato harpsichord. In addition, the expansive organ preludes in b, e, and c (BWV 544, 548, and 546) are cast in ritornello form; and in two cases (b and c) the episodes are fugal, forming a sharp contrast with the essentially homophonic ritornellos. The reverse is true of the massive fugal second movements of the solo violin Sonatas. Here the fugal expositions, with their copious multiple-stopping, which creates the impression of full texture, may be construed as ‘tutti’ ritornellos, whereas the single-line passage-work acts like solo episodes. It is rare in Bach for ritornello form to be deployed without some kind of overarching paragraph structure. Bipartite and binary form are both found (in BWV 544 and 1018 respectively), but by far the most common are the da capo (ABA) and reprise forms (ABA1) of Bach’s vocal arias. Despite their French name, the pre´ludes of the English Suites are Italianate

5 See Martin Geck, ‘Bachs Inventionen und Sinfonien im galanten Diskurs’, in M. Geck (ed.), Bachs Musik fu¨r Tasteninstrumente (Dortmund, 2003), pp. 159–80. 6 C. P. E. Bach, letter to J. N. Forkel, 7 Oct. 1774; BD III, No. 795.

212 p art i concertante movements in da capo form, in which the A-paragraph functions as a framing ritornello, while the central B-paragraph contains contrasting episodes in alternation with varied or abridged ritornello returns in keys other than the tonic. In the last two English Suites (those in e and d), the ritornellos are designed as fugal expositions, but this is primarily a matter of texture—the da capo–ritornello structure is unaffected by it. Fugal ritornellos become the norm in the fast second and fourth movements of the Six Sonatas, BWV 1014–19, where reprise and binary forms are used as well as da capo form. These movements share with the pre´ludes of the English Suites that crucial moment, so characteristic of the concerto, when the opening ritornello (paragraph A) reaches a tonic full-close, after which paragraph B opens with sharply contrasting material along the lines of a concerto episode. In the fast movements of the Brandenburg Concertos (BC), ritornello form is similarly deployed within an overall reprise (BC 1 nos. 1 and 3, BC 2 no. 1, BC 4 no. 3, BC 5 no. 1, and BC 6 no. 1) or da capo structure (BC 4 no. 1, BC 5 no. 3, and BC 6 no. 3). In most of the reprise-form movements, dominant-key material at the end of the first paragraph returns in the tonic at the end of the third, after the manner of Bach’s reprise-form arias. Two of the da capo movements, BC 4 no. 1 and the fugal BC 5 no. 3, open with an extensive compound ritornello made up of two or three smaller units, each of which might have functioned as a complete ritornello in a more modest structure. The finales of Concertos Nos. 4, 5, and 6, at the start of the B-paragraph, exhibit a similar sharp contrast to that described earlier in connection with the suites and sonatas. In the fugal finale of Concerto No. 4, for example, the polyphonic style of the opening ritornello/fugal exposition suddenly gives way to the bravura of the solo violin—an extreme conjunction of opposites—but the two elements are combined thereafter. Such combining of opposites is a striking example of that integration which, whether stylistic or thematic, was an overriding concern for Bach in his instrumental and keyboard music of this period. In the outer movements of the Brandenburg Concertos, for example, the episodes more often than not develop material that has been stated in the ritornellos. And in the fugal Allegros of the Sonatas for violin and obbligato harpsichord, a contrasting, episodic theme is introduced in the middle paragraph which first alternates with ritornello material (above all, the fugue subject), after which the two themes are gradually integrated.7 Thus stark opposition increasingly gives way to a satisfying, logical concordance. Another form of integration close to Bach’s heart is that which concerns old and new styles. The WellTempered Clavier I and the Clavieru¨bung I both owe much in conception to his predecessors J. C. F. Fischer and Johann Kuhnau respectively. In Bach, however, style is modernized, compositional technique becomes considerably more sophisticated, and the dimensions of the individual movements belong to a different order of magnitude. Nonetheless, reference is not infrequently made in Bach’s preludes and 7 This process of integration is studied in Hans Eppstein, Studien u¨ber J. S. Bachs Sonaten fu¨r ein Melodieinstrument und obligates Cembalo (Uppsala, 1966).

c on c lusi o n

213

fugues to Fischer’s themes, and old genres still make their mark, albeit in an updated form—for example, the North German multi-sectional praeludium with its toccatastyle and fugal constituents in the Praeludium in Eb (BWV 857) and the Toccata in E minor (BWV 830 no. 1). Bach had little opportunity to compose sacred music in Co¨then—we know of only one such work (BWV Anh. I 5, whose music is lost). However, he regularly wrote serenatas—a form of secular cantata cultivated by Stradella, Alessandro Scarlatti, Fux, and Caldara—for New Year’s Day and for Prince Leopold’s birthday. These take the form of dialogues between mostly allegorical characters, hence the preponderance of duet writing among their arias, recitatives, and choruses. Dance rhythms are also of common occurrence; indeed one cantata, No. 194a, is even conceived as a vocal version of a French ouverture-suite. Five of these secular works (BWV 66a, 134 a, 173 a, 184 a, and 194 a) are known to have been parodied to form sacred cantatas during the early Leipzig years. Bach must have considered the courtly idiom suitable for the church, since he took few steps to modify it—even a Lutheran chorale, the most obvious mark of sacred style, is not added in every case. Leipzig secular cantatas of the 1720s take two chief forms: the lyrical ‘cantata’ proper, either for solo voice (BWV 204 and 210 a) or for several singers (BWV 36c); and the dramma per musica (BWV 249a, 205, and 207). This was a common term for opera at the time, but in Leipzig it denoted a mini-opera, complete with plot and dramatis personae but without staging. The characters might be allegorical figures, such as Fortune, Gratitude, Diligence, and Honour (BWV 207), but more often they are drawn from Greek or Roman mythology (BWV 249a and 205), as in contemporary opera seria. Finally, the Trauer Music or tombeau for the Electress of Saxony and Queen of Poland, BWV 198 of 1727, is a special case. It was performed in the university church, but not in the context of a service, hence its secular text—an ode by J. C. Gottsched. Bach ignored Gottsched’s ode form, however, and set it as a large-scale cantata in two parts, alternating recitatives and arias within framing choruses. In tone and idiom the work is not dissimilar to parts of the St Matthew Passion, first performed during the same year. It is fitting, then, that two years later the outer choruses were parodied alongside extracts from the Passion in the lost funeral music for Prince Leopold of Anhalt-Co¨then (1729). No less appropriate is the reuse of both framing choruses, together with all the arias, in the lost St Mark Passion of 1731. No more perfect convergence between sacred and secular could be imagined: it was natural for Bach to use the same music for the death of the revered queen, protector of the faith (‘Glaubenspflegerin’; no. 7 line 3), and for that of the Saviour of the world. In the field of church music Bach was clearly aware of the demanding nature of his own compositions. In a memorandum to the Leipzig town council of 15 August 1736 he pointed out that ‘die musicalischen Kirchen Stu¨cke so im ersteren Chore gemachet werden, u. meistens von meiner composition sind, ohngleich schwerer und intricater sind, weder die, so im anderen Chore’ (‘The concerted pieces that are performed by the first choir, which are mostly of my own composition, are incomparably harder and

214

par t i

more intricate than those sung by the second choir’).8 Furthermore, a major challenge for him, now that Neumeister had transformed the church cantata into ‘ein Stu¨ck aus einer Opera’ (‘a piece from an opera’),9 was to find new ways of reconciling sacred and secular styles, while at the same time fulfilling the condition of his Leipzig appointment that his church music should ‘not make an operatic impression, but rather incite the listeners to devotion’.10 The audition cantatas, BWV 22 and 23, show Bach feeling his way towards a compromise between the progressive, opera-influenced and conservative, ecclesiastical styles. The choral third movement of Du wahrer Gott und Davids Sohn, BWV 23, which originally formed the finale, recalls the Co¨then serenatas in its rondeau form with duet episodes, its dance rhythm, and its homophonic texture. Later Bach appended an additional movement as finale, a highly elaborate setting of the German Agnus Dei, Christe, du Lamm Gottes. The return to this chorale, which had been quoted instrumentally in a previous movement (the recitative no. 2), recalls a standard Weimar procedure of Bach’s and significantly shifts the balance of the work in an ecclesiastical direction. The second audition cantata, Jesus nahm zu sich die Zwo¨lfe, BWV 22, places an ecclesiastical frame of biblical text and chorale around the operatic forms of aria and recitative, a frame that would become standard in the cantatas of Cycle I. Moreover, the impressive opening movement incorporates two modes of treatment that would recur regularly during Cycle I and beyond: the fugal setting of biblical words and the use of the bass voice as vox Christi, as in traditional Passion settings. The main weight of the Cycle I cantatas lies in the great opening chorus with biblical words. In keeping with the great authority of the texts, the time-honoured methods of fugue and motet are considered most appropriate for setting them. In some of the earliest cantatas of the cycle (Nos. 75, 76, 105, and 46), a concertante introduction in ritornello form acts as ‘prelude’ to a fugue, constructed in some cases according to the permutation principle, as in some of Bach’s earliest cantatas. As a whole, the movement acts as a musical correlative to the biblical theme that generates the entire cantata: ‘the poor’ (No. 75), ‘the glory of God’ (No. 76), fear of divine judgement (No. 105), and ‘lamentation’ (No. 46). At a somewhat later stage, concertante and fugal elements are integrated, giving rise to such schemes as rit[ornello]—prelude—fugue—postlude— (rit.) (Nos. 69 a, 40, and 65) or rit.—prelude—fugal exp. I—rit.—fugal exp II—rit. (Nos. 136, 67, and 104). The prelude and postlude are often variants of the ritornellos with added vocal parts, and even the fugue subject might be derived from the ritornello, so that the whole edifice hangs together. In a few cases (Nos. 179, 64, and 144) the opening chorus consists of nothing more than a motet-fugue in alla breve time with colla parte instruments and continuo, written in Bach’s most conservative polyphonic style. Bach returned to similar motet-style or concertante-fugal modes of biblical-text setting in the non-chorale-based cantatas of Cycles II and III.

8 9 10

BD I, No. 34; NBR, No. 183. Preface to his Geistliche Cantaten statt einer Kirchen-Music (1700). BD I, No. 92; NBR, No. 100.

c on c lusi o n

215

The Lutheran chorale is subordinate to the biblical text in Cycle I and in related cantatas from Cycles II and III. The chorale-finale is normally a simple four-part setting, though in twelve of the earlier cantatas from Cycle I it is embedded in a more or less elaborate instrumental setting, as in the finale of the audition cantata No. 22. The inner movements might include an additional plain four-part chorale or else a solo chorale cantus firmus setting with instrumental ritornellos (BWV 95, 166, 86, 37, and 44). On several occasions the biblical-text setting of the opening movement is combined with a complete instrumental chorale quotation, played either in four-part harmony (BWV 25) or in strict canon (BWV 77 and 48). Only in two cases (BWV 60 and 190) is the biblical text combined with a sung chorale. In Cantata No. 190, second movement, the chorale is troped by recitative, and this also applies to those few cases in which the opening text of a Cycle I cantata is chorale-based rather than biblical (Nos. 138, 95, and 73). The ‘chorale cantata’ of Cycle II is clearly anticipated in the Cycle I cantata Warum betru¨bst du dich, mein Herz, BWV 138, in which the first, second, and sixth movements all feature the chorale of that name. In the chorale-cantata genre, the texts of all movements are based on one and the same hymn (albeit paraphrased in the inner verses), so that the chorale replaces biblical words as the chief source of ecclesiastical authority. As in Cycle I, the main weight lies in the opening chorus, whose vocal parts consist of the chorale cantus firmus (normally in the soprano) and more or less elaborate accompanying parts. This chorale texture is set against the background of an independent, concertante instrumental texture, derived from the opening ritornello. This form, which Bach had hardly evolved before Cycle II, is capable of encompassing a wide range of styles—for example, those of the French ouverture (Cantata No. 20), the Italian concerto (No. 7), the Lullian chaconne (No. 78), the cantus firmus motet (Nos. 2, 38, and 121), and a modernized motet type of Bach’s own with partially independent instrumental parts (Nos. 135 and 101). Like the biblical-text chorus, the chorale-chorus typically creates a vivid tone-picture of the primary image of the text—for example, transience (No. 26), tribulation (No. 3), a visionary death scene (No. 8), or Christ at the River Jordan (No. 7). The concluding chorale is normally a plain four-part setting, as in Cycle I, but in Was willst du dich betru¨ben, BWV 107, it is sung against a ritornello-based instrumental background, as in the Cycle I cantatas Nos. 138 and 109. In Cantata No. 107 (as in the first movement of No. 68), this background draws on the rhythms of the siciliana, the most frequently occurring dance rhythm in Bach’s sonatas and concertos. The cantata is exceptional not only for its chorale-finale in siciliana rhythm: it is also the only Cycle II composition that takes the form of a chorale cantata per omnes versus; in other words, the original text of the chorale is preserved throughout. This type, which Bach had employed in the very early Cantata No. 4, albeit in association with a relatively antiquated style, recurs in the Cycle III cantata No. 137 and soon afterwards in No. 129 (1727). It would then become standard in the chorale cantatas of the 1730s.

216 p art i In the chorale cantatas of Cycle II, where a chorale verse is retained unaltered in an inner movement, the chorale melody is sung as a solo cantus firmus with instrumental ritornellos and accompaniment (fourth movement of BWV 178, 114, and 92; second of BWV 113), as in similar movements from Cycle I. One form of chorale treatment that does not occur in Cycle II, or at least not in the form described here, is the chorale-aria, of which precedents may be found in the St John Passion, Versions I (‘Mein teurer Heiland’, no. 32) and II (‘Himmel reiße’). This species, in which aria-ritornello form and the Bar form of the chorale are coordinated, comes into its own in two of the dialogue cantatas of Cycle III, BWV 49 (no. 6) and 58 (nos. 1 and 5). It is later employed in Cantata 158 (no. 2), which might have been composed around the same time, and in the second movement of two cantatas from the Picander Cycle, Nos. 156 and 159. The arias and recitatives that form the majority of the inner movements of Bach’s sacred cantatas are for the most part barely distinguishable from those of his secular cantatas except in the nature of their texts. The da capo form of contemporary opera is predominant, alongside the reprise form (ABA1) that Bach himself evolved—more sophisticated than da capo form in that paragraph A typically modulates to the dominant and therefore has to be tonally adjusted upon its return (A1) so that it may close in the tonic. Bach introduced secular dance rhythms of all kinds in the arias, flying in the face of contemporaries such as G. E. Scheibel, who in his Zufa¨llige Gedancken von der Kirchen-Music of 1721 claimed that such rhythms were inappropriate in church music.11 For Bach this convergence of sacred and secular explains why he found it so easy to parody secular occasional works to form sacred compositions that could become permanent items in the church-music repertoire. A particularly striking example is the aforementioned Co¨then secular cantata BWV 194a, whose sacred parody Ho¨chsterwu¨nschtes Freudenfest (BWV 194) was first used for an organ consecration during Bach’s first year in Leipzig, then revived on Trinity Sunday 1724—the last cantata performance of Cycle I. The work takes the form of a French ouverturesuite for vocal and instrumental ensemble in which every aria displays a different dance rhythm. The chorale added at the end of each half is the only specifically sacred feature of the work. In Cycle I sacred and secular styles are frequently placed side by side. An extreme example is the Purification cantata Erfreute Zeit im neuen Bunde, BWV 83, in which the concerto-style opening aria, with its solo violin, horns, oboes, and strings, gives way to a setting of the German Nunc Dimittis, sung to the traditional chant as cantus firmus and accompanied by canonic strings and continuo. In the penultimate movement of the Cycle II cantata Nimm von uns, Herr, du treuer Gott, BWV 101, sacred and secular styles are not merely juxtaposed but combined. In the opening ritornello and

11 Gottfried Ephraim Scheibel, Zufa¨llige Gedancken von der Kirchen-Music, wie sie heutiges Tages beschaffen ist (Frankfurt and Leipzig, 1721); Eng. trans. as ‘Random Thoughts about Church Music in Our Day’, introd. and trans. by Joyce L. Irwin, in Carol K. Baron (ed.), Bach’s Changing World (Rochester, NYand Woodbridge, 2006), pp. 227–49 (see 241).

c o n c lusi o n

217

in the vocal section that follows, the chorale cantus firmus is played or sung in counterpoint with a florid theme in siciliana rhythm, a combination that proves to be the main content of the movement. On Easter Sunday 1725, in the context of Cycle II, Bach performed two cantatas at opposite ends of the sacred–secular spectrum, namely Christ lag in Todes Banden BWV 4, and Kommt, fliehet und eilet, BWV 249. The early Cantata No. 4 (c. 1709), whose revival brought the chorale cantatas of Cycle II to a close, belongs to the strict per omnes versus type and is saturated throughout by the old melody associated with Martin Luther’s hymn. Cantata No. 249, on the other hand, was parodied from a secular dramma per musica (BWV 249a), performed before the Duke of Weißenfels only about five weeks previously. In Leipzig at that time, as already noted, the dramma per musica was equivalent to one act of an opera. Few changes were made when the highly theatrical Weißenfels piece was parodied for Easter beyond the libretto, in which the four shepherds of the original became four disciples. With its sung plot and dramatis personae, the Easter cantata, both in its 1725 version and in its oratorio version of about 1735, can only be described as a sacred dramma per musica—something that Bach attempted at no other stage of his career. In Cycle III, in the Picander Cycle, and in the intervening period (altogether 1725–9), the impact of secular music on sacred reaches a new level with the wholesale importation of some of Bach’s existing concerto or ouverture movements into numerous cantatas, the solo part normally being played on obbligato organ. Again, Scheibel would have objected in the strongest terms, for he argued that the quantity of instrumental music in church cantatas should be reduced to a minimum.12 In most cases the movements concerned act as immense sinfonias. In three cases, however, the concerto or ouverture movement that has been borrowed is transformed into a chorus or aria (BWV 110 no. 1, 169 no. 5, and 146 no. 2), and all three are among Bach’s most effective adaptations. The fugato of a French ouverture acts as an utterly convincing musical portrayal of the ‘laughter’ of the psalm text in Cantata No. 110; a siciliana of great beauty represents the ‘world’ in the alto aria from No. 169 (‘Die in me, world and all your love’); and in No. 146 the pathos of an Adagio slow movement in the minor mode with added vocal parts illustrates the biblical words ‘Through much tribulation must we enter into the Kingdom of God’ (Acts 14: 22). Another late importation from the secular sphere is the solo cantata. Bach occasionally cultivated the Italianate secular variety (BWV 202, 203, and 204), and in Weimar he had transferred it to the sacred domain in settings of two texts by Lehms (BWV 199 and 54). He revived this sacred variety, which can hardly be distinguished from the secular ‘parent’ type apart from its text, in the last phase of Cycle III (1726–7), first reverting to further texts by Lehms (BWV 170 and 35) and then turning to texts by other authors (BWV 169, 56, 55, 52, 82, and 84). The secular origin of the genre is clearly reflected in these compositions: no biblical texts are included and the Lutheran

See n. 11.

218 p art i chorale plays a minimal part—it is absent from Cantatas Nos. 170, 35, and 82, and elsewhere is represented only by a plain four-part setting at the end. In addition, the solo voice parts are often operatic in the considerable demands they make on the singer’s vocal technique and interpretative skills. The introduction of obbligato organ (BWV 170, 35, and 169) and of concerto movements (BWV 35, 169, and 52) confirms the impression of works that incline to the secular sphere more wholeheartedly than most of Bach’s other Leipzig sacred compositions. Unlike the solo cantata, the dialogue cantata, which also plays a significant role in Cycle III, traditionally belongs to the sphere of sacred music. Certain pre-Weimar and Weimar cantatas included movements in dialogue form (BWV 106 nos. 6–7, 21 nos. 7–8, 172 no. 5, and 152 no. 6). But the first complete cantata devoted to sacred dialogue was the remarkable Cycle I cantata No. 60, O Ewigkeit, du Donnerwort, which takes the form of a ‘Dialogus zwischen Furcht und Hoffnung’ (‘Dialogue between Fear and Hope’). The Jesus–Soul dialogue—the type represented in Cycle III (BWV 57, 32, 49, and 58)—flourished in seventeenth-century Germany in the hands of Hammerschmidt, Ahle, Briegel, Schu¨tz, and Buxtehude, but Bach had to update it by introducing operatic and concertante features in order to render it suitable for congregations of the late 1720s. He seems very much at home here in attempting a compromise between traditional, ecclesiastical and modern, secular styles. Accordingly, alongside chorale-arias and vox Christi movements we encounter, for example, an aria in giguepassepied rhythm (BWV 57 no. 7) or an operatic love-duet (BWV 49 no. 3). Although Bach’s two great Passion settings of the 1720s—those according to St John and St Matthew—belong to the liturgical oratorio-Passion genre, with its strict adherence to the biblical narrative, they owe much to the highly secularized, theatrical Passion-oratorio genre cultivated by Brockes, Hunold, and others. In particular, in certain movements from both Passions Bach adopts the dialogue format often employed by Brockes—a dialogue no longer between Jesus and the Soul but between the allegorical figure ‘Tochter Zion’ (Daughter Zion) and a group of ‘Die Glaübigen’ (the Faithful). This explains the responsorial form of the two bass arias with chorus from the St John Passion, ‘Eilt’ (no. 24) and ‘Mein teurer Heiland’ (no. 32). It also explains the dialogue format (Chorus I and II) of the St Matthew Passion as a whole and the question–answer sequence of the type ‘Kommt!—Wohin?—nach Golgotha’ (‘Come—Where?—to Golgotha’) that plays a significant part in both Passions. In the St Matthew Passion, in particular, the dialogue element derived from the theatrical Passion-oratorio is largely responsible for the dramatic power and immediacy of the Picander–Bach commentary on the Gospel narrative. Bach is concerned not just with bringing the Passion narrative to life, however, but with integrating it by musical means, which he must have viewed as a necessity for a work spread over such a vast canvas. With this in mind he employs varied returns of specific chorales and turbae (crowd choruses) as effective means of integration in both Passions. The emphasis in the St John Passion is on the integrating function of turbae; in the St Matthew Passion, on that of the chorales, but the essential principle is the same in both cases.

PA R T I I The middle Leipzig years: 1729–1739

This page intentionally left blank

II.1 Introduction

By 1730 Bach was no longer content with his situation in Leipzig. He quarrelled with the local council about his teaching duties at the Thomasschule1 and showed dissatisfaction with the provisions for local church music in a memorandum entitled ‘Kurtzer, iedoch ho¨chstno¨thiger Entwurff einer wohlbestallten Kirchen Music; nebst einigem unvorgreiflichen Bedencken von dem Verfall derselben’ (‘Short but most necessary Draft for a well-appointed Church Music, with certain modest Reflections on the Decline of the same’),2 dated 23 August 1730. Two months later, on 28 October, Bach wrote to an old schoolfellow, Georg Erdmann, listing his complaints and seeking an alternative post:3 Da aber nun (1) finde, daß dieser Dienst bey weitem nicht so erklecklich als mann mir Ihn beschrieben, (2) viele accidentia dieser station entgangen, (3) ein sehr theu¨rer Orth u. (4) eine wunderliche und der Music wenig ergebene Obrigkeit ist, mithin fast in stetem Verdruß, Neid und Verfolgung leben muß, als werde geno¨thiget werden mit des Ho¨chsten Beystand meine Fortun anderweitig zu suchen. (But since 1) I find that the post [at Leipzig] is by no means so lucrative as it was described to me; 2) I have failed to obtain many of the fees pertaining to the office; 3) the place is very expensive; and 4) the authorities are odd and little interested in music, so that I must live amid almost continual vexation, envy, and persecution; accordingly I shall be forced, with God’s help, to seek my fortune elsewhere.)

Rather than moving from Leipzig, however, Bach struck out in new directions. He continued to write in the sacred cantata and Passion genres, though to a far more limited extent than before, but he also cultivated two large-scale church-music genres that were new to him as a composer, the oratorio and the Missa (Kyrie and Gloria). The church was no longer the only major focus of his activity as a composer and performer, however. In March 1729 he became director of the Leipzig Collegium musicum (music society) that had been founded by Telemann in 1701. In this capacity Bach gave weekly concerts throughout the year at Gottfried Zimmermann’s coffee house, indoors during the winter and outdoors (in the coffee garden) during the

1 2 3

BD II, Nos. 280–1; NBR, No. 150. BD I, No. 22; NBR, No. 151. BD I, No. 23; NBR, No. 152.

222 p a r t i i summer. At these so-called ordinaire Concerten he performed works by other composers, including Handel, Steffani, Locatelli, and Johann Bernhard Bach,4 alongside compositions of his own—secular cantatas, concertos, ouverture-suites, sonatas, and no doubt keyboard works. Bach not only retained his Leipzig post, which after all possessed considerable advantages, but he also sought honorary positions elsewhere. Probably in 1729 Duke Christian of Weißenfels, whose birthday Bach had celebrated with a cantata performance in 1713 and 1725 (BWV 208 and 249a respectively), appointed him non-resident Court Capellmeister, a position Bach had previously held at Co¨then. In addition, after visiting Dresden in September 1731 to give an organ recital at the Sophienkirche and no doubt to attend the premiere of Johann Adolph Hasse’s opera Cleofide,5 Bach applied for a court title there from the Elector of Saxony. With this in mind he dedicated his newly composed Missa, BWV 2321—the Kyrie and Gloria of what would eventually become the B minor Mass—to the Elector on 27 July 1733, mentioning his dissatisfaction with conditions at Leipzig:6 Ich habe einige Jahre und bis daher bey denen beyden Haupt-Kirchen in Leipzig das Directorium in der Music gehabt, darbey aber ein und andere Bekra¨nckung unverschuldeter weise auch iezuweilen eine Verminderung derer mit dieser Function verknu¨pfften Accidentien empfinden mu¨ßen, welches aber ga¨nzlich nachbleiben mo¨chte, daferne Ew. Ko¨nigliche Hoheit mir die Gnade erweisen und ein Praedicat von Dero Hoff-Capelle conferiren . . . und ich offerire mich . . . iedesmahl auf Ew. Ko¨niglichen Hoheit gna¨digstes Verlangen, in Componirung der Kirchen Musique sowohl als zum Orchestre meinen unermu¨deten Fleiß zu erweisen (For some years and up to the present moment, I have had the Directorium of the Music in the two principal churches in Leipzig, but have innocently had to suffer one injury or another, and on occasion also a diminution of the fees accruing to me in this office; but these injuries would disappear altogether if Your Royal Highness would grant me the favour of conferring upon me a title of Your Highness’s Court Capelle . . . I offer myself . . . to show at all times, upon Your Royal Highness’s most gracious desire, my untiring zeal in the composition of music for the church as well as for the orchestra)

It was not until three years later, on 19 November 1736, that Bach’s request was granted and he received the title of ‘Electoral Saxon and Royal Polish Court Compositeur’.7 Bach paid homage to the Elector not only through the dedication of the Missa but through the composition of a series of secular cantatas for the ruler and members of his family, performed at the extraordinaire Concerten (special, occasional concerts) of

4 Handel: Armida abbandonata, HWV 105; Steffani: Ouverture La Tempeˆte from Il zelo di Leonato (1691); Locatelli: Concerto grosso in F minor, No. 8 from XII Concerti grossi a quattro e a cinque . . . Opera prima (Amsterdam, 1721); and Johann Bernhard Bach: five Ouvertures. See Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992), VBN I/H/2, I/S/3, I/L/1, and I/B/4–8. 5 See Christoph Wolff, ‘Anmerkungen zu Bach und “Cleofide”’, and Reinhard Strohm, ‘Johann Adolph Hasses Oper “Cleofide” und ihr Vorgeschichte’, both in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld (Kassel, 1988), pp. 167–9 and 170–6 respectively. 6 7 BD I, No. 27; NBR, No. 162. BD II, No. 388; NBR, No. 190.

introd uction

223

the ‘Bachische’ Collegium musicum during the 1730s and early 1740s.8 Most of them belong to the dramma per musica type—equivalent to one act of an opera. Since they contain some of Bach’s finest music, it is hardly surprising that in some cases he wanted to raise them above the level of occasional music—often restricted to a single performance—to that of regularly performable repertoire music. This involved parody, the re-texting of existing music, and all of Bach’s large-scale sacred works of the 1730s, plus several cantatas, originated in this way. Three of the drammi per musica for the Electoral House of Saxony, BWV 213–15, were plundered for the choruses and arias of the Christmas Oratorio, Parts I–V; and the Ascension and Easter Oratorios also had secular models (BWV Anh. I 18, 196; BWV 249a), as did Cantatas Nos. 30 and 36 (BWV 30a and 36c). The foremost meditative movements of the lost St Mark Passion (BWV 247, 1731) were drawn from the Trauer Music (BWV 198, 1727), which shows Bach’s keen awareness of what was appropriate: music originally written for the death of the Queen of Poland, already reused in the Funeral Music for Prince Leopold of Anhalt-Co¨then (BWV 244a, 1729) alongside music selected from the St Matthew Passion, could serve without incongruity for the commemoration of the death of Christ himself. In other cases, the parody model might be sacred rather than secular. Part VI of the Christmas Oratorio is based on a lost cantata of about 1734, believed to have been written for Michaelmas (29 September),9 a big occasion in Leipzig due to the influx of large numbers of visitors for the Michaelmas Fair. Likewise, all known parody models for the five Missae, BWV 2321 and 233–6, are movements from Bach’s repertoire of sacred cantatas, built up over the years since his arrival in Leipzig in 1723. In addition to this compositional activity in relation to the church music of the 1730s, much of it involving parody, it has recently become clear that Bach came to rely heavily during this period on the church music of his contemporaries. He had a cantata by Telemann, Machet die Tore weit (TVWV I:1074), performed on Advent Sunday (28 November) 1734, perhaps to relieve his burden in the run-up to the Christmas Oratorio. And this may not have been an isolated instance.10 Above all, however, two complete cantata cycles by Bach’s colleague Gottfried Henrich Sto¨lzel, Capellmeister at Saxe-Gotha, were evidently performed at the principal churches in Leipzig: the cycle Das Saiten-Spiel des Hertzens in the church year 1735–6, and the cycle Das Namenbuch Christi at some other time during the 1730s.11 Sto¨lzel’s cantatas were technically more straightforward than Bach’s and could therefore be entrusted to the

BWV Anh. I 11–12, BWV 213, 214, 215, 205a, 206, 207a, Anh. I 13, and 208a. See Andreas Glo¨ckner, ‘Eine Michaeliskantate als Parodievorlage fu¨r den sechsten Teil des Bachschen Weihnachts-Oratoriums?’, BJ 86 (2000), pp. 317–26. 10 At least six other Telemann cantatas (besides those mentioned in Part I Ch. 1) might have been performed by Bach or a deputy: see Kirsten Beißwenger, J. S. Bachs Notenbibliothek (Kassel, 1992), VBN II/ T/1–6. 11 For the ﬁrst cycle, see Marc-Roderich Pfau, ‘Ein unbekanntes Leipziger Kantatentextheft aus dem Jahr 1735—Neues zum Thema Bach und Sto¨lzel’ and Peter Wollny, ‘ “Bekennen will ich seinen Namen”— Authentizita¨t, Bestimmung und Kontext der Arie BWV 200: Anmerkungen zu J. S. Bachs Rezeption von Werken G. H. Sto¨lzels’, both in BJ 94 (2008), pp. 99–122 and 123–58 respectively. For the second cycle, see 9

224 p a r t i i direction of a deputy during Bach’s absence. This would leave him free for other undertakings, such as journeys to other cities or guest performances. Bach’s high regard for Sto¨lzel’s music is evinced by the inclusion of the latter’s Partita in G minor in the Clavierbu¨chlein for Wilhelm Friedemann Bach of 1720, and of the aria ‘Bist du bei mir’ (formerly attributed to Bach), from Sto¨lzel’s opera Diomedes, in the Clavierbu¨chlein for Anna Magdalena Bach of 1725. Not only were complete cantata cycles of Sto¨lzel’s performed in the two main Leipzig churches, but his Passion-oratorio Ein La¨mmlein geht und tra¨gt die Schuld—a setting of the libretto Der Glaübigen Seele Geistliche Betrachtungen ihres leidenden Jesu (Gotha, 1720)—was performed at Good Friday Vespers at the Thomaskirche, Leipzig in 1734.12 This is striking and incontrovertible evidence that by this date the Leipzig authorities accepted within the Good Friday service not only works of the ecclesiastical oratorio-Passion type, which includes a full rendition of the Gospel narrative (the type to which Bach’s St John, St Matthew, and St Mark Passions belong), but also works of the theatrical Passionoratorio type, which omit the Gospel account in favour of a poetic meditation upon it. The earliest texts of this kind include C. F. Hunold’s Der Blutige und Sterbende Jesus (Hamburg, 1704) and B. H. Brockes’s Der fu¨r die Su¨nde der Welt Gemarterte und Sterbende Jesus (Hamburg, 1712), which was set to music by Keiser, Handel, Telemann, and Mattheson. Bach’s willingness for Sto¨lzel’s Passion-oratorio to be performed at the Good Friday Vespers raises the question whether either or both of the two lost Bach Passions might have belonged to this genre. One can imagine, for example, a setting by Bach of Picander’s Erbauliche Gedancken auf den Gru¨nen Donnerstag und Charfreytag u¨ber den Leidenden Jesum of 1725. In any event, during the 1730s Bach not only composed a new Passion of the traditional kind, with the biblical narrative sung by the tenor Evangelist in recitative—the St Mark Passion of 1731—but he also revived his two earlier Leipzig Passions in revised versions. Version III of the St John Passion (already performed in earlier versions in 1724 and 1725) was given on Good Friday (11 April) 1732—a version characterized chiefly by the elimination of references to St Matthew’s Gospel, presumably to avoid an overlap with the St Matthew Passion, which had been performed a few years previously (probably in 1727 and 1729). The definitive version of the St Matthew Passion—that of the autograph score (Berlin, P 25)—was performed for the first time on Good Friday (30 March) 1736. Just as the chief vocal works of the period originated through parody, so the most prominent instrumental works arose through the arrangement of earlier compositions, many of which are lost. Of Bach’s fourteen harpsichord concertos, BWV 1052– 65, three are based on his surviving violin concertos (BWV 1054, 1058, and 1062), one

Andreas Glo¨ckner, ‘Ein weiterer Kantatenjahrgang Gottfried Heinrich Sto¨lzels in Bachs Auffu¨hrungsrepertoire?’, BJ 95 (2009), pp. 95–115. 12 See Tatiana Shabalina, ‘“Texte zur Music” in Sankt Petersburg: Neue Quellen zur Leipziger Musikgeschichte sowie zur Kompositions- und Auffu¨hrungsta¨tigkeit J. S. Bachs’, BJ 94 (2008), pp. 33–98 (esp. 77–84).

introd uction

225

on the fourth Brandenburg Concerto (BWV 1057), and one on Vivaldi’s Concerto in B minor for four violins from L’estro armonico, Op. 3 No. 10 (BWV 1065). Only one concerto was originally written for harpsichord (BWV 1061); all the others were probably adapted from lost concertos that Bach had written at an earlier period for melody instrument/s, most often violin or oboe. By specializing in the harpsichord concerto during the 1730s Bach made himself one of the pioneers of a new genre that would eventually sweep Europe, partly through the advocacy of his own sons and pupils. Characteristically, Bach in all probability had a practical purpose in mind: he wanted to create a vehicle within which to exhibit not only his own keyboard skills but those of his eldest sons, Wilhelm Friedemann and Carl Philipp Emanuel, and of outstanding pupils such as Johann Ludwig Krebs. This perhaps explains why the concertos for three or four harpsichords were apparently composed first, not long after Bach became director of the Collegium musicum (BWV 1063–5: c. 1730), followed by the concertos for two harpsichords (BWV 1060–2: c. 1733–6). For the contribution of the talented Bach sons would diminish and eventually cease when they left home and took up posts of their own in other cities. In 1739, after a two-year break, Bach became director of the Collegium musicum once more, and the solo harpsichord concertos, BWV 1052–9, whose autograph dates from about 1738, might have been intended for Bach himself to play in the context of a new series of concerts. Another vital new aspect of Bach’s musical activities in the 1730s is publication. He had made a tentative start in the late 1720s, testing the water by publishing a single partita at a time, roughly at the rate of one per annum. By the early 1730s, however, he felt confident enough to publish all six partitas in a collected edition, Clavieru¨bung [I] (Leipzig, 1731). Two further volumes in the Clavieru¨bung series then followed at fouryearly intervals: Part II (1735), which juxtaposed the popular national styles of the day in the Italian Concerto and French Overture; and Part III (1739), a liturgical collection of organ music containing a series of catechism chorales and a Missa (Kyrie and Gloria)— an organ equivalent of the four vocal Missae (BWV 233–6) that Bach composed around the same time. Publication inevitably gave him a much larger and more widespread public for his harpsichord and organ music than he had ever enjoyed before. It is clear from a comment made by Bach in his ‘Short but most necessary Draft’ that by 1730 he had become conscious of the gulf between old and new styles and of the necessity to adapt to new ideas in composition and performance alike. He reflects that, ‘nun aber der itzige status musices gantz anders weder ehedem beschaffen, die Kunst u¨m sehr viel gestiegen, der gusto sich verwunderens-wu¨rdig gea¨ndert, dahero auch die ehemalige Arth von Music unseren Ohren nicht mehr klingen will’ (‘Now . . . the state of music is quite different from what it was, since our artistry has increased very much, and the taste has changed astonishingly, and accordingly the former style of music no longer seems to please our ears’).13 Despite this acknowledgement, Bach’s

See n. 2.

226 p a r t i i former pupil Johann Adolph Scheibe, writing from the perspective of the Enlightenment, with its emphasis on the imitation of nature and the guidance of reason, famously criticized the great composer in 1737 for writing in what would then have been regarded as an old-fashioned style:14 He [takes] away the natural element in his pieces by giving them a turgid and confused style, and he [darkens] their beauty by an excess of art . . . Everything that one thinks of as belonging to the method of playing, he expresses completely in notes; and this not only takes away from his pieces the beauty of harmony but completely covers the melody throughout. All the voices must work with each other and be of equal difficulty, and none of them can be recognized as the principal voice . . . Turgidity has led [him] from the natural to the artificial . . . one admires the onerous labour and uncommon effort, which, however, are vainly employed, since they conflict with Nature.

Essentially, Scheibe accuses Bach’s music of being too artiﬁcial and complicated, of lacking natural grace and simplicity. These last-named attributes, alongside the primacy of singing melody, were sought by leading progressive composers such as Telemann, Hasse, and Graun (all greatly admired by Scheibe), who subscribed to the galant, pre-Classical style of the day. It was this musical style, together with its attendant concept of good taste, that was prevalent in progressive circles in the Hamburg, Berlin, and Dresden of the 1730s.15 The defence of Bach, Unpartheyische Anmerkungen (‘Impartial Observations’),16 published in the following year by his friend Johann Abraham Birnbaum, who taught rhetoric at the University of Leipzig, and perhaps compiled with the composer’s assistance, is effective but inordinately long and out of proportion to the original offence. In the exemplar of Scheibe’s Critische Musikus recently discovered at Jena, the composers criticized, left anonymous by Scheibe, are identiﬁed in the hand of Bach’s distant cousin, the Weimar town organist Johann Gottfried Walther. The inserted names include G. H. Sto¨lzel, C. G. Gerlach (music director of the Leipzig Neukirche), Johann Schneider (organist of the Leipzig Nikolaikirche), J. G. Go¨rner (Leipzig University music director), J. S. Bach, and C. F. Hurlebusch. It is clear from these annotations that Bach was neither the only prominent composer taken to task nor the most sharply criticized.17 That Bach was felt to be capable of composing according to the spirit of the times is clear from Lorenz Mizler’s reply to Scheibe,18 in which he says:

BD II, No. 400; NBR, No. 343. See George J. Buelow, ‘In Defence of J. A. Scheibe against J. S. Bach’, Proceedings of the Royal Musical Association, 101 (1974–5), pp. 85–100. 16 BD II, No. 409; NBR, No. 344. 17 See Michael Maul, ‘Johann Adolph Scheibes Bach-Kritik. Hintergru¨nde und Schaupla¨tze einer musikalischen Kontroverse’, BJ 96 (2010), pp. 153–98. 18 BD II, No. 336; NBR, No. 346. 15

introd uction

227

Anyone who heard the music that was performed by the students at the Easter Fair in Leipzig last year [1738], in the Most High presence of His Royal Majesty of Poland, which was composed by Capellmeister Bach, must admit that it was written entirely in accordance with the latest taste, and was approved by everyone. So well does the Capellmeister know how to suit himself to his listeners.

It has been suggested19 that here Bach might have been attempting to imitate the popular style of Hasse, Telemann, Fasch, and Graun. The work to which Mizler refers, however (BWV Anh. I 13), has not survived, though there is plentiful evidence of Bach’s embracing ‘modern’ styles in his works of the late 1720s and 1730s. Prominent examples include compositions written in the progressive sub-genres of the concert en ouverture (BWV 1067) or Sonate auf Concertenart (BWV 525–30, 1029, 1030, and 1032), as well as certain arias from the great Missa of 1733, particularly ‘Christe eleison’ and ‘Laudamus te’—the latter perhaps inspired by the singing of Faustina Bordoni. It is entirely characteristic that these ‘modern’ operatic arias are each followed by an alla breve fugue in pseudo-Renaissance style, Kyrie II and ‘Gratias agimus tibi’ respectively. One forms the impression that the whole known history of music at the time is within Bach’s reach, with the result that any style, whether ancient or modern, could be drawn upon whenever he considered it appropriate.

By Hans-Joachim Schulze, ‘Bach’s Secular Cantatas: A New Look at the Sources’, Bach, 21/1 (1990), pp. 26–41.

II.2 Clavieru¨bung II–III and other harpsichord, organ, and lute works

Clavieru¨bung II–III Title

Principal source

Place and date of edition

¨ bung Zweyter Theil der Clavier U ¨ bung Dritter Theil der Clavier U

Original edition Original edition

Nuremberg, 1735 Nuremberg, 1739

Four years after he had published the First Part of the Clavieru¨bung in a collected edition (1731), Bach followed it up with the Second Part, the ‘Zweyter Theil der Clavier ¨ bung’ (1735). This publication shares certain preoccupations with its predecessor— U above all, the juxtaposition of the contemporary French and Italian styles. In Part I the keyboard dances received French or Italian titles to clarify their stylistic orientation; Part II contains two large-scale compositions, one in each of the two national styles. At root they do not belong to keyboard genres but rather to the principal instrumental ensemble genres of the two nations, adapted for keyboard: ‘einem Concerto nach Italiaenischen Gusto’ (the Italian Concerto, BWV 971) and ‘einer Ouverture nach Franzo¨sischer Art’ (the French Overture, BWV 831). Moreover, for the first time Bach now specifies precisely the type of keyboard instrument required, namely ‘ein Clavicymbel mit zweyen Manualen’ (‘a harpsichord with two manuals’). He uses piano and forte indications to refer to the upper and lower manuals of a large harpsichord, enabling him to imitate tutti/solo contrasts and to differentiate clearly between solo and accompanying parts. Since the earliest stages in Bach’s career he had written works in which the whole real or imaginary texture of an instrumental ensemble was taken into the two hands of a single keyboard player. In doing so he was not acting alone but rather responding to a trend that manifested itself in German music around 1700. From about that time Lullian ouverture-suites were not merely transcribed for harpsichord but newly composed for the instrument, notably J. C. F. Fischer’s Les Pie`ces de clavessin, Op. 2, of 1696. In the first decade of the new century, the young Bach himself contributed to this trend in his Ouverture-suites in F major and G minor, BWV 820 and 822. Later, in Clavieru¨bung I, Bach prefaced Partita No. 4 in D, BWV 828 (1728) with an Ouverture,

cl avieru¨ b ung i i

229

and that might have given him the idea of including an entire ouverture-suite in his next published collection. In addition, it was not uncommon at that time to play whole Italian ensemble concertos on solo organ or harpsichord. This was done, for example, by the blind Dutch organist J. J. de Graaf, whom Bach’s patron Prince Johann Ernst must have heard in Amsterdam and of whom he no doubt brought back reports to Bach in Weimar. Around that time, c. 1713–14, Bach transcribed at least twenty-one Italian or at least Italianate concertos by Vivaldi, A. and B. Marcello, Torelli, Telemann, and others for solo harpsichord or organ (BWV 972–87 and 592–6). Two large-scale concerto-like works of his own composition also date from the Weimar years: the Fantasia and Fugue in A minor, BWV 944, and the Prelude and Fugue in the same key, BWV 894. Not long before the publication of Clavieru¨bung II, a clear parallel to the Italian Concerto was conceived, namely the Concerto in C, BWV 1061, which at first (c. 1732/3) was intended for two harpsichords without accompaniment; only at some unknown later date were ripieno strings and continuo added. The Italian Concerto, like the later Goldberg Variations, counts as one of the most perfect of all Bach’s large-scale keyboard works. Even J. A. Scheibe, well known as one of Bach’s most vociferous critics, described it as ‘a perfect model of a well-designed solo concerto’.1 We do not know exactly in what form it originated,2 but some sort of link with the Partitas of Clavieru¨bung I is not beyond the bounds of possibility. According to a contemporary press announcement,3 that collection was originally to have included seven partitas rather than six. This has been dismissed as an error, but it seems too coincidental that Bach’s models, Clavieru¨bung I and II by Johann Kuhnau (1689 and 1692), each contained seven partitas. Moreover, a seventh partita would have completed Bach’s key scheme—a partita on each degree of the diatonic scale, as in Kuhnau—simultaneously rising and falling from tonic to dominant (Ex. 1). The seventh partita would thus have been in the key of F major. Since Bach’s next published work, the Italian Concerto, is also in F, the question arises whether he originally intended the first movement of that work to form the prelude to the seventh partita. The decisions to abandon the seventh partita and to expand the existing movement into a complete three-movement concerto might have been closely interrelated. In accordance with its relatively late date, the Italian Concerto belongs among Bach’s most ‘modern’ contributions to the genre. This is manifest not only in the somewhat modish thematic writing, as in the cut-off, double-appoggiatura cadences in the opening theme of the first movement, but also in the exceptionally clear differentiation between ritornellos and episodes in the same movement. The episodes are distinguished not only thematically but in texture—often treble-dominated with

BD II, No. 463; NBR, No. 331. An early version in the hand of J. C. Oley survives; see Walter Emery and Christoph Wolff, Krit. Bericht, NBA V/2 (1981), pp. 14 and 39–47. 3 In the Leipziger Post-Zeitungen, 1 May 1730; BD II, No. 276. 2

230 c l a v i e r u¨ bung i i –i ii an d o t h e r w o r k s Ex. 1

Tonics of the six and planned seventh keyboard Partitas chordal accompaniment in galant style—and in dynamics (mainly right-hand forte and left-hand piano, whereas the ritornellos are forte throughout). Key change periodically takes place by repeat of the ritornello headmotive (bb. 1, 53, 139), a feature of the early concerto that we have often had occasion to notice in Bach.4 The most dramatic part of the movement occurs in the later stages where the tonic is being reestablished. A build-up over a dominant pedal (b. 124) leads one to expect the tonic and its associated ritornello theme, but instead there is a diversion into an episode (b. 129). The expected tonic ritornello then does occur briefly (b. 139), but it is soon diverted into a new episode (b. 147). Only after that does the full tonic return occur (b. 163)—an exact reprise of the original ritornello (bb. 1–30). The Andante is a bipartite movement (AB) with rhyming cadences, followed by a coda. It is not articulated by ritornellos; instead, a florid ‘violin’ solo of great beauty and expressive power is underpinned throughout by a variable ground bass. This enables the ‘solo’ part to roam free, unfettered by motivic or thematic constraints. The Corellian graces of the solo are by their very nature pseudo-improvisatory and hence largely exclude sequence or repetition. The concept of the florid right-hand solo with two left-hand supporting parts was developed in the Partitas of Clavieru¨bung I, particularly in No. 4 in D, and became a favoured texture in Bach’s slow movements. In this Andante the same manual distinction applies as in the episodes of the first movement: right-hand forte and left-hand piano. In the Presto-finale, the episodes are led as much by the left hand as the right, so that left-hand forte and right-hand piano is no less frequent than the reverse. Indeed, the parts are often interchanged between the two hands. Like the opening Allegro, the movement falls into an overall ABA1 reprise structure, in which B full-closes in the mediant, as in a da capo aria. This middle paragraph culminates in a large episode/ ritornello complex (bb. 104–50) such as we encounter frequently in Bach’s concertos. It includes three brief ritornellos (in keys IV, vi, and iii), surrounded by episodic material. Then a transition takes place—in a manner characteristic of Bach in the 1730s—to a varied reprise of paragraph A. Only the headmotive of the ritornello is heard at this stage, followed by a reverse-order reprise of the first two episodic formulations, which then leads to an exact and full final ritornello return. The French Overture may have originated not long before its publication—an early version in the key of C minor, BWV 831a, was copied out by Anna Magdalena around 1734. In the first movement, Ouverture, Bach gives a fine imitation of the highly 4

See Vol. I of this study, pp. 144–5.

c l a v i e r u¨ bung i i

231

intricate, profusely ornamented French style. Yet several features may be identified as Germanic or even Bachian, such as the motivic use of demisemiquaver-upbeat or dotted-rhythm figures. The central fugue, like that of the Ouverture from Partita No. 4, is cast in a species of ritornello form in which the ritornellos are constructed as fugal expositions. The episodes, with their soloistic figuration, provide a very clear contrast. The expectation that episodes will vary but ritornellos stay much the same is deliberately thwarted: the first episode (b. 47) recurs exactly but for key (b. 77), but it is followed by a fugal ritornello based on the countersubject rather than the subject itself. The Courante necessarily belongs to the French type, like those of Partitas Nos. 2 and 4; but unlike them, it is made up largely of free figure-work, with little in the way of theme or motive. It is thus closer to the native French style as imitated by Bach in the Courantes of the English Suites. The most individual feature is the cross-rhythm ostinato which, together with the virtually exact rhyming-close, helps to unite the two strains. The Gavotte, Passepied, and Bourreé are each made up of two dances, to be played alternativement (ABA). It is interesting to note that the key relation between the main dance and its ‘trio’ differs each time—Gavotte: trio in relative major; Passepied: in tonic major; Bourreé: in tonic. Only the last corresponds with Bach’s later practice (French Suites, Partitas); the use of a different key or mode for the trio belongs to an earlier conception (English and Cello Suites) that Bach revived here, presumably for the sake of maximum variety. The Sarabande lacks the hymn-like melody of those of the English Suites, but the texture is similar in its very rich fourpart harmony. It stands out, however, for its motivic quality: the opening figure, with its traditional second-beat stress, is constantly imitated between treble and bass. The Gigue, with its dotted rhythms in 6/8, belongs to the quintessentially French canarie type. The finale, entitled ‘Echo’, exhibits the dance-like 2/4 metre that Bach used much in the late 1720s to 1730s and which might have been considered fashionable at the time. The four-bar theme acts like a mini-ritornello (bb. 1, 13, 33, 45; keys i, v, III–iv, VI), but the passages that follow are continuations rather than episodes—the first marked by varied echoes (bb. 5 and 37; as distinct from the exact echoes of the cadential phrase at bb. 29 and 69), the second by change of manual (bb. 17 and 49). The second strain follows very largely the same course as the first. The Third Part of the Clavieru¨bung, published in 1739, bears no obvious relation to Parts I and II. Yet it was clearly designed as the culmination of a three-part series. Parts I–III appeared at four-year intervals—1731, 1735, 1739—and are progressive in their instrumental requirements: Part I for single manual, Part II for double manual, and Part III for double manual plus pedal.5 For Part I Bach does not specify the instrument required, but Part II, as we have seen, is designed ‘vor ein Clavicymbel mit zweyen Manualen’ (‘for a harpsichord with two manuals’), and Part III ‘vor die Orgel’. In other respects the three title pages are similar in their wording. In particular, in 5 See Albert Clement, Der dritte Teil der Clavieru¨bung von Johann Sebastian Bach: Musik, Text, Theologie (Middelburg, 1999), pp. 5–6.

232 c l a v i e r u¨ bung i i– i i i an d o t h e r w o rk s all three cases Bach states that the music was composed ‘denen Liebhabern zur Gemu¨ths-Ergo¨tzung’ (‘for music lovers, to delight their spirits’). For Part III, however, no doubt in recognition of the more restricted audience for organ pieces of this kind, he adds: ‘und besonders denen Kennern von dergleichen Arbeit’ (‘and especially for connoisseurs of such work’). It was formerly assumed that the contents of Clavieru¨bung III, all of which were evidently written with the printed edition in mind, were composed not long before publication in 1739. More recently, however, it has been suggested that the work might have undergone a lengthy and far-reaching pre-publication history going back to 1735, the publication date of Part II.6 This has since been conﬁrmed by studies of the original edition.7 The evidence of the engraving process, insofar as it can be reconstructed, points to an original conception of the work on a relatively modest scale and a huge expansion at a later stage. All this must have taken place between 1735 and the publication of the work at Michaelmas 1739. The original version contained only the Missa (Kyrie and Gloria), BWV 669–77, and the six pedaliter catechism chorales, BWV 678, 680, 682, 684, 686, and 688. At a later stage Bach added the six manualiter catechism chorales (BWV 679, 681, 683, 685, 687, and 689), the four Duetti, BWV 802–5, and the Praeludium and Fuga in E♭, BWV 552, whose two movements were physically separated in the printed volume in order to act as an overall frame for the whole work. The contents of the revised, published version are as follows: Prelude Kyrie

Gloria

Law Credo Paternoster Baptism Confession Communion

ped. ped. ped. ped. man. man. man. man. ped. man. ped. man. ped. man. ped. man. ped. man. ped. man. ped. man.

Praeludium Kyrie Christe Kyrie Kyrie Christe Kyrie Allein Gott in der Ho¨h do. do. Dies sind die heilgen zehn Gebot do. Wir glauben all an einen Gott do. Vater unser im Himmelreich do. Christ, unser Herr do. Aus tiefer Not do. Jesus Christus do.

BWV 552/1 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689

PRELUDE

MISSA

CATECHISM I

CATECHISM II

See Christoph Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), p. 206. ¨ bung III: The Making of a Print; with a Companion Study of the By Gregory Butler, Bach’s Clavier-U Canonic Variations on ‘Vom Himmel hoch’, BWV 769 (Durham and London, 1990). 7

clavieru¨ bung i i Duetti

Postlude

man. man. man. man. ped.

Duetto I Duetto II Duetto III Duetto IV Fuga

802 803 804 805 552/2

233

DUETTI

POSTLUDE

Of the original contents, the Missa may be viewed as an organ counterpart to the four vocal Missae that Bach composed around the same time (c. 1738); and the six pedaliter catechism chorales represent a summary of all the main articles of the Lutheran Faith, possibly assembled in preparation for the three Leipzig Reformation Festivals that took place in 1739.8 The subsequent expansion of the work might have been undertaken partly in the interests of greater accessibility, for the original contents are often complex and difficult, and they must have seemed austere and forbidding to all but the most intrepid of professional organists. Despite Mizler’s view of the work as ‘a powerful refutation of those who have made bold to criticize the compositions of the Honourable Court Composer’ (1740),9 the early version seems rather a confirmation of Scheibe’s opinion of his ex-teacher’s works as turgid, difficult, and obscure.10 The compositions that were added later, on the other hand—the manualiter chorales, the Duetti, and the Praeludium and Fuga in E♭—are on the whole easier to play and more ‘modern’ in style, introducing ‘natural’ galant elements that contrast markedly with the older ‘artificial’ style of the more complex polyphonic pieces. The additional pieces not only modernized the collection and lightened its tone, but they also gave it greater consistency. The Kyrie and Gloria of the Missa were already represented by both pedaliter and manualiter versions. It made sense, then, to present each of the six catechism chorales in manual and pedal versions too. Similarly, if a non-choralebased pedaliter piece was to be included as introduction and conclusion—the Praeludium and Fuga in E♭—then it was reasonable that it too should be given a manualiter counterpart, hence the set of four Duetti. In its final form the work juxtaposes ‘ancient and modern’ styles in accordance with Bach’s simultaneous backward- and forward-looking preoccupations during the 1730s. For the older styles he was presumably guided by Palestrina, Frescobaldi, Scheidt, de Grigny, and Lotti, among others. Two books, in particular, might have helped to determine the structure and contents of Clavieru¨bung III: Frescobaldi’s Fiori musicali (Venice, 1635), which Bach had copied out in 1714; and de Grigny’s Premier livre d’orgue (Rheims, 1699), copied by him around 1709/12.11 Among his

See Peter Williams, The Organ Music of J. S. Bach (rev. 2nd edn, Cambridge, 2003), pp. 388 and 391; all references are to this edn unless otherwise stated. 9 BD II, No. 482; NBR, No. 333. 10 See Williams, The Organ Music of J. S. Bach, p. 394. 11 J. A. Birnbaum mentions Palestrina, Lotti, and de Grigny in his defence of Bach against the criticisms of J. A. Scheibe; see BD II, No. 409, and NBR, No. 344. Regarding Bach’s copies of the Frescobaldi and de Grigny prints, see Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1991), VBN I/F/2 and I/G/3.

234 c l a v i e r u¨ bung i i –i i i an d o t h e r w o rk s contemporaries, writing in the 1730s, there is clear evidence of Bach’s engagement with G. F. Kauffmann’s Harmonische Seelenlust (1733–6), C. F. Hurlebusch’s Compositioni musicali (1734–5), and J. G. Walther’s Allein Gott in der Ho¨h sei Ehr (1736).12 The Missa opens with one of the most awe-inspiring of Bach’s mature organ compositions—the pedaliter Kyrie, a composite piece made up of the threefold prayer Kyrie–Christe–Kyrie, each with a German trope relating to the Trinity. The four-part Kyrie I and Christe are to be played ‘a` 2 clav. et ped.’, which allows the plain long-note cantus firmus to be accorded its own independent manual. It is presented in the soprano part (Kyrie I), then in the tenor (Christe), and finally in the bass as the pedal foundation of a climactic five-part organo pleno setting (Kyrie II). All three sections of the Kyrie are composed in Bach’s strictest stile antico, a style he developed from the study of Palestrina and his eighteenth-century imitators, such as Fux, Zelenka, and Caldara,13 though the cantus firmus pieces of Frescobaldi and Scheidt probably contributed here too. The subject of the accompanying parts is a diminution of the first two chorale lines, which is employed strictly, direct and inverted, as the basis of stretto fugue. The underlying melody is not a ‘classic’ Lutheran chorale but the old modal Kyrie chant, hence the ‘modal’ impression and unfixed tonality of Bach’s setting, which creates a certain mystical aura, not least in the unworldly last line of Kyrie II (bb. 54b ff.). The four-part alio modo settings of the Kyrie, for manuals only, are based on the first line of the chant, though Kyries I and II extract only its first three notes. In Kyrie I this theme, combined with a regular countersubject in quavers, is woven into an imitative texture. In Christe and Kyrie II, however, the theme forms the basis of stretto fugue, as in the pedaliter settings, though now on a miniature scale. It is noteworthy that these manualiter settings are modal-influenced and hence indeterminate in key, but since they are not cantus firmus settings their modality is no longer guided by the chant, as it was in the pedaliter Kyrie. Bach presents three settings of the German Gloria Allein Gott in der Ho¨h sei Ehr, preserving the Trinitarian symbolism of the Kyrie, although the threefold structure is here no longer a liturgical requirement, as it was in the Kyrie. Indeed, whereas each Kyrie–Christe–Kyrie sequence forms a composite whole, the three Gloria settings are entirely independent compositions. They were clearly conceived as a set of three, however: their tonics form a rising sequence of major keys: F, G, A; and they are designed for manuals only (F), with pedals (G), and for manuals only (A), so that the largest piece is centrally placed. The first manual setting (in F) places its ‘Canto fermo in alto’, surrounded by imitative and decorative treble and bass parts. These are 12 These publications are discussed in connection with Clavieru¨bung III by John Butt, ‘Bach and G. F. Kauffmann: Reflections on Bach’s Later Style’, in D. R. Melamed (ed.), Bach Studies 2 (Cambridge, ¨ bung III, pp. 4–16; and by Williams, The Organ Music of J. S. Bach, 1995), pp. 47–61; by Butler, Bach’s Clavier-U p. 388. 13 See Christoph Wolff, Der Stile antico in der Musik Johann Sebastian Bachs: Studien zu Bachs Spa¨twerk (Wiesbaden, 1968), esp. 17 ff.

clavieru¨ bung i i

235

derived from the opening two-part fugal exposition, which is in turn based on a decorative paraphrase of the first chorale line. The second manual setting (in A) is a ‘Fughetta’ with two expositions on different subjects, clearly set off by tonic cadences (bb. 7 and 15–16). The first subject is a derivative of the first chorale line in disjunct, staccato quavers; the second is based on the second line, but now conjunct and legato. At the conclusion (b. 16) the two subjects are combined just once in the upper parts. The big central Gloria setting (in G) is written ‘a` 2 clav. et ped.’, but the manuals and pedal function differently in Stollen and Abgesang. In the Stollen, the pedal has a purely accompanying role, so that the manuals carry both the cantus firmus and the fugal texture (based on a decorative paraphrase of the first chorale line). The Abgesang is quite different in structure. The pedal shares in the presentation of the cantus firmus, and the accompanying parts are no longer based on the complete fugue subject, which recurs only at the end, but rather on certain motives drawn from it. For sheer variety of chorale treatment within a single composition, this piece can hardly be matched except in the early stages of Bach’s composing career.14 As already noted, the earlier version of Clavieru¨bung III contained both the Kyrie– Gloria Missa and the six pedaliter catechism chorales. The title page refers only to these original contents: ‘bestehend in verschiedenen Vorspielen u¨ber die Catechismus und andere Gesaenge, vor die Orgel’ (‘consisting of various Preludes on the Catechism and other Hymns, for the Organ’). Quite why the Missa is described merely in terms of preludes ‘on other hymns’ is not at all clear. However, there are close links between the two sets of pieces, the Missa and the catechism chorales. In particular, the Trinitarian symbolism of the Kyrie and Gloria is carried further in the grouping of the pedaliter catechism chorales in sets of three. Each set has a central organo pleno setting, flanked on either side by a setting ‘a` 2 clav. et ped.’. The framing chorales of the first set, Dies sind die heilgen Zehn Gebot and Vater unser im Himmelreich, are both complex pieces in five-part texture with ‘Canto fermo in canone’—underlining the ABA symmetry of the group. The Mixolydian mode of the Ten Commandments chorale, reflected in the lack of key signature for a piece in G, causes a certain tonic–subdominant ambivalence. However, the pregnant motive of bb. 5–6, much used thereafter alongside its chromatic counterpoint, lends the music a certain ‘contemporary’ aspect, offsetting the antiquated effect of the modality. The French-style Paternoster chorale, despite its transposed Dorian signature (two sharps in E minor), sounds decidedly ‘modern’, with its constantly varied rhythms and ubiquitous Lombard figures. The opening two-part fugal exposition, based on a florid paraphrase of the first chorale line, recurs sufficiently often to be heard as a ritornello. The centrepiece of the first set, the Credo chorale Wir glaüben all an einen Gott, is not a cantus firmus setting but a four-part fugue based on a variant of the first chorale line.

See Vol. I, pp. 72–96.

236 c l a v i e r u¨ bung i i –i i i an d o t h e r w o rk s The great power of the piece is derived in no small measure from its inexorable ostinato, played throughout in the pedal as an elemental background to the fugal expositions. The framing chorales of the second set, the Baptism chorale Christ, unser Herr, zum Jordan kam and the Communion chorale Jesus Christus, unser Heiland, are both ‘2 Clav. et Ped.’ pieces with cantus firmus in the pedal. In both cases, however, the pedal cantus part functions as tenor, not bass, since the true bass is located in the lowest manual part. Both chorales have exceptionally bold ritornello themes with large intervals. However, whereas the Baptism chorale has three manual parts, the Communion chorale has only two, which allows Bach to adopt the light, figurative style that so often characterizes chorale trios from Pachelbel onwards. The centrepiece of the second set, the penitential chorale Aus tiefer Not, also forms the culmination of the catechism chorales altogether. Perhaps for this reason it reverts to the style of Kyrie II, the culmination of the Kyrie series: both are massive organo pleno pieces in stile antico with cantus firmus in the pedals. But whereas the texture of Kyrie II is five-part, that of Aus tiefer Not is six-part—four manual parts and double pedal (with cantus in the upper pedal part). Each cantus line is preceded by a full fugal exposition on the same line, plain and in standard note-values, until the last line, which is decorated by syncopation and passing-notes. For the later, definitive version of Clavieru¨bung III Bach added the manualiter versions of the six catechism chorales (each placed after its pedal counterpart) and the non-chorale-based pieces, namely the four Duetti and the Praeludium and Fuga in E♭. In the first set of three catechism chorales, the manualiter versions do not share the ABA pattern of their pedal equivalents. The Commandments and Credo chorales are both fughettas, the one four-part with an Italianate gigue rhythm, the other three-part and in French style with dotted rhythms and demisemiquaver upbeats. The Paternoster setting, on the other hand, is a cantus firmus chorale in four voices (with cantus in the treble), which in scale and structure resembles the typical Orgelbu¨chlein chorale, save only for the absence of pedal. Thus it possesses neither introduction and episodes nor conclusion. The three imitative accompanying parts, based on the opening semiquaver scale figure and its inversion, provide a strictly motivic background to the chorale throughout. In the second set of catechism chorales, the alio modo, manuals-only settings fall in with the ABA plan of their larger, pedaliter equivalents. Thus the first and third (Baptism and Communion) are fugues, whereas the middle chorale (Confession) is a cantus firmus setting. The Baptism chorale Christ, unser Herr is constructed as an inversion- or counter-fugue, based on the plain, unadorned first line of the chorale. In the penitential Aus tiefer Not, each line of the cantus firmus (in long notes in the soprano part) is prefaced by a stretto-inversion fugue based on the same line, plain but in diminished note-values. The last of the manualiter chorales, Jesus Christus, unser Heiland (Communion), is significantly dignified with the title ‘Fuga’ rather than fughetta. For this complex four-part stretto fugue is the most extended and profound of the manualiter chorales, forming a fitting culmination to the set.

clavieru¨ b ung i i

237

If the Praeludium and Fuga are regarded as separate pieces, in line with their placement at opposite ends of the collection, the chorale-free pieces number six (Praeludium, Fuga, and four Duetti), which corresponds with the number of Kyrie sections and of catechism chorales, with and without pedals. Since these groups include both large pedal settings and smaller manual counterparts (as does the Gloria), it seems logical to view the Duetti as the manual counterparts to the pedaliter Praeludium and Fuga. Indeed, it seems likely that this was the justification for their inclusion in the first place. The Duetti form a carefully organized set. Their tonics form an ascending series, with two major keys flanked by two minors: e–F–G–a. Furthermore, all four pieces are built on the same principles of fugue, canon, and double counterpoint. The strictness with which these techniques are employed often gives them a certain severity. Only in the G major Duetto, No. 3, do we encounter a welcome relaxation. The obvious precedent for the strict two-part counterpoint of these pieces lies in the Inventions (BWV 772–86, 1722–3) which, however, are imitative rather than fugal. Closer still is the two-part Fuga 10 in E minor from The Well-Tempered Clavier I, whose second half is essentially a double-counterpoint inversion of its first half. Such strict, consistent double counterpoint on a large scale informs the similarly chromatic Duetto I in the same key of E minor, where the second paragraph of the Bar form (AA1B) is a reprise of the first in double counterpoint and in new keys. It is not only the strict structuring of the piece that impresses but also its rich and florid figure-work. In Duetto II in F da capo form is amalgamated with fugue, as in some of Bach’s concerto movements. Indeed, the fugue subject opens with a tonic triad, a standard opening for a concerto (as in Bach’s E major Violin Concerto, for example), and the A-section of the ABA structure may be construed as a ritornello as well as a fugal exposition. Within the central B-section returns of the fugue subject act like mini-ritornellos. Recurring episodes are strictly canonic—in one case (b. 38) creating an oft-noted bitonal effect—as are the stretto entries of the subject in the middle paragraph. Duetto III in G, perhaps the lightest and most attractive of the set, is far less strict than Nos. I and II. Little use is made of canon or double counterpoint, the latter being restricted to the standard inversion of the subject–countersubject combination. The concluding Bsection of the AA1B Bar form, which it shares with Duetto I, contains no further full subject entries but instead repeatedly plays on its figures. Duetto IV in A minor returns to the concept of double counterpoint as the chief constituent of fugue, affecting not only the expositions but all the episodes too. The Duetti share the principle of fugue with their larger pedaliter counterparts. This applies not only to the concluding Fuga in E♭ but also to the opening Praeludium in the same key. For the second episode of this massive piece (b. 71) is in effect a threepart fugue on a Spielfuge subject for manuals only. A very substantial and extended development of this fugue takes place towards the end (b. 130), now with pedals. Copious other ingredients also contribute to this richly inventive prelude, which in this respect mirrors the contents of the collection as a whole, as Bach no doubt intended. Italian concerto and French overture, deliberately kept apart in

238 c l a v i e r u¨ bung i i –i i i an d o t h e r w o rk s Clavieru¨bung II, are here united. The massive opening ritornello in five-part texture (bb. 1–32) is couched in a celebratory ouverture style with dotted rhythms, which for Bach has the connotation of ‘beginning’. After a tonic full-close (bb. 31–2) there is a very clear new departure, as is customary in Bach’s concertante, ritornello-based structures. The simple chords of the first episodic theme (b. 33) and the florid, syncopated treble (with plain accompaniment) of the second (b. 41) illustrate Bach’s ‘modern’ style, as do the long multiple appoggiaturas of the ritornello theme. If anywhere in the collection Bach was trying to show that his style was not as outmoded as Scheibe made out, this prelude might have been the place. The great Fuga a 5 that concludes the collection represents the opposite stylistic extreme, at least in its first paragraph, namely the stile antico that Bach had already employed in the pedaliter Kyrie and in the six-part Aus tiefer Not. The Fuga a 5 is a triple fugue in three sections (more Trinitarian symbolism!), each based on its own subject (S I, II, III). S I, a neutral soggetto with a sequence of rising 4ths, as in the prelude in the same key from The Well-Tempered Clavier I, is worked in a five-part exposition and then in stretto. S II, two bars of sequential quavers, forms the subject of a manuals-only middle section. First treated in direct form, it is later inverted, after which it is combined with S I. Subject II then drops out, so that all three subjects are never combined. The pedals return for the final section, in which an exposition of the lively, rhythmic S III is followed by an exhaustive treatment of the combination S I + III.

Miscellaneous keyboard works Title

Earliest source/s

Scribe, date

Fantasia in A minor, BWV 904/1 Fuga in A minor, BWV 904/2 Fantasia and Fuga, BWV 906

Berlin, P 804 Berlin, P 288 Bethlehem, Pa. Dresden, SLB Leipzig UL Berlin, P 528

J. P. Kellner, post-1727 J. P. Kellner, 1727–38/40 Autograph, c. 1729 Autograph, c. 1738 Anon. 306, post-1750 Anon., post-1750

Fantaisie sur un rondeau, BWV 918 Six Pre´ludes (fu¨r Anfa¨nger auf dem Clavier), BWV 933–8

This is a motley group of pieces, all of which, however, exhibit the hallmarks of Bach’s mature style, even though none of them can be accurately dated. The largest of them, BWV 904, consists of two movements, fantasia and fugue, which were probably not brought together to form a pair until about 1800.15 Consequently, they are considered here as entirely separate compositions. The Fantasia is an alla breve piece, entitled in the chief source (see table) ‘Fantasia in A mol pro Cembalo’. It is thus speciﬁcally intended for harpsichord. The structure is a highly regular and symmetrical ritornello form. All four ritornellos (in keys i, v, iv, and i) are virtually identical 12-bar periods;

See Uwe Wolf, Krit. Bericht, NBA V/9.2 (2000), pp. 189–90.

miscellaneous keyboard works

239

and each of the three episodes is built consistently on a different group of figures, derived from the ritornello: dactyl, conjunct-quaver, and broken-chordal figures. The only mature Bach harpsichord piece that can be compared with this structure is the Praeambulum that introduces Partita No. 5 in G, BWV 829 (1730). The fugue is found in another Kellner source (see table), under the title ‘Fuga in A mol a 4 voc: manualiter’, which possibly points to the organ as the intended instrument. Like many of Bach’s Leipzig fugues, it is tripartite. The closest parallel among them is the Fuga in F, BWV 540 no. 2, since both this and the A minor piece are double fugues of the same formal type—A: double exposition of the first subject (S I); B: double exposition of S II; C: exposition of the combination S I + II. The headmotive of the powerful first subject is metrically diminished in b. 3, and this form, whether direct or inverted, furnishes much of the subsidiary material. S II is based on the chromatic 4th, variants of which are found in a number of other Leipzig fugues (BWV 548, 537, 906, and 997). Here it is mostly treated in stretto, and its cadential suspension is combined with a very strong and distinctive motive that (alongside its quaver counterpoint) fills the texture of the middle paragraph. The same motive continues to play a significant part in the concluding paragraph, alongside the combination of the two subjects in double counterpoint (bb. 61, 68, and 74). The cogency of the overall design, not only in its double-fugue structure but in its motivic detail, lends credibility to its placement in the mature Leipzig years. The Fantasia and Fuga in C minor, BWV 906, originally consisted only of a ‘Fantasia per il cembalo’ (‘Fantasia for harpsichord’, c. 1729). As in the organ works BWV 540, 546, and 562, the fugue was added at a considerably later date (c. 1738). The Fantasia, a fiery, virtuoso, Scarlattian ‘sonata’ in rounded binary form, exhibits close links with the keyboard Partitas of Clavieru¨bung I: the multiple appoggiaturas recall Partita No. 6 in E minor; and the hand-crossing figures, the Giga from Partita No. 1 in B♭. The movement might have been labelled ‘fantasia chromatica’, for the main theme outlines a chromatic descending 4th, and one of the chief motives (b. 5), a figure much used later, both direct and inverted (bb. 14 and 21), reverses its direction at speed. As so often in Bach’s mature binary form, the return of the tonic key at b. 34 coincides with the reprise of the main theme. The three-part fugue that follows in the later version was clearly designed to pick out some of the main elements of the fantasia—above all, its chromaticism, for the fugue subject is a decorated, rising chromatic 4th, and some of Bach’s strangest chromatic harmony occurs in the first two episodes (bb. 7–8 and 11–16; Ex. 2). The hand-crossing passage in the second paragraph (b. 38) also directly recalls that of the fantasia (b. 9). The fugue is possibly incomplete, but it is a perfectly reasonable supposition that paragraph B should be followed by a da capo of A, in which case the last chord would lead into bar 3. As a da capo fugue, the piece would then be comparable with the E minor organ fugue BWV 548 no. 2 and with the lute fugues in C minor and E♭, BWV 997 no. 2 and 998 no. 2. Three of these fugues (all but the E♭) have a chromatic subject, and all four have a semi-fugal or non-fugal middle section

240 c l a v i e r u¨ bung i i –i ii an d o t he r w o r k s in shorter note-values. Above all, in the fugue under discussion the ﬁrst subject entry of the da capo is accompanied in full texture at the end of paragraph B (b. 46), so that B and the return of A are dovetailed. This is exactly what happens in two of the other fugues mentioned (BWV 548 no. 2 and 998 no. 2). The middle section consists of a phrase in two-part texture in the tonic, followed by its free doublecounterpoint inversion in the dominant. This passage, seemingly unrelated to the fugue, nonetheless includes a reference to the diminished subject in pseudo-stretto (bb. 36 and 42; Ex. 3).

Ex. 2

Fuga in C minor, BWV 906 no. 2, bb. 7–8

Ex. 3

a) Fuga in C minor, bb. 36–7, treble only (bass omitted)

b) Fuga in C minor, subject

A third fantasia, the Fantaisie sur un rondeau, BWV 918, is little known but very remarkable and of considerable length. It cannot be dated on the basis of the sources. It is worth noting, however, that the association of rondeau form with a specific dance—Gavotte en rondeau (BWV 1006) or Passepied en rondeau (BWV 810)— belongs to the period around 1720. Only later did Bach detach the rondeau from specific dance rhythms, as in the Partita in C minor, BWV 826 (1727) or the Ouverture in B minor, BWV 1067 (early version: c. 1730). In addition, the two-part counterpoint in flowing semiquavers recalls another fantasia from this period: that which opens the Partita in A minor, BWV 827 (1725). And the strictness of the two-part contrapuntal and thematic treatment brings to mind the four Duetti from Clavieru¨bung III (1739). The rondeau theme (A) alternates with three episodes (B, C, and D) thus:

miscellaneous keyboard works Section: Material: Key: No. of bars:

A aba1 c–E♭–c 12

B cd c–g–c–f–c 16

A a c–E♭ 4

C efcgda1 e♭–g 48

A a c–E♭ 8

D fhi c–c 32

241 A aba1 c–E♭–c 12

The framing rondeau statements are identical—12 bars (4 + 8) of binary dance form— but the inner ones are restricted to the first strain, which on the second occasion is extended by imitation. The title suggests that this rondeau theme was Bach’s original invention—it exhibits too many Bachian fingerprints to have been borrowed from another composer. It is possible that Bach then wrote episodes of a shorter, simpler, and more periodic nature than those that have come down to us. The title ‘fantaisie’ would then derive from the last stage of expansion, which perhaps involved in-depth thematic development of the rondeau theme and of its constituent phrases and motives. In the course of this development much use is made of canon, double counterpoint, and Stimmtausch (exchange of parts). For example, three of the most significant episodic formulations, c, d, and f (bb. 13, 20, and 36), return subsequently in double-counterpoint inversion (bb. 44, 68, and 88). The title ‘fantaisie’ is perhaps to be understood in terms of a free contrapuntal searching into the developmental possibilities of the rondeau theme. The Six Pre´ludes (fu¨r Anfa¨nger auf dem Clavier), BWV 933–8, cannot be dated from the sources, all of which stem from the late eighteenth or early nineteenth centuries. Certain ornaments, such as the shake plus turn in No. 1, b. 4, belong to C. P. E. Bach’s style rather than his father’s; and the note e3 in No. 4, b. 46, lies outside Bach’s normal keyboard compass.16 Such observations might point to unauthorized intervention at some period. In any event, there is no certainty that the six pieces were collected together to form a set, still less conceived as such, by Bach himself. On the other hand, the diatonic key scheme, proceeding upwards by step from C, is fully in line with Bach’s habits, as the following cycles illustrate:17 1. Five Preludes 2. Three Preludes 3. Seven Preludes 4. Four Preludes 5. Fifteen Praeambula 6. Five Preludes and Fugues 7. Four Fughettas 8. Six Preludes

BWV 924, 926–7, 930, 928 BWV 924a, 925, 932 BWV 846–7, 851, 850, 855, 854, 856 BWV 939–42 BWV 772–86 BWV 870a, 899–902 BWV 872, 871, 876, 875 BWV 933–8

C–d–F–g/F C–D–e C–c–d–D–e–E–F C–d–e/a C–d–e–F–G–a–b C–d–e–F–G C–c–D–d C–c–d–D–E–e

The ﬁrst set of preludes listed was written out by Bach in 1720 (except BWV 927, which was added later by his son) in the Clavierbu¨chlein of that year as a series of 16 See Alfred Du¨rr, ‘Tastenumfang und Chronologie in Bachs Klavierwerken’, in A. Du¨rr, Im Mittelpunkt Bach (Kassel, 1988), pp. 220–31 (esp. 224). 17 Cf. also the German tradition of prelude collections from 1639 onwards as outlined by Thomas Synofzik, ‘ “Fili Ariadnaei”: Entwicklungslinien zum Wohltemperierten Klavier’, in S. Rampe (ed.), Bach: Das Wohltemperierte Klavier I (Munich and Salzburg, 2002), pp. 109–46.

242 c l a v i e r u¨ bung i i –i ii an d o t h e r w o r k s composition and keyboard exercises for the recipient of the book, his eldest son Wilhelm Friedemann. The second set was added to the same book later (c. 1725/6) by W. F. Bach in imitation of his father. The third set, entered in the Clavierbu¨chlein around 1721, comprises early versions of preludes that were later incorporated in The Well-Tempered Clavier I (1722). The four Preludes of the fourth set are clearly beginners’ pieces, but whether they had anything to do with Bach is another question—no composer is named in the only source. The fifteen Praeambula of the fifth set are early versions of the Inventions, entered in the Clavierbu¨chlein in 1722. The earliest source of the sixth set, the Five Preludes and Fugues, dates from c. 1726/7; around 1740 four of the ten pieces were adapted for inclusion in The Well-Tempered Clavier II (1739–42). The seventh set, the four Fughettas of c. 1738/9, were likewise revised and transposed for the same collection. It would hardly be surprising, then, if the Six Preludes originated as a set with the key scheme shown in the table under no. 8. Certain stylistic correspondences between the six pieces suggest that they might have been composed around the same time. When that was we cannot be sure, but certain indications in the text point to the 1730s. Binary dance form with repeats is rare in non-dance-style keyboard pieces before 1739: it occurs only in the Praeludium in G, BWV 930, composed in the style of a corrente, in the Praeludium in B minor from The Well-Tempered Clavier I, in the Invention in E, and in the presumably late Praeludium in G, BWV 902 no. 1. This fine prelude is fully comparable with those of The Well-Tempered Clavier II, of which no fewer than ten are cast in binary dance form with repeats. The same form, albeit on a smaller scale, is employed in all six of the Pre´ludes ‘fu¨r Anfa¨nger auf dem Clavier’, which perhaps points to a date closer to Part II than Part I of The Well-Tempered Clavier. Other features of the set recognizably belong to the galant style that Bach cultivated increasingly during the 1730s: clear periodic phrasing, double appoggiaturas at cadences, a specific cadential formula (No. 1, b. 2 and No. 4, bb. 2–3), and homophonic textures— the melodic interest is often largely restricted to the treble, to which the other part/s provide little more than accompaniment. In his biography of Bach,18 after discussing the touch exercises prescribed by Bach in keyboard lessons, Johann Nicolaus Forkel added: If [Bach] found that anyone, after some months of practice, began to lose patience, he was so obliging as to write little connected pieces, in which those exercises were combined together. Of this kind are the six little Preludes for Beginners . . . He wrote them down during the hours of teaching and, in doing so, attended only to the momentary want of the scholar. But he afterwards transformed them into beautiful, expressive little works of art.

18 Ueber Johann Sebastian Bachs Leben, Kunst und Kunstwerke (Leipzig, 1802); modern edn in BD VII, pp. 9–89; Eng. trans. in NBR, pp. 417–82.

two lute works

243

Forkel, whose account of the origin of the Six Pre´ludes was probably based on information supplied by one of the Bach sons, was in no doubt that, as with all Bach’s didactic works, their artistic value transcends their educational purpose.

Two lute works Title

Earliest source/s

Scribe, date

Suite in C minor, BWV 997

Berlin, P 650 Leipzig MB Tokyo, Ueno Gakuen

J. F. Agricola, 1738/41 J. C. Weyrauch, date? Autograph, c. 1735

Prelude pour la Luth. o Cembal. in E♭, BWV 998

Two of Bach’s ﬁnest solo instrumental works of the mid-to-late 1730s, BWV 997 and 998, were written for lute and/or Lautenwerk (lute-harpsichord). In Leipzig around 1740 Bach’s pupil Johann Friedrich Agricola saw and heard a lute-harpsichord designed by the composer and made by Zacharias Hildebrandt.19 However, the two compositions might equally have been occasioned by a visit to Bach’s home in 1739 by two celebrated lutenists from Dresden, Sylvius Leopold Weiss and Johann Kropffgans.20 The two works in question, the Suite in C minor, BWV 997, and the Prelude, Fugue, and Allegro in E♭, BWV 998, are in many ways remarkably alike and might have been composed in fairly close temporal proximity. Both open with a preludeand-fugue sequence and then continue with one or two movements in binary dance form. Accordingly, they might be regarded as free suites, such as Handel included in his Suites de pie`ces (London, 1720). Some years earlier, around 1730, Bach had adapted an entire dance suite—Cello Suite No. 5, BWV 1011—for the lute. But the Suite in F minor, BWV 823, which has recently been linked to the lute works of 1735/40,21 draws only the sarabande and gigue from the classical suite, as does the C minor Suite (BWV 997). The three movements of the F minor work—Pre´lude in chaconne form, Sarabande en rondeau (with a single couplet or episode), and Gigue in the style of a canarie—possibly represent the closest approach Bach ever made to pure French style. Returning to the Suites in E♭ and C minor, their preludes are both constructed in concerto form with short ritornellos, not unlike the harpsichord Fantasia in A minor, BWV 904 no. 1. The ritornello of the C minor Prelude (bb. 1–5) consists of a rhythmic ostinato ﬁgure over a moving bass, clearly differentiated from the recurring episodic material that follows. That of the E♭ Prelude modulates up a 5th in all but its last statement (bb. 1, 6, 14, 25, and 42); the episodes are based on the same material, so that 19

BD III, No. 744; NBR, No. 358e. BD II, No. 448; NBR, No. 209. 21 ¨ berlegungen zu Bachs Suite f-moll BWV 823’, in M. Geck (ed.), Bachs Musik fu¨r By Pieter Dirksen, ‘U Tasteninstrumente (Dortmund, 2003), pp. 119–31. Dirksen gives a complete facsimile reproduction of the source, J. P. Kellner’s copy in Berlin, P 804. 20

244 c l a v i e r u¨ bung i i –i ii an d o t h e r w o r k s the movement is all of a piece. The arpeggiated figuration recalls the preludes to the Cello Suites, one of which (No. 5) Bach had arranged for lute (or Lautenwerk) some years before (c. 1730). The following fugues are not only tripartite, like many Bach fugues from the late 1720s and 1730s, but are cast in da capo form (ABA), like the E minor organ fugue BWV 548 no. 2 and the C minor harpsichord fugue BWV 906 no. 2. There are further links between the four fugues: all but the E♭ lute fugue have chromatic subjects; the middle section is largely in diminution values, creating a very strong contrast with the framing sections; and, except in the C minor lute fugue, the last subject entry of the B-section also forms the first of the da capo, creating an admirably smooth join between the sections. Of the two lute compositions, only the C minor has a slow movement—a Sarabande notable for its striking headmotive, imitated at the lower octave, a motive familiar from the Sarabande of the A major English Suite (BWV 806, before 1717), the conclusio of the St Matthew Passion (1727), and elsewhere (see Part I Ch. 4, Ex. 9). The finales of the two lute compositions might almost have been designed with deliberate contrast in mind. The C minor is a French-style Gigue with the dotted rhythms of the canarie, followed by a double (variation). It is cast in rounded binary form: the tonic return coincides with the thematic reprise. The E♭ finale is also structured in binary dance form, but it inclines rather to the Italian style. Though somewhat giga-like, it makes no allusion to a specific dance rhythm, hence the noncommittal title ‘Allegro’.

Miscellaneous organ works Title

Earliest source/s

Scribe, date

Pastorella in F, BWV 590 Praeludium et Fuga in C, BWV 547 Valet will ich dir geben, BWV 736

Berlin, P 287/5 Berlin, P 274/1 Leipzig MB Poel.mus.ms.39

J. P. Kellner, post-1727 J. P. Kellner, post-1730 J. C. Kittel, c. 1780

The Pastorella in F, BWV 590, is unique among Bach’s organ works. Singular features include the four-movement form, otherwise found only in Bach’s sonatas; the restriction of pedals to the ﬁrst movement only; and the key scheme, F–a, C, c–f (half-close), F, which is clearly designed to create smooth joins between the movements. The earliest source, Johann Peter Kellner’s copy, dates from after 1727, and the use of treble rather than soprano clef in the upper stave points to a composition date after 1725.22 The piece has been linked to the South German/Italian tradition of Christmas pastorales—referring to the shepherds of the Nativity story—for organ in several

22 See Russell Stinson, The Bach Manuscripts of Johann Peter Kellner and His Circle (Durham and London, 1989), pp. 111 and 105–6.

miscellaneous organ works

245

movements by Frescobaldi, Zipoli, Georg Muffat, and others.23 Its style, however, particularly in the inner movements, represents Bach at his most galant and forwardlooking. The C major second movement is galant not only in its syncopated rhythms but in the use of binary dance form without a speciﬁc dance rhythm, as in the rondeau from BWV 918. In general, the style is not dissimilar to that of certain movements from the keyboard Partitas. The C minor slow movement no. 3 is perhaps one of the most purely Italianate pieces that Bach ever wrote. Its purely homophonic texture, with treble melody in triplet semiquavers accompanied by repeated-quaver chords, seems to belong rather to the middle than the early years of the eighteenth century. The true pastorale is the opening movement, with its compound metre and its long pedal points. The ﬁnale is a three-part fugue in gigue rhythm, with the subject inverted after the double bar and the direct subject returning at the end (as in the E minor Partita, BWV 830, of 1725). There is good reason to believe that the Praeludium et Fuga in C, BWV 547, might date from the later Leipzig years, perhaps around 1739. No autograph survives, but the sophisticated contrapuntal techniques of the fugue suggest that it originated at a time when Bach was preoccupied with such procedures. Two particularly close parallels are often mentioned. Firstly, much of the opening fugal exposition is virtually identical, note for note, with the Fughetta super Allein Gott in der Ho¨h sei Ehr, BWV 677, from Clavieru¨bung III (1739).24 The correspondence is close enough to suggest that Bach might have had one piece in mind when he composed the other (Ex. 4). Secondly, the later stages of the fugue have often been compared with those of Fuga 2 in C minor from The Well-Tempered Clavier II (c. 1739/40). In both cases the augmented subject, combined with direct and inverted forms of it in standard notevalues, brings the fugue to a powerful climax; and in both cases the bass part delivers a sequence of two or three subject entries at different pitches. The Praeludium may be heard in ritornello form, in which case the harmonically static opening theme constitutes the ritornello (bb. 1–7, 13–19, 48–59, and 80–8). The piece is entirely monothematic, however, which precludes much in the way of ritornello–episode contrast—the episodes are based throughout on the rich array of motives presented in the opening ritornello, where they are given in canonic imitation at the octave over a strong ostinato bass. This mode of treatment governs the episodes, where the opening theme is imitated in three voices and at various intervals (bb. 8, 20, 31, and later equivalents). The Fuga, like so many of Bach’s mature Leipzig fugues, is tripartite—A: direct subject only; B: direct plus inverted subject; C: direct and inverted plus augmented subject. Both A and B are for manuals only—the pedals enter only with the augmented subject in C. The concise,

23 See George Stauffer, ‘Bach’s Pastorale in F: A Closer Look at a Maligned Work’, Organ Yearbook, 14 (1983), pp. 44–60 (esp. 49–52); and Williams, The Organ Music of J. S. Bach, p. 197. 24 See the comparison in Williams, The Organ Music of J. S. Bach, p. 116, and in Stinson, The Bach Manuscripts of Johann Peter Kellner, pp. 115–17.

246 c l a v i e r u¨ bung i i –i ii an d o t h e r w o r k s Ex. 4

a) Fuga in C, BWV 547 no. 2, bb. 1–3

b) Allein Gott in der Ho¨h sei Ehr, BWV 677, bb. 1–3 (staccatos omitted)

epigrammatic subject is capable of modulating, which gives rise to extraordinary harmonic effects in the later stages of the fugue (bb. 56–8). When the direct and inverted forms of the subject begin to interact (b. 34), they are presented in a stretto sequence of six entries, with rectus answered by inversus or vice versa. The powerful direct, augmented pedal entries (bb. 49 and 51) are accompanied by a plethora of (mostly inverted) entries in the manual parts. When the augmented subject is inverted, however (bb. 59 and 62), the manual parts are free, presumably in order to set in relief what follows the tonic full-close (b. 66): a coda in which the subject is presented in stretto in all four manual parts, the first pair inverted and the second pair direct, over a tonic pedal. Bach wrote two large organ arrangements of the chorale Valet will ich dir geben, both with cantus firmus in the pedals. The earlier of the two, BWV 735a, dates from the period of comparative youth, the first decade of the eighteenth century; the later, BWV 736, on the other hand, is undoubtedly one of Bach’s most mature chorale-based compositions. The exuberant manual figuration, perhaps illustrating the thoughts of heaven in the text, 25 is based strictly on three six-note motives, identical in rhythm and differing only in melodic shape. The first of these (upbeat to b. 1) is identical with a motive that was employed by Georg Friedrich Kauffmann in his setting of Komm, heiliger Geist (Harmonische Seelenlust, 1733–6).26 Bach seems to have been impressed by this collection of organ chorales, for it had a demonstrable influence on his own subsequent writing in the field. This suggests a likely date for Valet will ich dir geben of the mid-1730s or later. Another comparison perhaps lends further support to this

Ulrich Meyer, quoted by Williams, The Organ Music of J. S. Bach, p. 482. See the comparative music examples in Williams, The Organ Music of J. S. Bach (1st edn), vol. ii, pp. 284–5. 26

miscellaneous organ works

247

Ex. 5

a) Valet will ich dir geben, BWV 736, b. 27

b) Praeludium in B♭, BWV 890 no. 1, bb. 5–6

Ex. 6

Valet will ich dir geben, original chorale (line 1) and Bach’s paraphrase dating: the Praeludium in B♭ from The Well-Tempered Clavier II is not only similarly gigue-like and in a cognate metre (12/16; BWV 736: 24/16), but it is also strictly motivic on the basis of a six-note ﬁgure (and its inversion), which is identical with the second motive from the organ chorale (upbeat to bb. 25 ff.; Ex. 5). The three dominant motives are built up to form pre-cantus paraphrases of the chorale lines (lines 1–3: bb. 1–6; line 5: bb. 25–6; line 7: b. 44; line 8: bb. 54–6; Ex. 6). There is no ritornello in the manual parts, but a passage of sequential imitation, based on the opening motive, recurs at intervals in variant forms (bb. 7–8 a, 33, 39b, 49, and 56); and, still more important, variants of one and the same cadential phrase form a clearly audible link between all six main cadences (bb. 11–12, 23–4, 35–6, 42–3, 51–2, and 58–9).

II.3 The harpsichord concertos and other instrumental works

Two ouvertures Title

Earliest source/s

Scribe, date

Ouverture in D, BWV 1068 Ouverture in B minor, BWV 1067

Berlin, St 153 Berlin, St 154

Part-autograph, 1731 Part-autograph, c. 1738/9

Bach’s Ouvertures in D (BWV 1068) and B minor (BWV 1067) both belong to a late, hybrid version of the genre, the concert en ouverture (concerto-overture), which was cultivated by Telemann and others in the late 1720s and 1730s, and described by J. A. Scheibe in Der critische Musicus (1740).1 Bach must have become acquainted with this sub-genre by 1730 at the latest, for in that year he and some associates copied out the performing parts of a number of ouvertures by his cousin Johann Bernhard Bach, including the Ouverture in G minor (VBN I/B/7),2 which includes a part for ‘violino concertino’. These ouvertures would have been performed by the student Collegium musicum, which was directed by Bach from spring 1729 till summer 1737. According to C. P. E. Bach and J. F. Agricola (J. S. Bach obituary, 1754), ‘Johann Bernhard wrote many ﬁne ouvertures in the manner of Telemann’.3 There is, of course, a strong Italianate, concertante element in the earlier type of ouverture, to which Bach’s Ouvertures in C and D (BWV 1066 and 1069) belong, namely the use of ritornello form in the introductory movement. But the concert en ouverture introduces an additional concertante element in the form of a solo instrument—usually violin, but occasionally ﬂute or recorder. Scheibe makes it clear that the soloist is not expected to exhibit the same degree of virtuosity as in a true concerto. Nor, judging by the surviving examples, does there need to be a solo part

1 See Joshua Rifkin, ‘The “B-minor Flute Suite” Deconstructed: New Light on Bach’s Ouverture BWV 1067’, and Steven Zohn, ‘Bach and the Concert en ouverture’, both in G. G. Butler (ed.), Bach Perspectives 6 (Urbana and Chicago, 2007), pp. 1–98 and 137–56 respectively. 2 VBN, in Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992), pp. 221–400. Bach’s copies of Johann Bernhard’s ouvertures are listed under VBN I/B/4–8. 3 BD III, No. 666 , p. 81; NBR, No. 306 , p. 298.

two ouv ertures

249

in every movement. In other words, the original French flavour of the genre has to be maintained. All these conditions are observed in the Ouverture in D, BWV 1068. Some original parts of this ouverture date from early 17314—not long after the copies of the Johann Bernhard Bach ouvertures—and must have been used for a performance by the Collegium musicum. This version, scored with trumpets and drums as well as two oboes, seems to have been based on a lost original for strings and continuo only,5 whose date can only be guessed. It seems unlikely to have been composed before about 1725, however, since the earliest ouvertures of this type by Telemann, the most prolific exponent and possibly founder of the sub-genre, date from around 1725. Just as in Bach’s earlier ouvertures, the central vite section of the introductory movement has fugal ritornellos. Here, however, the anapaestic subject is combined with three clearly differentiated countersubjects according to a permutation scheme. The true ‘concerto moment’ arrives when the ritornello cadences in the tonic (b. 41) or relative minor (b. 70) and is followed by a concertante episode for the first violin, accompanied by lower strings which are at first non-thematic but later develop the fugue subject, thereby binding ritornello and episode together. The first violin’s figuration is soloistic in character and might well have been intended for a single player.6 If this is true of the ouverture proper (the first movement), it must have been true also of the following Air (popularly known as the ‘Air on a G string’), in which the first violin has a florid, cantabile melody of great beauty. The inner parts to some extent share in the elaboration, while the continuo provides a strong underpinning of the freely evolving upper lines with its pseudo-ground bass. The French dances that follow—alternativement Gavottes, a Bourreé, and a Gigue—lack the concertante violin element and essentially differ little from those of Bach’s earlier ouvertures. Bach’s other concert en ouverture, that in B minor, BWV 1067—in its revised version for flute, strings, and continuo—almost certainly dates from 1739 and might have been occasioned by Bach’s return to the directorship of the Collegium musicum in October of that year after standing down for two years.7 It has recently been shown, however,8 that it was based on an earlier composition, a lost ouverture in A minor for concertante violin, strings, and continuo. The first movement in its surviving version is strikingly similar to that of Johann Bernhard Bach’s Ouverture in G minor (mentioned earlier), as is the opening movement of Bach’s Sonata in B minor for flute and obbligato harpsichord, BWV 1030 (Ex. 1). This suggests that Bach might have composed the A minor original version of BWV 1067 with his cousin’s ouverture in 4 According to Andreas Glo¨ckner, ‘Neuerkenntnisse zu J. S. Bachs Auffu¨hrungskalender zwischen 1729 und 1735’, BJ 67 (1981), pp. 50 and 71. 5 ¨ berlieferung: Bemerkungen zur Ouvertu¨re BWV 1068’, BJ 83 See J. Rifkin, ‘Besetzung, Entstehung, U (1997), pp. 169–76. 6 ¨ berlieferung’, p. 175. As Rifkin points out in ‘Besetzung, Entstehung, U 7 See Rifkin, ‘The “B-minor Flute Suite” Deconstructed’, p. 49. 8 By Rifkin, ‘The “B-minor Flute Suite” Deconstructed’. Werner Breig, however, believes that the original version was scored for strings and continuo without solo instrument; see his ‘Zur Vorgeschichte von Bachs Ouvertu¨re h-Moll BWV 1067’, BJ 90 (2004), pp. 41–63.

250 t h e h a r p s i c h o r d c o n c e r t o s e t c .

Ex. 1

a) Johann Bernhard Bach, theme from 1st movement of Ouverture in G minor

b) Theme from 1st movement of Sonata in B minor, BWV 1030

c) Theme from 1st movement of Ouverture in B minor, BWV 1067

mind, which in turn points to an origin in 1730/1, not long after Bach had taken over the Collegium musicum (spring 1729) and performed his relative’s ouvertures (1730). It is impossible to judge what the original violin part was like, for Bach himself wrote out the flute part of the transposed version and might have made many alterations in the process. The string and continuo parts, however, were copied by anonymous scribes and therefore must have been left unchanged apart from the transposition.9 This suggests that any changes Bach made to the solo part did not go deeply into the substance of the music but mainly took the form of increased elaboration—one of the most characteristic features of Bach’s revision procedures. As in the D major Ouverture, the solo instrument in true concertante fashion leads the episodes that intervene between the ritornellos of the fast, fugal middle section of the introductory movement. In addition, the two ouvertures have in common the manner in which the accompaniment to the solo part is at first free but later based on the fugue subject or a figure thereof, knitting together the episodes and the ritornello. The soloist has no independent role in the dotted-rhythm introduction, but in the lentement conclusion it provides a descant to the theme for the first six bars. The role of the soloist in the following dances varies. It lacks an independent part in the Sarabande and Menuet, and in the gavotte-like Rondeau it detaches itself from the ensemble only in five bars from the last episode (bb. 32–6). The Bourreé and Polonaise, however, both take the form ABA (Bourreé I, II, I; Polonaise, double, Polonaise), which enables the solo instrument to take the limelight in the middle section. And in the concluding Badinerie (= ‘banter’)—one of those modish, dancelike pieces in 2/4 time that occur quite frequently in Bach’s vocal and instrumental music of the late 1720s and 1730s—the solo instrument leads the ensemble throughout and is provided with plentiful opportunities for a glittering display of virtuosity. 9

As Rifkin concludes in ‘The “B-minor Flute Suite” Deconstructed’, pp. 7–10.

tw o v i o li n c o n ce r tos

251

By comparison with Bach’s earlier ouvertures, that in B minor is especially notable for the thematic nature of the bass line and its interaction with the first violin and/or solo instrument. In what is perhaps the most overtly thematic of Bach’s dottedrhythm introductions, the theme occurs only twice in the treble (bb. 1 and 11) but five times in the bass (bb. 2, 4, 6, 8, and 20). The same theme returns in a triple-time variant (now in all four voices) in the lentement conclusion. In the Rondeau, the bass is given a strong counter-theme; indeed, it could be argued that the true inventio is a double theme in the outer parts. Both strains of the Sarabande are in canon at one bar and the lower 5th between the outer parts (relaxed only at the final cadences). In Bourreé I the bass has a true ostinato as a counterweight to the treble melody. In the Polonaise, which belongs to the mazurka type occasionally cultivated by Bach around 1730, the lentement staccato melody is transferred to the bass in the double, while the soloist has a very ornate counter-melody in short note-values. The opening quaver figure of the Menuet is constantly interchanged between treble and bass. Finally, in the Badinerie the bass has a free inversion of the solo theme; later, however (b. 6), it takes up the solo theme itself as counterpoint to a new repeated-note motive in the solo part.

Two violin concertos Title

Earliest source/s

Scribe, date

Violin Concerto in A minor, BWV 1041 Two-Violin Concerto in D minor, BWV 1043

Berlin, St 145 Krako´w, St 148

Part-autograph, 1730 Part-autograph, 1730/1

The Ouvertures in D (BWV 1068, revised version) and A minor (BWV 1067, original version) are not the only Bach compositions with concertante violin that might have been performed by the Collegium musicum around 1730–1, not long after he took over the directorship. For the original performing parts of the Concertos in A minor for solo violin (BWV 1041) and in D minor for two violins (BWV 1043) also date from 1730 or thereabouts.10 Whether the presumed Collegium musicum performances were the original ones is not known, but it seems likely. Both compositions exhibit a late tendency of Bach’s to mask divisions in the interests of continuity. Thus, in the ﬁrst movement of the A minor Concerto, it is not the ABA1 reprise form that is unusual but rather the modulatory link into the reprise (bb. 117–22). The slow movement represents an advance on that of the E major Violin Concerto (BWV 1042). Both are based on an ostinato theme, which functions as a variable ground bass during the solos. But in the E major Concerto, the theme supplies framing ritornellos only, whereas in the A minor it furnishes a complete ritornello structure. As in the ﬁrst movement, an overall reprise structure (ABA1) is clearly evident and a modulatory 10

According to Glo¨ckner, ‘Neuerkenntnisse’, pp. 49–50 and 71.

252

the harps ichord concertos etc.

link at the end of the middle section (keys g–c, bb. 30–6) masks the join to the reprise. In the finale, fugue and ritornello form are combined within an overall ABA da capo form, as in the finale of the fifth Brandenburg Concerto. The analogy between the two finales goes further, since both are cast in the rhythm of a gigue. As in the first two movements, middle section and reprise are closely interlinked, here by means of a false reprise: the ritornello theme is anticipated in the tonic (bb. 91–3) before an episodic diversion and a powerful perfidia passage lead into the reprise proper (b. 117). The opening Vivace of the D minor Concerto, like the finale of the fourth Brandenburg, is an amalgamation of fugue, ritornello, and reprise structures. As in the first movement of the A minor Concerto, however, there is a subtle join between the middle paragraph and the reprise. The modulatory, developmental passage from the first episode returns (b. 69 = 30), then the episodic parallel 10ths theme (b. 77 = 26), and finally the ritornello theme itself (b. 85). The celebrated Largo is similar in structure to the Vivace, albeit on a much smaller scale; and once again the middle section and reprise are joined, here by means of a modulatory link (bb. 38b–40, B♭–c– d–F). In the finale, as in that of the A minor Concerto, the ritornello return (b. 134) is anticipated at the end of the middle section (b. 123), prompting the return of a powerful sequential passage built on a cadential tutti figure from the opening ritornello (b. 127 = 41). The steps Bach takes in the A minor and D minor Violin Concertos to mask divisions and preserve continuity apparently represent a late development in his concerto writing and suggest that the two compositions might have been written in fairly close succession around 1730, presumably for the Collegium musicum.

Concertos for three or four harpsichords Title

Earliest source/s

Scribe, date

Concerto in D minor for three harpsichords, BWV 1063 Concerto in C major for three harpsichords, BWV 1064 Concerto in A minor for four harpsichords, BWV 1065

Berlin, Am.B.67

J. F. Agricola; orig. c. 1730

Berlin, Am.B.68

J. F. Agricola; orig. c. 1730

Berlin, St 378

Anon., c. 1730

These three compositions appear to have been the earliest of Bach’s fourteen harpsichord concertos. They most likely originated around 1730, shortly after Bach had taken over the directorship of the Collegium musicum.11 The presence of his two eldest sons in the family home till 1733 (Wilhelm Friedemann) and 1734 (Carl Philipp Emanuel) probably explains why he was writing for three or four harpsichords at this time. F. K. Griepenkerl, in his preface to the ﬁrst edition of the three-harpsichord

See Rudolf Eller and Karl Heller, Krit. Bericht, NBA VII/6 (1976), pp. 26–7, 61–2 , and 89.

concertos for three or four harpsichords

253

concertos, remarks that the D minor (BWV 1063) owes its origin ‘to the fact that the father wished to give his two eldest sons the opportunity to develop in all types of performance’.12 Griepenkerl must have obtained this information from his teacher J. N. Forkel, who in turn no doubt acquired it from the Bach sons themselves. By extension, presumably the same role in the musical education of Bach’s two eldest sons would have been adopted by the C major and A minor Concertos (BWV 1064–5). Bach himself would presumably have played Cembalo I, Wilhelm Friedemann Cembalo II, and Carl Philipp Emanuel Cembalo III. In the four-harpsichord concerto, Cembalo IV might have been played by Bach’s next son Johann Gottfried Bernhard or possibly by his pupil Johann Ludwig Krebs. The Concerto in A minor for four harpsichords was arranged by Bach from Vivaldi’s Concerto in B minor for four violins from his L’estro armonico, Op. 3 No. 10 (Amsterdam, 1711). Bach had, of course, fashioned numerous solo harpsichord concertos (senza ripieno) on the basis of this collection during his Weimar years, but it is not a foregone conclusion that he would still value it at this much later stage in his career. The three-harpsichord Concerto in D minor—except for the slow movement, which is believed to have been imported from elsewhere—was probably adapted from an original in the same key for two violins, strings, and continuo. There are strong grounds for concluding that this lost original was composed by Bach.13 Above all, the characteristics of the concerto form that Bach inherited from Torelli, Albinoni, and Vivaldi are present but modified according to his own propensities, such as contrapuntal writing and motivic or thematic development. The first movement is not only cast in Bach’s typical frame form (ABA1) with reverse-order reprise, but the unisono ritornello theme is presented in octave canon as early as the first episode (b. 15) and inverted in the bass of the second episode (b. 43). The Alla Siciliana is cast in binary dance form with varied repeats, a structure occasionally employed by Bach and eventually inherited by his son Carl Philipp Emanuel.14 As so often in Bach, the Allegro-finale unites ritornello form with fugue. In two cases (bb. 101 and 167) the forte ritornello/fugue subject (stated twice in different keys) is prefaced by a piano variant in canonic stretto. In the first movement, the virtuoso writing is largely confined to Cembalo I, which Bach himself no doubt played; and it is Cembalo I that leads the double-like varied repeats in the siciliana. In the finale, on the other hand, each of the three harpsichords receives its own extended solo (bb. 41, 73, and 139), presumably because the required technical standard is not as high as that of the first movement. 12 ‘wahrscheinlich dem Umstande, daß der Vater seinen beiden a¨ltesten So¨hnen, W. Friedemann und C. Ph. Emanuel Bach, Gelegenheit verschaffen wollte, sich in allen Arten des Vortrags auszubilden’; quoted in Eller and Heller, Krit. Bericht, NBA VII/6 , p. 26. 13 As argued convincingly by Eller and Heller, Krit. Bericht, NBAVII/6 , pp. 29–31. The theory of an original in D minor for two violins is set out in the same critical report, p. 28 , and amplified by Karl Heller, ‘Eine Leipziger Werkfassung und deren unbekannte Vorlage: Thesen zur Urform des Konzerts BWV 1063’, in U. Leisinger (ed.), Bach in Leipzig—Bach und Leipzig (Hildesheim, 2002), pp. 89–108. 14 This form is discussed by Ulrich Siegele, Kompositionsweise und Bearbeitungstechnik in der Instrumentalmusik Johann Sebastian Bachs (Neuhausen and Stuttgart, 1975), pp. 45–6.

254

t h e ha r p s i c h o r d c o n c e r t o s e t c .

It is widely accepted that the Concerto in C for three harpsichords, BWV 1064, was adapted from a lost concerto in D for three violins.15 Bach’s authorship of the original version, which is believed to date from the late Weimar or early Co¨then years,16 is unquestioned. The opening movement is cast in Bach’s standard ABA1 reprise form. With his love of thematic combination, he opens the ritornello with a double theme: a Vivaldian unisono, as in the D minor Concerto, but here combined with a galant theme with syncopations, repeated notes, and triplets. The galant theme, however, is purely decorative; it is the unisono theme that is subsequently developed—in bass sequence (bb. 21 and 78), for example, and in an eight-bar modulatory sequence in which it is constantly interchanged between the three soloists (bb. 91–8). The Adagio, one of Bach’s most profound slow movements, needs space in which to expand, hence its full ritornello form. The ritornello is again built on the principle of thematic combination: a one-bar bass theme is combined with a violin theme made up of a seven-note motive and its inversion. The bass theme underpins all the episodes as basso quasi ostinato, a standard feature of Bach’s usual slow-movement form. From the third episode onwards (b. 18) the ostinato theme is inverted and/or enhanced in interval; and the combination of this with the expressively decorated harpsichord parts and the string chords, spiced with appoggiaturas, plumbs depths rarely reached even in Bach’s slow movements. The ritornello of the alla breve finale again combines themes (bb. 1–3)—one of which is a severely plain soggetto—before introducing further motives in the following sequential tutti (bb. 4–9). Each harpsichord takes the limelight in turn in this movement (bb. 59, 102, 141), which confirms the overall impression that they are more evenly matched than in the D minor Concerto. Only the brilliant perfidia passages are restricted to Cembalo I (first movement, b. 107b; finale, b. 70). This might suggest that the concerto originated somewhat later than the D minor—after the Bach sons had further developed their technique.

Concertos for two harpsichords Title

Earliest source/s

Scribe, date

Concerto in C, BWV 1061 Concerto in C minor, BWV 1062 Concerto in C minor, BWV 1060

Berlin, St 139 Berlin, P 612 Berlin, St 136

A. M. and J. S. Bach, c. 1732/3 Autograph, 1736 J. C. Altnickol et al.; orig. c. 1736

The concertos for two harpsichords appear to have originated a little later than those for three or four harpsichords, which might possibly be connected with the three eldest Bach sons leaving home one after another in the early-to-mid 1730s. The earliest 15

A reconstruction of the lost original is given by Wilfried Fischer in NBA VII/7 (1970), pp. 103–38. Gregory Butler proposes Co¨then, 1717/18; see his ‘Toward a More Precise Chronology for Bach’s Concerto for Three Violins and Strings, BWV 1064a’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke (Witten, 1997), pp. 235–47. 16

c o nc e rt os fo r tw o harp si cho r ds

255

of these compositions seems to be the Concerto in C, BWV 1061.17 This work is exceptional in several ways: it seems to be the only one of Bach’s fourteen harpsichord concertos that was originally written for the instrument; and the version of its earliest source, Anna Magdalena Bach’s copy of around 1732/3, is a concerto senza ripieno— that is, without the accompaniment of strings and continuo. Several of the Brandenburg Concertos belong to this type—Nos. 3 and 6 together with the original version of No. 2—but a closer parallel is found in cases where the entire concerto texture is entrusted to a single harpsichord, as in the Weimar concerto transcriptions and in the Italian Concerto from Clavieru¨bung II (1735). The last-named work is particularly relevant, since it was conceived for solo harpsichord and dates from around the same time as the two-harpsichord Concerto in C. At a later date, Bach added ripieno parts to the C major Concerto, which bring it into line with the other two-harpsichord concertos.18 With its fresh, sparkling material, full of great charm, this is without doubt one of Bach’s finest concertos. In the opening paragraph (bb. 1–61)—largely in the tonic and dominant—the themes of ritornello and episode are kept clearly distinct. In the central ‘development’ (bb. 61–86), however, they are intermingled. Here, moreover, ritornello and episode are conjoined within a single, unbroken period. As in a da capo aria, a mediant cadence and hiatus precede the tonic return (b. 87), which opens with ritornello material only (bb. 87–99) to clarify its reprise function. Thereafter, however, themes of ritornello and episode consort once more; and in the final period (bb. 141–66) episode merges into ritornello. In the ‘Adagio ovvero Largo’ that follows, it is no longer possible to speak of ritornellos or episodes. This is a frame form, but the frame is not a ritornello, as often elsewhere, but a melodic period in the tonic (bb. 1–11 and 53b–63). Its two themes—an expressive melody in free siciliana rhythm and an answering phrase in sequential imitation (from b. 7b)—provide all the material for the central part of the movement. The concluding Fuga is built on an ample subject with a threefold structure of headmotive a, sequential consequent b, and tail-figure c. This ritornello-like design occurs not infrequently among Bach’s Weimar fugue subjects, but the closest parallel in melodic and rhythmic shape appears to be the fugue subject from Reincken’s Toccata in G (Ex. 2). Perhaps Bach’s invention was sparked off by the Reincken piece, or perhaps the Fuga was intended as a tribute to the Hamburg composer. As often in Bach, fugue and ritornello form are united, and there

17 See the date of the original performing parts in the table. According to Hans-Joachim Schulze, the concerto might have been composed for Dresden, alongside the Missa (BWV 2321), in the early summer of ¨ berlieferung und Chronologie’, in P. Ahnsehl et al. (eds.), 1733; see his ‘J. S. Bachs Konzerte: Fragen der U Beitra¨ge zum Konzertschaffen J. S. Bachs, Bach-Studien 6 (Leipzig, 1981), pp. 9–26 (esp. 11–12). 18 The two versions and their relationship are discussed in detail by Karl Heller, ‘Zur Stellung des Concerto C-Dur fu¨r zwei Cembali BWV 1061 in Bachs Konzert-Oeuvre’, in W. Hoffmann and A. Schneiderheinze (eds.), Bericht u¨ber die Wissenschaftliche Konferenz zum V. Internationalen Bachfest der DDR Leipzig 1985 (Leipzig, 1988), pp. 241–52. See also Karl Heller and Hans-Joachim Schulze, Krit. Bericht, NBA VII/5 (1990), pp. 91–5.

256

the harpsichord concertos etc.

Ex. 2 etc.

a) Theme from ﬁnale of Concerto in C, BWV 1061

b) Jan Adam Reincken, theme from Toccata in G

is a very clear demarcation between fugal ritornellos and figural episodes until the last episode (bb. 112–34), which functions rather like a development section, treating fugal and episodic themes alike. The two C minor Concertos, BWV 1060 and 1062, appear to be of slightly later origin. One of them, BWV 1062, was adapted in 1736 from the Concerto in D minor for two violins,19 and the other, BWV 1060, might have originated around the same time as a companion piece. Its original version does not survive, but it has long been thought that it might have been a violin and oboe concerto of Co¨then origin.20 The opening Allegro exhibits a type of concerto form often cultivated by Bach in which episodic material plays a relatively small part compared with ritornello themes. Even in the first episode, the chief episodic theme already alternates with the ritornello headmotive. Twice in the later course of the movement, a ritornello proper (bb. 33–6 and 71–4) is prefaced by a period that shares the characteristics of ritornello and episode (bb. 23–32 and 61–70). Here we form the impression of a false ritornello which is episodically diverted before the entry of the true ritornello—a potent dramatic device. The final complex (bb. 89–110) is devoted to ritornello themes throughout. The ‘Largo [ovvero] Adagio’ is fugal on the basis of a pastorale-like subject, which, both in the opening and middle paragraphs, is answered first at the 5th and then at the octave—the latter in a key-confirming passage that leads to a submediant (b. 10) or mediant (b. 23) full-close. The same passage confirms the tonic return in the concluding paragraph. The Allegro-finale falls into a relatively straightforward ABA1 reprise structure, often found in Bach’s finales, in which the outer paragraphs are almost entirely devoted to ritornello material. The major element of contrast takes place after the dominant cadence that closes the first paragraph (bb. 67–8): the middle paragraph then opens with a brilliant, extended perfidia episode for Cembalo I, accompanied by the headmotive of the ritornello. At the end of the central paragraph, this passage

19 See the facsimile of the autograph score, Berlin P 612: J. S. Bach: Konzert c-Moll fu¨r zwei Cembali und Streichorchester, BWV 1062, Sonate A-Dur fu¨r Flo¨te und Cembalo, BWV 1032, ed. H.-J. Schulze (Leipzig, 1979). 20 On the nature of the original version see Joshua Rifkin, ‘Verlorene Quellen, verlorene Werke: Miszellen zu Bachs Instrumentalkomposition’, in Geck and Breig (eds.), Bachs Orchesterwerke, pp. 59–75 (esp. 61–5).

concertos for solo harpsichord

257

returns on a dominant pedal, acting as dominant preparation for the reverse-order reprise of the opening paragraph.

Concertos for solo harpsichord Title

Earliest source/s

Scribe, date

Six Harpsichord Concertos: No. 1 in D minor, BWV 1052 No. 2 in E, BWV 1053 No. 3 in D, BWV 1054 No. 4 in A, BWV 1055 No. 5 in F minor, BWV 1056 No. 6 in F, BWV 1057 Concerto in G minor, BWV 1058 Concerto in D minor, BWV 1059

Berlin, P 234

Autograph, c. 1738

Berlin, P 234 Berlin, P 234

Autograph, c. 1738 Autograph, c. 1738

Around 1738 Bach began an opus of solo harpsichord concertos.21 This is clear from the inscription ‘J.J.’ (‘Jesu juva’) at the top of the first page of the autograph manuscript. He completed the first concerto, BWV 1058, an arrangement of the A minor Violin Concerto transposed down to G minor, then wrote the first nine bars only of a Concerto in D minor, BWV 1059, before abandoning the opus. He might have been dissatisfied with the relatively simple mode of transcription employed in the G minor Concerto. In addition, upon reflection he might have felt that he wanted to make a more brilliant, virtuoso start to the opus. Therefore he began again with the Concerto in D minor, BWV 1052, which amply fulfils this condition, resembling Vivaldi’s ‘Grosso Mogul’ Concerto, RV 208 (arranged by Bach as the Organ Concerto in C, BWV 594) in its extreme virtuosity. He then proceeded to complete an opus made up of a standard set of six compositions, headed ‘J.J.’ as before, but now inscribed at the end ‘Finis. S.D.Gl.’ (‘Soli Deo gloria’), Bach’s customary sign of completion. All six concertos are transcriptions,22 but only in two cases are the originals extant: No. 3 is based on the E major Violin Concerto (BWV 1042), and No. 6 on Brandenburg Concerto No. 4 in G (BWV 1049). Nos. 1 and 5 are based on lost violin concertos, and No. 4 on a lost concerto for oboe d’amore (the original of No. 2 has not been 21 The history of Bach’s solo harpsichord concertos has been reconstructed by Werner Breig; see his ‘J. S. Bach und die Entstehung des Klavierkonzerts’, Archiv fu¨r Musikwissenschaft, 36 (1979), pp. 21–48; ‘Zur Chronologie von J. S. Bachs Konzertschaffen: Versuch eines neuen Zugangs’, Archiv fu¨r Musikwissenschaft, 40 (1983), pp. 77–101; ‘Zum Kompositionsproceß in Bachs Cembalokonzerten’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld: 61. Bachfest der Neuen Bachgesellschaft, Duisburg 1986 (Kassel, 1988), pp. 32–47; ‘Composition as Arrangement and Adaptation’, in J. Butt (ed.), The Cambridge Companion to Bach (Cambridge, 1997), pp. 154–70 (esp. 165–70); and Krit. Bericht, NBA VII/4 (2001). 22 As are all but one (BWV 1061) of Bach’s concertos for one to three harpsichords. Eight of the twelve originals are lost. A coherent but in the last analysis unverifiable theory regarding their date of origin has been put forward by Siegbert Rampe and Dominik Sackmann, Bachs Orchestermusik: Entstehung, Klangwelt, Interpretation (Kassel, 2000).

258

the harpsichord concertos etc.

established). The date c. 1738 suggests that these concertos might have been composed (or arranged) in preparation for the second phase of Bach’s directorship of the Collegium musicum, which began in October 1739 and continued till 1741 at the earliest. It has been pointed out, however,23 that Bach’s lengthy stay in Dresden in May 1738 would have provided another suitable occasion for the performance of the concertos. The most likely soloist is, of course, Johann Sebastian Bach himself, though other names have been put forward from time to time. Bach opened the set with one of his greatest concertos, No. 1 in D minor, which conveys a sense of huge elemental power. This is in no small measure due to the inexorable concentration on the minor-mode tonic and dominant in the opening complex of both outer movements (Allegro, bb. 1–27; finale, bb. 1–41). These keys are juxtaposed with little transition, creating a somewhat modal, archaic effect. The key notes are, of course, violin open strings, which is highly relevant to the original form of the concerto: in the middle paragraph of both outer movements, brilliant bariolage episodes take place in the minor keys of D and A (finale) or A and E (first movement), exploiting the open strings; and here again the keys are starkly juxtaposed rather than prepared. This relentless adherence to the minor mode produces the effect that, when the motto theme of the first movement does enter briefly in the major (F, bb. 40–1; C, bb. 91–2; B♭, bb. 134–5), it appears like a sudden, unexpected shaft of light. Yet another minor key, the subdominant G minor, plays a significant role: it is the key of the central Adagio, and in both outer movements it bridges the join between the central and concluding paragraphs. The first two movements are both framed by Vivaldian unisono ritornellos; and indeed there is something Vivaldian about the extended passages of unbroken tonic and dominant that have been described. Where Bach pursues his own agenda is, above all, in the thematic or motivic use of ritornello material in the episodes. This developmental purpose had to be pursued alongside another: that of trying to achieve the maximum brilliance in the solo harpsichord part. In the opening Allegro, these aims result in an alternation between free, virtuoso episodes, including perfidia and cadenza (bb. 62–90 , 95–103 , 109–12 , 136–71), and those in which ritornello themes are developed, mostly in the ripieno parts (for example, bb. 28–39 , 42–55 , and 113–33). In both outer movements, the A-paragraph of the ABA1 reprise structure contains mainly ritornello-based episodes, and the free, virtuoso display of the soloist begins at the start of the middle paragraph. In the first movement, Bach three times enhances the drama by employing perfidia as a form of preparation for an approaching ritornello (bb. 82 , 95 , and 162). No less dramatic are the diversionary tactics employed at the end of both outer movements: what purports to be the final tonic ritornello enters briefly (Allegro, b. 172; finale, b. 224) but is diverted into an episode before the arrival of the ritornello proper. In the central Adagio, the unisono ritornello not only

By Schulze, ‘J. S. Bachs Konzerte’, pp. 9–26.

c o n c e r t o s f o r s o l o h a r p s i c ho r d

259

serves as a frame around the four solo periods, but acts as a ground bass during them, underpinning the florid decoration of the solo harpsichord part. The more genial, lyrical Concerto No. 2 in E has in common with the D minor the adaptation of its original model for use in sacred cantatas in the late 1720s.24 Neither the key nor the solo instrument of that original model can be established, though a woodwind instrument seems likely—perhaps oboe or oboe d’amore. It is possible that the original dates from Bach’s early Leipzig years (1723–6).25 In both outer movements he employs da capo aria form (ABA), as he does quite frequently in other concerto movements. Typical of his procedure is that, although the solo harpsichord has its own episodic theme in paragraph A (b. 9), it almost invariably leads to or is combined with ritornello material, whereas the solo episodes of B are largely free of such constraints. The dance-like finale is remarkable for its ritornello theme: not only is the headmotive in close, three-part canonic imitation, but its rhythm occurs six times in succession, with various forms of melodic decoration. The new solo theme that follows is treated like a vocal motto: briefly stated, answered by an equally brief ritornello return, then restated in extended form as the first episode proper. Again, the first major contrast with the ritornello material takes place in paragraph B, which opens with a striking chromatic theme, quite distinct from anything that has been heard up to this point. As in the standard Scarlattian da capo aria, the middle paragraph ends with a full-close in the mediant g♯, and a hiatus ensues before the tonic return. The beautiful, haunting slow movement is cast in the rhythm of a siciliana (and hence so titled), a dance that Bach employed in several concerto slow movements (BWV 1055, 1061 , 1063), as well as in various cantata arias from the later 1720s. As in the slow movement of the D minor Concerto, the ritornello forms a frame around the central paragraph, which here consists of two solo periods (bb. 7 and 19). Each period opens with a solo siciliana melody—quite distinct from that of the ritornello which, however, is subsequently varied by the soloist. Concerto No. 3 in D is an effective harpsichord arrangement of the E major Violin Concerto (BWV 1042). The original model for Concerto No. 4 in A, on the other hand, is lost, though it has long been accepted that it was probably for oboe d’amore.26 The opening Allegro is exceptionally rich in themes, largely of a light, athletic character. The ritornello returns in full only at the end of the first paragraph A (b. 33) and at the end of the whole movement (b. 89). Elsewhere, we hear only its headmotive, or ‘motto’, which introduces three successive episodes in the middle 24 The first two movements of Concerto No. 2 in E were used in Cantata 169 , nos. 1 and 5 (1726) and the third movement in Cantata 49 , no. 1 (also 1726). The first two movements of Concerto No. 1 in D minor were used in Cantata 146 , nos. 1–2 (c. 1728) and the finale in Cantata 188 , no. 1 (also c. 1728). 25 According to Gregory G. Butler, ‘J. S. Bach’s Reception of Tomaso Albinoni’s Mature Concertos’, in D. R. Melamed (ed.), Bach Studies 2 (Cambridge, 1995), pp. 20–46. 26 This theory of Donald Francis Tovey’s is now generally accepted by scholars; see his ‘Concerto in A major for Oboe d’amore with strings and continuo’, in D. F. Tovey, Essays in Musical Analysis, vol. ii (London, 1935), pp. 196–8. Schulze, ‘J. S. Bachs Konzerte’, pp. 13–15 , gives solid grounds for dating the original oboe d’amore concerto in the year 1721.

260 t h e h a r p s i c h o r d c o n c e r t o s e t c . paragraph B (bb. 49 , 57 , and 65). In the concluding paragraph A1 (b. 79), the motto in the tonic purports to be the full ritornello return but is diverted by an episode before the tonic return proper—a rhetorical device that we have already seen enacted in other concertos. In the first episode a new solo theme enters, which however leads straight back to the motto theme; the two themes are subsequently combined in counterpoint (bb. 42–3 and 82–3). The slow movement is a siciliana, yet less obviously so than that of Concerto No. 2 , hence the non-committal heading ‘Larghetto’. Both slow movements are constructed in frame form, but here the outer tonic ritornellos are built on the traditional lamento bass, with its chromatic descent through a 4th. This lamento theme often returns in the harpsichord bass, underpinning the righthand solo, which is largely free, though it does contain some elements of reprise. The opening ‘motto’ of the finale is unusually florid for a tutti theme, with its rapid demisemiquaver scale figures; and the solo part, which begins with a quite new theme, displays a galant mixture of different note-values. The ritornello, as rich in themes as that of the opening Allegro, is greatly abridged in all its internal returns, often consisting of motto only (bb. 57 , 79 , and 95). The ritornello that concludes the middle paragraph of the ABA1 structure is diversified by an interpolated episode. A full-close in the mediant c♯ precedes the tonic return, as in the standard da capo aria. Concerto No. 5 in F minor is probably based on a lost violin concerto in G minor, whose slow movement was for the harpsichord version replaced by a movement originally in F major, drawn from a lost oboe concerto. This movement had already been adapted in 1729 to form the sinfonia of the cantata Ich steh mit einem Fuß im Grabe, BWV 156.27 The ritornello of the opening movement is strikingly similar to that of the C minor Concerto, BWV 1060: in both cases, the motto is immediately repeated in a different key, and its abruptly cut-off ending is echoed by the soloist/s. Similar phrase-ends occur in the Presto-finale of the F minor Concerto, here echoed by pizzicato strings. This passage is later expanded, and the interval of the cadence figure enlarged (5th–6th–7th), a device that adds enormous power to the central dominant and concluding tonic ritornellos (bb. 99 and 203). Both outer movements are exceptional in structure. The opening movement is highly condensed by comparison with Bach’s usual ritornello structures: the framing paragraphs A and A1 consist of ritornello or troped ritornello only, and the middle paragraph B of three episodes, divided only by entries of the motto theme. The dance-like finale is bipartite: two corresponding paragraphs are surrounded by central and framing ritornellos. The Adagio lacks

27 See Joshua Rifkin, ‘Ein langsamer Konzertsatz Johann Sebastian Bachs’, BJ 64 (1978), pp. 140–7. According to Rifkin, the sinfonia and concerto versions had a common source—the slow movement of a lost oboe concerto in D minor, whose outer movements were in 1726 employed as sinfonias in Cantata No. 35. Bach began to arrange this oboe concerto for solo harpsichord (BWV 1059), but discontinued the arrangement after only nine bars; see Werner Breig, ‘Bachs Cembalokonzert-Fragment in d-Moll (BWV 1059)’, BJ 65 (1979), pp. 29–36. Regarding the origin of the F minor Concerto in a lost violin concerto in G minor, see Breig, ‘Zur Werkgeschichte von Bachs Cembalokonzert BWV 1056’, in Geck and Breig (eds.), Bachs Orchesterwerke, pp. 265–82.

s on a t a s i n c o n c e r t o s t y le

261

ritornellos. Instead, a variable ground bass underpins a lovely, simple, Italianate melody, still no doubt close to its original forms as in the cantata version, but here expressively and idiomatically decorated for the harpsichord. When seeking a fitting culmination for his set of six solo harpsichord concertos, it is not hard to understand why Bach chose to adapt the fourth Brandenburg Concerto to form Concerto No. 6 in F. It is undoubtedly one of the finest of all his concertos and, with the exception of the original model of Harpsichord Concerto No. 1 , it possesses the most brilliant solo violin part. By framing the set with these two concertos, Bach placed the virtuosity of the harpsichordist in the two most prominent positions. The nature of Brandenburg No. 4 as a concerto grosso might have been a disadvantage, but Bach turns it into an asset. Compared with the pure string ripieno of the first five concertos, the contrasted tone of the two recorders acts as an enhancement. Although they still play a very prominent role, both in ritornellos and in episodes, their importance is somewhat diminished to the extent that the harpsichord’s (compared with that of the original violin) is increased. Thus, in episodes originally for two recorders and bass (first movement, bb. 165 and 293; finale, b. 159), an entirely new obbligato harpsichord part is added. And in the slow movement, the concertino parts, originally for two recorders and solo violin, are now played entirely by the solo harpsichord, while the recorders join the ripieno. Again, the harpsichord has a new, florid treble part at one point (bb. 55–8) in counterpoint with a thematic bass. Despite the brilliance of the original violin part, there is still scope for enhancement by diminution in the harpsichord part: running passages are often spiced with, or even replaced by, shorter note-values (first movement, b. 83; finale, bb. 41 and 219). All told, the balance between concerto grosso and solo concerto, fairly equal in the Brandenburg version, is here shifted decisively in favour of solo harpsichord.

Sonatas in concerto style Title

Earliest source/s

Scribe, date

Six Sonatas for organ: No. 1 in E♭, BWV 525 No. 2 in C minor, BWV 526 No. 3 in D minor, BWV 527 No. 4 in E minor, BWV 528 No. 5 in C, BWV 529 No. 6 in G, BWV 530 Sonata in G minor for viola da gamba and obbligato harpsichord, BWV 1029 Sonata in B minor for ﬂute and obbligato harpsichord, BWV 1030 Sonata in A for ﬂute and obbligato harpsichord, BWV 1032

Berlin, P 271

Autograph, c. 1730

Berlin, St 163

C. F. Penzel, 1753

Berlin, P 975

Autograph, c. 1736/7

Berlin, P 612

Autograph, c. 1736/7

262

the harpsichord concertos etc.

The Sonate auf Concertenart (sonata in concerto style) was described by Johann Adolf Scheibe in his Der critische Musicus (2nd edn; Leipzig, 1745).28 The chief criterion is the adoption of the three-movement concerto form (fast–slow–fast), rather than the fourmovement form of the sonata da chiesa. Bach had made a move in this direction in the last of his Six Sonatas for violin and obbligato harpsichord (BWV 1014–19) from the early 1720s, whose first two and last movements form a concerto-like frame within which a suite-like succession of movements is inserted. Above all, the work opens with an Allegro that has much in common with the form and style of the concerto-Allegro, and this sets the tone for the sonata as a whole. In the sonatas listed in the table from around 1730 , Bach took the further step of reducing the overall design to the threemovement, fast–slow–fast scheme characteristic of the Vivaldian concerto. It is significant that concerto style is for Bach associated with sonatas in which the keyboard instrument is emancipated from its traditional role as a continuo instrument and instead takes two of the three obbligato parts, the third being allocated to a melody instrument (violin, flute, or gamba). This permits a degree of virtuosity characteristic of concertante-style writing but not normally expected of the trio sonata. In the six Organ Sonatas Bach goes still further, allocating the entire three-part texture to the keyboard player alone. There is an obvious analogy with his composing entire concertos for solo harpsichord (BWV 971) or for two harpsichords (BWV 1061) in the early 1730s. The Organ Sonatas, completed around 1730 , have strong associations with the chamber music that Bach was writing around that time; and their overall design as a standard set of six sonatas recalls that of the Sonatas for violin and obbligato harpsichord of the early 1720s. Their distinctive character as organ music should not be overlooked, however. As trios ‘a` 2 Clav. et Ped.’ (for two manuals and pedal), they have precedents in the French organ music of Boyvin, Du Mage, de Grigny, and Raison, with which Bach is known to have been acquainted. And he had already made contributions to the Lutheran tradition of chorale trios during the Weimar years (BWV 655a and 664 a). It is hard to imagine a better exercise for the aspiring organist than the trio ‘a` 2 Clav. et Ped.’, with its demand for obbligato playing by right hand, left hand, and feet simultaneously. This surely explains why, according to J. N. Forkel, ‘Bach composed [the six Organ Sonatas] for his eldest son Wilhelm Friedemann, who by practising them had to prepare himself to become the great performer on the organ that he afterwards was’.29 The dual character of the sonatas as organ and chamber music is to some extent reflected in their prehistory. The outer movements of Sonata No. 1 might have been 28 Facsimile reprint (Hildesheim, 1970), pp. 675–83. For a full discussion of this hybrid genre, see Jeanne R. Swack, ‘On the Origins of the Sonate auf Concertenart’, Journal of the American Musicological Society, 46 (1993), pp. 369–414. 29 ‘Bach hat sie fu¨r seinen a¨ltesten Sohn, Wilh. Friedemann, aufgesetzt, welcher sich damit zu dem großen Orgelspieler vorbereiten mußte, der er nachher geworden ist’; BD VII, p. 80; Eng. trans. in NBR, pp. 417–82 (see 471–2).

sona tas in concerto style

263

adapted from a lost trio sonata in B♭ for recorder, oboe, and continuo;30 and Sonata No. 4 might have originated as a trio sonata for oboe, gamba, and continuo.31 In the late 1720s Bach expanded two of his organ preludes and fugues by incorporating a trio movement in central place: the Largo from Sonata No. 5 became the central slow movement of the Praeludium et Fuga in C, BWV 545, and the finale of Sonata No. 4 became the centrepiece of the Praeludium et Fuga in G, BWV 541.32 In addition, certain movements from the set of sonatas are found in the context of cantatas or concertos. The opening Adagio–Vivace sequence from Sonata No. 4 became the Sinfonia to Part II of the cantata Die Himmel erza¨hlen die Ehre Gottes, BWV 76, in 1723—in its supposedly original chamber-music version for oboe d’amore, viola da gamba, and continuo. And the ‘Adagio e dolce’ from Sonata No. 3—in an instrumental version for flute, violin, and harpsichord—became the slow movement of the Concerto in A minor, BWV 1044, probably in the 1740s. It is clear, however, that Bach was already working on the organ-trio movements before the set of six sonatas materialized: the Andante from Sonata No. 4 exists in two early versions as an independent organ trio; and Bach’s pupil J. C. Vogler copied out the opening Andante of Sonata No. 3 as a self-contained organ trio, probably in December 1729. 33 The three sonatas that evidently existed in some form before their inclusion in the set, Nos. 1 , 3 , and 4 , are essentially composed according to sonata rather than concerto principles. All three employ fugue, canon, and long-range interchange of parts to establish complete equality between the two manuals, to which the pedals are subordinate, having a supporting bass part that is only intermittently thematic. All three movements of Sonata No. 1 in E♭ are freely fugal in a manner that resembles the three-part Sinfonias (BWV 787–801) of 1722/3. The outer movements are built on instrumental themes of great vitality, whereas the central Adagio is a siciliana in all but name, and its theme is supported by the same bass as the siciliana from Harpsichord Concerto No. 2 in E (BWV 1053). The exposition of the first movement alternates with a substantial episode (bb. 11–21 and 36–50) which, however, offers no concerto-style contrast, since it is built on the first bar of the subject (albeit combined with a new countersubject). The Adagio slow movement and Allegro-finale are both cast in binary dance form, though in the latter case without dance rhythm; and in both movements the theme is inverted after the double bar. Free inversion of the siciliana theme robs it of all its expressive power, whereas the strict inversion of both subject and countersubject in the finale is completely successful, inaugurating a second strain

30 According to Klaus Hofmann, ‘Ein verschollenes Kammermusikwerk Johann Sebastian Bachs: Zur Fassungsgeschichte der Orgelsonate Es-Dur (BWV 525) und der Sonate A-Dur fu¨r Flo¨te und Cembalo (BWV 1032)’, BJ 85 (1999), pp. 67–79. 31 According to Pieter Dirksen, ‘Ein verschollenes Weimarer Kammermusikwerk Johann Sebastian Bachs? Zur Vorgeschichte der Sonate e-Moll fu¨r Orgel (BWV 528)’, BJ 89 (2003), pp. 7–36. 32 See Dietrich Kilian, ‘Dreisa¨tzige Fassungen Bachscher Orgelwerke’, in M. Geck (ed.), Bach-Interpretationen (Go¨ttingen, 1969), pp. 12–21, and the same author’s Krit. Bericht, NBA IV/7 (1988), pp. 84–6. 33 Kilian, Krit. Bericht, NBA IV/7 , pp. 80–1 and 74–6.

264

the harps ichord concertos etc.

that is in every detail a close variant of the first, preserving its phrase structure and order of events. In this first sonata, the pedals are often allotted simplified or shortened versions of the themes. In Sonata No. 3 in D minor, on the other hand, the pedals are given a purely supporting, athematic role, which perhaps suggests a somewhat earlier conception of the organ trio. The thematic material, on the other hand, is fashionably galant, with its triplet figures and double appoggiaturas, reminiscent of the Sarabande from the A minor Partita (BWV 827) of 1725 (Ex. 3). Both outer movements are pseudo-fugal entities in da capo aria form (ABA), hence the impression of a fresh start with new material at the beginning of paragraph B (Andante, b. 50; Vivace, b. 37). No true concertante effect is created, however, and the Andante tempo of the first movement already distances it from the concerto-Allegro. The ‘Adagio e dolce’, a dance-style movement in 6/8 and in rounded binary form, was later adapted to form a concerto slow movement (BWV 1044 no. 2), as we have seen, but its opening gambit is perfectly attuned to the style of the Andante first movement, suggesting that they might have belonged together ab initio. Interestingly, the galant, homophonic antecedent is answered by a consequent in strict canonic imitation—Bach clearly had no compunction in uniting the two styles.

Ex. 3

a) 1st movement of Organ Sonata No. 3 in D minor, BWV 527, bb. 21–3

b) 2nd movement of the same sonata, bb. 1–2

c) Sarabande from Partita in A minor, BWV 827, bb. 1–2

sona tas in concerto style

265

The Adagio–Vivace sequence that opens Sonata No. 4 in E minor recalls not so much the instrumental concerto as the cantata sinfonia. It was, of course, used as such in 1723 (BWV 76 no. 8), a version scored for oboe d’amore, viola da gamba, and continuo. Moreover, there is a precedent among Bach’s Weimar cantatas—the Sinfonia to Tritt auf die Glaubensbahn, BWV 152 (1714), in which a small chamber group (recorder, oboe, viola d’amore, viola da gamba, and organ) play a highly decorated four-bar Adagio (not so called, but to be assumed) followed by an Allegro fugue. In the organ sonata, too, a decorated four-bar Adagio leads to a pseudo-fugal fast movement (here Vivace). But both here and in the following Andante, the subject is answered at the octave or unison rather than the 5th. In other respects the two movements are designed on fugal principles: both comprise three expositions (tonic, dominant, and subdominant), each of which contains two subject entries, one for each manual. These identical features of design surely indicate that the two movements originated together. The main difference is that the Andante—which, incidentally, resembles the two B minor Adagios that formed the fourth movement of different versions of the Sonata in G, BWV 1019—introduces a sharply contrasting episode with pseudo-cross-string figuration (bb. 11 and 28) in alternation with the expositions. The finale, unlike the two preceding movements, is a true fugue with answers at the 5th. Furthermore, whereas the subjects of the Vivace and Andante are restricted to the manual parts, that of the finale occurs in the pedals (bb. 21 and 80), giving rise to a three-part fugue. In the finely wrought subject Bach unites old and new—the implied chromatic-4th descent of long-standing tradition and the somewhat modish triplet figures and trilled appoggiaturas. The three sonatas Nos. 2 , 5 , and 6 might have been composed with inclusion in the set of six in mind, for there is no evidence that they existed beforehand. They are a great deal closer to the form and style of the concerto than the three sonatas considered so far. The major difference concerns the character of the opening movement, which is no longer fugal but concertante in style and ritornello-based in form. Thus the opening Vivace of Sonata No. 2 in C minor begins with a striking homophonic theme built on plain tonic and dominant harmony. Fugal and canonic textures, so well adapted to the organ trio, occur in the contrasting episodes (bb. 8b, 22b, 38b, and 46b). The opening Allegro of Sonata No. 5 in C is a ritornello structure, incorporated within an overall ABA da capo scheme, as so often in Bach’s fast concerto movements. The initial motto theme is like a brief solo answered by tutti. After the tonic cadence that closes paragraph A (b. 50), a fresh start is made with a fugato on a new Spielfuge subject, and shortly afterwards (bb. 68–71) a dialogue takes place between the A and B themes. In the opening Vivace of Sonata No. 6 in G, written in a fashionable 2/4 metre, the impression of a tutti is created by the unison of the two manual parts. Then, after the tonic cadence that closes the ritornello, the episodic theme (b. 21) creates the effect of an accompanied solo. A further concertante feature is the perfidia-like repetition of a broken-chordal figure for 16 bars (bb. 37, 85, and 137).

266

the harpsichord concert os etc.

On the whole, the slow movements differ less than the opening ones from their equivalents in the (presumably) earlier sonatas. The beautiful theme of the Largo from Sonata No. 2 might have graced a concerto slow movement, but strangely, after its initial tonic statement and dominant counterstatement, it nowhere returns. The Largo of Sonata No. 5 is pseudo-fugal in construction, like its equivalent in the first sonata, but the overall form is no longer binary but rather da capo, as in the first movement. After the exposition of the extremely florid subject in paragraph A, pseudo-cross-string figuration creates the effect of a concerto episode. The Lente of Sonata No. 6 is a siciliana, as in No. 1 , but it is also related to the other compoundtime slow movements, those of Nos. 3 and 5—all three have in common their extremely intricate, delicate embellishment. Both sicilianas are pseudo-fugal within binary dance form, in this case rounded—an exact subdominant reprise starts at b. 25. It is interesting to note that the harmonic framework of the theme, with its implied chromatic-4th descent, is virtually identical with that of the finale of Sonata No. 4 (both movements are in the key of E minor). The finales of Sonatas Nos. 2 , 5 , and 6 are fugues, like those of Nos. 1 , 3 , and 4 , but certain aspects of them are related to the concertante style of their first movements. In the Allegro-finale of Sonata No. 2 , a double exposition of the alla breve subject stands in lieu of ritornello. A second subject, always treated independently, then enters as a new departure after a tonic full-close (b. 59). This somewhat modish charactersubject, with its kinetic recurrence34 and Scarlattian freakishness, is responsible for the extreme contrast, also found in many a concerto episode at this point. The equation ‘fugal exposition = ritornello; fugal episode = concerto episode’, so often part of Bach’s thinking, is no less clear in the Allegro-finale of Sonata No. 5. Here, the initial exposition (bb. 1–28) may be construed as the ritornello, after which a fresh start is made in the episode at bar 29, with its trill figure in three-part imitation. Later, the headmotive of the fugue subject, treated in bass sequence, is combined with exchanges of the trill theme in the manual parts. The similarly fugal Allegro-finale of Sonata No. 6 is, like many Bach concerto movements, cast in ABA1 reprise form, and the concerto-style contrast between the middle paragraph and its surroundings is as strong as in his instrumental da capo movements. After a tonic full-close, marking the end of paragraph A (b. 18), a new departure is made with the fugal exposition of a playful new subject, albeit with links to the old, in the relative minor. In his Sonaten auf Concertenart for viola da gamba (BWV 1029) or transverse flute (BWV 1030 and 1032) and obbligato harpsichord, Bach adopts not only the threemovement, fast–slow–fast design of the concerto but also many details of its form and style, as in Organ Sonatas Nos. 2 , 5 , and 6. Since there is reason to believe that these are the latest of the six Organ Sonatas, the three chamber works might have been composed afterwards, perhaps around 1730. Certainly there is no evidence that Bach

A coinage of Arthur Hutchings’s; see his The Baroque Concerto (London, 1959; 3rd edn, 1973), pp. 43–4.

sonata s i n c oncerto s tyle

267

cultivated the ‘sonata in concerto style’ (apart from an isolated, tentative move in that direction in BWV 1019) before the organ sonatas. It is possible, then, that all his sonatas of this type belong to the Collegium musicum period (1729–41), whether or not they were performed at Collegium concerts. For the gamba sonata no original sources survive to confirm or refute this dating, but the autograph scores of the two flute sonatas both date from about 1736/7. All three sonatas might have been preceded by older versions, but this can be verified only in the case of the B minor Flute Sonata: a harpsichord part has survived that belongs to an earlier version in G minor, BWV 1030a.35 Of all the movements from these three sonatas, the opening Vivace of the Viola da gamba Sonata in G minor most closely resembles a concerto-Allegro, both in style and form. The headmotive has much in common with that of the first movement from the third Brandenburg Concerto and also with the first solo theme from the finale of the D minor Harpsichord Concerto, BWV 1052 (see Part I, Ch. 3 , Ex. 1b). Moreover, the cadential theme from the ritornello (bb. 7–8) recalls the secondary theme (bb. 9ff.) from another ‘concerto-Allegro’ of chamber-music proportions, the first movement of the Sonata in G for violin and obbligato harpsichord, BWV 1019 (Ex. 4). There the opening theme receives a dominant counterstatement with interchanged parts, ensuring that both instruments are immediately brought into prominence. Here the entire tonic ritornello is stated twice, first by the gamba and then (b. 11) by the harpsichord (with a new gamba counter-melody). Two large internal periods (bb. 35–52 and 64–81) begin with ritornello material but also include episodic passages, in accordance with the older concerto-Allegro design of Torelli and Albinoni. Also characteristic of Albinoni, and often borrowed by Bach, is the statement of the ritornello headmotive three or four times in succession, each time in a different key (bb. 35 , 64 , and 95). In the first two cases, due to this repetition, the ritornello headmotive brings about almost all the modulation within the period concerned, whereas the intervening episode (bb. 53–63) is, after its first two bars, tonally stable. Thus the later concept of tonally stable ritornellos and modulating episodes is not altogether relevant here. Four tonic or dominant returns of the headmotive at crucial stages in the discourse (bb. 9 , 35 , 73 , and 95) are given special emphasis by means of partial of complete unisono treatment, imitating the effect of a tutti in a large instrumental ensemble. The Adagio, cast in binary dance form with repeats, is without doubt one of Bach’s most remarkable instrumental slow movements. Throughout the first strain a French sarabande (harpsichord, right hand) is combined in counterpoint with a highly embellished Italian Adagio (gamba) over an ostinato harpsichord bass. The

35 Berlin P 1008 (c. 1770/80) in the hand of Johann Friedrich Hering, who belonged to the Berlin circle of C. P. E. Bach. According to Klaus Hofmann, this version was probably adapted from a lost original—a trio for lute, violin, and string bass that might date from c. 1727 to 1732. See K. Hofmann, ‘Auf der Suche nach der verlorenen Urfassung: Diskurs zur Vorgeschichte der Sonate in h-Moll fu¨r Querﬂo¨te und obligates Cembalo von J. S. Bach’, BJ 84 (1998), pp. 31–59.

268 t h e ha r p s i c h o r d c o n c e r t o s e t c .

Ex. 4

a) 1st movement of Sonata in G minor, BWV 1029, bb. 7–8 (gamba only)

b) 1st movement of Sonata in G, BWV 1019, bb. 9–10 (harpsichord RH only) instrumental roles are then reversed for four bars (bb. 13–16), after which gamba and harpsichord share both sarabande and Adagio material for the remainder of the movement. In the Allegro-finale, a concertante fugue in ritornello form, an extreme contrast is set up between the fugal ritornellos and the long intervening episodes, in which a lovely cantabile melody (labelled as such) is accompanied by an ostinato figure and supporting bass (bb. 19 , 37 , 69 , 91 , 104). On each occasion the episode is repeated with interchanged upper parts. Close parallels may be found in the central episode of both outer movements from the fifth Brandenburg Concerto (Ex. 5) and also at the equivalent point in the second movement of the Sonata in E for violin and obbligato harpsichord, BWV 1016 (b. 63). Such resemblances do not necessarily have chronological significance, for the Brandenburg Concertos and the Sonatas for violin and obbligato harpsichord are very likely to have been current during the Collegium musicum years and Bach could very easily have picked up ideas from them when working on new compositions in the 1730s. The Sonata in B minor for flute and obbligato harpsichord, BWV 1030, is without question the greatest of all Bach’s flute compositions. This is in no small measure due to its expansive ritornello form, as well as its extremely rich thematic material and the contrapuntal elaboration with which it is treated in the opening Andante. The initial theme, with its syncopation and kinetic recurrence, is not dissimilar to concerto themes by Vivaldi and others. It is particularly close, however, to contemporaneous ouverture themes by Bach and his second cousin Johann Bernhard (see Ex. 1). This

Ex. 5

a) Finale of Sonata in G minor, BWV 1029, bb. 19b-22 (gamba only)

b) Finale of Brandenburg Concerto No. 5 in D, BWV 1050, bb. 79–85 (ﬂute only)

sonata s in c oncerto s tyle

269

theme and the other material of the opening ritornello (bb. 1–20) is not merely restated in subsequent ritornellos but varied or developed (bb. 35 , 40 , 44 , 53 , 75 , 109), a procedure that often involves canonic imitation. Moreover, the middle ritornello (b. 33) is expanded by incorporating the chief episodic phrase (from b. 21), which is also now subjected to canonic imitation (b. 53). Conversely, the following episode (b. 59) incorporates two brief ritornello returns (opening theme only) in keys VI and iv (G and e). The concluding ritornello (b. 80) follows exactly the same course as the middle one for seven phrases (bb. 80–101 = 33–54), but then a substantial interpolation of episodic material takes place (bb. 102–8) before ritornello phrases bring the movement to a close. The ‘Largo e dolce’ is a siciliana (and was so called in the early version) in binary dance form with repeats. The flute here has the florid solo part to itself, while the harpsichord has a fully notated continuo accompaniment. The rather modish syncopated, repeated-note figure in the flute part (bb. 5 and 13) is extracted from the keyboard accompaniment (bb. 2 and 4). The bipartite finale presents two metrically contrasting treatments of essentially the same material: a Presto fugue in alla breve time, ending in a tonic half-close, gives way to a 12/16 gigue in binary dance form with repeats. The gigue is based on a subtle thematic variant of the fugue subject (Ex. 6). The Sonata in A, BWV 1032, for the same combination of instruments,36 is neither as distinguished in thematic material nor as satisfying in form as its B minor counterpart. The latter consideration might explain why Bach evidently excised a substantial portion of the opening Vivace,37 presumably replacing it with a fresh working-out of the material on an inserted sheet of paper that is now lost. The harpsichord ritornello itself is satisfactory enough—an eight-bar period clearly articulated into Vordersatz (bb. 1–2), Fortspinnung (bb. 3–6), and Epilog (bb. 7–8). Not only this

Ex. 6

Finale of Sonata in B minor, BWV 1030: a) Theme of Part I (bb. 1–4, ﬂute only) b) Theme of Part II (bb. 84–5, ﬂute only)

36 In its ﬁnal form, but according to Michael Marissen it might be based on a lost trio in C for recorder, violin, and continuo; see his ‘A Trio in C major for Recorder, Violin and Continuo by J. S. Bach?’, Early Music, 13 (1985), pp. 384–90 , and ‘A Critical Reappraisal of J. S. Bach’s A major Flute Sonata’, Journal of Musicology, 6 (1988), pp. 367–86. 37 See the facsimile edition, ed. H.-J. Schulze (for full details see n. 19).

270 t h e ha r p s i c h o r d c o n c e r t o s e t c . three-phase structure but also the kinetic recurrence in the Vordersatz belongs to the idioms of the concerto. The flute then enters with a two-bar episodic theme, but what follows is a mere alternation of episodic and ritornello themes—Bach seems unable to free himself from either. Moreover, it seems perverse that the expected dominant modulation (b. 15) is followed by an immediate return to the tonic. The next modulation (circle of 5ths, E–b–f♯) is carried out by repeating the opening ritornello theme in different keys, a procedure that works well in the G minor Gamba Sonata but palls here due to the overuse of the theme beforehand. On arrival at the relative minor f♯, Bach uses only its opening figure in the bass (b. 46) to introduce an entirely new development. Despite the motivic link with the ritornello (compare bb. 3 and 49), however, its two- and three-part canonic imitation sounds somewhat arid, being insufficiently motivated by what has preceded it. The slow movements of the A major and B minor flute sonatas are alike in metre and tempo mark—6/8 and ‘Largo e dolce’—which may be connected with the preparation of their autograph manuscripts around the same time (about 1736/7). However, the highly unusual key relationship between the inner and outer movements of the A major Sonata—tonic major/minor (A–a–A)—suggests that the slow movement might have been imported from elsewhere.38 It is cast in a simple bipartite scheme (AA1), with much play on the contrast between the slurred-semiquaver and abrupt-quaver figures of its first two bars. The sonata closes with a concertante fugue in the 3/8 dance rhythm of the passepied. As usual the opening fugal exposition, including its internal episodes, constitutes the ritornello. What then happens recalls the fugal Allegros from the Sonatas for violin and obbligato harpsichord: a tonic fullclose at the end of the ritornello is followed by a fresh start in a new key (the relative minor f♯) and with new material. The threefold imitation here at the unison and octave provides relatively little contrast, however, since its descending scale figure recalls the last bar of the subject (b. 8) and the cadential phrase derived from it (bb. 16– 22). The true contrast comes with the second episode (b. 118), whose theme moves in regular quavers, includes diminished intervals, and is accompanied by long trills. This episodic theme then alternates with the original fugue subject (b. 174) in order to highlight the contrast between them before the concluding ritornello (b. 209).

38 Klaus Hofmann’s theory is that it originally formed the central slow movement of a trio in B♭, alongside the outer movements of Organ Sonata No. 1 , BWV 525. See his ‘Ein verschollenes Kammermusikwerk J. S. Bachs’, pp. 67–79.

II.4 Sacred and secular: vocal works II

Secular cantatas Title, occasion

Earliest source/s

Scribe, date

Der Streit zwischen Phoebus und Pan, BWV 201 Hercules auf dem Scheidewege, BWV 213, birthday of Saxon prince To¨net, ihr Pauken, BWV 214, birthday of queen Preise dein Glu¨cke, BWV 215, anniversary of coronation Schleicht, spielende Wellen, BWV 206, birthday of Augustus III

Berlin, P 175, St 33a

Autograph, part-autograph, 1729

Berlin, P 125, St 65

Schweigt stille, plaudert nicht, BWV 211 [Coffee Cantata] Non sa che sia dolore, BWV 209 Angenehmes Wiederau, BWV 30a, homage cantata

Berlin, P 141, Vienna ¨ NB, SA.67.B.32 O Berlin, P 135 Berlin, P 43/1, St 31

Autograph, part-autograph, for 5 Sept. 1733 Autograph, part-autograph, for 8 Dec. 1733 Autograph, part-autograph, for 5 Oct. 1734 Autograph, part-autograph; orig. for 7 Oct. 1734, perf. 7 Oct. 1736 Autograph, part-autograph, 1734

Berlin, P 41/1, St 91 Berlin, P 139, St 77 Berlin, P 42, St 80

J. N. Forkel et al.; orig. 1734? Autograph, part-autograph, for 28 Sept. 1737

Bach had cultivated the dramma per musica—equivalent to a single-act opera—in Leipzig in the mid-1720s (BWV 249a, 249b, 205, 207, and 193 a), those works being written for speciﬁc events connected with the Elector of Saxony, the Duke of Weißenfels, the Governor of Leipzig, or Leipzig University. In 1729, having taken over the directorship of the Collegium musicum, Bach was able to compose a dramma per musica according to his own agenda. The result was Der Streit zwischen Phoebus und Pan (The Dispute between Phoebus and Pan), BWV 201, which might have been performed on the occasion of Bach’s debut as director in the spring of 1729. Its text concerns the subject closest to Bach’s heart, namely the art of music itself. Picander’s libretto is based on Ovid’s account (Metamorphoses, Book XI) of the Greek myth that tells of the musical contest between Phoebus and Pan. In his new role as music director Bach here seized the opportunity to present to the Leipzig public his credo as a creative artist. His views are exhibited forcibly in the contrast between the arias of the two main protagonists. Phoebus’ aria, no. 5, represents the mature style of Bach himself, and in order to present it in the best possible light he lavishes all his resources

272 sacred a nd s ecula r: vocal works ii upon its composition. It is extraordinarily rich and dense, not only in texture and instrumentation—transverse flute, oboe d’amore, muted strings, and continuo—but also in theme, motive, and embellishment, bearing out the view that Bach’s music is at its most expressive when it is most florid.1 The melodic shape of the headmotive, with its rising 6th from the dominant, is amongst Bach’s most characteristic, perhaps deliberately imparting a highly personal stamp to the composition. He had employed it first (among his extant works) in the opening movement of Sonata No. 4 in C minor for violin and obbligato harpsichord, BWV 1017 (c. 1722), which shares with Phoebus’ aria its Largo tempo and subdominant sequential repeat, though not its siciliana rhythm. Other examples of this characteristic melodic opening are to be found in three cantatas from the Leipzig Cycle III (1726–7), ‘Bleibt, ihr Engel’ (BWV 19 no. 5), ‘Stirb in mir’ (BWV 169 no. 5), and ‘Ich habe genung’ (BWV 82 no. 1), as well as in ‘Erbarme dich’ from the St Matthew Passion (no. 39, 1727). Phoebus’ aria thus belongs in exalted company: like the other arias quoted, it is not only highly personal but decorative, refined, and deeply felt (see Ex. 1, which also includes examples not cited here). Pan’s aria, no. 7, on the other hand, though popular and immediately appealing in its melodic style, is relatively coarse with its plain, simple texture and scoring (unison violins and continuo) and its straightforward theme and lack of ornament. There is also a strong comic element in it: Pan’s style is ridiculed in the shaking figure of the ritornello (bb. 5–8) and in the repeated notes for the word ‘wackelt’ (‘shakes’; bb. 41–4). In the middle paragraph, Pan in turn ridicules Phoebus’ style. His objection that ‘if the note sounds too laboured and the mouth sings with restraint, it arouses no mirth’ is set in alla breve, minor-mode counterpoint with chromatic lines and swift modulation. In the interests of a balanced view, Bach appears to be allowing certain aspects of his own style to be exposed to ridicule. For the same reason, perhaps, relatively coarse arias in Pan’s style and refined arias in Phoebus’ style are equal in number and alternate throughout the drama. Tmolus’ F♯ minor aria with solo oboe d’amore, no. 9, represents the intricate, florid, serious style of Phoebus, as does Mercury’s E minor aria with two transverse flutes, no. 13. On the other hand, the very first aria in the score, Momus’s continuo aria in G, no. 3, mimics Pan’s superficial style, as does Midas’ aria, no. 11, in which he declares that ‘to my two ears, Pan’s singing is incomparably fine’. In the boldest satirical stroke of the whole drama, this statement is accompanied by the braying of an ass in unison violins, anticipating Midas’ reward of ass’s ears in the following recitative. Before his aria, however, Midas had followed Pan in raising popular objections to the refined style of Phoebus, who ‘makes it far too florid’, whereas Pan’s ‘most lovely mouth sang easily and unforced’. Bach was clearly secure enough in his own style to let these objections (penned by Picander) stand. Moreover, in the recitatives nos. 2 and 8, Pan’s pipes and

A characteristically perceptive remark of Donald Francis Tovey’s.

se c ular ca ntat as

273

Ex. 1 Largo

a) Theme of 1st movement from Sonata in C minor, BWV 1017 (c. 1722)

b) Theme of 2nd movement from Liebster Gott, wenn werd ich sterben, BWV 8 (1724) Adagio

c) Theme of 5th movement from Es erhub sich ein Streit, BWV 19 (1726)

d) Theme of 5th movement from Gott soll allein mein Herze haben, BWV 169 (1726)

e) Theme of 3rd movement from Ich armer Mensch, BWV 55 (1726)

f) Theme of 1st movement from Ich habe genung, BWV 82 (1727)

g) Theme of ‘Erbarme dich’ from St. Matthew Passion, No. 39 (1727) Largo

h) Theme of Phoebus’s aria from ‘Phoebus and Pan’, BWV 201, no. 5 (1729) Lente

i) Theme of Lente from Organ Sonata No. 6 in G, BWV 530 (c. 1730) Adagio

j) Theme of 3rd movement from Wachet auf, BWV 140 (1731)

274 sa cred and secular: vocal works ii

k) Theme of 5th movement from To¨net, ihr Pauken, BWV 214 (1733); subsequently parodied in Christmas Oratorio, No. 15 (1734)

l) Theme of 3rd movement from Non sa che sia dolore, BWV 209 (1734?) Siciliano

m) Solo theme of 2nd movement from Harpsichord Concerto No. 2 in E, BWV 1053 (c. 1738)

singing are not altogether rejected: ‘Pan sings for the forest; the nymphs he might well delight’. By implication, however, Phoebus aims ‘to please the gods’. We might take the forest and nymphs as referring to hoi polloi, and the gods to connoisseurs. Both popular and sophisticated styles are valid and have their own audiences. Bach’s satire is directed, above all, at those who would elevate the light style to the level of the serious style, or even above it (see nos. 3, 13, and 14). One thing is clear: there is no simple correlation between popular and modern styles, on the one hand, or between refined and older styles, on the other. This is clear from the concluding chorus which celebrates the cause of fine music in the most ‘modern’ style of the whole work, namely a homophonic, dance-like 2/4 with much use of syncopation—a manner much cultivated by Bach in the late 1720s and 1730s.2 In 1727 Bach had performed two cantatas for the Elector of Saxony (BWV Anh. I 9 and 193a, both now lost). But after he assumed the directorship of the Collegium musicum in 1729, works composed and performed in honour of the Elector and his family greatly increased in frequency. Nine such works are known from the period 1729–39, of which four are no longer extant (BWV Anh. I 11, 12, 13, and BWV 205a). The five compositions that survive, BWV 213, 214, 215, 206, and 207 a, all belong to the dramma per musica type and were newly written for the celebrations concerned, with the exception of Auf, schmetternde To¨ne, BWV 207a, which was parodied from the homage cantata No. 207 of 1726. The ‘Hercules’ and ‘Queen’s’ Cantatas, BWV 213 and 214, were composed in the autumn and winter of 1733 for the birthday of the Prince and Electress of Saxony 2 For example, in the Aria from Partita No. 4 in D, BWV 828 (1728) and in BWV 214 no. 7 (1733), parodied in the Christmas Oratorio, BWV 248, Part I, no. 8 (1734). Other examples are given in Doris FinkeHecklinger, Tanzcharaktere in Johann Sebastian Bachs Vokalmusik (Trossingen, 1970), pp. 142–3.

se c ular ca ntat as

275

respectively. Almost all of their choruses and arias were reused to new, sacred words in the Christmas Oratorio during the following winter (1734–5). In terms of style these two fine music dramas seem to continue where ‘Phoebus and Pan’ left off, as it were. The earlier of the two, Hercules auf dem Scheidewege (Hercules at the Crossroads), more obviously embraces the conventions of Baroque opera. Its arias include many of the standard types: a lullaby, no. 3, one of Bach’s finest (later to become ‘Schlafe, mein Liebster’ in the Christmas Oratorio, Part II); an echo aria, no. 5 (adapted from the lost BWV Anh. I 11, no. 7, of 1732); a ‘decision’ aria, no. 9, in which the hero forcefully rejects one course of action in favour of another (the violins are marked ‘unisoni e staccato’, giving the piece a far more vigorous affect than in the later Christmas version, ‘Bereite dich, Zion’); and finally, a love duet, ‘Ich bin deine, du bist meine’, no. 11, notable for its highly florid, ornamental melodic lines. This movement, like the homophonic ‘Decree of the gods’, no. 1, with its modish, cut-off cadence figures, comes across as relatively modern by comparison with the fugal aria no. 7, whose subject nonetheless illustrates the hovering of wings. The concluding ‘Chorus of Muses’ is a gavotte en rondeau, adapted from a Co¨then serenata (BWV 184a) that had been composed over ten years earlier. The Queen’s Cantata owes its exceptional quality at least in part to Bach’s typically keen response to the aural imagery of the text. In the opening chorus, drums, trumpets, strings, and later voices enter in turn in accordance with the anonymous text, a congruence lost in the later Christmas version (Part I no. 1). Not only the opening ritornello but the whole movement is exceptionally colourful in instrumentation and dazzling in its rhythmic diversity. The A major aria, no. 3 (not included in the Christmas Oratorio) is no less vivid in its response to the aural imagery of the text. The soprano, taking the part of Bellona, sings fervently, ‘Blow the well-bored flutes’ and ‘Ring out with exultant singing’ to the accompaniment of two traversi and pizzicato bass. The B minor aria, no. 5, is addressed by Pallas (alto) to the ‘devout Muses’, hence the highly personal form of the theme, a reminiscence of Phoebus’ aria (BWV 201 no. 5), not only in intervallic shape but in its on-thebeat start and its 3/8 metre without specific dance reference (cf. Ex. 1h and k). Within this aria the second and third solos describe ‘joy’ and ‘rejoicing’ in operatic coloratura, with extremely rapid ornamental notes and lengthy melismas. The D major aria, no. 7, for bass (Fama), trumpet obbligato, strings, and continuo is cast in a ‘modern’, dance-like 2/4 with syncopated rhythms, recalling the finale of ‘Phoebus and Pan’. The Queen’s Cantata concludes with an amalgam of binarydance, ritornello, and rondeau forms. Each of the four characters, Irene, Bellona, Pallas, and Fama, enters in turn, bringing his/her own contribution to the text and hence to the music. This explains why each voice enters with different music, rather than with imitation of the same theme—a procedure that has no textual justification in the Christmas version (Part III, opening chorus). The ‘Polish’ and ‘Rivers’ Cantatas, BWV 215 and 206, were both written for performance in October 1734 (the latter was postponed, however, till 1736), hence

276 sacred and secular: vocal works ii the occasional similarities between them. The Polish Cantata relies heavily on parody,3 having been assembled at the last minute. The opening double-choral movement, which later became the Osanna of the B minor Mass, was adapted from a lost cantata for the Elector of Saxony (BWV Anh. I 11, 1732). Its power is derived in large part from the unisono headmotive, which is later sung three times in three different keys a 3rd apart as an interjection from Choir II during the continuous singing of Choir I (bb. 59–69). At the return of the tonic key (b. 114) the entire ritornello returns with inbuilt vocal parts. The B-paragraph of this ABA da capo structure appears to have been composed specifically for BWV 215.4 As a whole the movement is a great chorus of praise and thanksgiving, differing only in scale from many such movements among Bach’s sacred cantatas. The equivalent movement of the Rivers Cantata, on the other hand, is a tone-picture of rivers flowing gently (piano) and then rushing swiftly (forte), a contrast built into the opening ritornello. Similar word-painting returns in the B minor aria no. 5, where the Elbe (tenor) sings of ‘every billow of my waves’, hence the flowing semiquaver figuration of the obbligato violin part. Exceptionally ‘modern’ in style are the tenor aria from Cantata 215 (no. 3) and the bass and alto arias from Cantata 206 (nos. 3 and 7). Indeed, the F♯ minor aria for the Danube (alto) from the Rivers Cantata (no. 7) sounds like a minor-mode adaptation of the tenor aria from the Polish Cantata (no. 3). In both cases the two-bar theme comes across as somewhat modish, with its syncopated rhythms and exact internal repeat. Only in Cantata 215 do we encounter a stock-in-trade operatic piece, the ‘rage’ aria no. 5, with its prestoconcitato semiquavers in the string parts. The last arias of the two cantatas are alike in their keen responsiveness to the text. The text of the soprano aria from Cantata 215 (no. 7) informs us that, rather than punishing his enemies, Augustus ‘repays malice with kindness’, a truly Christian quality, hence the absence of continuo and its replacement with a bassett for violins and violetta. This sacred element might explain why this aria alone was later selected from the cantata for inclusion in the Christmas Oratorio (Part V, no. 47). In the equivalent aria from Cantata 206 (soprano, no. 9), the exquisite scoring for a trio of transverse flutes with continuo is derived from the aural imagery of the text: ‘Do listen! The gentle choir of flutes gladdens the breast [and] delights the ear.’ The aria illustrates Bach’s serene G major mood, often evoked to characterize blessedness, peace, concord, or union (as in weddings). The finales of the two cantatas have much in common: both are dance movements in which binary, da capo, and rondeau forms are amalgamated.

3 See Werner Neumann, Krit. Bericht, NBA I/37 (1961), pp. 69–74, and Stephen A. Crist, ‘The Question of Parody in Bach’s Cantata “Preise dein Glu¨cke, gesegnetes Sachsen”, BWV 215’, in R. Stinson (ed.), Bach Perspectives [I] (Lincoln, Nebr., 1995), pp. 135–61. 4 According to Neumann and Crist (see n. 3).

secular cantatas

277

Secular cantatas for other occasions from the decade 1729–39 include a homage cantata for a lawyer, BWV 36b (c. 1735; transmitted in an incomplete form), parodied from BWV 36c of 1725; the lost BWV Anh. I 10 (1731) for the birthday of the Leipzig governor Count von Flemming; and three lost cantatas for Thomasschule celebrations in 1732 and 1734 (BWV Anh. I 18, 19, 210). The three cantatas that survive complete, BWV 211, 209, and 30a, all belong to different types. The Coffee Cantata, BWV 211, which was probably first performed in the summer of 1734, resembles a miniature comic opera. Its historical counterpart lies in the comic intermezzi—the forerunners of opera buffa—that were played between the acts of opera seria. Accordingly, the dramatis personae are no longer ancient Greek or Roman gods or allegorical characters, as in the dramma per musica, but rather representatives of Bach’s fellow Leipzig citizens. Much of the charm of Bach’s setting of Picander’s libretto lies in his vivid characterization of the strict, irascible father Schlendrian and his young coffee-loving daughter Liesgen. The dotted-rhythm motive, marked ‘con pompa’, in the continuo part of the opening secco recitative already characterizes Schlendrian as a pompous, self-important old fellow. And in the following aria, no. 2, he gives vent to the many annoyances—clearly illustrated by the ritornello theme—caused by his daughter’s disregard of his words. In his second aria, no. 6, the chromatic sequences of the ritornello, employed thereafter in quasi-ostinato fashion, probably stand not only for the obstinacy (as he sees it) of the girl but also for his frustration at her continued disobedience. In the B minor aria no. 4 Bach employs a light, delicate touch—staccato flute, continuo marked pizzicato and piano sempre—to characterize the girl Liesgen and her addiction to coffee. In her second aria, no. 8, in siciliana rhythm she shows herself capable of passionate feelings, aroused by thoughts of a potential lover and utterly eclipsing her devotion to coffee. The denouement, no. 10, brings the complete vocal-instrumental ensemble together in a bourreé, hence the binary dance form within an overall rondeau structure. The playful character of the music is perfectly in keeping with the cat and mouse—a metaphor for girls and coffee—of the text. The second of the three compositions under discussion is an Italian solo cantata Non sa che sia dolore, BWV 209. It is one of five extant Bach secular cantatas for solo voice, of which three have German words (BWV 202, 204, and 210) and two Italian (BWV 203 and 209). Certain sources of the text have been discovered, one of which dates from 1728/9, proving that the libretto cannot have originated before 1729.5 It has recently been suggested that the likeliest date for the composition and performance of the work is 1734, the same year as the Coffee Cantata.6 Both solo soprano and obbligato flute feature throughout the work. Indeed, so significant is the contribution of the flute that the sinfonia seems almost like the opening Allegro of a flute

5 The known sources of the text and their bearing on chronology are summarized by Alfred Du¨rr, The Cantatas of J. S. Bach, rev. Eng. trans. by Richard D. P. Jones (Oxford, 2005), pp. 923–4. 6 ¨ berlegungen zu der Kantate “Non sa che sia dolore” BWV 209’, BJ See Klaus Hofmann, ‘Alte und neue U 76 (1990), pp. 7–25.

278

sacred and secular: voca l works ii

concerto—Bach must have had a fine player at his disposal at the time.7 Characteristic of him in the sinfonia is not only the combination of ritornello and da capo form but the dominant role of the headmotive throughout the movement. In the E minor aria, no. 3, the rising-6th headmotive is a clear Bachian fingerprint (see Ex. 1), as is, once again, its dominant role thereafter. The dance-style finale no. 5 includes an unusual 7th-chord on the leading-note in its opening phrase (b. 3), as well as certain ‘modern’ Italianate features, such as the Lombard rhythm at the end of the ritornello (bb. 13–14), later taken up by obbligato flute. The third of the three secular cantatas from the 1730s, Angenehmes Wiederau, BWV 30a, returns to the dramma per musica type. As often in Bach’s works of this kind, the singers are cast as allegorical characters, Zeit (Time, soprano), Glu¨ck (Fortune, alto), Elster (the river of that name, tenor), and Schicksal (Destiny, bass). The work was written for performance in 1737 as a homage cantata for Johann Christian von Hennicke on taking possession of his fief at Wiederau, near Leipzig. In accordance with its late date, it is relatively ‘modern’ in style. Strong emphasis is placed throughout on the melodic line, on regularity of phrase structure, and on homophonic textures—counterpoint is, surprisingly for Bach, little in evidence. The opening chorus introduces fashionable syncopations into its dance-like 2/4 metre—one of Bach’s more progressive modes of rhythmic movement at this period. In accordance with its unspecified dance rhythm, it is cast in the binarywithin-rondeau form that Bach used elsewhere for finales, and indeed it returns as finale of the present work. Its modish syncopations return in the A major aria no. 5, a gavotte whose ritornello is appropriately cast in binary dance form with repeats. The texture of this ritornello is made up of pure melody and accompaniment, a foretaste of pre-Classical style. The scoring is exquisite: the melody is assigned to muted first violin/s in unison with transverse flute and is accompanied by pizzicato strings and continuo. All told, this remarkable aria finds Bach at his most bewitching. The B minor aria no. 7 returns to the dance-like 2/4 of the first movement, here in association with reverse dotting. The E minor aria no. 9 is the only movement written in an older style.8 The music of the last aria no. 11 had already been used in an earlier secular cantata (BWV 210a no. 8, 1729), but its dance rhythm, periodic phrasing, and treble-dominated texture render it perfectly suited to its context in the Wiederau cantata. During the 1730s Bach apparently no longer composed new cantata cycles, but only isolated cantatas for special occasions, such as weddings, council elections, or the Reformation Festival, or to fill gaps when no cantata was available for a specific Sunday in the church year. In some cases Bach produced new compositions by his customary method of parodying existing secular or sacred works.

7 One possibility is Lorenz Christoph Mizler, whose ﬂute playing is discussed at length by Joshua Rifkin, ‘The “B-minor Flute Suite” Deconstructed: New Light on Bach’s Ouverture BWV 1067’, in G. G. Butler (ed.), Bach Perspectives 6 (Urbana and Chicago, 2007), pp. 1–98 (see 50–3). 8 Cf., for example, BWV 69a no. 3 and BWV 7 no. 4 from Leipzig Cycles I and II.

sacred cantatas

279

Sacred cantatas Title, occasion

Earliest source/s

Scribe, date

Sei Lob und Ehr dem ho¨chsten Gut, BWV 117, wedding? Nun danket alle Gott, BWV 192, wedding? Jauchzet Gott in allen Landen, BWV 51, Trinity 15 Der Herr ist mein getreuer Hirt, BWV 112, Easter 2 Wir danken dir, Gott, BWV 29, council election Wachet auf, ruft uns die Stimme, BWV 140, Trinity 27 Schwingt freudig euch empor, BWV 36, Advent 1 Ich ruf zu dir, Herr Jesu Christ, BWV 177, Trinity 4 In allen meinen Taten, BWV 97, wedding?

Berlin, N.Mus.ms.34

Autograph, 1728/31

Berlin, St 71 Berlin, P 104, St 49

J. L. Krebs et al., 1730? Autograph, J. L. Krebs et al., for 17 Sept. 1730? Autograph, S. G. Heder et al., for 8 Apr. 1731 Autograph, part-autograph, for 27 Aug. 1731 J. L. Krebs et al., for 25 Nov. 1731 J. L. Krebs et al., for 2 Dec. 1731

New York PML, Leipzig TS Berlin, P 166, St 106 Leipzig TS Berlin, P 45/1, St 82

Berlin, P 116, Leipzig Autograph, part-autograph, for 6 July 1732 TS New York PL, Berlin, Autograph, F. C. S. MohrSt 64 heim, 1734 Was Gott tut, das ist wohlgetan, BWV 100, Berlin, P 159, St 97 Autograph, part-autograph, wedding? 1734/5 Wa¨r Gott nicht mit uns diese Zeit, BWV 14, Berlin, P 879, Leipzig Autograph, part-autograph, Epiphany 4 TS for 30 Jan. 1735 Ein feste Burg ist unser Gott, BWV 80, Berlin, P 177 J. C. Altnickol; perf. 1735? Reformation Festival Es ist das Heil uns kommen her, BWV 9, Washington LC, Autograph, Anon. Ve et al., Trinity 6 Leipzig TS for 17 July 1735 Gott ist unsre Zuversicht, BWV 197, wedding Berlin, P 91 Autograph, 1736/7 Freue dich, erlo¨ste Schar, BWV 30, Berlin, P 44/1, St 31 Autograph, part-autograph, St John’s Day c. 1738

The main content of Cantata No. 120, for example—two arias and a chorus—was used repeatedly in different contexts and with different words: in a wedding cantata of 1729 (BWV 120a), in a cantata for the bicentenary of the Augsburg Confession in 1730 (BWV 120b), and finally in a council election service around 1742 (BWV 120). The wedding cantata BWV 120a is a late example of Bach’s habit, formed in the cantatas of Cycle III, of adapting a sinfonia from an existing instrumental work: the Sinfonia to Part II is adapted from the introductory Preludio of the Partita in E for solo violin (BWV 1006), with the solo part played on obbligato organ and a new string accompaniment. Two years later, in 1731, Bach used the same sinfonia as an introduction to the council election cantata No. 29, giving it a more festive sound by adding parts for trumpets and drums. In some cases of parody or adaptation several stages were involved before the definitive version was achieved. The small Weimar Lenten cantata BWV 80a, for example, was converted into a large-scale cantata for the Reformation Festival in two stages: first, by introducing a plain four-part chorale at the beginning (BWV 80b, 1728/ 31), and then by replacing it with a massive chorale fantasia (BWV 80, c. 1735). Similarly, the secular homage cantata BWV 36c of 1725 was not long afterwards (1726/30) parodied

280

sacred and secular: vocal works ii

to form the Advent cantata Schwingt freudig euch empor, BWV 36, which adhered closely to its secular model. Only later, in 1731, was it radically remodelled into a much larger work with a regular alternation of arias and chorales. Two of the most prominent cantata types of Cycle III, the solo cantata and the dialogue between Jesus and the Soul, are each represented only once among the late cantatas, but the works concerned are among the most celebrated of all Bach’s sacred cantatas: Jauchzet Gott in allen Landen, BWV 51, and Wachet auf, ruft uns die Stimme, BWV 140. Jauchzet Gott is a genuine ‘cantata’ for solo soprano, obbligato trumpet, strings, and continuo. It was originally written to be performed ‘in ogni tempo’ (at any time), but the brilliant concertante trumpet part seems to point to a festive occasion. The solo soprano part exceeds Bach’s normal requirements in range and technique, which suggests that it might have been intended for a female coloratura soprano. This in turn raises the possibility of an original performance outside Leipzig, and the birthday of Duke Christian of Saxe-Weißenfels on 23 February 1729 has been identified as a possible occasion.9 Bach later allocated the work to the Fifteenth Sunday after Trinity, and it might have been revived on that occasion in 1730. The text is a jubilant song of praise and thanksgiving for God’s support, coupled with a prayer for his future faithfulness. The instrumentation is unique for Bach but common in the Italian solo cantata as cultivated by (among others) Alessandro Scarlatti. The first movement opens in concerto style with a triadic, unisono headmotive, but the ritornello is then dominated by a tremolo figure that Bach had used in Weimar (BWV 80a, 1715) as a ‘victory’ motive. In accordance with his late style, the da capo form of the movement is modified in the interests of economy and continuity: a modulatory link joins the middle paragraph with the return, which also lacks its opening ritornello. The continuo aria no. 3 has a sequential ritornello whose main figure then underpins the vocal solos as basso quasi ostinato, a technique Bach had employed since his earliest arias. The da capo form of the movement, however, exhibits the relatively ‘modern’ feature of a linked return, as in the first movement. In the following chorale arrangement, no. 4, the cantus firmus is sung in plain long notes by the soprano. The Bar form of the chorale is articulated by ritornellos constructed in two-part fugue, to be played by the trio-sonata combination of two violins and continuo. The finale, which follows without a break, is a brilliant Alleluia for the entire ensemble—again fugal, but with expositions for soprano and trumpet alternating with expositions for strings. Wachet auf, ruft uns die Stimme, BWV 140, is not a pure dialogue cantata but a hybrid. In its framing and middle movements (nos. 1, 4, and 7) it is a chorale cantata, based on Philipp Nicolai’s well-known hymn of 1599. The intervening portions, however (nos. 2–3

9 ¨ berlegungen zu By Klaus Hofmann, ‘J. S. Bachs Kantate “Jauchzet Gott in allen Landen” BWV 51: U Entstehung und urspru¨nglicher Bestimmung’, BJ 75 (1989), pp. 43–54. Uwe Wolf, on the other hand, has found no archival evidence in support of this hypothesis; see his ‘J. S. Bach und der Weißenfelser Hof: ¨ berlegungen anhand eines Quellenfundes’, BJ 83 (1997), pp. 145–50. U

sacred cantatas

281

and 5–6), take the form of a dialogue in freely invented verse (though with numerous biblical quotations) between Jesus as Bridegroom (bass) and the Faithful Soul as Bride (soprano). The dialogue is introduced by a quasi-Evangelist’s narrative, sung in Passion style by the tenor in secco recitative (no. 2). A soprano–bass duet (no. 3) then expresses the longing of the soul for Christ—the mystical counterpart to a woman’s longing for her lover. Its theme belongs to the very personal type that Bach had two years earlier employed in Phoebus’ aria from Cantata No. 201 (see Ex. 1h and j). After the celebrated central chorale arrangement, ‘Zion ho¨rt die Wa¨chter singen’ (no. 4), an accompagnato for bass as vox Christi, in which Jesus takes the Soul as His bride, introduces the second dialogue movement, no. 6, a pseudo-operatic love duet in which earthly means are used to express heavenly love—the union of Christ and the Soul. Biblical-text settings are rare in this later period by comparison with Cycles I and III. A notable exception is the psalm setting that directly follows the sinfonia in Cantata No. 29, Wir danken dir (1731). It is on a massive scale, in keeping with the splendid introductory sinfonia (an ensemble arrangement of BWV 1006 no. 1). Bach sets the words ‘Wir danken dir, Gott, wir danken dir und verku¨ndigen deine Wunder’ (‘We thank You, O God, we thank You and declare your wonders’) from Psalm 75: 1 as an immense alla breve stretto fugue in motet style—a worthy counterpart to the great antiquity and authority of the text. Two years later, in 1733, Bach was to adapt the music to a Latin thanksgiving text, the ‘Gratias agimus tibi’ from the Dresden Missa that would eventually form the Kyrie and Gloria of the B minor Mass. The cantata and Missa versions of the movement probably derive from a common, pre-1731 original.10 Choruses to non-biblical texts are relatively rare in the Leipzig cantatas (unlike those of the Weimar period) and seem to require special justification. The opening movement of the wedding cantata No. 197 could be set as a chorus because its first line, ‘Gott ist unsre Zuversicht’ (‘God is our assurance’), is an exact quotation from Psalm 46: 1. The remainder of the text, however, is freely invented and lends itself to da capo aria form, which Bach duly employs. Within the da capo ritornello structure, fugue is incorporated in accordance with the venerable psalm quotation, which is set to the headmotive of the fugue subject. Cantatas Nos. 30 and 36 were both parodied from secular cantatas, which explains why they both begin with a non-biblical chorus. In both cases, the librettist endeavoured to keep close to the original secular text, so that the source of Bach’s inspiration remains clear. The opening chorus of Cantata No. 36 retains the original first line, ‘Schwingt freudig euch empor’ (‘Soar up joyfully’), which explains the character of Bach’s setting, but it is now ‘cheerful voices in Zion’ that rise up to the stars, rather than ‘good wishes’ (on the birthday of a teacher). In Cantata No. 30, the trumpet choir of the festive original (BWV 30a, 1737) was felt to be unsuited to St John’s Day and consequently omitted. The key words ‘rejoice’, ‘prosperity’, and ‘foundation’ are common to both works, but in the secular original they concern ‘pleasant Wiederau’; in the sacred parody, the ‘redeemed host’. 10

According to Joshua Rifkin, notes to his recorded performance of the B minor Mass, and Christine Fro¨de, Krit. Bericht, NBA I/32.2 (1994), pp. 41–2 and 51–5.

282

sacred and secular: vocal works ii

Seven chorale cantatas survive from the 1730s—Nos. 117, 192, 112, 177, 9, 97, and 100—of which all but one (No. 9) belong to the per omnes versus type in which the hymn text is preserved unaltered throughout. Bach had employed this venerable type in the early Easter cantata No. 4 and in three cantatas of the mid-1720s, Nos. 107 (Cycle II, 1724), 137 (Cycle III, 1725), and 129 (post-Cycle III, 1727). Only in Cantata No. 4, however, were the internal hymn verses set as chorale arrangements; in the three later cantatas of this type they were set in up-to-date fashion as arias and recitatives, and this also applies to the chorale cantatas of the 1730s. In view of the inherent unsuitability of the hymn texts for operatic forms, however, da capo form is absent among these late works, reprise form occurs only once (BWV 97 no. 7), and the characteristic Bar form of the chorale is found frequently—whether or not the chorale melody is quoted—alongside bipartite and motet-derived forms. The opening chorale-chorus is invariably composed along the lines of its equivalents in Cycle II. The finale, however, is not necessarily Bach’s usual plain four-part chorale. The finale of Cantata No. 192, for example, is cast in the Bar form of the chorale, which is sung by the soprano in plain long notes, accompanied by florid lower vocal parts, a type of setting normally reserved for the introductory choralechorus. The gigue rhythm of the framing ritornellos and its associated melody recall the finale of Bach’s Ouverture in D (BWV 1068), which was performed around the same time (early 1731). Two of the later chorale cantatas, Nos. 177 and 100, lack recitatives altogether, underlining the gulf between the strict chorale cantata and the pseudo-operatic type that Bach otherwise cultivated. In addition, the melodic invention in the inner movements is not always entirely free, but instead is sometimes governed by the chorale melody. In Cantata No. 117, for example, the repeated notes that form the headmotive of the chorale are heard not only in the chorale arrangements (nos. 1, 4, and 9) but in three other movements too (nos. 3, 5, and 7). And in three arias (BWV 192 no. 2, 177 no. 2, and 100 no. 4) the main vocal theme is a paraphrased version of the first line of the chorale melody. Dance rhythms, common in the arias of the 1720s, are now relatively rare, with the partial exception of the siciliana. The soprano aria with obbligato flute from Cantata 100 (no. 3), for example, is graced by a beautiful, simple, sequential siciliana melody in 6/8 dotted rhythms. Otherwise, dance rhythms tend to give way to a newer, more galant mode of rhythmic movement, often in 2 or 2/4 time and characterized by syncopation and internal repetition (examples are BWV 100 no. 4 and 112 no. 4). Bach comes still closer to the progressive styles of the day in the soprano aria from Cantata 97 (no. 8), with its short phrases divided by rests, its appoggiatura figures at cadences, its variety of notevalues, including triplets, and its simple bass and slow harmonic rate. On the other hand, Bach returns to older styles no less frequently—for example, in the duet from Cantata 100 (no. 2), where strict canonic imitation is employed throughout in a Corellian trio texture over a walking-quaver bass. Four of the most important cantatas from the 1730s, Nos. 140, 36, 14, and 80, might be described as semi-chorale cantatas. Cantata No. 140, Wachet auf, is in four of its

sacred cantatas

283

seven movements (nos. 2–3 and 5–6), as we have already seen, a dialogue between Jesus and the Faithful Soul. In its outer and central pillars (nos. 1, 4, and 7), however, it is a chorale cantata, setting all three verses of Nicolai’s hymn Wachet auf, ruft uns die Stimme. The framing movements are magnificent specimens of Bach’s standard chorale-chorus and four-part chorale types respectively, whereas the celebrated central movement, ‘Zion ho¨rt die Wa¨chter singen’ (no. 4), is a chorale trio with tenor cantus firmus in largely plain, standard note-values and an obbligato string part derived from the melodious opening ritornello. Just a week after Cantata No. 140, Bach performed a revised and enlarged version of the Advent cantata Schwingt freudig euch empor, BWV 36. It was probably influenced by the semi-chorale-cantata structure of Wachet auf, since in both cases three movements—here the three newly composed ones, nos. 2, 6, and 8—are based on the same chorale. Bach sets the first and last verse, as well as an intermediate verse (v. 6), of Luther’s great Advent hymn Nun komm, der Heiden Heiland (1524). The first verse (no. 2) is set as a chorale duet with continuo ritornellos. Each chorale line is given in a lightly decorated form and treated imitatively in a manner reminiscent of the old seventeenth-century chorale concerto. Verse 6 (no. 6) is more typical of Bach’s chorale arrangements for small ensemble: ritornellos for two imitative oboes d’amore alternate with the chorale cantus firmus, sung by the tenor in plain long notes. The work ends with a plain four-part setting of the last verse of Luther’s chorale. In this definitive 1731 version of Schwingt freudig a uniquely satisfying balance is achieved between older sacred and newer secular forms and styles—a balance that seems to reflect the first verse of the chorale, ‘All the world marvels’ (‘Des sich wundert alle Welt’) at the Saviour’s birth. Some years later, in 1735, Bach returned to the semi-chorale cantata in Wa¨r Gott nicht mit uns diese Zeit, BWV 14. It is based on Martin Luther’s metrical version of Psalm 124 (1524), of which verses 1 and 3 are retained in their original wording in the outer movements—a chorale-chorus (no. 1) and a plain four-part chorale (no. 5)— while verse 2 is paraphrased in a tenor recitative (no. 3). The aria texts are free (nos. 2 and 4), so that the overall structure, with its symmetrical ‘pillar’ design, closely resembles that of Wachet auf (BWV 140). The opening chorus is a grand chorale motet, in which the cantus firmus is purely instrumental (for unison of horn and two oboes), as in three Cycle I cantatas (Nos. 77, 25, and 48). Variants of each chorale line are extensively treated in counter-fugue before, during, and after their cantus firmus statement. The product is an imaginative rethinking of the antiquated chorale-motet form and one of the outstanding contrapuntal achievements among Bach’s chorale-choruses. Finally, a semi-chorale cantata was achieved by adapting the Weimar cantata Alles, was von Gott geboren (BWV 80a, 1715) to form a cantata for the Reformation Festival based on Luther’s great hymn Ein feste Burg ist unser Gott (1528/9). The process began around 1728/31, but only a fragment of that intermediate version survives (BWV 80b). In the definitive version, perhaps from the mid-to-late 1730s, all four verses of Luther’s chorale are incorporated (movements nos. 1, 2, 5, and 8), interspersed with two freetexted recitative–aria pairs drawn from the Weimar original. The analogy with the

284

s a c r e d a n d s e c u l a r : v o c a l wo r k s i i

overall form of Cantatas Nos. 140, 36, and 14 is clearly evident. The chorale-aria no. 2 opened the original Weimar cantata, from which it differs only in that the chorale cantus firmus is no longer purely instrumental but sung by the soprano (doubled by oboe) to the words of verse 2. The gigue-like chorale-chorus no. 5 is unique in its scoring for ‘4 voci in unisono’, an effect so rare in Bach’s vocal works as to create a tremendously powerful effect. The finale is a plain four-part chorale, as was the opening movement of the intermediate version. But in the definitive version this opening was replaced by a magnificent chorale motet with instrumental cantus firmus, a clear counterpart to the opening movement of Cantata No. 14, which suggests a date of origin for this movement, and hence for the final version of the whole cantata, of some time around or shortly after 1735. The chorale motet from Ein feste Burg has the additional refinement that the instrumental cantus firmus is in canon (as in the Cycle I cantatas Nos. 77 and 48)—at the octave and half-bar between unison oboes and violone with organ. As in Cantata No. 77, the presence of the chorale cantus firmus at the top and bottom of the texture surely symbolizes the almighty power of God, reflecting the first line of the text, ‘A mighty fortress is our God’. The only late chorale cantata of the Cycle II type—hymn text preserved in the framing movements but paraphrased in the inner ones—is Es ist das Heil uns kommen her, BWV 9. For the other late chorale cantatas Bach presumably had no librettist available and therefore had to resort to the per omnes versus type. In the case of Cantata No. 9, however, the text might have been already at hand in 1724–5, the year of Cycle II, but for some reason Bach had been unable to set it at that time. It is interesting to note that all three recitatives (nos. 2, 4, and 6), which paraphrase nearly all the inner verses of the hymn, are for bass voice, which allows them to preserve a sermon-like continuity. The opening chorus is light, sparkling, and brilliant—rather like ‘Ehre sei dir, Gott, gesungen’, newly composed around the same time for Part V of the Christmas Oratorio. The two arias represent contrasting types, both of frequent occurrence in Bach’s vocal works. The tenor aria no. 3 is illustrative: the words ‘We were already sunk too deep’ give rise to a ritornello-cum-vocal theme built on the image of sinking into the abyss. The soprano–alto duet no. 5, on the other hand, is symbolic: the strict, authoritative Lutheran doctrine of the text, ‘Lord, look not at good deeds, but at the heart’s strength of faith’, is represented by strict canon at the lower 5th and upper 4th between transverse flute and oboe d’amore, and later by double canon between the pairs of voices and instruments.

Passion and motet Text, occasion

Earliest source/s

Scribe, date

St Mark Passion, BWV 247, Good Friday O Jesu Christ, meins Lebens Licht, BWV 118, funeral

(lost) Scheide collection Priv. poss., Basel

(?); for 23 Mar. 1731 Autograph, 1736/7 Autograph, 1746/7

p a s s i o n a n d m o t et

285

It is well known that the Lutheran view of kings as God’s representatives on earth enabled Bach to use the same music for royalty as for divinity. A particularly clear example is the aria ‘Kron und Preis gekro¨nter Damen’ (‘Crown and prize of crowned ladies’), referring to the Queen of Poland (BWV 214 no. 7, 1733), which in the following year became ‘Großer Herr, o starker Ko¨nig’ (‘Great Lord, O mighty King’), referring to the divine ruler (Christmas Oratorio, Part I no. 8, 1734). By the same token identical music could be used for the death of princes or kings as for the crucifixion of the divine king Jesus Christ. This observation is highly relevant to the background history of the St Mark Passion, BWV 247, whose music is lost (the only copy of the score, in the hand of Franz Hauser, was destroyed in the Second World War), though certain movements may be recovered on the basis of Picander’s libretto (Leipzig, 1732), which is extant. Four years before its completion, on 11 April 1727, the St Matthew Passion had probably received its first performance; and later in that same year Gottsched’s Trauer-Ode (mourning ode) for Queen Christiane Eberhardine had been set to music by Bach (BWV 198) in a similar style and with much the same lofty tone. The St Matthew Passion was revived on 15 April 1729, only a few weeks after ten of its movements had been re-texted for use in the funeral music for Bach’s erstwhile employer Prince Leopold of Anhalt-Co¨then (BWV 244a, lost) alongside the framing choruses of the Trauer-Ode. Two years later the St Mark Passion was first performed, on 23 March 1731. The framing choruses of the Trauer-Ode now became the exordium and conclusio of the new Passion, and all of the arias from the ode (nos. 3, 5, and 8) were reused in the Passion with new words. It is clear from the surviving libretto that the St Mark Passion was an oratorioPassion along the same lines as the St John and St Matthew Passions. Accordingly, there were three components: biblical narrative, hymn verses, and Picander’s madrigalian verse. The biblical text, Mark 14–15, would have been sung as recitative by the Evangelist (tenor), Jesus (bass), and other characters. It would also have included numerous turbae, or crowd choruses. This entire biblical foundation of the work is lost beyond recovery. The hymn verses, of which there are sixteen as in the St Matthew Passion, would no doubt have been sung mainly as four-part chorales, though it is possible that one or two received more elaborate settings, especially that which concludes Part I, ‘Ich will hier bei dir stehen’ (no. 23). The choice of chorales strikingly overlaps with that of the other two Passions. Verse 5 of Reusner’s In dich hab ich gehoffet, Herr occurs in both the Matthew (no. 32) and Mark settings (no. 5), as does verse 1 of Gerhardt’s Befiehl du deine Wege (Matthew no. 44, Mark no. 28). The same author’s O welt, sieh hier dein Leben occurs in all three Passions (John no. 11, Matthew nos. 10 and 37, Mark no. 7). Above all, the most prominent chorale of each of the earlier Passions, that which contributes most to the character of the work, recurs in the St Mark Passion: Paul Stockmann’s Jesu Leiden, Pein und Tod (John nos. 14, 28, and 32; Mark no. 21); and Paul Gerhardt’s O Haupt voll Blut und Wunden (Matthew nos. 15, 17, 54, and 62; Mark nos. 23 and 30).

286 s a c r e d a n d s e c u l a r : v o c a l w o r k s i i Picander’s poems were sung as choruses—nos. 1 and 46 (first and last movements), set to the equivalent movements of the Trauer-Ode, nos. 1 and 10—and as arias, of which there are only six, compared with eight in the St John Passion and no fewer than fifteen in the St Matthew. Three of the arias were, as already noted, borrowed from the Trauer-Ode (nos. 9, 17, and 24). A fourth, ‘Falsche Welt, dein schmeichelnd Ku¨ssen’ (no. 19), might quite plausibly have been borrowed from the opening alto aria of the Weimar cantata Widerstehe doch der Su¨nde, BWV 54.11 But there is no general agreement as to Bach’s model for the other two arias (nos. 34 and 42), and it is perfectly possible that their music was newly composed for the Passion. Bach is now known to have revived the work in 1744, incorporating two additional arias, one in each half.12 In terms of its contemplative movements the work would then have been comparable with the St John Passion. It differs from the other two Passions, of course, in the large part played by parody, but this should not mislead us into supposing it to have been inferior. If we recollect that two of Bach’s greatest sacred masterpieces, the Christmas Oratorio and the B minor Mass, also relied to a large extent on parody, then we have to admit the possibility that the St Mark Passion was fully equal to its two predecessors, in which case its loss is utterly deplorable and in no way to be mitigated by any necessarily partial and speculative attempts at so-called ‘reconstruction’.13 The single-movement O Jesu Christ, meins Lebens Licht, BWV 118, is in essence a chorale motet, hence Bach’s designation ‘motetto’. Each line of the chorale O Jesu Christ, verse 1 (anonymous, 1608) is presented as a long-note cantus firmus in the soprano part while the lower voices accompany in traditional fashion with imitation of a plain or lightly decorated form of the same line in standard note-values (minims in alla breve metre). Chorale motets of this kind are not uncommon among Bach’s chorale-choruses.14 The role of the instruments in O Jesu Christ, however, strictly speaking places the work outside the motet genre. In the earlier version of about 1736/7, the voices are accompanied by two ‘litui’ (generally thought to be B♭ alto horns) plus the ‘solemn music’ choir of cornett and three trombones. This instrumental choir is largely colla parte during the chorale lines, but the litui have their own accompaniment figures. Above all, the instrumental ensemble is allotted a substantial prelude and postlude (based on a decorated version of chorale line 2), extracts and variants of which provide brief episodes between the chorale lines. Thus a ritornello structure, albeit remote from concertante style, frames both

11 As suggested by Friedrich Smend, ‘Bachs Markus-Passion’, BJ 37 (1940–8), pp. 1–35; repr. in F. Smend, Bach-Studien: Gesammelte Reden und Aufsa¨tze, ed. C. Wolff (Kassel, 1969), pp. 110–36. 12 See Tatiana Shabalina, ‘“Text zur Music” in Sankt Petersburg—Weitere Funde’, BJ 95 (2009), pp. 11–48 (esp. 30–6). 13 Daniel Melamed concludes that a proper reconstruction of the St Mark Passion cannot be achieved; see his ‘Parody and Reconstruction: The St. Mark Passion, BWV 247’, in D. R. Melamed, Hearing Bach’s Passions (Oxford, 2005), pp. 97–110. 14 Examples include BWV 38 no. 1 (Cycle II) and 28 no. 2 (Cycle III).

magnificat and missa

287

the chorale and its individual lines. Occasional precedents may be found among Bach’s sacred cantatas for this species of chorale motet with semi-independent instrumental parts, notably the opening chorale-chorus of the Cycle II cantatas Nos. 135 and 101. Bach appears to have extracted this hybrid motet-ritornello form from such cantata movements and applied it here to a single-movement occasional work, no doubt written for a funeral. The work may be regarded as a special piece of Trauermusik, intended for an occasion of great solemnity. The instrumentation suggests outdoor performance, perhaps during a burial or funeral procession. When the work was revived about ten years later (c. 1746–7) the trombone choir was replaced by strings and continuo, which indicates an indoor performance.

Magniﬁcat and Missa Title

Earliest source/s

Scribe, date

Magniﬁcat in D, BWV 243 Missa in B minor, BWV 2321 Missa in A, BWV 234

Berlin, P 39 Berlin, P 180, Dresden SLB Darmstadt HLB, Berlin, St 400 Darmstadt HLB Berlin, P 15/2 Berlin, P 15/3 Berlin, P 70/2 Darmstadt HLB

Autograph, c. 1732/5 Autograph, part-autograph, 1733 Autograph, J. C. F. Bach et al., c. 1738, 1748/9 Autograph, c. 1738 J. C. Altnickol; orig. c. 1738 J. C. Altnickol; orig. c. 1738 Anon., post-1750 C. G. Sander, post-1800

Missa in G, BWV 236 Missa in G minor, BWV 235 Missa in F, BWV 233 Kyrie eleison—Christe, du Lamm Gottes, BWV 233a

There seems to be a close connection between Bach’s two major pieces of Latin church music of the early 1730s, the Magnificat in D, BWV 243, and the Missa in B minor, BWV 2321. Around the time when the Missa was composed (1733), Bach undertook a major revision of the Magnificat he had composed during his first year in Leipzig (BWV 243a), transposing it down a semitone from E♭ to D and altering numerous details. And it is quite possible that this revised version, like the Missa, was intended for the Dresden court. Both works are written for five voices (SSATB) rather than the four voices that Bach normally employed. The instrumental ensemble is essentially the same in both cases: three trumpets and drums, two flutes, two oboes, strings, and continuo (including bassoons). And the autographs of the two works are written on paper of the same type.15 In the case of the Magnificat, the watermark points to a date between 1732 and 1735. The Missa, of course, is known to date from 1733 (its dedication is dated 27 July); and a suitable occasion for the performance of the revised Magnificat arose in that very year. On 1 February the Elector, Friedrich Augustus I (nicknamed ‘Augustus the Strong’) died in Warsaw. His death led to an extensive period of national mourning. The performance of concerted church music then resumed on 2 July, the

Wisso Weiss and Yoshitake Kobayashi, NBA IX/1 (1985), No. 121 (p. 93).

288 sa cred and secular: vocal works ii Feast of the Visitation, when the Magnificat (Luke 1: 46–55) formed part of the Gospel reading. It has been suggested that Bach’s revised Magnificat might have been performed on this occasion.16 If so, little over three weeks would have intervened between the performance of the Magnificat and the dedication of the Missa, in which case Bach might well have been working on the two compositions simultaneously. Parallels between the two compositions are not restricted to the make-up of the vocal and instrumental ensemble. The festive trumpets-and-drums key of D major is common to the Magnificat and the ‘Gloria’ of the Missa. Both works are made up of twelve movements (albeit on a much smaller scale in the Magnificat), alternating between solos and choruses. And details of scoring in the revised Magnificat can be more or less matched in the Missa: the obbligato for oboe d’amore in ‘Quia respexit’, no. 3, as in ‘Qui sedes’ (Missa, no. 10); the muted strings in ‘Et misericordia’, no. 6, as in ‘Domine Deus’ (Missa, no. 8); and the pizzicato cello in ‘Esurientes’, no. 9, as in ‘Domine Deus’ again. Other alterations to the Magnificat are matters of detail: in terms of substance Bach seems to have been content for the most part with the work as it was heard during his first year in Leipzig. The solo vocal parts are often to some extent further decorated. The climactic minor-9th pause chord towards the end of the chorus ‘Omnes generationes’ (b. 24) is softened by an appoggiatura, and the same interval is circumvented in the tenor solo ‘Deposuit potentes’ (bb. 4, 18, and 57)—by this stage Bach might have found it somewhat harsh and crude. The cantus firmus in ‘Suscepit Israel’, no. 10, is now played by unison oboes rather than first trumpet, and the former bassett for unison upper strings is replaced by basso continuo. Finally, in the doxology, no. 12, the imitative ‘gloria’ passages in triplets, formerly sung unaccompanied, are now supported by continuo pedal points played tasto solo. The Christmas laudes are of course omitted, but they were in any case an optional extra that could be left out whenever the Magnificat was performed on an occasion other than Christmas. Bach’s three large-scale liturgical works of the early 1730s—the St Mark Passion of 1731, the Missa of 1733, and the Christmas Oratorio of 1734—have in common the considerable part played by parody in their composition. They differ, however, in the circumstances of their performance. The Passion and oratorio were both written for performance during Leipzig services—on Good Friday and at Christmas respectively. The Missa, on the other hand—the Kyrie and Gloria that later formed the basis of the B minor Mass (1748/9)—was conceived in the first place for performance in Dresden, though it might have been revived in Leipzig at some later date. The performing parts were dedicated to the new Elector of Saxony Friedrich Augustus II on 27 July 1733.17 The great work must have been performed at a special festive service in Dresden— hence the exceptionally large dimensions and the use of trumpets and drums—

16 See Hans-Joachim Schulze, preface to facs. edn of the Dresden Missa (n. 18) and his ‘The B minor Mass: Perpetual Touchstone for Bach Research’, in P. Williams (ed.), Bach, Handel, Scarlatti: Tercentenary Essays (Cambridge, 1985), pp. 311–20. 17 BD I, No. 27; NBR, Nos. 161–2.

magnificat and missa

289

perhaps to mark the accession of the new Elector after the mourning period for his father (1 February to 2 July 1733).18 There is special significance in Bach’s offering Latin church music to the Elector of Saxony in 1733. For although the citizens of Dresden were Lutheran, the court itself was Catholic and therefore the music at the Hofkirche, the court church, was exclusively Latin-texted. Settings of the Mass Ordinary formed the central focus of musical activity in relation to the church, and an impressively wide-ranging repertoire was gradually amassed that included not only Mass settings by the resident composers Heinichen, Zelenka, and Hasse, but also by Fux and Caldara of Vienna, Lotti and Vivaldi of Venice, A. Scarlatti and F. Durante of Naples, and even the revered Palestrina of a bygone age. The Ordinary of the Mass would have been far from alien to Bach and other Lutheran composers of the day. For the five main parts of the Ordinary—Kyrie, Gloria, Credo, Sanctus, and Agnus Dei—had been accepted in the Formula missae of 1526 and in the Augsburg Confession of 1530. And it had been stipulated that there should be no changes to the public celebration of the Mass except for the addition of German hymn paraphrases, such as Allein Gott in der Ho¨h sei Ehr for the Gloria or Wir glauben all an einen Gott for the Credo.19 This was accepted in the Saxon Agenda of 1539, with the result that all five parts of the Mass Ordinary were used in the main Leipzig service, the Hauptgottesdienst, in Bach’s day. As far as settings of the Mass are concerned, then, there was at that time much common ground between Lutheran and Catholic composers. This applies above all to the abbreviated Mass, the Kyrie and Gloria (all of Bach’s five Mass settings of the 1730s, BWV 232–6, belong to this ‘Missa’ type). Due to the increasing size of concerted settings of the Mass, it became the custom within the Lutheran Church—and often in the Catholic Church too—to perform only the Kyrie and Gloria in polyphonic settings during services, leaving the Credo, Sanctus, and Agnus Dei to monophonic rendition. The contents of Bach’s music library show clearly that long before 1733 he had studied and performed Mass settings by numerous seventeenth- and early eighteenthcentury composers. During the Weimar years (1708–17) he had copied out Masses or Mass movements by M. G. Peranda, J. Baal, J. C. Pez, and J. L. Bach. And during his first ten years in Leipzig (1723–33) he had supplemented the Pez and J. L. Bach materials and copied out Masses by F. Durante, J. H. von Wilderer, and A. Lotti.20 It is significant that all of these composers, with only one exception (J. L. Bach), were Catholics—the divide between the Lutheran and Catholic confessions tends to be broken down in this sphere. It is also significant that, with the exception of Baal’s Mass

Schulze gives detailed arguments to support his view that the Missa might have been performed in Dresden before the dedication. See his preface to J. S. Bach: Missa h-Moll BWV 2321–Faksimile nach dem Originalstimmensatz der Sa¨chsischen Landesbibliothek Dresden (Neuhausen and Stuttgart, 1983). 19 See Robin A. Leaver, ‘Music and Lutheranism’, in J. Butt (ed.), The Cambridge Companion to Bach (Cambridge, 1997), pp. 35–45 (esp. 42–4). 20 See Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992), VBN I/P/3 (Peranda), I/B/1 (Baal), I/P/6 (Pez), I/B/18 (J. L. Bach), I/D/2 (Durante), I/W/2 (Wilderer), and I/L/2 (Lotti).

290 s ac r e d a n d s e c u l a r : v o c a l w o r k s i i in A, all of these Mass settings belong to the ‘Missa’ type (Kyrie and Gloria only) that Bach was to cultivate himself. He had already composed a number of Latin church works before 1733: the Kyrie in F, BWV 233a (Leipzig, 1723–5); the two Sanctus settings, BWV 237–8 (Leipzig, 1723); the original E♭ version of the Magnificat, BWV 243a (Leipzig, 1723); and the six-part Sanctus, BWV 232111 (Leipzig, 1724), later incorporated in the B minor Mass. But the great Dresden setting of the Kyrie and Gloria, composed in 1733, was evidently his first Missa. There are many indications that Bach was acquainted with the Dresden Mass repertoire and its normal performance conditions when he set about the composition of the 1733 Missa. Indeed, the Missa seems to have been written deliberately in such a way as to conform with local Dresden expectations. The very choice of the Missa format (Kyrie and Gloria only) is significant in this regard, for not only was it the usual type of Mass setting in the Lutheran services at Leipzig, but it was also frequently employed in the Catholic services at the Dresden Hofkirche. Furthermore, in structuring the Missa Bach adopted the larger and more progressive of the two types of concerted Mass setting that are encountered in the Dresden Mass repertoire. In the smaller and older type, common in Venice, Vienna, and other Catholic centres, the Kyrie and Gloria were each presented as a continuous whole, composed in contrasting sections according to the import of the words and with inbuilt tutti/solo contrasts. Bach’s Missa, however, belongs to the larger, ‘Neapolitan’ type of concerted Mass setting, so called because it was frequently cultivated by composers active in Naples, such as Alessandro Scarlatti and Francesco Durante. This type is closely related to contemporary opera. The Mass is no longer set as a continuous series of interconnected sections, but rather as a series of more or less independent movements—a mixture of arias and choruses, often employing contemporary da capo and ritornello procedures. Bach’s Missa greatly exceeds those of the Dresden Hofkirche in duration. Yet it seems very likely that he had the first-rate singers and players of the court Capelle in mind when he composed it. Any minor discrepancies between his score and local performing conditions in Dresden could be ironed out by Zelenka or some other local musician, as was customary at the time.21 The Kyrie is a threefold Greek prayer for divine mercy, used in the liturgy of the Christian Church since about the fourth century. Bach sets each of the three parts in a separate movement as shown: Part: Movement: Key:

Kyrie I chor. b

Christe duet D

Kyrie II chor. f♯

21 Regarding Mass composition in Dresden, see Wolfgang Horn, Die Dresdner Hofkirchenmusik 1720–1745: Studien zu ihren Voraussetzungen und ihrem Repertoire (Kassel, 1987); Janice B. Stockigt, ‘Consideration of Bach’s Kyrie e Gloria BWV 232I within the Context of Dresden Catholic Mass Settings, 1729–1733’, in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass (Belfast, 2007), vol. i, pp. 52–92; and George B. Stauffer, Bach: The Mass in B minor (New York, 1997), pp. 16–23.

magnificat and missa

291

This scheme, the general character of the first movement, and even certain thematic details appear to be indebted to the Missa in G minor by the Palatine Electoral Court Composer Johann Hugo von Wilderer, which Bach had copied out before 1731.22 The autograph of Bach’s three Kyrie movements is a fair copy, which invites speculation over lost parody models.23 However, the Adagio introduction, which sets the tone for the great cry of mercy that follows, was clearly an afterthought—it has all the signs of a first draft and must have been composed directly into the score.24 The bass line appears to be an adumbration of the Kyrie I theme that follows, and the top line (soprano I doubled by woodwind) has been construed as an embellished cantus firmus based on the Kyrie from Luther’s Deutsche Messe of 1524.25 The main Kyrie I that follows—possibly parodied from a lost original in C minor26—is one of the most monumental movements in the entire Missa. The manner in which Bach unites fugal and ritornello modes of structuring allows him to achieve vast proportions: Paragraph: Function: Key: Ensemble:

A Rit. b instrs

Exp. b vv

Rit. f♯ tutti

Epis. A instrs.

Exp. b tutti

Rit. b tutti

The element of contrast between ritornello and episodes, so vital in a concerto movement, is deliberately avoided here. For Kyrie I is intended to be all of a piece— one great, unified cry of humanity to its divine maker. Bach achieves this by assimilating, to a large extent, the ritornellos and fugal expositions. Thus whereas the opening ritornello is purely instrumental, the five-part vocal ensemble participates fully in the second and third ritornellos, as it does in the intervening fugal expositions. Moreover, the ritornello itself includes three entries of the fugue subject (bb. 5, 7, and 22). The fugal expositions may thus be heard as expansions of the fugal component of the ritornello. This has enormous repercussions for the overall structure of the movement, for it allows the second and third ritornellos to emerge imperceptibly out of the preceding exposition, as if they were simply a continuation of it. If exposition and ritornello are heard as a unit in this way, we find that an overall bipartite form (AA1) is superimposed on the ritornello form. These two vast paragraphs are joined by a brief instrumental episode, which provides a welcome element of contrast, since it contains the only major-mode entry of the subject. The only 22

Bach’s indebtedness to Wilderer is illustrated by Christoph Wolff, ‘Origins of the Kyrie of the B minor Mass’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass., and London, 1991), pp. 141–51. 23 See Robert L. Marshall, ‘The Mass in B minor: The Autograph Scores and the Compositional Process’, in his The Music of Johann Sebastian Bach (New York, 1989), pp. 175–89. 24 According to Marshall, ‘The Mass in B minor’, pp. 175–89. 25 See Wolff, ‘Origins of the Kyrie’, p. 147. 26 According to Rifkin, notes to his recording of the Mass.

292

sacred and secular: vocal works ii

significant cadences occur at the end of the ritornellos, which both confirms the impression of an overall bipartite structure and also contributes to the very broad, virtually seamless sweep that Bach aims to achieve. A common feature of the Dresden Mass repertoire is that movements in a ‘modern’ operatic style were frequently juxtaposed with choruses written in a traditional style derived from Renaissance polyphony. In the Kyrie, for example, ‘Christe eleison’ was often set as an operatic love duet, whereas the following Kyrie II was often a choral fugue, written in the stile antico.27 This is exactly what happens in Bach’s setting. Not long before the composition of the Missa, around 1727/31, he had set ‘Christe eleison’ as a duet (BWV 242), to be incorporated in Francesco Durante’s Missa in C minor. Now he sets those words in a similar fashion in his own Missa of 1733. The style represents Bach at his most galant.28 Whereas the second main vocal phrase (b. 14) is imitative, the first is written in mellifluous parallel 3rds, which in operatic duets customarily represent the union of the two lovers. In Bach’s sacred music such parallel 3rds often have a beatific connotation, reminding one of the state of blessedness that arises from oneness with the divine. From the galant style of the ‘Christe’ duet Bach moves to the opposite extreme of strict stile antico in Kyrie II.29 The reduction to four voices (SATB) suggests that the movement might have existed in some form before the Missa was composed.30 It is a motet-style alla breve fugue with voices doubled by instruments, written in pseudoRenaissance polyphony, though the poignant use of the flat supertonic, which forms a link with Kyrie I, and the recurring chromatic 4th in the bass are key features of Baroque style. After a double exposition of the flat-supertonic subject, a dynamic second subject enters (b. 31), in marked contrast with its relatively static predecessor. Herein lies a major difference between this movement and Kyrie I, which represented a single state of the soul, unruffled by changes or contrasts of any kind. For in Kyrie II the second subject injects a new note of urgency into the smooth surface of the fugue. Whereas the first half of the fugue is occupied with a double exposition of the main subject, the second half alternates four-part strettos of the subsidiary subject with two-part strettos of the main subject. The overlapping of the vocal entries creates a sense of increasing intensity as the pleas for mercy are multiplied.

As Stauffer observes in Bach: The Mass in B minor, pp. 58–9 and 61. As pointed out by R. L. Marshall, ‘Bach the Progressive: Observations on His Later Works’, in his The Music of J. S. Bach, pp. 23–58 (see 41–2). 29 On this style and Bach’s adoption of it see, above all, Christoph Wolff, Der Stile antico in der Musik Johann Sebastian Bachs: Studien zu Bachs Spa¨twerk (Wiesbaden, 1968). Paul Walker emphasizes the important part played in the stile antico movements of the Missa (and indeed in the Mass of 1748/9) by the example of the ‘father of contrapuntists’, Johann Theile, who in turn owed much to Zarlino’s treatise Le istitutioni harmoniche of 1558. See P. Walker, ‘Bach’s Use of Fugue in the stile antico Vocal Writing of the B-minor Mass’, in Tomita et al. (eds.), Understanding Bach’s B-minor Mass, vol. ii, pp. 368–86. 30 As pointed out by John Butt, Bach: Mass in B minor (Cambridge, 1991), p. 45. Other writers have reached the same conclusion. 28

magnificat and missa

293

Unusually for Bach the two contrasting subjects are invariably treated in alternation rather than combined.31 The text of the Gloria is made up of two components: the ‘angelic hymn’ from Luke 2.14—‘Gloria in excelsis’ and ‘Et in terra pax’—and the liturgical hymn that opens with the words ‘Laudamus te’, which was added by the early Church. Bach sets the angelic hymn as a separate complex and the liturgical hymn in a symmetrical pattern of three complexes thus: Gloria in excelsis Et in terra pax Laudamus te Gratias agimus tibi Domine Deus Qui tollis Qui sedes Quoniam Cum Sancto Spiritu

1 2 3 4 5 6 7 8 9

chor. chor. aria chor. duet chor. aria aria chor.

D G–e–D A D G–b b b D D

angelic hymn praise thanksgiving Father and Son miserere miserere Trinity

liturgical hymn

The tonal centres of the Kyrie and Gloria, B minor and D major respectively, and their musical associations represent archetypal states of the soul at opposite ends of the spectrum. B minor represents the plea for mercy and has associations with sorrow, sinfulness, and penitence; its relative major D, on the other hand, stands for praise, thanksgiving, and the divine glory. Yet in the interests of depth and verisimilitude both Kyrie and Gloria have their tonal opposites contained within them. The Kyrie has its central ‘Christe’ in the key of D, representing the comfort offered by Christ; and the Gloria returns to B minor for ‘Qui tollis’ and ‘Qui sedes’, which clearly refer back to the Kyrie, both including the prayer ‘miserere nobis’ (‘have mercy upon us’). The angelic hymn is set as two great choruses, to be sung without a break.32 The biblical context of the words is the account of the Nativity in Luke 2, hence no doubt Bach’s later reuse of the same music in a Latin cantata for Christmas Day (BWV 191, 1745). The ritornello structure of the ﬁrst chorus, ‘Gloria in excelsis Deo’, is remarkably similar to that of Kyrie I, suggesting composition in close temporal proximity. It is on a smaller timescale, however, and lacks the fugal element: A1

A Section: Material: Key: Ensemble:

Rit. abc D instrs.

Epis. ab D–A vv.-instrs.

Rit. abc A tutti

Epis. aa1 A–e–b–D instrs.-vv.

Rit. abc D tutti

This occasionally happens in his Leipzig organ fugues, e.g. in the Fuga in C minor, BWV 537 no. 2. The possible parody origin of these movements is noted by Rifkin (see n. 26), Butt (n. 30), and Stauffer (n. 21). 32

294 s a c r e d a n d s e c u l a r : vo c a l w o r k s i i The principal theme a is a triadic blaze of glory, played initially by two imitative trumpets and eminently well suited to the ‘multitude of the heavenly host’ who sing these words in Luke’s Gospel. As in Kyrie I, the main theme serves for the choral episodes as well as the ritornellos. There is a further structural parallel: since the important cadences occur only at the end of the ritornellos, we hear superimposed on the alternating ritornello scheme an overall bipartite, AA1 stucture, so that after the instrumental introduction the music runs continuously in two broad sweeps. It has been pointed out that the style of the chorus, with its 3/8 dance rhythm, is closely related to that of the first movement of a number of secular cantatas from the years immediately before or after the 1733 Missa: BWV 201 (1729), 206 (1734), 214 (1733), and 215 (1734).33 Two of these movements were subsequently parodied with sacred words: BWV 214 no. 1 became the opening movement of the Christmas Oratorio (1734); and BWV 215 no. 1, the Osanna of the B minor Mass (1748/9). It is perfectly possible, then, that the great Gloria chorus began life as a secular piece, perhaps in praise of an earthly king rather than the heavenly King and for four voices rather than five. It might have formed the A-paragraph of a lost ABA da capo movement,34 for all the comparable secular movements cited previously are cast in da capo form. The second chorus of the angelic hymn, which follows without a break, is designed as a sort of vocal prelude-and-fugue. In the ‘prelude’ the trumpet choir is silent and an aura of peace evoked in accordance with the text, ‘Et in terra pax hominibus bonae voluntatis’ (‘And on earth peace to men of good will’). The conjunct quavers slurred in pairs, the parallel 3rds and 6ths, and the long-held pedal notes all help to create a vividly pastoral atmosphere. Pastorales were then associated with Christmas due to the biblical story of the shepherds abiding in the fields—the very context of the angelic hymn in Luke 2. The subject of the fugue incorporates the two main themes of the ‘prelude’—the gentle, stepwise ‘Et in terra pax’ theme and the more rhythmic cadential phrase on ‘bonae voluntatis’. The wonderful music built on this subject is structured as a permutation fugue—the subject a and the three countersubjects b, c, and d follow one another in each voice in turn, giving rise to a variety of different combinations. The fugue alternates between tonally stable expositions, all centred around the tonic D, and modulatory, homophonic episodes (bb. 37 and 60), which return to the kind of antiphonal exchanges between vocal and instrumental groups that opened the ‘prelude’. The first complex of the liturgical hymn consists of an aria of praise, ‘Laudamus te’, and a chorus of thanksgiving, ‘Gratias agimus tibi’. A greater contrast than between these two movements could hardly be imagined. ‘Laudamus te’ is quite the most ‘modern’ and overtly operatic aria in the entire Missa, whereas ‘Gratias agimus tibi’ is a choral fugue, written in a traditional alla breve style. The sequence of these two movements thus harks back to ‘Christe’ and Kyrie II in its immediate juxtaposition of

See Butt, Bach: Mass in B minor, p. 46.

According to Rifkin, notes to his recording.

magnificat and missa

295

strikingly progressive and conservative styles. In the aria ‘Laudamus te’ Bach summons up his most lavish resources of embellishment as a worthy offering of praise to the divinity. It has been suggested35 that the highly florid, operatic style of the solo soprano part might have been conceived with Faustina Bordoni in mind, the famous Venetian mezzo-soprano who had become the leading prima donna of the age. Bach was almost certainly present at her Dresden debut in 1731 in the opera Cleofide by her husband Johann Adolf Hasse. Both her mezzo-soprano range and her known expertise at coloratura would have equipped her ideally for the part.36 Bach is extremely lavish in this reprise-form movement (ABA1), not only in his embellishment of the solo soprano and obbligato violin lines but also in the sheer profusion of his invention. In the opening ritornello, which generates the rest of the movement, each of the six phrases introduces new kinds of figuration. And when the solo soprano enters she seems determined to outdo the solo violin in the brilliance of her coloraturas. The great chorus ‘Gratias agimus tibi’ is one of only two movements in the Missa whose parody model survives (the other being ‘Qui tollis’). It is based on the setting of Psalm 75: 1 from Wir danken dir, BWV 29, no. 2 (1731).37 Very little change was required, for not only are the German and Latin texts both concerned with thanksgiving but they are very close in wording. In the context of the Missa, after the extremely ornate setting of ‘Laudamus te’, the plainness of the ‘Gratias agimus tibi’ theme is startling by comparison. Like Kyrie II, this movement is a four-part alla breve fugue in motet-style polyphony with the voices doubled by instruments. Both movements employ stretto as their main structural principle and both are based on two distinct subjects. However, whereas stretto and the second subject both made a late entry in Kyrie II, ‘Gratias’ is structured as a double stretto fugue throughout: both subjects are treated in stretto from the very outset. Stretto expositions of the two subjects alternate within an overall tripartite scheme as shown: Exposition A SI

Middle Section B SII

A1 SI

B1 SII

Conclusion A2 SI

B2 SII

A3 SI

In the middle section, strettos of the principal subject are linked to form massive ediﬁces of ﬁrst nine (4 + 5) and then thirteen (3 + 6 + 4) entries. The two subjects, unlike those of Kyrie II, are attached to different words. The plain, arch-shaped theme on ‘Gratias agimus tibi’ has a grand simplicity to represent the act of thanksgiving. By contrast, the more animated theme on ‘propter magnam gloriam tuam’ includes a

By Marshall, ‘Bach the Progressive’, p. 42, taking up a suggestion of Arthur Mendel’s. J. J. Quantz’s description of Faustina’s singing is quoted by Charles Burney, A General History of Music (London, 1776–89); see Stauffer, Bach: The Mass in B minor, pp. 72–3. 37 Or, to be more precise, ‘Gratias agimus’ and ‘Wir danken dir’ might have had a common source, according to Rifkin and Fro¨de (see n. 10). 36

296

sacred and secular: vocal works ii

melisma in quavers to represent God’s glory. The voices are doubled throughout by instruments in motet style, as in Kyrie II, except for the trumpet choir which doubles the vocal entries sparingly for special climactic effect and to some extent has an independent role. The central complex of the liturgical hymn comprises three movements that flow directly into one another: ‘Domine Deus’, a duet that addresses God the Father and the Son; ‘Qui tollis’, a choral prayer for mercy, addressed to Christ as the sacrificial Lamb of God; and ‘Qui sedes’, a solo prayer for mercy addressed to Christ in glory. Significantly, the texts relating to the Father and the Son are set in the major mode (the subdominant G), whereas the two ‘miserere’ prayers are set in the minor (the relative minor b). For ‘Domine Deus’ Bach returns to the operatic love-duet idiom of ‘Christe’, but here the theme is the mutual love of Father and Son. Their relationship is symbolized by dual texting: the first two lines that refer to the Father and the third and fourth that refer to the Son are sung simultaneously. The two voices with different texts signify the separate identity of the two Persons, while their simultaneity signifies their oneness within the Godhead. The unity of Father and Son is further underlined by the parallel motion of the two voices in 3rds, 6ths, and 10ths—a common device in operatic duets to signify the union of the happy couple. In addition, a warm, rich instrumental colour from transverse flute, muted upper strings, and pizzicato cello and violone is employed alongside the ‘beatific’ key of G major to conjure up the blissful union between Father and Son. Bach so often uses similar means to express connubial bliss in the arias and duets of his wedding cantatas that the lost original model of ‘Domine Deus’ might be sought in those circumstances. After the unclouded bliss of the majormode A-paragraph, the key turns to the minor for paragraph B, where the two voices unite in a single text relating to the sacrificial Lamb of God. The expected da capo (presumably present in the parody model) is omitted, not just to preserve continuity and avoid unnecessary repetition but also because the words addressing the Lamb of God flow directly into the following prayer for mercy, ‘Qui tollis peccata mundi’. In order to underline this point Bach goes out of his way to smooth the join between the two movements. Not only is the B minor tonality of ‘Qui tollis’ anticipated in the last part of the duet but the concluding phrase of the latter begins with vocal imitation of the very same notes that are about to open ‘Qui tollis’ (Ex. 2). ‘Qui tollis’ was parodied from the opening chorus of Schauet doch und sehet, BWV 46, written ten years earlier during Bach’s first year in Leipzig (1 August 1723). The movement is a great lament, based on German words from Lamentations 1: 12 which in English translation read: ‘Behold and see if there be any sorrow like my sorrow, which has been inflicted on me.’ The Latin ‘Qui tollis’ text, however, is a prayer for mercy. It is fitting that the lament and the prayer about the ‘sins of the world’ should be sung to the same music in view of the New Testament context of the lament in Cantata No. 46: in the Gospel reading for the day (Luke 19: 41–8) Christ laments the sins of Jerusalem. In the Latin version Bach greatly enhances the original theme,

m a g n i f i c a t a nd m i s s a

297

Ex. 2 Sop. 1

Alto

gnus De

li - us

tris.

Qui

Ten.

gnus De - i,

tol - lis pec

Qui

li - us Pa

tol - lis pec - ca

mun

tris.

mun -

Gloria from Dresden Missa of 1733: join between end of ‘Domine Deus’ (bb. 93–5) and beginning of ‘Qui tollis’ (vocal parts only)

Ex. 3

Schau - et

doch

und

se - het, ob

ir - gend - ein Schmerz sei

wie mein Schmerz

a) Vocal theme from 1st movement of Schauet doch und sehet, BWV 46

Qui

tol - lis pec - ca

mun - di

mi - se - re - re

no - bis

b) Same theme as remodelled for ‘Qui tollis’ of Dresden Missa already highly expressive, mainly by rhythmic differentiation (Ex. 3). The integral sinfonia of the original is dropped in the Missa, not for any inherent unsuitability but rather for the sake of economy and continuity. The movement falls into three broad paragraphs (bb. 1, 14, and 28), each opening with an almost strict canonic exposition of the main theme. After six bars the texture is decorated by an ornate duet for two ﬂutes, derived from the original sinfonia. The richly dissonant harmony, with its many 7th and 9th suspensions, helps to convey the feeling of a prayer uttered de profundis. The coda (b. 42) ends in the dominant to allow ‘Qui sedes’ to emerge directly out of it; and as in the join between ‘Domine Deus’ and ‘Qui tollis’, there is a motivic link as well as a tonal one (Ex. 4). ‘Qui sedes’, an alto aria with obbligato oboe d’amore, strings, and continuo, is essentially a prayer for mercy, like ‘Qui tollis’, hence the use of the same key of

298

s a c r e d a n d s e c u l a r : v o c a l wo r k s i i

Ex. 4 Sop.

Ob. d’amore

de - pre - ca - ti - o

nem no

stram

Dresden Missa: join between end of ‘Qui tollis’ (bb. 47–50, soprano only) and beginning of ‘Qui sedes’ (oboe d’amore only) B minor, relating back to the prayer of similar import in the Kyrie. In ‘Qui tollis’ the appeal was to Christ the sacrificial Lamb of God, hence the tone of sorrow. In ‘Qui sedes’, however, the appeal is to Christ in glory, with the result that the tone is more optimistic and a dance-like rhythm is heard. A striking feature is the cadential echo (bb. 4 and 16), in which forte staccato chords are echoed by piano legato chords. When the echo occurs in the voice it invariably falls on ‘Patris’ (‘Father’). It is likely, then, that the loud, detached chords represent the relatively stern figure of God the Father; the soft, smooth chords, the more yielding, compassionate figure of God the Son. The echo would then represent Christ as a reflection of the Godhead, though with His own distinctive character. The third and last complex of the liturgical portion of the Gloria is a hymn in praise of the Trinity, made up of two movements, the aria ‘Quoniam to solus’ and the chorus ‘Cum Sancto Spiritu’. In order to emphasize their essential unity, these two movements are composed in the same metre, 3/4, and in the same key, the overall tonic D major. Moreover, they are sung without a break—the aria cadences directly into the chorus. The ‘Quoniam’, written in the style of a polonaise,38 is set as a bass aria in reprise form (ABA1), in which the voice is accompanied by one of the most unusual instrumental combinations of any Bach aria: solo horn, bassoon duet, and continuo. Thus all the vocal and instrumental parts are in the bass register, with the single exception of the horn, which sounds an octave lower than written, in the tenor/alto register. This unique instrumentation seems to have a symbolic significance. The third clause of the text reads: ‘Tu solus altissimus, Jesu Christe’ (‘Thou only art the Most High, Jesus Christ’). Accordingly, the solo horn as representative of Christ stands out alone as the ‘most high’ participant in the ensemble. The attributes of Christ with which the text is concerned are his holiness, lordship, and exalted status as Son of God. Consequently, the horn opens with a boldly majestic theme in striding crotchets. Against this horn theme the two bassoons play their own counter-theme in ‘beatific’ 3rds. This, in conjunction with the bass pitch of the instruments, may symbolize the beatitude of lowly mortals under the rule of Christ in glory. 38 The popularity of the polonaise in Dresden at the time, its use in ceremonial music and in Mass settings, and its association with sovereignty are explored by Szymon Paczkowski, ‘On the Role and Meaning of the Polonaise in the Mass in B minor by J. S. Bach as Exemplified by the Aria “Quoniam tu solus sanctus”’, in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass (Belfast, 2007), vol. i, pp. 33–51.

m a g n i f i c a t a n d mi s s a

299

Whereas ‘Quoniam’ addresses Christ, the Second Person of the Trinity, ‘Cum Sancto Spiritu’ praises the Holy Spirit and the Father (in that order), the Third and First Persons. This praise is enacted within a supremely jubilant and exhilarating chorus. Like Kyrie I, it is a ritornello structure with fugal episodes. Unlike in Kyrie I, however, there is no introductory instrumental ritornello: for the sake of immediacy the voices enter at the very outset. Many years ago it was suggested by Donald Francis Tovey that this movement (like ‘Qui tollis’) is an arrangement of an earlier piece, now lost, with its opening ritornello discarded, and this view is now generally accepted by scholars.39 Bach introduces a most striking contrast between the fully scored ritornellos—homophonic in texture and with full participation of trumpets and drums—and the more lightly scored fugal episodes in which the trumpet choir is silent. Since the words are the same, this contrast is presumably created for musical reasons only. Indeed, after the sheer massiveness of the opening ritornello with inbuilt choral parts, a certain lightening of the texture was a musical necessity. The reduction to a single voice with continuo at the start of each fugue (bb. 37 and 81) and the process of adding fugal entries one by one till a full ﬁve-voice texture is achieved twice produces an extremely impressive build-up that culminates in the following ritornello. The great tutti with sustained voices and motivic instruments at bars 5–8 is later extended (b. 25) to incorporate a powerful harmonic sequence over a stepwise bass descent, familiar from Partitas Nos. 4 and 5, BWV 828 and 829 (Ex. 5). This sublime passage occurs within each ritornello at the words ‘in gloria Dei Patris’—the heavens open, as it were, to reveal a great vision of God the Father. In the fugal episodes, on the other hand, the key words are ‘cum Sancto Spiritu’ (‘with the Holy Spirit’), which open the fugue subject in sprightly fashion. The fugal episodes are thus to be heard as vivacious songs in praise of the Holy Spirit. In the dedication letter that accompanied the Missa of 1733, Bach asked the Elector of Saxony to grant him a title in ‘Your Highness’s Court Capelle’, promising in return to exercise ‘my untiring zeal in the composition of music for the church as well as for the orchestra’.40 On 19 November 1736 his wish was granted and the title of ‘Compositeur to the Royal Court Orchestra’ bestowed upon him.41 It may be assumed that he then fulﬁlled his promise of composing church and orchestral music for the Dresden court. Since it was a Missa that had secured him the post and the four Missae BWV 233–6 date from around 1738, not long after the appointment, it is reasonable to suppose that they too might have been composed for the Dresden court. Bach’s continued interest in the Mass during the interim is evinced by his copying out the six Masses from the collection Acroama missale by Giovanni Battista Bassani42 around 1735. Like the Missa of 1733, those of about 1738 were assembled largely by parody of

39 40 41 42

See D. F. Tovey, Essays in Musical Analysis, vol. v (London, 1937), pp. 34–5. BD I, No. 27; NBR, No. 162. BD II, No. 388; NBR, No. 190. See Beißwenger, J. S. Bachs Notenbibliothek, VBN I/B/48.

300

s a c r e d a n d s e c u l a r : v o c a l wo r k s i i

Ex. 5

glo

ri - a De - i

tris,

in (gloria)

a) Finale of Dresden Missa, ‘Cum Sancto Spiritu’, bb. 25–30 (vocal parts and continuo only)

b) Gigue from Partita No. 4 in D, BWV 828, bb. 89–94

c) Praeambulum from Partita No. 5 in G, BWV 829, bb. 29–31 (cf. also bb. 83–6 with their climactic 6–4–2 chord, as in the two previous examples) existing movements.43 In this case, however, the parody models mostly survive: the majority of the movements are drawn from Leipzig church cantatas of Cycles I (1723–4) and III (1725–7). This in itself casts doubt on the theory that the four Missae were intended primarily for Leipzig use, for the movements concerned would have been performed regularly in Leipzig within their original cantata context whenever the cycle concerned was revived. It is hard to believe that Bach would have transplanted 43 A detailed comparison between parody and original in the four Mass settings is given in Georg von Dadelsen, ‘Anmerkungen zu Bachs Parodieverfahren’, in W. Rehm (ed.), Bachiana et Alia Musicologica: Festschrift Alfred Du¨rr zum 65. Geburtstag (Kassel, 1983), pp. 52–7.

m a gn i f i c a t a n d m i s s a

301

the music other than for use outside Leipzig. The parody models overlap between the four Missae: movements from the same cantata are in three cases (BWV 79, 102, and 179) used in several Mass settings. This, in conjunction with their related conception and overall form, strongly suggests that the four Missae were composed as a series in fairly close temporal proximity to each other.44 The only surviving evidence of their intended order within the series is found in the complete copy by Bach’s pupil and son-in-law J. C. Altnickol (Leipzig, 1744/8), which gives them in the order A–G–g–F (BWV 234, 236, 235, 233). The Missae were formerly undervalued as works hastily put together from existing music, regardless of suitability. Careful inspection of the autograph sources (keys A and G only) and of the music itself in relation to its parody models, however, creates a different impression. It becomes clear that the original German-texted movements were revised with great care to render them suitable for their new liturgical function, and that creating a valid musical correlative for the Latin texts of the Missa (Kyrie and Gloria) was the first requirement. All four Missae have their overall form in common: a single-movement Kyrie made up of three parts—Kyrie I, Christe, Kyrie II (except in key G)—and a Gloria whose outer choruses (‘Gloria in excelsis’ and ‘Cum Sancto Spiritu’) frame three arias with various dispositions of the text. The scoring is restricted to four-part choir (SATB), strings, a pair of woodwinds (key F: plus two horns), and continuo. Thus in terms of performing forces as well as length and number of movements, the four Missae are far more modest than their great predecessor of 1733. This is no doubt a reflection of the different occasions for which they were intended. Whereas the 1733 Missa was no doubt written for an event of quite exceptional significance, probably connected with the accession of the new Elector of Saxony, the Missae of around 1738 were more likely written for performance on the Sundays and feast days of the church year—they were not necessarily tied to a particular occasion but could be revived at will. In the Missa in A there is a very clearly audible link between the Kyrie and Gloria. The solo episodes of ‘Gloria in excelsis’ recall Kyrie I in metre, rhythmic movement, and mode of figuration—gently slurred dotted rhythms in triple time, played by two transverse flutes moving largely in parallel. These passages, more than any others, determine the inimitable character of this Mass setting. ‘Christe’, unusually set as a recitative a 4, recalls the penultimate movement of the Christmas Oratorio, Part VI, first performed about three years earlier (6 January 1735). Both are quasi-fugal recitatives based on a modulating subject (similar in shape), which is passed from one voice to another at the interval of a 4th or 5th (Ex. 6). The oratorio movement is based on a single subject, however, whereas the ‘Christe’ builds a permutation scheme on three subjects, the first of which is also treated in stretto. A similar permutation scheme is found in Kyrie II, though the recitative-fugue of the ‘Christe’ is here

See Emil Platen and Marianne Helms, Krit. Bericht, NBA II/2 (1982), pp. 14–19.

302 sa cred and secular: vocal works ii

Ex. 6

Chri - ste e - lei - son

- lei

e - lei

son

e - lei - son, Chri - ste, Chri - ste

e -

son

a) Opening of ‘Christe’ from Missa in A, BWV 234, B and Tentries (held string chords omitted)

Was will der Höl - len Schrek - ken

nun

Was will uns Welt und

Sün - de

(tun)

b) Christmas Oratorio, Part VI, penultimate movement (No. 63), S and T entries

succeeded by a dance-fugue. ‘Gloria in excelsis’ presents a fourfold alternation of ‘Vivace e forte’ tuttis and ‘Adagio e piano’ solos. This alternating structure of sharply contrasted thematic material was parodied from the aria-chorus ‘Friede sei mit euch’ from the Easter cantata Halt im Geda¨chtnis Jesum Christ, BWV 67 (no. 6), one of the finest achievements of Bach’s first year in Leipzig. The Adagio passages in cantata and Missa are united by their reference to peace: Christ’s ‘Friede sei mit euch’ (‘Peace be with you’) becomes the angels’ ‘Et in terra pax’ (‘And on earth peace’). The Vivace passages, however, formerly associated with the devil’s rage, now become the angels’ song of praise. Yet no incongruity is perceptible here: the concitato string writing and the associated vocal parts are perfectly capable of illustrating either sentiment. The first solo movement, ‘Domine Deus’, whose parody model is not known, is a minor-mode trio whose solo violin part is better suited to God the Son (lines 3–6)

m a g n i f i c a t a nd m i s s a

303

than to the Father (lines 1–2). ‘Qui tollis’ was parodied from the Cycle I cantata Siehe zu, daß deine Gottesfurcht nicht Heuchelei sei, BWV 179 (no. 5), but radically revised in the process. In particular the continuo part was replaced by a bassett for unison upper strings, presumably to symbolize the loss of a foundation in God caused by ‘the sins of the world’. The music is equally suited to its German and Latin texts, both of which represent an appeal for mercy on account of human sinfulness. ‘Quoniam tu solus sanctus’ was parodied from the Reformation Festival cantata Gott der Herr ist Sonn und Schild, BWV 79 (no. 2), of 1725. Only a generalized spirit of praise is common to the German and Latin texts, but that is sufficient to render the movement perfectly appropriate in its new context. Numerous changes were made for the Missa version. In particular, the main theme in both its vocal and instrumental forms is subjected to further elaboration. The choral finale ‘Cum Sancto Spiritu’ is parodied from the first movement of the Cycle I cantata Erforsche mich, Gott, und erfahre mein Herz, BWV 136. The original German text, from Psalm 139: 23, bears little relation to the Latin Gloria text. Nevertheless, the music is well suited to both the sense and the speech-rhythms of the Latin words. Two important changes were made. The movement was prefaced by a new slow introduction, which provided a modulatory link with the preceding ‘Quoniam’, aiding continuity; and the framing ritornellos of the cantata movement were excised. Both of these alterations are in line with Bach’s parodying procedures in the 1733 Missa. Built into the overall conception of the Missa in G is that stark juxtaposition of motet and concertante styles that has already been encountered in the Missa of 1733. Kyrie–Christe–Kyrie are here integrated within a single continuous movement, a counter-fugue (inversion fugue) in which fugal expositions of the main subject (Kyrie) alternate with canonic or imitative episodes on a secondary subject (Christe). The two subjects are united in the conclusion. The movement is written in a traditional motet style with alla breve metre and colla parte instruments, though independent continuo. It is parodied from the opening chorus of Siehe zu, daß deine Gottesfurcht nicht Heuchelei sei (BWV 179), whose text (from Ecclesiasticus 1: 34) reads: ‘See to it that your fear of God be not hypocrisy, and do not serve God with a false heart.’ Though the theme of hypocrisy is far removed from the Kyrie, Bach’s chromatic descent on ‘falschem’ (‘false’) is utterly convincing when transferred to the plea for mercy of ‘eleison’. After the motet-style Kyrie, ‘Gloria in excelsis Deo’ is one of Bach’s greatest concertante choruses, parodied from the first movement of Gott der Herr ist Sonn und Schild (BWV 79). The original psalm text (Psalm 84: 11), ‘God the Lord is sun and shield’, yields music eminently well suited to the new Gloria text. The movement is somewhat diminished by the omission of the original horns and timpani from the instrumental ensemble. The immense integral sinfonia is retained in full, but in an audacious stroke the horn duet with which it originally began becomes a soprano–alto duet to the text ‘Gloria in excelsis Deo’. ‘Gratias agimus tibi’ is parodied from the Cycle I cantata Warum betru¨bst du dich, mein Herz, BWV 138, no. 4. The original

304 s a c r e d a n d s e c u l a r : v o c a l w o r k s i i

Ex. 7

a1.) Opening violin theme of ‘Gratias agimus’ from Missa in G, BWV 236 (parodied from BWV 138 no. 4, 1723) 2.) Opening horn theme of ‘Quoniam’ from Dresden Missa of 1733

Gra

ti - as

gi - mus

ti - (bi)

b 1.) ‘Gratias agimus’ from Missa in G, first bass entry, bb. 21–2 (continuo omitted) 2.) Subsidiary theme from Sinfonia No. 13 in A minor, BWV 799, bb. 21–2(upper parts omitted) German text, ‘Auf Gott steht meine Zuversicht’ (‘In God lies my confidence’), is replaced by Latin words of thanksgiving, to which Bach’s music is equally well suited. The headmotive of the ritornello theme in its revised form is identical with that of the ‘Quoniam’ from the 1733 Missa (Ex. 7a). Moreover, the revised solo bass part has a florid variant of the ritornello theme reminiscent of the haunting subsidiary theme from the Sinfonia in A minor, BWV 799 (Ex. 7b). The soprano–alto duet ‘Domine Deus’ is drawn from the same cantata as the opening chorus of the Gloria, namely BWV 79 (no. 5). The German text is a prayer not to be forsaken; the Latin, a prayer for mercy. The minor-mode duet writing, with much recourse to parallel 3rds and 6ths, creates a plaintive, pleading tone appropriate for both texts. For the Missa version Bach prefaces the movement with a four-bar continuo introduction which, however, does not recur thereafter. The unison violins’ ritornello is utterly transformed for the Missa, becoming far more ornate and deeply expressive. The ‘Quoniam’ is parodied from the same cantata as the Kyrie, BWV 179 (no. 3). Originally a fully scored piece, it is here scaled down to tenor, oboe, and continuo, perhaps on account of the repeated word ‘solus’. The beautiful, intricately wrought obbligato originally represented the outward beauty that hides inner ‘filth’. In the later version, however, without any incongruity musical beauty is taken literally to represent the Son of God. The finale is prefaced by a new six-bar slow introduction to the words ‘cum Sancto Spiritu’. The following fast movement is parodied from the psalm chorus that opens the Cycle III cantata Wer Dank opfert, der preiset mich, BWV 17. Being a chorus of praise and thanksgiving, it is perfectly suited to its new

m a g n i f i c a t a nd m i s s a

305

context. The extensive opening ritornello of the German version is omitted altogether, a measure familiar from Bach’s parody process in the 1733 Missa. In addition, each fugal entry is now prefaced by a tutti homophonic refrain to the words ‘in gloria Dei Patris’. The Missa in G minor, BWV 235, opens with a parody of one of Bach’s greatest biblical-text choruses, the introductory movement of the Cycle III cantata Herr, deine Augen sehen nach dem Glauben, BWV 102. The tripartite structure of this movement renders it ideal for the threefold prayer of the Kyrie. That the original biblical text (Jeremiah 5: 3) deals with impenitence—virtually the opposite of the Kyrie text—hardly affects the suitability of the music for the Kyrie, though the two fugue subjects inevitably lose their illustrative connotation. The three prayers of the Kyrie each receive quite different settings, the second and third being fugal but on different subjects. All three, however, are incorporated within an overall ritornello structure that guarantees unity. Thus motet and concertante principles are brought into play simultaneously. ‘Gloria in excelsis’ is adapted from the madrigalian-text chorus that opens the Cycle III cantata Alles nur nach Gottes Willen, BWV 72. As so often in these Missa parodies, the opening instrumental ritornello is omitted for the sake of concision and to maximize the impact of the sudden entry of the word ‘Gloria’. The liturgical hymn of the Gloria (from ‘Gratias agimus tibi’ to ‘Cum Sancto Spiritu’) is a parody of the Cycle III cantata Es wartet alles auf dich, BWV 187, virtually in its entirety, encompassing the opening chorus and all three arias, and omitting only the inappropriate recitatives and chorale. The original model of the ‘Gratias’ is a vox Christi solo, whose music is no less fitting for its new context. The third vocal solo (b. 81) is extended from eight to 20 bars. An even greater expansion takes place in the second and third paragraphs of the reprise-form ‘Domine Fili’. ‘Qui tollis’ is a bipartite movement (AB) with the greatest imaginable contrast in metre, tempo, and general character between its two sections. The stately dotted rhythms of the A-section, originally associated with the thought that ‘God takes care of all life’, now seem to portray the majesty of the Son enthroned in heaven (‘Qui sedes ad dexteram Patris’), though the florid oboe melismas perhaps illustrate his gentler attributes. The quick B-section in 3/8 was apt for its original text, ‘Retreat, you cares’, but it is less easy to see how it is justified in relation to the Gloria text ‘For You only are holy/the Lord/the most high Jesus Christ’. The finale is an ABA1 reprise structure with outer concertante paragraphs framing a central fugue. As in the opening chorus of the Gloria, the original opening ritornello is omitted for the sake of immediacy and concision. The opening vocal phrase, originally sung twice, is now sung three times in different keys in order to accommodate the three portions of text—‘Cum Sancto Spiritu’, ‘In gloria Dei Patris’, and ‘Amen’—and to provide a tonal link with the E♭ tonic of the previous movement (E♭–c–g). As far as we know, the Kyrie from the Missa in F, BWV 233, is the only movement from the four Missae that is derived from an original with the same text, namely the Kyrie in F, BWV 233a. This is one of Bach’s most unusual creations—a motet-style alla breve piece for five voices (SSATB) and continuo with double (Latin and German)

306 s a c r e d a n d s e c u l a r : vo c a l w o r k s i i cantus firmus. The outer vocal parts both carry liturgical chants: soprano I sings the German Agnus Dei, ‘Christe, du Lamm Gottes’ (1528), while the bass sings the Latin Kyrie from Luther’s German Litany (1529). Since these two chants are based on the same plainsong, Psalm Tone I, a loosely imitative relationship between them often arises. The non-cantus firmus carrying voices are strictly fugal according to the following scheme: Kyrie I: exposition of subject rectus Christe: exposition of subject inversus Kyrie II: exposition of subject rectus + inversus It is not known when Bach composed this remarkable piece (all original sources are lost). The Mu¨hlhausen and Weimar periods have been suggested, but neither seems plausible since no real parallels may be found among Bach’s compositions at those early stages of his career. As a set of three variations on the German Agnus Dei, the nearest parallel is the movement that Bach employed as the finale of both the Quinquagesima cantata No. 23 (1723) and Version II of the St John Passion (1725). A parallel for the use of the German Agnus Dei in combination with a second cantus firmus may be found in the opening movement of another Quinquagesima cantata, No. 127 of 1725. Thus the first Leipzig years, 1723–5, seem a more likely place to look for the origin of the Kyrie/Agnus in F, BWV 233a. For the Missa version the vocal parts were reduced to four, the Agnus Dei chant was no longer sung but played as an instrumental cantus firmus by two horns and two oboes in unison, and the voices were doubled by colla parte strings and bassoon. As in the G major Missa, a sharp contrast is heard between the older motet style of the Kyrie and the more modern, concertante style of the Gloria. ‘Gloria in excelsis Deo’, of which a parody model is assumed but not known, falls into an overall ABA1 reprise structure that incorporates both ritornello and fugue. The opening horn theme is remarkably similar to the initial trumpet theme of the equivalent movement from the 1733 Missa. The structure of ‘Domine Deus’ was probably reduced from the original ABA da capo form of the presumed parody model to the bipartite form AB— we have had occasion to notice similar truncation procedures in the 1733 Missa. A triadic, majestic ritornello/bass solo theme conveys the almighty power of God the Father, whereas the contrasting pathos of minor keys and a more florid style are employed for God the Son. The movement ends in the dominant G in preparation for the G minor of ‘Qui tollis’, which follows without a break. ‘Qui tollis’ and ‘Quoniam’ form a contrasting pair—the one ‘Adagio’, the other ‘Vivace’—parodied from the two arias of the Cycle III cantata BWV 102 (nos. 3 and 5). The original model of ‘Qui tollis’ is one of Bach’s most remarkable arias, ‘Weh der Seele’, a graphic portrayal of the lost soul, whose tone is set by the wayward, irregular, and rhapsodic ritornello for solo oboe and continuo. Such music injects a note of anguish into the setting of the Latin text, with its reference to the ‘peccata mundi’ (‘sins of the world’) and its

oratorio

307

prayer ‘miserere nobis’ (‘have mercy on us’). In the ‘Quoniam’, a parody of the aria ‘Erschrecke doch’, there is no apparent link with the original text, which deals with the fear of divine retribution. However, this subject matter does not appear to affect the ritornello but only the free vocal entry with its graphic setting of ‘Erschrecke doch!’ (‘Fear, then!’), which in the later version is completely remodelled in a style appropriate to the Latin words. The finale, ‘Cum Sancto Spiritu’, is parodied from the first movement of the Christmas cantata Dazu ist erschienen der Sohn Gottes, BWV 40 (Cycle I), which, however, is radically restructured for its new purpose. As in ‘Domine Deus’, the ABA1 form is reduced to AB, though here by omitting the opening A-section. In addition, the opening ritornello is reduced to a quarter of its original length. All this takes place in the interests of concision, but problems remain. The movement is tonally centred on the key relation I/IV (tonic/subdominant), but it is by no means clear that the tonic is sufficiently freed from the subdominant to be fully re-established at the close. Moreover, the very brief ritornello that remains from the restructuring enters only twice (bb. 1 and 72) and its material otherwise plays no part in the proceedings. For these reasons the movement cannot really be regarded as one of Bach’s most successful adaptations.

Oratorio Title, occasion

Earliest source/s

Scribe, date

Weihnachts-Oratorium (Christmas Oratorio), BWV 248, Christmas Day to Epiphany Himmelfahrts-Oratorium (Ascension Oratorio), BWV 11, Ascension Day Oster-Oratorium (Easter Oratorio), BWV 249, Easter Sunday

Berlin, P 32, St 112

Autograph, part-autograph, for 25 Dec. 1734 – 6 Jan. 1735

Berlin, P 44/4, St 356

Autograph, part-autograph, for 19 May 1735 Autograph, part-autograph, 1735?

Berlin, P 34, St 355

Bach’s three oratorios belong to a series of large-scale sacred works from the 1730s, including the St Mark Passion and the ﬁve Missae, that originated to a large extent through the parodying of existing compositions. Although the oratorios superﬁcially resemble cantatas—the six parts of the Christmas Oratorio are each hardly longer than a cantata and are similarly made up of choruses and chorales, recitatives and arias—they were clearly modelled not on the cantatas but on the Passions. In conjunction with his librettist—presumably Picander—Bach conceived the idea of applying the framework of the oratorio-Passion to three of the four Lutheran High Feasts—Christmas, Easter, and Ascension45—so that each of the main events in the life of Christ and his Apostles should be celebrated in a major composition of 45 The question has been raised whether Bach might also have written a Whit Oratorio, i.e. one for the fourth Lutheran High Feast, which is now lost; see Du¨rr, The Cantatas of J. S. Bach, p. 44.

308

sacred and secular: vocal works ii

imposing dimensions. Length was restricted, however: whereas the Passions could extend to two or three hours due to conditions at the Good Friday Vespers, the oratorios had to be performed in place of a cantata at the Sunday-morning Hauptgottesdienst (main service), during which little over half an hour could be devoted to concerted music. In the case of the Christmas Oratorio Bach circumvented this restriction by dividing the work into six parts, each to be performed on a different occasion during the Christmas season. In Passion and oratorio alike, the foundation of the work for Bach was the Gospel narrative as declaimed in secco recitative by the tenor Evangelist. In both cases other characters participate where necessary—in the Christmas Oratorio, an angel or Herod in recitative (nos. 13 and 55) and the angels, shepherds, or wise men in turbae, or crowd choruses (nos. 21, 26, and 45). And in both cases other kinds of text—hymn strophes and madrigalian verse—are ancillary to the central biblical narrative. These secondary texts are sung in a similar fashion in the two genres: freely composed verse as choruses, accompanied recitatives, ariosos, and arias; hymn texts as four-part chorales or more elaborate chorale arrangements. Furthermore, as in the Passions, the recurrence of certain chorale melodies is employed as a means of creating aural recognition and coherence. Herzlich tut mich verlangen, for example, is the first and last chorale in the entire Christmas Oratorio; Luther’s Vom Himmel hoch binds together Parts I and II (nos. 9, 17, and 23); and Gelobet seist du, Jesu Christ, also by Luther, performs the same function in Parts I and III (nos. 7 and 28). The Christmas Oratorio, the first, grandest, and most important of the three oratorios, was composed for performance on the three successive Christmas feast days in 1734 (Parts I–III: 25–7 December) and on New Year’s Day, the Sunday after New Year, and Epiphany in 1735 (Parts IV–VI: 1, 2, and 6 January). This performance schedule is reflected in the music. Parts I–III, for the three Christmas feast days, exhibit the overall key scheme D–G–D, with festive trumpets and drums in the tonic D framing the pastoral, subdominant second part, with its pronounced woodwind scoring. Parts IV–VI, performed in the New Year, exhibit a quite different key scheme, F–A–D. Part IV, performed on New Year’s Day, represents a new departure, both in key (F being unrelated to the overall tonic D) and in the participation of horns. Part V, which introduces the Wise Men, balances the Shepherds of Part II both in key (II: a 5th below the tonic; V: a 5th above) and in its woodwind scoring. Finally, Part VI, performed on the Feast of the Epiphany, returns to the festive D major with trumpets and drums of Parts I and III. This concluding part was adapted from a lost sacred cantata, BWV 248/VIa, which might have been composed for Michaelmas (29 September) 1734.46 In Parts I–V, the biblical-text settings and chorales were newly composed for the oratorio, as were the sinfonia and the accompanied recitatives. The madrigalian choruses and arias, however, with only two exceptions (nos. 31 and 46 According to Andreas Glo¨ckner, ‘Eine Michaeliskantate als Parodievorlage fu¨r den sechsten Teil des Bachschen Weihnachts-Oratoriums?’, BJ 86 (2000), pp. 317–26.

or ato ri o

309

51), were parodied from secular cantatas written about a year earlier (late 1733) for the Electoral House of Saxony. Six movements are drawn from the Hercules Cantata, BWV 213 (nos. 4, 19, 29, 36, 39, 41), four from the Queen’s Cantata, BWV 214 (nos. 1, 8, 15, 24), and one from the Polish Cantata, BWV 215 (no. 47). Since almost all of the madrigalian movements in Cantatas Nos. 213 and 214 were adapted, and in view of the short interval between original and parody, we cannot exclude the possibility that these pieces were already earmarked for parody at the time of their composition. We are here presented with a classic case of Bach’s bringing occasional music, which had served its purpose at its original performance, into permanent form. The festive Part I celebrates the birth of the Christ child. The opening chorus of praise in da capo form was transferred with little change beyond the words from the Queen’s Cantata of the previous year. The focus of praise is no longer the Queen of Poland but the divine ruler who has brought about the Incarnation. The Christmas text is less specific than the original, which mentions drums, trumpets, and strings in the order of their entry in Bach’s setting. The alto aria ‘Bereite dich, Zion’, no. 4, has a quite different affect from its parody model (Hercules Cantata, no. 9), where Hercules indignantly rejects ‘depraved Pleasure’. In the Christmas version, Zion is exhorted to prepare ‘with tender desire’ to see the ‘fairest, dearest’ Saviour. In the original, the obbligato violins’ part is marked ‘unisoni e staccato’, imparting a vigorous, forceful character to the music. In the adapted version, on the other hand, the quavers are slurred in pairs, giving a graceful, minuet-like character to the obbligato part. ‘Er ist auf Erden kommen arm’, no. 7, is a chorale trope, a type often cultivated in Bach’s cantatas of the 1720s. Ritornellos surround and accompany the chorale lines, sung by soprano, which alternate with bass accompagnato. The music has a pastoral flavour— it follows an account of Jesus’s birth among farm animals—not least due to its scoring for two oboes d’amore and continuo. The bass aria ‘Großer Herr’, no. 8, a da capo movement written in a ‘modern’ dance-like 2/4 with syncopated rhythms, is adorned by a splendid trumpet obbligato, which originally (Queen’s Cantata, no. 7) illustrated Fama’s promise to carry the queen’s renown throughout the whole world but now stands for divine sovereignty. There is one significant shortcoming of the parody: the paradox of the Godhead’s embracing at once the ‘mighty King’ of all creation and the baby Jesus, who ‘must sleep in a hard crib’, is nowhere reflected in Bach’s setting. The trumpet-choir refrains of the concluding four-part chorale refer not to the words—verse 13 of Luther’s Vom Himmel hoch—but rather to the opening chorus, creating a festive frame for Part I as a whole. Here, then, structural considerations take precedence over the text. The story of the encounter of shepherds and angels in Part II elicits some of the finest, most varied, and specifically seasonal music of the entire oratorio, including the pastoral sinfonia, no. 10, the cradle song, no. 19, and the angelic hymn, no. 21. The sinfonia, with its substantial woodwind participation and its dotted rhythms in compound time, essentially takes the form of a pastorale. However, it has long been thought that an angel choir participates too (flutes and strings) in antiphony with the

310

sacred and secular: vocal works ii

rustic choir of shepherds (oboes).47 The music falls into an overall reprise form (ABA1), each paragraph consisting of ritornello plus episode. In the recitative, no. 13, the angel is sung by soprano; in no. 16, however, by the tenor Evangelist—an inconsistency for which there is no obvious explanation. The music of the tenor aria ‘Frohe Hirten’, no. 15, works equally well for Pallas and the Muses celebrating the queen’s birthday with ‘spontaneous enthusiasm’ (BWV 214 no. 5) and for the ‘joyful shepherds’ hastening to see the newborn child, though one misses Bach’s usual style of characterizing the word ‘eilt’ (‘hasten’). The extremely florid melismas in the second and third solos render this aria perhaps the most operatic in the whole oratorio. The alto aria ‘Schlafe, mein Liebster’, no. 19, is a lullaby, just like its parody model in the Hercules Cantata (no. 3), where Wollust (Pleasure) tries to lull the young hero to sleep in order to divert him from the true path. Since temptation from virtue cannot be illustrated in musical terms, there is no difficulty in using the same music as a cradle song for the Christ child. The angelic hymn ‘Ehre sei Gott’, no. 21, is a motet-style movement with a quite different setting for each of its three text-phrases: imitative (‘Glory be to God’), invertible counterpoint (‘peace on earth’), and fugal/canonic (‘goodwill towards mankind’). The three themes and their working-out are briefly restated in the same order, giving rise to the overall form ABC, A1B1C1. The finale has much in common with that of Part I: both comprise a four-part choral setting of the chorale melody Vom Himmel hoch with instrumental refrains. Here, however, the refrains are played by woodwind and allude to the pastoral sinfonia, not merely in scoring but in thematic material, returning to the rustic oboes’ theme that conjures up the shepherds with their pipes. Part III is linked to Part I in its festive trumpets-and-drums scoring and to Part II as a continuation of the shepherds’ story. As Part I had begun with the first movement of the Queen’s Cantata, so Part III begins with its finale. This is cast in binary dance form with repeats, common in the finales of Bach’s secular cantatas but rare in his first movements. During the solo episodes that intervene between the instrumental strain and its tutti repeat, each line of text is sung to different music by tenor, soprano, and alto in turn. In the parody model, these solos were sung by different characters, each with their own words, but this raison d’eˆtre no longer exists in the Christmas version and for Bach the non-imitative entries sound strangely unconvincing. The chorus of shepherds, no. 26, cannot be differentiated in style from the turba choruses of the Bach Passions. The swift obbligato for unison violins and flutes no doubt signifies the haste of the shepherds as they journey to Jerusalem. The soprano–bass duet ‘Herr, dein Mitleid’, no. 29, was originally a love duet between Hercules and Virtue (BWV 213 no. 11) with much writing in parallel 3rds, 6ths, and 10ths in keeping with its amorous theme. In the Christmas version the subject is no longer love but divine mercy and compassion; hence the justification for the operatic love-duet style is no longer clearly

Alfred Du¨rr, following Albert Schweitzer; see Du¨rr, The Cantatas of J. S. Bach, p. 120.

o r a t o r io

311

apparent. The alto aria ‘Schließe, mein Herze’, no. 31, is possibly the only newly composed aria in the entire oratorio. The very inward theme of the text, ‘Enclose, my heart, this blessed miracle [the Incarnation] firmly within your faith’, might have been deeply personal for Bach, hence perhaps the great care he lavished on this movement, even down to its detailed articulation marks. Part III concludes not with a four-part chorale but with a reprise of the opening chorus; the intention was clearly to give it a festive frame, just like Part I and indeed the entire triptych (Parts I–III) for the three Christmas feast days. Part IV, which opens the second half of the oratorio, introduces an entirely fresh colouring by virtue of its F major tonality and its use of horns. Like Part I, it is concerned with Jesus himself—in this case, with his name. The original subject of the opening chorus, the pagan ‘son of the gods’ (‘Go¨ttersohn’) Hercules (BWV 213), now becomes the Christian ‘Son of God’ (‘Gottes Sohn’). The ‘caressing’ and ‘alert’ figures with which Bach describes the gods’ caring for and watching over the young Hercules are in the Christian version successfully transferred to the gestures of kneeling ‘with thanks’ and ‘with praise’. The graceful dance rhythms are equally well suited to the caring and prayerful texts. Movements nos. 38–40 form a central triptych concerned with the name of Jesus: chorale trope—aria—chorale trope. The Stollen of Rist’s Jesu, du mein liebstes Leben, verse 1 (1642), are sung before the aria and the Abgesang after it. The general form of the chorale trope is similar to that of Part I no. 7: the chorale is sung as soprano arioso and the troping recitative as bass accompagnato. Here, however, the bass continues to sing during the chorale lines, forming a soprano–bass duet. Moreover, the chorale is in this case unusually florid and might be an original melody of Bach’s. The echo aria that interrupts the chorale is parodied from Cantata 213 no. 5, where Hercules asks ‘Faithful Echo’ to confirm his decision in favour of virtue. Here, a convention of Baroque opera converges with a tradition of church poetry and music that reaches far back into the seventeenth century.48 In the aria Jesus echoes the thoughts of the Christian at the same soprano pitch. The tenor aria ‘Ich will nur dir zu Ehren leben’, no. 41, adapted from BWV 213 no. 7, illustrates an often unavoidable pitfall of parody, namely the change from precise image, clearly reflected in the music, to generalized sentiment. Thus in the secular original the melismas represent ‘hovering’ and ‘climbing’ on wings; in the sacred version, they are merely sung to the words ‘leben’ (‘live’) and ‘Kraft’ (‘strength’). Parts IV and VI, the framing parts of the second half of the oratorio, both present full instrumental ritornello structures as a framework for their concluding chorales, a type of chorale-finale relatively rare in Bach’s Leipzig church music. In Part IV a clear link is established between the choralefinale and the central chorale-trope: both are based on Rist hymns (in this case, verse 15 of his Hilf, Herr Jesu, laß gelingen of 1642) and in both cases the melodies, which have certain features in common, are thought to have been composed by Bach himself. 48 See Ernst Koch, ‘Tro¨stendes Echo: Zur theologischen Deutung der Echo-Arie im IV. Teil des Weihnachts-Oratoriums von J. S. Bach’, BJ 75 (1989), pp. 203–11.

312 s a c r e d a n d s e c u l a r : v o c a l w o r k s i i Whereas Parts II–III are concerned with the Shepherds, their equivalents in the second half, Parts V–VI, are devoted to the Wise Men. Bach originally planned to open Part V with the gavotte-finale of the Hercules Cantata (BWV 213 no. 13), but in the event he might have considered it too lightweight. Consequently he composed a new chorus that exhibits a more complex and weighty structure: the A-section of the da capo scheme is itself tripartite (aba1) and incorporates both fugue and ritornello with Choreinbau (choral insertion). The chorus of Wise Men, no. 45, corresponds with the chorus of Shepherds in Part III (no. 26), both belonging to the same type as the turba choruses of the Passions. Indeed, it has been suggested that the chorus of Wise Men might have been parodied from the lost St Mark Passion.49 The Wise Men seek the Christ child in a bipartite chorus, of which each half is followed by a commentary in the form of an alto accompagnato giving the inner viewpoint, ‘Seek Him in my breast’. The original secular version (BWV 215 no. 7) of the bass aria no. 47 has a bassett to symbolize Augustus’ divine quality of repaying ‘malice with kindness’. Since there is no such motivation for a bassett in the Christmas text, Bach restores the usual basso continuo, simultaneously transposing the movement down a 4th from b to f♯. The lower pitch is better suited to the ‘dark thoughts’ that Christ is called upon to illuminate. The terzetto, no. 51, for which no parody model is known, is remarkably close in several respects to the alto aria no. 31, which was newly composed for the oratorio. Above all, in both cases Bach employs a thematic style with syncopated rhythms in a dance-like 2/4 that is found in some of his most ‘fashionable’ music. The terzetto is notable for its dramatic conception: the alto repeatedly tries to silence the soprano–tenor duet. In paragraph A of the ABA1 reprise structure, the duet partners ask, ‘When will Christ appear?’, to which the alto replies, ‘Be silent; He is already here’. The duet partners then accept this message in paragraph B, which makes nonsense of the varied reprise A1 that follows. A possible explanation is that the movement was parodied from a lost source in which the reprise was justified by the text.50 Bach reserved one of his most powerful sacred compositions for Part VI of the oratorio, performed on the Feast of the Epiphany. This final part was adapted not from secular cantatas, like Parts I–V, but from a complete sacred cantata (only the Evangelist’s recitatives and the intermediate chorale are new).51 The opening chorus, no. 54, is one of the strongest, most massive and well integrated of all Bach’s cantata choruses. It is a combination of fugue, ritornello, and reprise forms (ABA1), in which A + B and A1 each consist of an expanded ritornello reprise. The soprano aria no. 57 is exceptional in being laid out as a complete dance. Not only is the dance rhythm that of a polonaise but the whole aria is cast in an expanded binary dance form: essentially,

49 See Ortwin von Holst, ‘Turba-Cho¨re des Weihnachts-Oratoriums und der Markuspassion’, Musik und Kirche, 38 (1968), pp. 229–33. 50 Du¨rr believed that the movement was a parody; see his The Cantatas of J. S. Bach, p. 172. 51 This conclusion rests on the evidence of the original performing parts, Berlin St 112.

oratorio

313

each strain is first played then sung, after which both strains are played in succession as a concluding ritornello. The tenor aria no. 62 is a dance-like piece in 2/4, a type sprinkled fairly liberally throughout the oratorio (nos. 8, 19, 31, 47, 51, and 62). The penultimate movement, no. 63, sums up the sentiments of Part VI—‘What will the world and sin do to us, since we rest in Jesus’s hands?’—in a quite exceptional form: a thematic, imitative recitative for four voices (SATB) and continuo. The only real parallel is to be found in the Christe from Bach’s Missa in A, BWV 234. The concluding chorale, like that of Part IV, is embedded within an elaborate, independent instrumental texture, but this time on a far larger scale. The splendid ritornello is led by a brilliant obbligato part for first trumpet. The chorale melody, Herzlich tut mich verlangen, harks back to the first chorale in the oratorio, no. 5, which employed the same melody, albeit to words from a different hymn. As a result of this magnificent finale not only does Part VI have a powerful frame but so too does the entire oratorio. The Ascension Oratorio Lobet Gott in seinen Reichen, BWV 11, was probably composed for performance on 19 May 1735—that is, during the same church year as the Christmas Oratorio. It is hardly surprising, therefore, that the two works have many features in common: the festive opening chorus with trumpets and drums, the Evangelist’s secco recitative, the intermediate four-part chorale, the meditative accompagnati, the semi-dramatic treatment of biblical characters (here the ‘two men in white’), and the elaborate chorale-finale. The Ascension Oratorio, however, had to be performed on a single occasion, with the result that it corresponds in length and design with only one part of the Christmas Oratorio. The closest analogy is with Part VI, on which it might even have been modelled. Both compositions open with a powerful, festive chorus with trumpet choir, cast in ABA1 reprise form, though here without the fugal element in keeping with its secular parody model (the lost BWV Anh. I 18, no. 1),52 which in turn explains the ‘modern’ dance-like 2/4 metre with Lombard rhythms. In both cases the finale consists of a chorale embedded in an elaborate instrumental setting, derived from a ritornello of great power and of considerable extent. And in both finales a minor-mode chorale melody (here Von Gott will ich nicht lassen) is incorporated in a major-mode setting. Finally, in both cases the soprano aria (Ascension no. 8, Christmas no. 57) is not only in dance rhythm—here that of a 3/8 minuet—but in binary dance form, in this case within an overall da capo scheme. In the Ascension movement both the voice and the obbligato instruments, unison flutes and oboe, are pitched high and the continuo is replaced by a bassett for unison violins and viola. Presumably this represents the ascended Christ and His ‘glances of grace’ (‘Gnadenblicke’). Both this movement and the alto aria no. 4 were parodied from a lost secular wedding cantata, Auf! Su¨ß entzu¨ckende Gewalt (BWV Anh. I 196), with libretto by Gottsched, performed on

52 Or possibly a birthday cantata for J. W. C. Dumpff, with text by Picander; see Hans-Joachim Schulze, ‘Johann Sebastian Bachs Himmelfahrts-Oratorium und Picanders Geburtstagkantate fu¨r “Herrn J. W. C. D.”’, BJ 95 (2009), pp. 191–9.

314

s a c r e d a n d s e c u l a r : v o c a l wo r k s i i

27 November 1725. The alto aria is less well known than the much later and more radically altered adaptation of the wedding piece that Bach employed as the Agnus Dei of the B minor Mass (c. 1748/9). In the Christmas and Ascension Oratorios alike, the oratorio-Passion format is applied to other major church festivals. The Easter Oratorio, however, is fundamentally different. It is best described as a sacred dramma per musica. Its music originated in a secular dramma per musica—the lost pastoral cantata Entfliehet, verschwindet, entweichet, ihr Sorgen, BWV 249a, composed for the birthday of Duke Christian of Saxe-Weißenfels on 23 February 1725. Only just over five weeks later, on 1 April 1725, a sacred parody was performed—the Easter cantata Kommt, gehet und eilet, BWV 249. Perhaps in 1735 a revised version of this cantata was presented as the Easter Oratorio. This is the only known case in which the operatic nature of the secular original was fully maintained in the sacred parody. A plot that honoured Duke Christian became one that hailed the risen Christ. The shepherds and shepherdesses of ancient mythology became the disciples and followers of Jesus. Above all, the plot continued to be put into the mouths of the dramatis personae, as on the operatic stage, rather than narrated by an Evangelist. No concession is made to Lutheran tradition by adding biblical-text movements or chorales. It seems surprising that the Leipzig church authorities accepted such an overtly secular work. Yet there seem to have been at least four performances during Bach’s lifetime. Perhaps the work was viewed in terms of the old custom of scenic representation of the Easter story. Bach presumably saw its dramatic form as a justification for subsuming it under the oratorio genre. He was clearly prepared to overlook the fact that it bore no relation whatsoever to his two previous oratorios. The first two movements, [Allegro] and Adagio, form a concertante sinfonia, of which the duet (later chorus) no. 3 perhaps stands in lieu of finale. However, the old view that these three movements originated in a lost concerto is no longer accepted by scholars.53 The duet/chorus no. 3 is equally suited to the secular words ‘Flee, vanish, fade, your cares’ and to the sacred ‘Come, hasten, and run [to find the resurrected Jesus]’. The lovely, florid soprano aria with obbligato flute, no. 5, is so full of fine detail that it aptly expresses Doris’s words in the pastoral cantata ‘A hundred-thousand compliments [to Duke Christian] well up now in my breast’; but its connection with Mary’s ‘O soul, your spices shall no longer be myrrh’ remains elusive. Again, the tenor aria no. 7, one of Bach’s loveliest cradle songs, with its charming accompaniment of muted violins doubled by recorders at the upper octave, is perfectly suited to Menalcas’ ‘Rock yourselves to sleep, you satiated sheep’ but less so to Peter’s ‘My deathagony shall be but a slumber, O Jesus, through your napkin’. In the alto aria no. 9, with its concertante ritornello—half tutti and half oboe d’amore solo—the secular text,

53 See J. Rifkin, ‘Verlorene Quellen, verlorene Werke: Miszellen zu Bachs Instrumentalkomposition’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke (Witten, 1997), pp. 59–75 (esp. 74, n. 57); and S. Rampe and D. Sackmann, Bachs Orchestermusik: Entstehung, Klangwelt, Interpretation (Kassel, 2000), p. 466 n. 2.

oratorio

315

Sylvia’s ‘Come then, Flora, come quickly’, and the sacred, Mary Magdalene’s ‘Tell me quickly where to find Jesus’, satisfactorily converge. The choral finale, which originally wished Duke Christian ‘good fortune and well-being’, becomes a song of praise and thanksgiving to the divine Lord. It is a bipartite movement (AB), close in structure and rhythmic movement to the Christmas Sanctus of 1724 that was later incorporated in the B minor Mass. The principal section is cast in binary dance form with repeats, common in Bach’s secular finales. After an 18-bar episode there is no return, however, but rather new music in a different metre and tempo. This is unfortunately too short to balance the main section and delays the tonic return till too close to the end.

II.5 Conclusion

At the outset of Bach’s first term as director of the Collegium musicum, in 1729, Der Streit zwischen Phoebus und Pan (BWV 201) shows him defining his refined style against what he saw as the relatively coarse style advocated by certain contemporaries. This enterprise demanded a high degree of self-confidence, and that is clearly exhibited in his compositions of the 1730s. In the field of the cantata we observe a large measure of continuity with the late 1720s: the solo cantata, whether sacred or secular (BWV 51 and 209); the sinfonia made out of imported instrumental music and featuring obbligato organ (BWV 120a and 29); the dialogue between Jesus and the Faithful Soul (BWV 140); and the chorale cantata per omnes versus which, having being employed sporadically in the mid-1720s (BWV 107, 137, and 129), now became Bach’s standard type. However, new cantata types are also introduced into his repertoire: the secular comedy, which resembles a miniature comic opera (BWV 211); and the ‘semi-chorale cantata’, as it might be termed, which includes an entire chorale text (or, at the very least, the two outer verses plus one inner verse) intermingled with madrigalian-text arias and recitatives (BWV 140, 36, 14, and 80). In genres other than the cantata Bach ventured still further into new territory in the 1730s: in the concert en ouverture, the Sonate auf Concertenart, the organ sonata, the harpsichord concerto, the Missa, and the oratorio. The concert en ouverture, a cross between French overture and Italian concerto, must have been considered fashionable in the 1730s, since it was cultivated by the leading progressive G. P. Telemann and described in some detail by arch-modernist J. A. Scheibe. It may be regarded as a manifestation of the vermischte Geschmack (‘mixed taste’), with which Bach is likely to have been acquainted at least since his Dresden visit for the contest with Marchand in 1717.1 He contributed two works to this sub-genre, the Ouverture in D, BWV 1068, and the Ouverture in B minor, BWV 1067. Scheibe also penned a description of the fashionable Sonate auf Concertenart, a cross between the sonata and the concerto. Here, not only is the standard three-movement form of the concerto applied to the sonata, but so too is the concertante style of writing normally associated with the large-ensemble genre. In three of these hybrid sonatas, BWV 1029, 1030, and 1032,

1 See Ulrich Siegele, ‘Bachs vermischter Geschmack’, in M. Geck and K. Hofmann (eds.), Bach und die Stile (Dortmund, 1999), pp. 9–17.

c o n cl usi on

317

Bach employs obbligato harpsichord in place of continuo, allowing concertante exchanges between keyboard and flute or viola da gamba. In the six Organ Sonatas, BWV 525–30, which belong to the same sub-genre, the entire three-part texture is allotted to a single player and one instrument. The parallel with Bach’s concertos of the same period for one or two harpsichords without accompaniment—the Italian Concerto, BWV 971 (1735) and the Concerto in C, BWV 1061a (1732/3)—is clear. Bach’s fourteen harpsichord concertos are historically significant as the antecedents (via those of his sons) of the piano concertos of the Viennese Classical period. The earliest of them seem to be those for three or four harpsichords (BWV 1063–5), which were probably composed around 1730, not long after Bach took over the Collegium musicum. The Concerto in A minor for four harpsichords, BWV 1065, was adapted from Vivaldi’s L’estro armonico, Op. 3 (Amsterdam, 1711), which had proved such a fertile source of inspiration for Bach during his Weimar period. All the other harpsichord concertos were apparently adapted from earlier Bach concertos for other solo instruments (chiefly violin and oboe), with the exception of the two-harpsichord Concerto in C, BWV 1061 (c. 1732/3), which originated as such—at first senza ripieno. The other concertos for two harpsichords (BWV 1060 and 1062) date from around 1736. The most important historically are the solo harpsichord concertos of about 1738. Bach seems to have made a preliminary attempt (BWV 1058–9) and then given up (BWV 1059 remained a torso) before embarking on a definitive set of six concertos, BWV 1052–7. Two crucial factors lie behind their composition: the virtuoso use of solo harpsichord in Brandenburg Concerto No. 5; and Bach’s transference of certain concerto movements to cantatas in the late 1720s, suitably adapted and usually employing obbligato organ. Of the solo harpsichord concertos, all three movements of Nos. 1 and 2 and the slow movement of No. 5 had already been adapted in this way. The obbligato-organ arrangements may thus be viewed as a preliminary stage in the development of the solo harpsichord concerto. In these concertos, by comparison with the original solo part for violin, oboe, or another treble instrument, the harpsichord part is consistently rendered more florid, brilliant, and idiomatic. The most uncompromisingly virtuoso concerto, that in D minor (BWV 1052), is fittingly placed at the head of the set. Significantly, it culminates in the Concerto in F, BWV 1057, a work that resembles Brandenburg Concerto No. 5 in its character as a cross between solo concerto and concerto grosso. It is, in fact, an adaptation of Brandenburg Concerto No. 4, in which the harpsichord not only replaces the original violin as soloist but also contributes to a concertino alongside two recorders. Bach cultivated two large-scale sacred genres for the first time in the 1730s: the Missa and the oratorio. He had already composed some Latin church music before then: a Kyrie (BWV 233a), three Sanctus settings (BWV 232III and 237–8), and the original E♭ version of the Magnificat (BWV 243a). But the great Missa in B minor, written for the Dresden court in 1733, seems to have been his first Mass setting. In composing it he must have been guided to some extent by the many Mass settings he already knew—by Peranda, Baal, Pez, J. L. Bach, Durante, Wilderer, Lotti, and others.

318 p art i i The composition of the Missa and the revision of the Magnificat (now in the key of D, BWV 243) were evidently carried out around the same time, which suggests that the latter might have been intended for Dresden too. In the 1720s, parody (the re-texting of existing compositions) had frequently been applied to Bach’s cantatas but not to his large-scale sacred works, the St John and St Matthew Passions. The Missa of 1733, on the other hand, like all Bach’s other major sacred compositions of the 1730s—the lost St Mark Passion, the four Missae of around 1738, and the three oratorios—seems to have originated to a large extent through the parody of existing compositions. Only two parody models survive, however: those of ‘Gratias agimus tibi’ and ‘Qui tollis’, whose originals are both sacred movements with German biblical texts (BWV 29 no. 2 and 46 no. 1 respectively). In the case of the four later Missae, most of the parody models survive: they are movements drawn from some of the finest cantatas of the Leipzig Cycles I and III, particularly BWV 67, 79, 102, 179, and 187. These works would have been performed regularly in the Leipzig churches, which suggests that the four Missae might have been written for performance elsewhere—perhaps Dresden, where Bach had recently been appointed Royal Court Composer (November 1736), having promised his ‘unermu¨deten Fleiß . . . in Componirung der Kirchen Musique’ (‘untiring zeal in the composition of music for the church’).2 The form of oratorio cultivated by Bach in the 1730s appears to have been his own invention. In essence, he applies fundamental aspects of the oratorio-Passion genre to three other major feast days in the church year—Christmas, Easter, and Ascension. Of the four Lutheran High Feasts only Whit is missing. Except in the Easter Oratorio, the format is identical with that of the Passions. The Gospel narrative is sung by the tenor Evangelist in secco recitative and by other characters—angels, shepherds, wise men, etc.—in recitative and turbae (crowd choruses). It is interspersed with a meditative commentary, made up of four-part chorales, or more elaborate chorale arrangements, and madrigalian verse, sung as accompagnato, ariosos, arias, and choruses. Most of the madrigalian choruses and arias were parodied from secular cantatas, which, in the case of the Christmas Oratorio, were written during the previous year for the Electoral House of Saxony. It is quite possible that Bach had the sacred parody in mind when he conceived the secular original. On the whole, the parodies are eminently successful. Only occasionally is one disturbed by the unavoidable change from precise image to generalized sentiment. The Easter Oratorio is an altogether different matter: it can only be described as a sacred dramma per musica, so close does it remain to its secular original, the pastoral cantata BWV 249a of 1725. The plot is sung by the dramatis personae, as in an opera, rather than narrated by a tenor Evangelist, and Bach makes no attempt to ‘sanctify’ it by incorporating biblical words or chorales. It remains an enduring mystery why he saw fit to link this composition with the Christmas and Ascension oratorios by giving it the same genre title of ‘Oratorium’. A recent study,

BD I, No. 27; NBR, No. 162. BD II, No. 388; NBR, No. 190.

c o n c lusi o n

319

however, sheds light on this and other problems connected with the three oratorios.3 Good reasons are given for dating the Easter Oratorio 10 April 1735—that is, shortly after that for Christmas (25 December 1734 to 6 January 1735) and shortly before that for Ascension (19 May 1735). The three compositions would then amount to a closely knit oratorio trilogy based on the life of Christ (hence the absence of a Whit oratorio). Much of the music is parodied from drammi per musica, which are essentially dramatic in character, each resembling a single act from an opera. The use of parody is no mere convenient device but a reflection of the character of oratorio as a sacred opera, a ‘musicalische Vorstellung einer geistlichen Historie’ (‘musical representation of a sacred story’), according to J. G. Walther (Musicalisches Lexicon, 1732). There are, moreover, clear analogies between the secular and sacred texts: the ‘Go¨ttersohn’ (‘divine son’) Hercules, who represents Crown Prince Friedrich of Saxony in Hercules auf dem Scheidewege, BWV 213, becomes ‘des Ho¨chsten Sohn’ (‘the Son of the Most High’) in the Christmas Oratorio; both are weak infants but have the potential to become men of great power and might for the benefit of their people. Finally, with regard to the entirely non-biblical text of the Easter Oratorio, on Good Friday 1734 Bach had performed G. H. Sto¨lzel’s Passion-oratorio Ein La¨mmlein geht und tra¨gt die Schuld, which likewise dispenses with the Gospel narrative. Perhaps there was a new climate of opinion in Leipzig church circles around this time according to which the spirit mattered more than the letter. Bach’s juxtaposition of the French and Italian styles in the solo violin Partitas (BWV 1002, 1004, and 1006) and in the keyboard Partitas of Clavieru¨bung I reaches its logical conclusion in Clavieru¨bung II of 1735, which contains a large-scale, multi-movement composition in each of the two national styles, the Concerto nach Italiaenischen Gusto (Italian Concerto, BWV 971) and the Ouverture nach Franzo¨sischer Art (French Overture, BWV 831). The chief ensemble genre of each nation is here transferred to the keyboard. A few years earlier Telemann had likewise published a collection that sought to compare and contrast the prevailing French and Italian styles, namely his Fantaisies pour le Clavessin (Hamburg, 1732/3). For Bach’s Clavieru¨bung II there are, of course, precedents among his earlier compositions: the French Overture has forerunners in the early ouverture-suites (BWV 820 and 822) and in keyboard Partita No. 4 (1728), with its ‘Ouverture’ introduction; and the Italian Concerto, in the Weimar concerto transcriptions and in the two-harpsichord Concerto in C senza ripieno, BWV 1061a. In other compositions of this period, the French and Italian styles are not so much juxtaposed as amalgamated in accordance with the vermischte Geschmack, notably in the concerts en ouverture BWV 1067–8, in which a ritornello-form introduction (main section of first movement) and concertante-style writing for a solo instrument are incorporated within a French ouverture-suite. Similarly, French-overture style

3 Christoph Wolff, ‘J. S. Bachs Oratorien-Trilogie und die große Kirchenmusik der 1730 er Jahre’, BJ 97 (2011), pp. 11–25.

320 p art i i is fused with Italian concerto form in the introductory Praeludium in E♭ from Clavieru¨bung III (1739). The keyboard and lute fugues of the 1730s, like the organ fugues of the late 1720s, tend to be tripartite in structure and exhibit the maximum possible contrast between the middle and outer paragraphs. In some cases this overall tripartite form is determined by the fugal structure. Thus the Fuga in A minor, BWV 904 no. 2, is a double fugue of the same type as the organ Fuga in F, BWV 540 no. 2: A S[ubject] I; B S II; C S I + II. The Fuga in E♭ from Clavieru¨bung III is a triple fugue that takes the form: A S I; B S II followed by S II + I; C S III followed by S III + I. Here the middle paragraph contrasts further by virtue of its manualiter texture, as in the organ fugues in F and B minor, BWV 540 no. 2 and 544 no. 2. The Fuga in C, BWV 547 no. 2, is cumulative on the basis of the direct, inverted, and augmented forms of its subject: A S direct; B S direct + inverted; C S direct + inverted + augmented. The two lute fugues, BWV 997 no. 2 and 998 no. 2, are constructed in da capo form, like the organ Fuga in E minor, BWV 548 no. 2. So too is the harpsichord Fuga in C minor, BWV 906 no. 2, assuming that the autograph has been correctly interpreted (it was formerly believed to be a fragment). In all four cases the middle paragraph forms a very strong contrast with its surroundings due to the introduction of continuous motion in shorter note-values coupled with more informal treatment of the fugue subject. In instrumental and vocal music alike, Bach continued to make fruitful use of a highly characteristic structure in the 1730s: that in which fugue and ritornello form are united within an overall da capo (ABA), reprise (ABA1), or bipartite (AA1) scheme. The da capo-form ﬁnale of the A minor Violin Concerto (BWV 1041) and the repriseform ﬁrst movement of the D minor Concerto for two violins (BWV 1043) both possess fugal ritornellos, which contrast sharply with the brilliant violin writing of their solo/duo episodes. Other means of uniting ritornello form and fugue are employed in the vocal works. In Part VI of the Christmas Oratorio, performed on the Feast of the Epiphany (1735), the opening reprise-form chorus consists in the main of a series of vocal fugues based on themes drawn from the introductory instrumental ritornello, itself not fugal but concertante in style and texture. In the da capo-form opening movement of the wedding cantata Gott ist unsre Zuversicht, BWV 197 (1736/7), a vocal fugue with only a motivic link to the concertante ritornello forms the main content of the A-paragraph. The vocal music that follows is largely built into an instrumental ritornello return. Kyrie I from the 1733 Missa is similar in that paragraphs A and A1 of the bipartite (AA1) structure consist of a vocal fugue together with a ritornello return with inbuilt vocal parts. Here, however, Bach forges a very close thematic and textural link between ritornello and fugal exposition: since the ritornello itself is partly fugal, the exposition is heard as an expansion of it. A late development of great interest is Bach’s concern to mask divisions and preserve continuity in his ritornello structures, whether they fall into overall da capo, reprise, or bipartite schemes. An obvious example is Kyrie I (described in the previous paragraph) from the Dresden Missa of 1733, where the seamless joining of

c on c lusi o n

321

fugal exposition and ritornello (with inbuilt vocal parts) twice gives rise to a paragraph of great breadth and power. Another manifestation of this tendency of the 1730s is that smooth transitions are often used to conceal the join between the middle paragraph and the reprise in da capo and reprise-form movements. In the A minor Violin Concerto, for example, a modulatory link at the end of the middle paragraph masks the join with the varied reprise in both the first and second movements; and something very similar takes place in the equivalent movements of the D minor Concerto for two violins. A clear instance of such smooth transition in the keyboard music of the same period occurs in the reprise-form finale of the Italian Concerto from Clavieru¨bung II of 1735 (bb. 150–2). In the harpsichord Fuga in C minor, BWV 906 no. 2, and the lute Fuga in E♭, BWV 998 no. 2, both in da capo form, an effective transition is made by purely thematic means: the last subject entry of the middle paragraph doubles as the first of the reprise. Smooth transitions at the same point also occur in vocal da capo structures. In the concertante movement that opens the solo cantata Jauchzet Gott in allen Landen, BWV 51, for example, da capo aria form is modified in the interests of economy and continuity by the insertion of a modulatory link between the central B-paragraph and the return of A, which for the same reason dispenses with its opening ritornello. Bach was undoubtedly conscious of the necessity to write in an up-to-date style in the 1730s. This is clear from his own observation that ‘the former style of music no longer seems to please our ears’ (1730)4 and from Mizler’s remark that a secular cantata performed before the king in 1738 (BWV Anh. I 13) was ‘written entirely in accordance with the latest taste’.5 Although the work concerned is lost, Mizler’s comment applies to much of Bach’s surviving music of the 1730s. Scheibe’s unalloyed praise of the Italian Concerto, which he described as ‘a perfect model of a welldesigned solo concerto’,6 may well be due partly to its obvious galant qualities. Other conspicuous examples of Bach’s writing at this period in a relatively ‘modern’ style are the aforementioned Praeludium in E♭ from Clavieru¨bung III, with its periodic, homophonic, treble-dominated episode (b. 33); the Pastorella, BWV 590, particularly its two middle movements, with their singing treble and simple accompaniment; Organ Sonata No. 3 in D minor, BWV 527, with its triplet figures and appoggiatura cadences; and the Fantasia in C minor, BWV 906, with its brilliant Scarlattian keyboard writing and its empfindsam effects. In the late cantatas, a clear tendency to broad, song-like, easily intelligible melody has been observed.7 This is perhaps related to the contemporary emphasis on a singing style of melody in the international galant idiom. It is already apparent in certain arias from the Picander Cycle (for example, 4

‘Die ehemalige Arth von Music unseren Ohren nicht mehr klingen will’; BD I, No. 22; NBR, No. 151. BD II, No. 336; NBR, No. 346. 6 BD II, No. 463; NBR, No. 331. 7 ¨ berlegungen zu J. S. Bachs Kantatenschaffen nach 1730’, Beitra¨ge zur BachBy Andreas Glo¨ckner, ‘U Forschung, 6 (1987), pp. 54–64; repr. in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld (Kassel, 1988), pp. 64–73. 5

322

par t i i

BWV 149 no. 4, 159 no. 4, 174 no. 2, and 188 no. 2), but the same trend then carries on throughout the 1730s. A particularly beautiful late example is the aria ‘Schla¨fert allen Sorgenkummer’ from the wedding cantata Gott ist unsre Zuversicht, BWV 197 (1736/7), whose broad oboe d’amore melody, simply accompanied and later taken over by the alto voice, would hardly have been out of place in contemporary opera seria. Galant features often go hand in hand with this melodic style: numerous melodies are built on repeated syncopated rhythms (BWV 214 no. 7 = 248, Part I, no. 8; BWV 215 no. 3, 206 no. 7, 30a = 30 no. 5, 100 no. 4, and 248, Part V, no. 51) or incorporate reverse dotting (BWV 30a = 30 no. 8, 195 no. 3, and 11 no. 1). Elsewhere we encounter kinetic recurrence, a feature of concerto style taken up by galant composers (BWV 112 no. 4), or a melody ornamented with many small figures or passages (BWV 97 nos. 4 and 8), which Quantz regarded as a feature of galant melodic style. A sign of the times, already evident in the late 1720s, is the incidence of dance-like movements in rondeau or binary form that nonetheless lack a specific dance rhythm or title. Rondeaux of this kind occur in the Fantaisie sur un rondeau, BWV 918, in the Ouverture in B minor, BWV 1067 (no. 2), and earlier in keyboard Partita No. 2, BWV 826 no. 5 (1727). In addition, certain binary movements of the late 1720s and 1730s are written in a somewhat modish, dance-like 2/4 time which occurs in vocal and instrumental music alike.8 The finales of the two ouvertures in B minor (BWV 831 and 1067), Echo and Badinerie respectively, belong to this type, as does the finale of ‘Phoebus and Pan’ (BWV 201) and the framing movement of the Wiederau cantata and its parody (BWV 30a/30). Non-binary examples include the opening Vivace of Organ Sonata No. 6 in G, BWV 530, several arias from the Christmas Oratorio (nos. 8, 31, 51, and 62), and the opening chorus of the Ascension Oratorio (parodied from BWV Anh. I 18). The 2/4 metre tends to go hand in hand with such galant features as homophonic texture, regular phrase structure, syncopated rhythm, and occasionally reverse dotting. Bach showed no less interest at this stage, however, in exploring the musical styles of the distant past, as it must have seemed from his perspective. Clavieru¨bung III is designed ‘for connoisseurs’ (‘Kennern’) partly, perhaps, because its main contents, the Missa and the pedaliter chorales, create a somewhat austere impression. To some extent they preserve the modality of the old chants and chorales on which they are based; and several movements—Kyrie–Christe–Kyrie, Aus tiefer Not, and the first section of the concluding Fuga a 5—are composed in the stile antico, the eighteenth-century understanding of Renaissance style, derived in Bach’s case partly from Zarlino’s seminal Le istitutioni harmoniche (1558) as mediated by Johann Theile.9 8 See Doris Finke-Hecklinger, Tanzcharaktere in Johann Sebastian Bachs Vokalmusik (Trossingen, 1970), pp. 142–3. 9 The stile antico as the Baroque understanding of Palestrina’s style is explored in Christoph Wolff, Der Stile antico in der Musik J. S. Bachs: Studien zu Bachs Spa¨twerk (Wiesbaden, 1968). The Zarlino–Theile–Bach line is traced by Paul Walker, ‘Bach’s Use of Fugue in the stile antico Vocal Writing of the B-minor Mass’, in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass (Belfast, 2007), vol. ii, pp. 368–86.

c o nc l usi on

323

Bach’s subsequent expansion of the Clavieru¨bung to include simpler, manualiter chorales, the four Duetti, and the framing Praeludium and Fuga might have been undertaken in order to achieve greater accessibility and consequently a wider public for the work. It must have been with deliberate intent that not only in Clavieru¨bung III but in other compositions of the 1730s Bach so often placed traditional and modern styles side by side. The great Missa of 1733, for example, moves straight from the overtly galant style of the duet ‘Christe eleison’ to the strictest stile antico in the second Kyrie. Similarly, in the Gloria of the same Missa, ‘Laudamus te’, a ‘modern’ operatic aria well suited to a prima donna like Faustina Bordoni, is immediately followed by ‘Gratias agimus tibi’, a motet-style fugue whose lineage reaches far into the past. Contrasts of this kind recur in the later Missae: in those in G and F (BWV 236 and 233), for example, a motet-style fugal Kyrie in Bach’s strictest contrapuntal style is followed by a concertante Gloria in a relatively up-to-date style. At this period Bach not only demonstrated his ﬁrm intention of exploring all available styles, whether ancient or modern, but he also seemed to relish their coexistence within a single composition. For him the galant was no mere reaction against the strict or learned style, as it was for many composers of the day; nor did the new, modern, and rational in music have to be embraced at the expense of tradition. On the contrary, ancient and modern, strict and free, serious and pleasing—all could be brought together by the integrating power of Bach’s creative genius. This holistic attitude could lead to the most unexpected results, as when the Sarabande from the Ouverture in B minor, BWV 1067, is treated as a strict canon between the outer parts; when Vivaldian unisono and ‘modern’ galantstyle themes form a double subject in the opening movement of the three-harpsichord Concerto in C, BWV 1064; or when a French sarabande and an Italianate Adagio take place simultaneously in the slow movement of the Sonata in G minor for viola da gamba and obbligato harpsichord, BWV 1029.

This page intentionally left blank

PA R T I I I The late Leipzig years: 1739–1750

This page intentionally left blank

III.1 Introduction

Bach’s difﬁcult relations with the Leipzig authorities do not seem to have improved in later years. On 17 March 1739 he was informed that permission had been withdrawn for the Passion performance that was due to take place only ten days later, on Good Friday, 27 March. Bach replied: ‘It had always been so; he did not care, for he got nothing out of it anyway, and it was only a burden.’1 It is thought that the work concerned might have been the St John Passion, for at the end of the 1730s Bach was preparing a fair copy of that work and revising it in the process, but he suddenly broke off, perhaps when he heard that the performance was not to take place.2 Whether due to resentment over this incident or because his interests now lay elsewhere, Bach evidently composed no new sacred cantatas in the 1740s. Instead, he met the weekly requirement of concerted church music with revivals of cantatas from the Weimar and early Leipzig years. We know of some thirty such revivals,3 and it is very likely that there were many more of which there is no evidence. In a few cases substantial changes were made. The wedding cantata Dem Gerechten muß das Licht, BWV 195 (1727/32) was revived with enlarged forces around 1742 and then radically revised for another wedding around 1748/9. And the cantatas Gott, man lobet dich in der Stille, BWV 120 (1729) and Lobe den Herrn, meine Seele, BWV 69a (1723) were adapted for the annual council-election service in about 1742 and in 1748 respectively. Two new works, BWV 191 and 1083, again adapted from earlier sources, are of special interest as lying outside the cantata genre normally cultivated by Bach. Gloria in excelsis Deo, BWV 191, a Christmas cantata with Latin text, was perhaps written for a thanksgiving service in the Leipzig university church on Christmas Day 1745, held to mark the Peace of Dresden which brought an end to hostilities caused by the Prussian invasion of Saxony.4 For this composition Bach borrowed three movements from the Gloria of the Dresden Missa of 1733, the framing choruses, ‘Gloria in excelsis Deo’ and ‘Cum Sancto Spiritu’, and the duet ‘Domine Deus’. The magniﬁcent six-part Christmas

BD II, No. 439; NBR, No. 208. According to Arthur Mendel, Krit. Bericht, NBA II/4 (1974), p. 75. 3 BWV 8, 9, 10, 16, 29, 34, 40, 42, 47, 64, 69, 76, 82, 91, 96, 97, 100, 114, 120, 129, 137, 139, 168, 170, 177, 181, 185, 186, 187, and 195. 4 See Gregory Butler, ‘J. S. Bachs Gloria in excelsis Deo BWV 191: Musik fu¨r ein Leipziger Dankfest’, BJ 78 (1992), pp. 65–71. 2

328

par t i i i

Sanctus (BWV 232 no. III) was probably performed on the same occasion, which might have induced Bach to expand the Kyrie–Gloria Missa into a Missa tota (the so-called B minor Mass), incorporating all five divisions of the Mass Ordinary. Tilge, Ho¨chster, meine Su¨nden, BWV 1083, is an arrangement by Bach of Pergolesi’s Stabat Mater, set to a German paraphrase of the penitential Psalm 51 and dating from around 1746/7. Bach greatly enriched the vocal and instrumental parts, but he also showed much sensitivity towards the style galant of the Italian master. Despite the incident of 1739, Bach continued to mount an annual performance of the Passion on Good Friday, and all three known oratorio-Passions of his were revived in the 1740s—the St Mark Passion (of which only the text survives) in 1744,5 the St Matthew in about 1742 and perhaps again around 1743/6, and the St John in 1749—in Version IV, which in most respects returns to the original version. In addition, he continued to perform the Passion music of his contemporaries.6 The Passion-oratorio Ein La¨mmlein geht und tra¨gt die Schuld by Gottfried Heinrich Sto¨lzel, which Bach had performed in the Thomaskirche at Good Friday Vespers in 1734, finds a late echo in the aria ‘Bekennen will ich seinen Namen’, BWV 200, a transposed adaptation by the Leipzig composer of an aria from the Sto¨lzel Passion.7 The anonymous St Luke Passion (BWV 246), which Bach is thought to have performed in 1730, was apparently revived in 1743/6, with the inclusion of his own setting of the chorale Aus der Tiefen, BWV 246/40 a.8 The ‘Keiser’ St Mark Passion (in fact of uncertain authorship), already performed by Bach in 1713 and 1726, was revived by him in about 1743/8 in a pasticcio version, beefed up by seven arias from Handel’s Brockes Passion.9 The Handel Passion itself was copied out by Bach and his son Carl Philipp Emanuel in 1746/7 and 1748/9, presumably in preparation for a performance.10 Since the Sto¨lzel Passion had been accepted for liturgical performance, it is possible that the Handel too—a work of the same Passion-oratorio genre—was performed in church rather than in assembly rooms. Finally, there are indications that Bach might have possessed a score of Carl Heinrich Graun’s so-called ‘Kleine Passion’, Ein La¨mmlein geht und tra¨gt die Schuld (it opens with the same chorale strophe as the Sto¨lzel Passion already mentioned). If so, it is possible that during his last decade he might have performed not only this work but also the pasticcio Passion based on it, Wer ist der, so von Edom ko¨mmt, which 5 According to a recently discovered printed text; see Tatiana Shabalina, ‘“Text zur Music” in Sankt Petersburg—Weitere Funde’, BJ 95 (2009), pp. 11–48 (esp. 30–5). 6 See Andreas Glo¨ckner, ‘Bach and the Passion Music of his Contemporaries’, Musical Times, 116 (1975), pp. 613–16, and his ‘J. S. Bachs Auffu¨hrungen zeitgeno¨ssischer Passionsmusiken’, BJ 63 (1977), pp. 75–119. 7 See Peter Wollny, ‘“Bekennen will ich seinen Namen”—Authentizita¨t, Bestimmung, und Kontext der Arie BWV 200: Anmerkungen zu J. S. Bachs Rezeption von Werken Gottfried Heinrich Sto¨lzels’, BJ 94 (2008), pp. 123–58. Wollny prints the Bach and Sto¨lzel arias one after the other, permitting a direct comparison. 8 Bach’s autograph score of the chorale was discovered in 1966 and first published by Yoshitake Kobayashi, ‘Zu einem neu entdeckten Autograph Bachs—Choral: Aus der Tiefen’, BJ 57 (1971), pp. 5–12. Regarding the St Luke Passion, see Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992), VBN I/An/8, and the same author’s Krit. Bericht, NBA II/9 (2000), pp. 69–80. 9 See Beißwenger (n. 8), VBN I/K/2 and Krit. Bericht, NBA II/9, pp. 106–9. 10 Beißwenger (n. 8), VBN I/H/1.

introduction

329

incorporates a parodied Latin motet ascribed to Johann Kuhnau, two movements from a Palm Sunday cantata by Telemann, the opening movement of Bach’s Quinquagesima cantata Herr Jesu Christ, wahr Mensch und Gott (BWV 127), and an arioso that is thought to have been composed by Bach, ‘So heb ich denn mein Auge sehnlich auf ’ (BWV 1088)11 Bach is known to have revived three of his secular cantatas in a more or less modified form during the last ten or twelve years of his life. O! angenehme Melodei, BWV 210a, a homage cantata for Duke Christian of Saxe-Weißenfels (1729), was parodied to form the secular wedding cantata O holder Tag, erwu¨nschte Zeit, BWV 210, around 1738/41. The Hunt Cantata, BWV 208, also written for Duke Christian, was slightly reworded in 1742 to render it suitable for the name day of the Elector of Saxony (BWV 208a). And Bach’s musical credo Der Streit zwischen Phoebus und Pan (The Dispute between Phoebus and Pan), BWV 201 (1729), already revived in the late 1730s, was performed again in 1749 with textual changes that reflect a new object of ridicule—either Johann Gottlieb Biedermann, Rector of the Freiberg Gymnasium, whose negative attitude to music and musicians caused widespread indignation, or Count Heinrich von Bru¨hl, the Saxon Prime Minister, and his prote´ge´ Gottlob Harrer, who at Bru¨hl’s insistence was auditioned as Bach’s successor over a year before the latter’s death.12 The only newly composed secular cantata of the 1740s, the Peasant Cantata of 1742, shares with the Coffee Cantata (1734) its comic style and contemporary setting and characters, recalling the comic intermezzi of opera seria. Bach seems to relish the rustic background and peasant characters of this cantate [en] burlesque (as the librettist Picander termed it), making extensive use of existing folk songs and folk dances to provide the necessary local colour. Like Phoebus and Pan, the work includes a singing contest that contrasts Bach’s refined style with the coarse style of his detractors. Both works reveal a serious purpose behind the comedy, recalling Forkel’s words that ‘if . . . [Bach] sometimes composed and performed something gay and even jocose, his cheerfulness and joking were those of a sage’.13 Whereas in the 1730s Bach furthered his links with the Saxon capital Dresden, in the 1740s he also established connections with the Prussian capital Berlin. In 1738 his son Carl Philipp Emanuel had become harpsichordist to Friedrich, Crown Prince of Prussia; and when the prince became King Friedrich II (Frederick the Great) in 1740, C. P. E. Bach accompanied him to Berlin and to his palace in neighbouring Potsdam. Johann Sebastian made two visits to Berlin in the 1740s, partly no doubt to visit his son but also to make contact with court musicians whom he knew (or knew

See J. W. Grubbs, ‘Ein Passions-Pasticcio des 18. Jahrhunderts’, BJ 51 (1965), pp. 10–42; and Beißwenger (n. 8), VBN II/G/2 (Graun) and pp. 89–100 (pasticcio). A copy of the score, Berlin, Mus. ms. 8155, formerly thought to be in Altnickol’s hand, is now known to have been written by Johann Christoph Farlau (P. Wollny, BJ 88 (2002), pp. 36–47). 12 BD II, Nos. 583–4; NBR, Nos. 265–6. 13 J. N. Forkel, Ueber Johann Sebastian Bachs Leben, Kunst und Kunstwerke (Leipzig, 1802); repr. in BD VII, pp. 9–89 (see 89); Eng. trans. in NBR, pp. 419–82 (see 479).

330 p art i i i of) and admired. On the occasion of the first visit, in 1741, he might have composed the Flute Sonata in E, BWV 1035, for the flautist Michael Gabriel Fredersdorf, chamberlain to the king. The second visit, in May 1747, was, in terms of prestige and renown, perhaps the most significant occasion in Bach’s life. The Prussian king Frederick the Great welcomed him and provided a subject on which he was asked to improvise. Next day the king asked him to improvise a six-part fugue on a subject of his own choice. Needless to say, both tasks were carried out with such skill that all those present were filled with admiration and astonishment. Bach’s intention to compose and publish a work based on the ‘Thema Regium’ (Royal Theme) was already mentioned in the Berlin press only a few days later. And only about four months after Bach’s Potsdam visit, the Musicalisches Opfer (Musical Offering), dedicated to the monarch, was in print: copies were available at the Leipzig Michaelmas Fair by the end of September. The contents included not only two fugues and ten canons in the ‘learned’ style for which Bach had earned a formidable reputation, but also a more ‘modern’ trio sonata, featuring transverse flute as a tribute to the fluteplaying king. On the occasion of his Potsdam visit Bach must have relished the opportunity to make or renew acquaintance with some of the outstanding musicians from the court Capelle. In a letter to Forkel, C. P. E. Bach listed the musicians whom his father ‘esteemed highly in his last years’.14 It includes Fux, Caldara, Handel, Keiser, Hasse, the Graun brothers, Telemann, Zelenka, Benda, ‘and in general everything that was worthy of esteem in Berlin and Dresden’. The Graun brothers and Franz Benda were members of the Berlin court Capelle, as was Johann Joachim Quantz, an admirer of Bach’s who had moved from Dresden to Berlin in 1741. These prominent figures, alongside Hasse and Telemann, were among the leading lights of the ‘modern’ galant and empfindsam styles of composition, and Bach’s admiration of them signifies a genuine interest in music written ‘according to the latest taste’.15 After the publication of Clavieru¨bung III in 1739, Bach undertook a major retrospective of his keyboard (harpsichord and organ) works of the Weimar and Co¨then periods. This took two forms. Firstly, around 1739/42, he collected together and revised his large-format organ chorales from the Weimar years. This project was unfinished when he succumbed to his final illness and consequently had to be completed by his son-in-law Altnickol and another pupil. The result was the collection known today as the ‘Eighteen Chorales’. Secondly, during the same period he composed a sequel to The Well-Tempered Clavier (Co¨then, 1722), designed along exactly the same lines as the original. Such repetition is extremely unusual in Bach and may reflect the success of the first volume, not only among his sons and pupils but

Letter of 13 Jan. 1775; BD III, No. 803; NBR, No. 395. The phrase ‘nach dem neuesten Geschmack’ was used by Lorenz Christoph Mizler in 1739 to describe the lost cantata BWV Anh. I 13, performed on 28 Apr. 1738; see BD II, No. 336; NBR, No. 346; and Werner Neumann, Krit. Bericht, NBA I/37 (1961), pp. 97–102 (esp. 101). 15

introduction

331

further afield, and the consequent necessity for providing further material of a similar nature. The second volume moves in two directions at once, as it were: towards greater contrapuntal rigour in the fugues and towards a more up-to-date style in the preludes. Around the time when The Well-Tempered Clavier II was completed, Bach published the Goldberg Variations (Nuremberg, 1741), which are governed to a large extent by the principles of variation and canon, though the degree and kind of keyboard virtuosity possible on a large two-manual harpsichord are no less significant. The techniques of variation, canon, and fugue would then proceed to determine the character of most of the major keyboard and instrumental works from Bach’s last decade. In the Fourteen Canons, BWV 1087 (c. 1747/8), the bass of the Goldbergs, reduced to essentials, is subjected to numerous different forms of canonic variation, as is the Christmas chorale Vom Himmel hoch in the Canonische Vera¨nderungen (Canonic Variations), BWV 769 (1746/7). On a larger scale, The Art of Fugue, BWV 1080 (c. 1742–9), and the Musical Offering, BWV 1079 (1747), are both spacious monothematic structures in which fugues and canons (plus a trio sonata in the latter work) are built upon variants of a single theme. Bach was no doubt supported in the composition and publication of these late contrapuntal works by the recognition he received from several influential quarters around this time. The Prussian king’s invitation to him to play fugues extempore on a given subject at Potsdam in May 1747 was a tribute not only to his legendary skill at improvisation but also to his growing fame as a great master of strict counterpoint. A month later, in June 1747, Bach accepted an invitation to join Lorenz Christoph Mizler’s Correspondirenden Societa¨t der Musicalischen Wissenschaften (Corresponding Society of Musical Sciences), whose members included Handel, Telemann, Sto¨lzel, and Carl Heinrich Graun. Each member was required to submit to the society at least one musical or theoretical work per annum. Bach submitted his Canon triplex (Triple Canon) a 6 for his induction in 1747—the same piece appears in the oil portrait of 1746 by E. G. Haußmann that Bach also had to present to the society—and, in the same year, a presentation copy of the Canonic Variations on Vom Himmel hoch, ‘fully worked out’ (that is, with all the canons fully realized). He also presented to the society an exemplar of the original edition of the Musical Offering, presumably as his 1748 contribution.16 It is possible that The Art of Fugue was intended to be his 1749 submission. Since Mizler’s society was of a learned, intellectual persuasion, it seems likely that Bach would have found in it a sympathetic, appreciative audience for these late contrapuntal masterworks. Bach’s last great vocal work, the Mass in B minor,17 BWV 232 (1748–9), is in one sense retrospective, since it returns to the sublime Missa (Kyrie and Gloria) that Bach had written for the Dresden court in 1733, adding Credo, Sanctus, and Agnus Dei to

BD III, Nos. 665 and 666 (pp. 88–9); BD II, No. 557; NBR, Nos. 241, 306 (p. 307), and 247. Strictly speaking, of course, the Mass is in D major, but the nickname has proved persistent and is therefore used here too. 17

332

par t i ii

form a complete Mass, or Missa tota. To see, lying behind this great achievement, the purpose of producing a valedictory summation of his life’s work is pure sentiment. It is far more likely that the ever-practical Bach had a specific purpose in mind—perhaps a commission from Dresden or Vienna (see Part III Ch. 4). The newly composed portions of the Mass rely at least as much on parody (the re-texting of existing music) as the 1733 Missa. And much the same direct juxtaposition of old and new styles is encountered, particularly in the central Credo choruses that deal with the Incarnation and the Crucifixion. ‘Et incarnatus est’, undoubtedly one of Bach’s latest pieces, was apparently influenced by Pergolesi’s Stabat Mater, which had been parodied by the German composer only a few years previously. The following movement, ‘Crucifixus’, on the other hand, is the oldest piece in the entire Mass, dating back to Bach’s Weimar period (1714). Accordingly, it takes the time-honoured form of a chaconne over a lamento bass. Thus Bach’s long-standing tendency to look backwards and forwards in style within the same composition, particularly acute in the 1730s and 1740s, continued until the end of his life.

III.2 The Well-Tempered Clavier II, and other keyboard/organ works

Title

Earliest source/s

Scribe/publisher, date

Eighteen Chorales, BWV 651–68

Berlin, P 271

Das Wohltemperierte Clavier II (The WellTempered Clavier II), BWV 870–93 Aria mit verschiedenen Vera¨nderungen (Goldberg Variations), BWV 988 Verschiedene Canones (Fourteen Canons), BWV 1087 Die Kunst der Fuge (The Art of Fugue), BWV 1080 Einige canonische Vera¨nderungen u¨ber das Weihnachtslied: Vom Himmel hoch, BWV 769 Sechs Chora¨le von verschiedener Art (Schu¨bler Chorales), BWV 645–50

London BL Berlin, P 430 Original edition

Autograph (also J. C. Altnickol and anon.), c. 1739/42, 1746–50 J. S. and A. M. Bach, c. 1739–42 J. C. Altnickol, 1744 Nuremberg: B. Schmid, 1741

Paris BN

Autograph, c. 1747/8

Berlin, P 200 Original edition Berlin, P 271 Original edition

Autograph, c. 1742–6 Leipzig: Breitkopf, 1751 Autograph, c. 1746–7 Nuremberg: B. Schmid, 1747

Original edition

Zella: J. G. Schu¨bler, 1748/9

Eighteen Chorales The publication of Clavieru¨bung III in 1739 was followed by an intensive preoccupation on Bach’s part with organ and harpsichord music. During the period 1739–42 he not only compiled Part II of The Well-Tempered Clavier and composed (and published) the Goldberg Variations, but he also collected together and revised most of the Eighteen Chorales, BWV 651–68. Thirteen of them were entered in the autograph manuscript at this stage (BWV 651–63). Then some years later, around 1746/7, Bach added nos. 14 and 15 (BWV 664–5), together with the Canonic Variations on the Christmas chorale Vom Himmel hoch (BWV 769). No. 18 (BWV 668) was entered in an anonymous hand after Vom Himmel hoch, perhaps by way of appendix, most likely between April and July 1750 (Bach died on 28 July).1 Finally, Bach’s pupil and

1 The scribe concerned is known as Anon. Vr (Du¨rr) or Anon. 12 (Kast); for further information see Y. Kobayashi and K. Beißwenger, NBA IX/3 (Textband) (2007), No. 231. The dating of the autograph entries follows Kobayashi Chr, pp. 45 and 56–7, and the same author’s NBA IX/2 (1989), p. 207.

334 the we ll-temper ed clavier ii etc. son-in-law Johann Christoph Altnickol entered nos. 16 and 17 (BWV 666–7) on pages left blank by Bach after the composer’s death, between August 1750 and April 1751.2 That nos. 16–18 are written in a hand other than Bach’s does not necessarily imply that their inclusion was unauthorized. Bach must have possessed a collection of separate chorale manuscripts from which the autograph of the Eighteen Chorales was compiled, and it would no doubt have included nos. 16–18, which are closely linked to their predecessors in the set. It is more than likely that Bach’s associates were simply assisting the ailing (and later deceased) composer in completing the collection.3 All of the chorales are revised versions of Weimar or pre-Weimar compositions preserved in Walther/Krebs manuscripts of the period 1710–17,4 with one exception, namely no. 18 (BWV 668), which survives in no source earlier than 1750. Two versions are known: Wenn wir in ho¨chsten No¨ten sein, BWV 668a, which was printed in 1751 as an appendix to the original edition of The Art of Fugue; and Vor deinen Thron tret ich, BWV 668, the last item in the manuscript of the Eighteen Chorales. This entry is a fair copy of a revised version of the chorale, in keeping with Bach’s procedure throughout the manuscript. According to the editors of The Art of Fugue, Bach dictated the piece to a friend after he had gone blind (in other words, after March 1750). It is more likely, however, that the earlier version (BWV 668a) already existed at this time and it was the revised version (BWV 668) that Bach dictated.5 The change of title to Vor deinen Thron tret ich (‘Before your Throne I now appear’) would then reflect Bach’s consciousness of his approaching end. Not only the absence of a Weimar source for this organ chorale but also its motivic and fugal stringency point to an origin in Bach’s mature Leipzig years. Like no. 17 (BWV 667), however, it is based ultimately on an Orgelbu¨chlein chorale, Wenn wir in ho¨chsten No¨ten sein, BWV 641. Looking back at that arrangement, Bach appears to have been struck by the manner in which the accompaniment to all four chorale lines is based on the diminished four-note headmotive of the first line, direct and inverted. Consequently, he dropped the original ornamented chorale melody in favour of a plain cantus firmus and adopted a revised version of the chorale accompaniments from the Orgelbu¨chlein chorale. Moreover, their fundamental principle is applied to a substantial new introduction and new interludes: each line of the cantus is introduced by stretto fugue on the diminished chorale line, answered by inversion according to

2 For the dating of the contributions of Anon. Vr and J. C. Altnickol, see the preface to Peter Wollny’s facsimile edn of the Eighteen Chorales and the Canonic Variations on Vom Himmel hoch, BWV 769 (Laaber, 1999). 3 Russell Stinson, however, believes that No. 18 was added by Anon. Vr at Bach’s request but Nos. 16–17 by Altnickol of his own volition; see his J. S. Bach’s Great Eighteen Organ Chorales (Oxford, 2001), pp. 33–8. 4 The revision process is discussed in detail by Werner Breig, ‘The “Great Eighteen” Chorales: Bach’s Revisional Process and the Genesis of the Work’, in G. Stauffer and E. May (eds.), J. S. Bach as Organist (London, 1986), pp. 102–20. The Weimar (or, in some cases, pre-Weimar) versions of the chorales are discussed in Vol. I of the present study, pp. 224–5. 5 See Christoph Wolff, ‘The Deathbed Chorale: Exposing a Myth’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 282–94.

d as w ohlt emp erierte cl avier i i

335

the principle of counter-fugue. The last line also occurs twice in augmentation, direct and inverted, and entries of the fourth line continue beneath the long-held last note of the chorale. Many other allusions to the chorale lines may be traced within the fugal texture. Thus transformed, the composition is essentially a cantus ﬁrmus chorale of the Pachelbel type, like Nun danket alle Gott (BWV 657) and several others of the Eighteen Chorales. The closest parallel, however, is the four-part, alio modo setting of Aus tiefer Not (BWV 687) from Clavieru¨bung III (1739), where each chorale line is likewise introduced by a very thorough and substantial counter-fugue.6

Das Wohltemperierte Clavier II Since Bach was not in the habit of repeating himself, we have to ask what lay behind his decision to compile a second set of twenty-four preludes and fugues in all keys between about 1739 and 1742. During an interval between Clavieru¨bung I–III (1726–39) and the late keyboard publications (the Goldberg Variations and the Art of Fugue), around 1739/40, Bach began to undertake a retrospective of some of the finest keyboard and organ music of the Weimar and Co¨then periods. This involved not only revising the larger Weimar organ chorales to form the substantial collection that we know as the ‘Eighteen Chorales’; it also involved compiling another set of preludes and fugues along the lines of The Well-Tempered Clavier from the Co¨then years. In the absence of a title page to the London autograph, the manuscript within which the new collection was compiled, it has often been doubted whether Bach truly considered it to be a successor to the Co¨then collection. However, the title page of Altnickol’s 1744 manuscript, which incorporates Bach’s final readings and many entries in his own hand, removes all doubt. The work is here entitled ‘Des Wohltemperirten Claviers Zweyter Theil’ (‘Second Part of The Well-Tempered Clavier’), establishing clearly that the Co¨then and Leipzig collections are to be considered as Parts I and II of a single great magnum opus. It might be significant in this regard that Part II was completed in 1742, exactly twenty years after the completion of Part I in 1722. By no means all of the material that Bach incorporated in The Well-Tempered Clavier II (henceforth WTC II) was new: eleven pieces, nearly a quarter of the total number, survive in early versions whose sources date from the late 1720s and 1730s. One of Bach’s chief sources was a set of ‘V. Praeludien und V. Fugen von J. S. Bach’ in the key series C d e F G (BWV 870a, 899–902).7 However, only four pieces were selected from it—the Praeludium and Fuga in C and the fugues in F and G—and all four were radically revised: above all, the Praeludium in C and the Fuga in F were doubled in length and the latter was transposed to A♭. Another important source was

6 The two organ chorales are compared by Peter Williams, The Organ Music of J. S. Bach (rev. 2nd edn, Cambridge, 2003), pp. 384–5. 7 ¨ ber ein unbeachtetes Sammelwerk J. S. Bachs’, See Klaus Hofmann, ‘ “Fu¨nf Pra¨ludien und fu¨nf Fugen”: U in W. Hoffmann and A. Schneiderheinze (eds.), Bericht u¨ber die wissenschaftliche Konferenz zum V. Internationalen Bachfest der DDR (Leipzig, 1988), pp. 227–35.

336 the we ll-temper ed clavier ii etc. a set of four Fughettas in the key order C c D d, written out by Bach’s pupil J. F. Agricola around 1738/9. Partly by means of transposition, this set supplied the fugues in c, C♯, d, and E♭. With the help of such manuscripts it is possible to build up a picture of the early stages in the genesis of the WTC II. It appears that Bach started with a diatonic key order, that transposition was often used to cater for the more remote keys, that ‘modal’ key signatures were sometimes still used, that some of the fugues were originally entitled ‘fughetta’, that older compositions were on occasion used to fill gaps, and that in several cases existing pieces underwent a huge expansion to render them suitable for their new context. In all these respects the genesis of Part II appears to be remarkably similar to that of Part I. The London autograph itself was assembled in three layers between about 1739 and 1742.8 Layer 1, compiled jointly by Johann Sebastian and Anna Magdalena in about 1739/40, includes most of the more commonly used keys and comprises half of the total collection. Layer 2, compiled around 1740/1 by Bach alone, comprises preludes and fugues in the more remote keys: its purpose was clearly to fill the chromatic gaps in the key series of Layer 1. Finally Layer 3, written out by Bach himself in about 1741/2, adds the only two prelude-andfugue pairs that were still missing, those in C and A♭, both of which involved radical revision of existing pieces (except in the case of the Praeludium in A♭, which seems to have been newly composed). In general the preludes of Part II tend to be more thoroughly worked out and designed on a larger scale than those of Part I. Accordingly, while a large proportion of the preludes of Part I are written in a pseudo-improvisatory style, this applied to only two preludes from Part II, those in C and C♯, both of which existed in some form before the collection was compiled and were heavily revised upon inclusion. The Praeludium in C is written in a traditional preludial style, with freistimmig (freevoiced) texture, long pedal points, and little in the way of theme or motive. In its original form the piece sounded like an improvisation, but the subsequent inclusion of a subdominant reprise (bb. 20–8 = 5b–14a) substantially modifies that impression.9 The free but largely four-part texture suggests that Bach was seeking the maximum fullness of sonority in order to create a big, celebratory sound suitable for the opening of the work. A clear sign of the difference between the Co¨then and Leipzig collections is the incidence of the arpeggiated prelude: it occurs frequently in Part I, but there is only one specimen in Part II, namely the Praeludium in C♯. This piece was originally in the key of C and its first 24 bars were notated in the conventional shorthand form of five-part minim chords, to be arpeggiated freely by the player.10 On entering the piece

See Kobayashi Chr, pp. 45–6, and Don O. Franklin, ‘Reconstructing the Urpartitur for WTC II: A Study of the “London Autograph” (BL Add. MS 35021)’, in Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 240–78. 9 As pointed out by James A. Brokaw II, ‘The Genesis of the Prelude in C major, BWV 870’, in Franklin (ed.), Bach Studies [I], pp. 225–39. 10 The same form of notation as Bach used for the Praeludium in C from the WTC I in the Clavierbu¨chlein for W. F. Bach. The C major version of the Praeludium in C♯ from the WTC II was written out in Anna Magalena’s hand in Berlin, P 226 (c. 1739/40).

d as w ohlt emperierte cl avier i i

337

in the London autograph, Bach not only transposed it from C to C♯ but provided an elaborate realization of the chords, thereby removing it one step from its improvisatory origins. Of course, it is already removed from them by virtue of its bipartite structure: the arpeggiated prelude acts as an introduction to an Allegro fugato in 3/8 time and three-part texture. To a greater extent than in Part I, Bach often employs a strict two-part texture in the preludes of Part II, profiting from the example of the two-part Inventions. This connection is particularly clear in the Praeludium in D minor, one of the earliest pieces in the collection, having been radically revised at least twice and extended from 43 to 53 and finally to 61 bars.11 The two statements of the double theme, showing its invertible counterpoint (bb. 1–5, 6–8), the reverse-order dominant counterstatement (bb. 26–34), and the sequential episodic formulations with Stimmtausch or partsexchange (bb. 9 and 13) were all standard techniques in the Inventions. The Praeludium in B minor starts in a similar fashion, with the treble theme immediately transferred to the bass (b. 5), but its counterpoint at this point is new and further inventive new counterpoints are introduced subsequently (bb. 21 and 25). The Praeludium in F♯ also displays the characteristic two-part Invention texture, but on a far larger scale—perhaps an indication of relatively late date, for it is among the most mature and sophisticated of the Part II preludes. The opening dotted-rhythm theme functions rather like the headmotive of a ritornello, recurring periodically in different keys (bb. 17, 42, and 57). On the last occasion, the return of the tonic coincides with the thematic return. The second theme (b. 4), however, contains the two most significant motives and acts as the main source of development. Binary dance form plays a far greater part in Part II (ten preludes) than it does in Part I (one prelude).12 This is no doubt connected with Bach’s huge expansion of the form in the interim—above all, in the partitas of Clavieru¨bung I—but also with its emancipation from the dance in his sonatas and concertos. With one exception—the relatively short and light Praeludium in G, which is primarily figurative rather than subject-based—the two-voice binary pieces, those in c, d♯, e, and a, incline to the Invention type. All open with a brief subject in the treble which is immediately imitated at the octave in the bass. And in two cases—the preludes in c and a—that subject is combined with a regular countersubject ab initio, so that the octave imitation involves exchange of parts. The Praeludium in E minor, tightly constructed around the opening theme and its constituent motives, achieves length by means of the tonic reprise during the second strain (bb. 72–103) of the dominant-key portion of the first strain (bb. 23–48). The Praeludium in A minor is closer to the F minor 11 The 43-bar version is in J. C. Vogler’s hand in Berlin, P 1089 (c. 1727/31); the 53-bar version was copied by Anna Magdalena into P 226 (see n. 10) and into the London Autograph. The expansion to 61 bars was the product of Bach’s revision of the London Autograph. 12 ¨ ber Themenbildung und thematische Arbeit On the Part II preludes in this form, see Klaus Hofmann, ‘U in einigen zweiteiligen Pra¨ludien des Wohltemperierten Klaviers II’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld (Kassel, 1988), pp. 48–57.

338 the we ll-temper ed clavier ii etc. Sinfonia than to any of the Inventions. Both pieces are studies in strict invertible counterpoint, built on the chromatic 4th, whose implications yield music of great pathos. Moreover, in both cases the episodes are handled almost as strictly as the main thematic combination. The A minor Praeludium possesses the additional feature that the combined themes are melodically inverted after the double bar, after which melodic and contrapuntal inversion go hand in hand. It is possible that the piece has a symbolic significance: the chromatic key scheme of the entire work is distilled in its completely chromatic, ‘twelve-note’ thematic combination. Only one of the binary preludes is written in a three-part texture, that in E, but it bears little resemblance to the three-part Sinfonias. It may be viewed rather as a fresh take on the traditional preludial idiom from the vantage point of Bach’s most mature style. Material that might have been improvised by an organist around 1700 (bb. 1–4) receives the status of a theme, recurring at bar 5 in dominant counterstatement with interchanged parts and again after the double bar (b. 25). Patterns that might have been purely incidental to the improvising organist, such as the four-note figure imitated in bb. 1–2, are accorded a motivic function (bb. 9ff.). As often in the keyboard partitas, Bach expands the binary form and gives it coherence by introducing a distinct thematic nexus in the dominant key (bb. 18–24), which is recapitulated (in the reverse order) in the tonic at the end of the second strain (bb. 43–50). The remaining four binary preludes, those in D, f, g♯, and B♭, are all freistimmig (free as regards the number of voices), which not only allows enormous flexibility but is appropriate wherever, as here, melodic and figural writing is uppermost rather than contrapuntal line-drawing. The preludes in f and g♯ are like twin sisters: heartfelt slow movements in Bach’s most up-to-date idiom, their themes rich in the figure of repeated note plus appoggiatura to which he would later return in the more empfindsam pieces from the Musical Offering (Ex. 1).13 Both preludes end with a much abridged and varied reprise of the first strain, this time entirely in the tonic. The preludes in D and B♭ may also be regarded as twins. Both are in a gigue-like compound quadruple time, and both may be viewed as ‘sonatas’ in the Scarlattian sense—brilliant, idiomatic keyboard pieces, one of which (B♭) involves handcrossing. The remaining four freistimmig preludes, those in E♭, F, A♭, and B, are throughcomposed. In each case, they are laid out in three or four large paragraphs, articulated by structural cadences in interrelated keys. The opening four-bar theme of the Praeludium in E♭, with its melodic treble and ostinato bass, returns in its original form only in the third and last paragraph (bb. 50ff., at b. 61), which thus takes on a reprise function. The main content of the middle paragraph (bb. 25–50) is a huge

13 The ‘modern’, empﬁndsam aspects of the 3-part Ricercar from the Musical Offering are discussed at length by Werner Breig, ‘J. S. Bachs Leipziger Klaviermusik und das Prinzip “Empﬁndsamkeit” ’, in S. Schmalzriedt (ed.), Aspekte der Musik des Barock: Auffu¨hrungspraxis und Stil (Laaber, 2006), pp. 295–315 (esp. 304–12).

d a s wo h l t em p e r ie r t e c l a vi e r i i

339

Ex. 1

a) Incipit of Praeludium in F minor, BWV 881 no. 1

b) 2nd bar of Praeludium in G♯ minor, BWV 887 no. 1

c) Ricercar a 3 from the Musical Offering, bb. 113–14

[ ]

d) Sonata from the Musical Offering, 3rd movement (Andante), bb. 1–2 (figuring omitted) sequential, modulatory period (bb. 32–46) based on material drawn from the original answering phrase (b. 5). There are clearly audible links between the opening and closing phrases of the three paragraphs: the first two paragraphs share a rhymingclose, and the second and third both open with a variant of the main theme. The thematic material of the Praeludium in F has roots in the old preludial style in which melodic notes are constantly sustained to build chords, as in the French style luthe´. This traditional aspect, however, is belied by the exceptional regularity of the phraseand period-structure, a markedly progressive feature that forms one of the main stylistic distinctions between the Co¨then and Leipzig collections. In the Praeludium in B, among the most improvisatory in style of all the Part II preludes, Bach adopts a type of free texture already cultivated in certain dance movements from the sixth French Suite and the first Partita, a texture that moves freely in and out of two parts, three parts, or a single part divided between the hands. It is cast in ABA1 reprise form with a rhyming-close at the end of the two A-sections (bb. 15–17 and 44–6). The major

340 t he we ll -tem p e re d c lav ie r i i etc . surprise here is the decorated cantabile duet that emerges at the heart of the middle paragraph (bb. 23b–28), underpinned by the broken-chordal theme from bar 3, which has here turned into an accompaniment figure.14 The most spacious and impressive prelude from this group, the Praeludium in A♭, was the last to be entered in the London autograph and perhaps the last to be composed. It forms a sister-prelude to that in F♯: both are large-scale designs in the major mode and in 3/4 time with semiquaver motion against dotted rhythms. And in both cases recurrences of the main theme in different keys have the effect of ritornellos (here bb. 1, 17, 34, 50; keys I, V, vi, IV), whereas the second theme (here b. 7) yields the combined motives which, separate or together, form the chief material of subsequent development. Structural cadences in different keys bring the four paragraphs to a close (bb. 16, 33, 49, 77; keys V, vi, IV, I). The through-composed preludes include one in four-part texture (key g) and four in three parts (c♯, f♯, A, and b♭). The full four-part texture of the Praeludium in G minor, in conjunction with its unvaried dotted-semiquaver rhythms, belongs to an older style than most of the Part II preludes and suggest an earlier origin. This impression is reinforced by the relatively small scale of the piece, which recalls the preludes of Part I. The two periods are linked by a thematic return (b. 1 = 9) and a rhyming-close (b. 7 = 17b). Of the three-part preludes, only those in A and b♭ bear any resemblance to the three-part Sinfonias. The Praeludium in A is a pastorale in character, and as such has a certain affinity with the Sinfonia in E and with the Praeludium in E from the WTC I. Both of the WTC preludes are cast in a lyrical ABA1 reprise form with subdominant recapitulation. The Praeludium in B♭ minor is similar in form to that in A—both have a subdominant reprise with interchanged upper parts. But the minor-mode piece is built on a far larger scale and, like many of the Sinfonias, designed as an informal three-part fugue—the initial subject entry is already accompanied and the third voice enters with a free part. The Praeludia in c♯ and f♯ are both cantabile trios, but they have little else in common and neither has much connection with the three-part Sinfonias. The Praeludium in F♯ minor is a ‘magnificent stream of lyrical melody’,15 whose texture corresponds with that of numerous Bach slow movements from Clavieru¨bung I onwards: a florid treble with two left-hand supporting parts (the tenor imitation in bb. 2 and 31 is deceptive). The Praeludium in C♯ minor, no less rich an outpouring of florid cantabile melody, is nonetheless unique in Bach’s keyboard music: all three voices of the trio participate equally in the highly ornamented themes and their variants, creating a texture of remarkable intricacy and thematic intensity.

14 Thematic use of what we would regard as a mere accompaniment ﬁgure is quite common in Bach. A prominent example occurs in the Sinfonia in A, BWV 798, where the LH accompaniment to the RH dialogue in bb. 9–12 and 20–3 forms part of the theme itself (b. 2a). 15 Donald Francis Tovey, Commentaries in ABRSM edn of WTC II, ed. Richard D. P. Jones (London, 1994), p. 180.

d as w ohlt emperierte cl avier i i

341

The concept of a set of fugues that schematically illustrate the various features of contrapuntal technique is as far removed from Part II of The Well-Tempered Clavier as it is from Part I. Like the preludes, the fugues are primarily of interest as works of art, to be relished by students and skilled players alike, rather than as studies in composition. Accordingly, ten fugues in Part II—those in C, e, F, f, F♯, G, A, a, B♭, and b— make no use of the devices of strict counterpoint, but instead are built on subjects (and, in some cases, countersubjects) of strong and distinct character that lend the composition as a whole a highly individual flavour. Thus the main interest of the three-part Fuga in C lies in the arresting headmotive of the subject, which is repeatedly exchanged between the two upper parts in the first and fourth episodes (bb. 13 and 55). Further play on this motive follows in the concluding 16-bar period (b. 68), which was absent from the early version and added only when the piece was incorporated in the London autograph.16 No less elementary in construction is the three-part Fuga in F, a gigue-fugue devoted mainly to the development of the three figures of the subject (there is no regular countersubject). The following fugue, that in F minor, is built on a strikingly characterful subject whose repeated-note figure dominates the entire piece, partly through its sequential use in two episodic formulations (bb. 17 and 56). The first of these has a cadential function, occurring at the end of all four expositions (bb. 17–24, 33–40, 66–71, and 78–85) and thereby clarifying the overall structure. Four of these ‘character-fugues’, as they might be termed, introduce a new countersubject (Fuga in B minor) or a combined pair of countersubjects (G, A, B♭) in medias res, but the new material is then used regularly and contributes greatly to the character of the fugue. This is particularly true of the three-part Fuga in G, one of the most athletic and brilliant of the fugues from Part II, though of small extent, thereby betraying its early origin.17 The early version lacks the impressive dominant preparation for the concluding subject entry (bb. 53–64), and its countersubject consists of nothing more than repeated quaver chords. When revising the piece for the WTC II, Bach replaced the plain chords with a charming decorated suspension chain (b. 16), which then adorns both of the following entries (bb. 34 and 41), the second time with interchanged parts. Another two-voice countersubject of this kind is introduced in the three-part Fuga in A (b. 5) and is subsequently used regularly, if informally. Its dotted rhythms interlock with the syncopated rhythms of the subject to produce an attractively playful effect. In the three-part Fuga in B♭, every entry from bar 32 onwards is accompanied by two combined countersubjects that together make up a decorated suspension chain, as in the G major fugue. The B♭ fugue, however, is a cantabile piece in an overtly melodic style, hence its clear division into three paragraphs by means of prominent cadences (bb. 31–2 and 77–8) and its rhyming-close (bb. 29–32 = 90–3). In the three-part Fuga in B minor, a single countersubject, made up of two implied parts 16 17

The genesis of this 16-bar conclusion is discussed in detail by R. D. P. Jones in the ABRSM edn, p. 160. The prehistory of the fugue is discussed by R. D. P. Jones in the ABRSM edn, p. 182.

342 the we ll-temper ed clavier ii etc. that greatly enrich the texture, is introduced during the discourse (b. 29) and, together with its simpler episodic form (b. 32), proceeds to dominate the remainder of the fugue. Three character-fugues, in e, F♯, and a, have a regular countersubject ab initio that contributes almost as much to the character of the fugue as the subject itself. In the Fuga in E minor, maximum emphasis is placed on rhythmic diversity, a sign of the progressive ‘mixed’ style at that time. The subject falls into three sub-phrases, each with its own rhythm, to which the countersubject (bb. 7–12) adds its own distinctive rhythmic figures. The result is a comic, scherzo-like fugue that derives its character from the quixotic subject–countersubject combination. The version of the London autograph ends at bar 70; the remaining 16 bars are recorded primarily in Altnickol’s 1744 manuscript and represent one of the most important late revisions to the text of the WTC II. They form a magnificent peroration of boldly rhetorical character. The Fuga in F♯ is similarly inventive in the varied figure-work of its subject, which paradoxically opens with a cadential formula—trilled leading-note to tonic—before stressing the flat 7th degree, giving a hint of subdominant, and then closing with the repeated note plus appoggiatura of the empfindsamer Stil that Bach also employed in the preludes in the minor keys of f and g♯. This figure is then taken over into the countersubject. Its prominence in most of the episodes (from b. 24b onwards) lends the fugue a decidedly ‘modern’ aspect. The Fuga in A minor is notable, above all, for the startlingly original use it makes of traditional material. The headmotive of the subject, with its falling diminished 7th, is a cliche´ of Baroque fugue, but its consequent phrase is quite unexpected: a free diminution in staccato quavers. The three figures of the countersubject, all quite different in rhythm, have a ‘torrential vigour’ that contrasts markedly with the ‘giant strides’ of the subject.18 The traditional fugue subject is here divorced entirely from its usual affective connotations. The fugues in which the techniques of strict counterpoint are employed tend to be weightier and are often in four contrapuntal parts rather than three. Closest to the character-fugues are two successive minor-mode fugues, those in d and d♯, in which stretto and subject inversion are introduced only in the second half as a means of diversification. In the Fuga in D♯ minor, whose richly expressive harmony-counterpoint justifies the term fuga pathetica, the sole stretto (bb. 23b–27 a) forms the centrepiece of an unbroken succession of ten subject entries that constitutes the middle paragraph (bb. 15–35). The tonic full-close four bars from the end is followed by simultaneous direct and inverted forms of the subject by way of coda. In the Fuga in D minor, the semiquaver motion of the countersubject forms an effective contrast with the triplet semiquavers of the subject. The second half is largely made up of two strettos, the first based on the direct subject and the second on its inversion. Since the parts of the first stretto are inverted in the second, the latter constitutes a full contrapuntal and melodic inversion, a mirror image of the first stretto.

Tovey, commentary to ABRSM edn, p. 190.

d a s wo h l t em p e r ie r t e c l a vi e r i i

343

Three successive major-mode fugues, those in D, E♭, and E, form a triptych of stretto fugues in a traditional, pseudo-vocal style. The Fuga in D is divided into two halves by a mediant full-close (b. 27). This and the ensuing hiatus recall the da capo aria and produce a sense of reprise at the following tonic return. The first half alternates between strettos at one bar and at half a bar (bb. 5, 14, 21, and 22), whereas the second half is confined to strettos at the quarter-bar, producing an effect of intensification. The piece is strictly motivic in every bar on the basis of the two dominant figures of its subject. The Fuga in E♭ is relatively simple: after the initial exposition (bb. 1–30), a single form of stretto, at one bar and the lower 5th, is stated three times (bb. 30, 37, and 59), the second time with interchanged parts. The Fuga in E is based on a brief epigrammatic soggetto that Bach apparently borrowed from J. C. F. Fischer’s Ariadne musica of 1702. It is interesting to note that at this late stage Bach still returned to the collection that, perhaps more than any other, inspired the original Well-Tempered Clavier of 1722. The stylistic antecedents of the fugue, however, reach back much further—to the Palestrina style as viewed from the perspective of the early eighteenth century and as cultivated by Bach, Fux, Caldara, Zelenka, and others in the so-called stile antico.19 By comparison with the monothematic fugues in D and E♭, the E major is enriched not only by an important regular countersubject but by various manipulations of the subject itself—subject variation of this kind being an important but little noticed resource of Bach’s. The first, second, and fourth stretto expositions (bb. 9, 16, and 35) contain strettos of standard type—the plain subject in stretto at the 4th or 5th and at one bar or half a bar. But the third stretto exposition (bb. 23–34) is based on the varied subject, then on its diminution (b. 27), and finally on the diminution in inverted as well as direct forms (b. 30). The da capo aria-like mediant/tonic hiatus at the end of this exposition recalls the D major fugue. Manipulation of the subject also plays a significant role in two consecutive stretto fugues, those in c and C♯. In both cases the subject is not only inverted but also augmented (and in the C♯ fugue diminished too), a rare form of treatment in Part II as well as in Part I. In the Fuga in C minor, the first stretto (bb. 14–15) combines in counterpoint three different forms of the subject—original, augmented, and inverted. Shortly afterwards (bb. 19–23) these three forms in a different order are stated consecutively in the bass. The C♯ major is a stretto fugue ab initio—even the opening entries overlap. The headmotive, however, is often detached for separate stretto treatment, and a new consequent figure occurs in bb. 8–9 and is much used thereafter. The subject is throughout treated not only in stretto but in inversion and diminution, both of which occur as early as the opening period (bb. 2 and 5b–6a). Another form of treatment belongs to a late stage in the evolution of the fugue. Two early versions in C major are known20—a 19-bar fughetta, devoid of all semiquaver movement, and a

19 See Christoph Wolff, Der Stile antico in der Musik Johann Sebastian Bachs: Studien zu Bachs Spa¨twerk (Wiesbaden, 1968). 20 The 19-bar version is transmitted in Berlin P 563 in the hand of Johann Heinrich Michel, a Hamburg copyist of C. P. E. Bach’s (after 1768); the 30-bar version in P 595/4, in J. F. Agricola’s hand (c. 1738/9).

344 the we ll-temper ed clavier ii etc. 30-bar fugue that lacks bars 25–9. These five bars were clearly added when the fugue was entered in the London autograph. They represent a climactic exposition of the augmented subject, which occurs twice, headmotive only (bb. 25 and 27), surrounded by myriad direct and inverted entries in standard note-values. The Fuga in B♭ minor, among the most profound and highly organized fugues in the entire collection, employs the same contrapuntal techniques as those in C♯, c, d, and d♯, namely stretto and inversion, but in a considerably more logical and systematic fashion. It falls into three large sections: 1. exposition of the direct subject, followed by stretto on the direct subject; 2. exposition of the inverted subject, followed by stretto on the inverted subject; 3. stretto on the direct and inverted forms of the subject combined. The finely wrought subject is combined with a chromatic countersubject (bb. 5–8), which contributes greatly to the pathos of the fugue and is inverted whenever the subject itself is inverted. The episodes are as highly organized as the expositions. Episode 5 (b. 37), for example, includes five sequential steps of a triplecounterpoint combination, including three different permutations. The ultimate tour de force occurs at bar 96, the last of the three strettos involving the direct and inverted subjects. Here, the combination is enhanced by doubling both forms of the subject in 3rds and 6ths, which produces a totally thematic four-part texture. Two consecutive four-part fugues in the latter half of the collection, those in g and A♭, are based throughout on the principle of invertible counterpoint. The themes concerned are the subject and one or two regular countersubjects which are present from the outset of the fugue. In the Fuga in G minor, the two lively main themes, containing a wealth of significant figures, are already combined in double counterpoint at the octave in the opening exposition (bb. 5 and 13). The second exposition (b. 28) then presents them in double counterpoint at the 12th, which creates a powerful sequence of 7ths between the parts concerned. This is immediately followed by two combined entries in double counterpoint at the 10th (bb. 32 and 36). The potential implications of this species are revealed in the third exposition (b. 45): contrapuntal inversion at the 10th allows the subject to be doubled in 3rds (b. 45), then in 6ths (b. 51), and finally both subjects to be doubled in 3rds simultaneously (b. 59). The richness and power of this scheme is much enhanced by the judicious use of different keys to colour the various subject entries. In the Fuga in A♭, the rhythmic subject is combined from the outset with two regular countersubjects: a slow chromatic-4th descent and a sequence of continuous semiquavers. These three themes are combined in a triple counterpoint that is heard in five of its six possible permutations (bb. 6, 8, 22, 24, and 35) as well as in a variant combination (b. 37). The early version, whose chief source dates from around 1727/31, was in the key of F and only 24 bars long.21 Thus, of the present bipartite (AB) fugue of 50 bars (24 + 28), only A belongs to the original content; B (bb. 24ff.) was added when the piece was prepared for inclusion

It is transmitted by J. C. Vogler in Berlin, P 1089.

d a s wo h l t em p e r ie r t e c l a vi e r i i

345

in the London autograph. All the variety of key of the third exposition (bb. 24ff.) and the remarkable Neapolitan harmony of the fourth (bb. 41ff.) are thus products of a huge expansion, undertaken to render the fugue suitable for its new context. Double fugue is represented by three pieces in the collection—the fugues in c♯, g♯, and B—and triple fugue by one, that in f♯. In the double fugues, a second subject enters in the course of the fugue and is then combined with the first subject in invertible counterpoint. In the Fuga in C♯ minor, the first two expositions (bb. 1 and 16) are devoted solely to the Spielfuge subject, and the third (b. 24) to its inversion. Only in the second half is a new, partially chromatic subject presented (fourth exposition, b. 35), though adumbrations of it have already been heard (bb. 3, 5, 18, 20, and 30). The fifth and sixth expositions (bb. 48 and 61) are then devoted to its combination with the original subject, which is heard not only in double counterpoint at the octave (bb. 61 and 66) but at the 12th (bb. 48 and 55). The final entry (b. 67b) presents a new combination in which the second subject enters half a bar later than before. The Fuga in B is divided into three paragraphs by clear cadences (key V, bb. 26– 7; IV, bb. 59–60). The first paragraph is devoted to the exposition of the neutral soggetto-like subject alongside its syncopated countersubject. In the second paragraph (b. 27) it becomes clear that this is to be a relatively informal double fugue. The second subject, which here replaces the original countersubject, consists merely of a coiling sequence of quavers and is not treated alone first, as would be normal in a double fugue, but is immediately combined with the first subject. This combination (bb. 27 and 48) is also inverted at the 12th (bb. 35, 42, and 53). The third paragraph (b. 60) immediately introduces a new combination of the two subjects, the second one entering at a different point. It continues with their separate entry (bb. 71 and 75) before reverting to the original combination (b. 85) and the 12th inversion (b. 93). The Fuga in G♯ minor represents a more formal, stricter type of double fugue—a double exposition of the diatonic subject in flowing quavers (bb. 1–32, 33–60), followed by a middle section entirely devoted to the exposition of the chromatic second subject (bb. 61–96) before the two subjects are combined for the remainder of the fugue (bb. 97 ff.). As in the C♯ minor fugue, the new subject is already adumbrated in the counterpoints to the original subject: the second and fourth subject entries of the opening exposition are accompanied by a chromatic figure that will later form the second half of the new subject. A similar strictness and formality inform the Fuga in F♯ minor, a triple fugue of the classic type in which each subject receives its own exhaustive exposition before it is combined with the other subject/s. Thus the opening paragraph, which cadences in the relative major A at b. 20, is devoted to the original subject, whose triadic headmotive, direct or inverted, saturates the surrounding voices. The second paragraph, which cadences in the dominant C♯ minor at b. 36, gives a very full stretto exposition of the second subject, with its trilled dotted-rhythm headmotive, before combining it twice with the original subject (bb. 29 and 34). The third paragraph, which closes in the submediant D at bar 50, is entirely devoted to exposition of the third subject, whose flowing, sequential semiquavers inject a new

346 t he we ll-tempe red clav ier ii etc. vitality into the fugue. The concluding paragraph begins by recalling the original subject alone (b. 52)—although the semiquaver ﬂow continues unbroken—presumably to bring it clearly to mind before further combinations follow. The great climax of the fugue is then reached—the triple counterpoint of the three subjects, which is heard in three different permutations. Bach’s twin aims of study and delectation are perfectly balanced in this piece, a demonstration fugue and a great work of art rolled into one.

Aria mit verschiedenen Vera¨nderungen (Goldberg Variations) The Goldberg Variations, BWV 988, perhaps the most perfect of all Bach’s large-scale keyboard works, were ﬁrst published in 174122 and probably composed around 1740— thus during the compilation of the WTC II. These two great works together cover the three main principles of the keyboard and instrumental music of Bach’s last decade, namely fugue, variation, and canon. The title of the variation set reads: ¨ bung bestehend in einer ARIA mit verschiedenen Veraenderungen vors Clavicimbal Clavier U mit 2 Manualen. Denen Liebhabern zur Gemu¨ths-Ergetzung verfertiget . . . Nu¨rnberg in Verlegung Balthasar Schmids (Keyboard Practice, consisting of an Aria with Divers Variations, for the Harpsichord with two Manuals. Composed for Music Lovers, to delight their Spirits . . . Nuremberg, published by Balthasar Schmid)

The wording of the title page is thus similar to that of Clavieru¨bung I–III, and the speciﬁcation of two-manual harpsichord recalls Clavieru¨bung II. Therefore it seems reasonable to suppose, as is often assumed, that Bach intended the variations to form Part IV of the series. But it is curious that no reference is made to Part IV on the title page, and there are other causes for uncertainty as to whether the variations truly belong to the series or are designed to stand alone.23 Another thorny issue is the intended occasion of the variations. There is no dedication that might have conﬁrmed Forkel’s story that they were written for the young J. G. Goldberg, a pupil of Bach’s, to play to Count von Keyserlingk, Russian ambassador to the Dresden court.24 On the contrary, the lack of dedication casts doubt on the story. It has been suggested, however, that Bach might have presented a copy of the newly published work to Keyserlingk when he stayed at the count’s Dresden home in November 1741.25 Forkel adds that Bach ‘had hitherto considered [variations] as an ungrateful task on account of the constant sameness of the fundamental harmony’.26 This may well be true, for before about 1740 Bach seems to have composed no secular variation sets

Regarding the publication date, see Gregory G. Butler, ‘Neues zur Datierung der Goldberg-Variationen’, BJ 74 (1988), pp. 219–23. 23 See Albert Clement, Der dritte Teil der Clavieru¨bung von Johann Sebastian Bach: Musik, Text, Theologie (Middelburg, 1999), pp. 4–10. 24 BD VII, pp. 64–5; NBR, pp. 464–5. 25 See C. Wolff, Bach: Essays on His Life and Music, p. 213. 26 BD VII, pp. 64–5; NBR, pp. 464–5.

goldber g variations

347

since the Aria variata (BWV 989), which most likely dates from the first decade of the eighteenth century. Not surprisingly, considering the vast gulf between their dates of origin, the two sets have little in common beyond the use of a decorated aria as theme. The stylistic gap between them is immense. Furthermore, the early variation set preserves not only the harmonic framework of the theme but often its melodic outline too. The late set, on the other hand, is founded only on the bass line of the aria, together with its implied harmonies. The beautiful decorated, sarabande-like Aria, written in Bach’s most progressive, galant style,27 assumes importance, however, through its statement both before and after the variations, which gives it the function of a frame. The Aria is quintessentially Bachian and there is no solid foundation for the old view that it might have been composed by someone else. Its bass, the foundation of the intervening variations, often changes to a certain extent, but at root it consists of four statements of an eight-bar pattern, producing a 32-bar binary structure (16 + 16) and cadencing after every eight bars in keys I, V, vi, and I. This bass and its implied harmonies provide a clear, strong, stable, and symmetrical basis for the variations. The two statements of the Aria enclose a series of thirty variations (it is thought that an early version might have contained only twenty-four).28 Within the variation framework Bach systematically explores the principles of 1. stylistic diversity; 2. keyboard virtuosity; and 3. strict counterpoint. Hence the thirty variations are organized in ten groups of three, each group consisting of character-piece, study, and canon (only the first and last groups are slightly differently ordered). The character-pieces, including the Aria which obviously belongs to this type, are: Part I: Aria, Var. 2 [Trio], 4 [Stretto], 7 Al tempo di Giga, 10 Fughetta, 13 [Cantabile] Part II: Var. 16 Ouverture, 19 [Trio], 22 Alla breve, 25 Adagio, 30 Quodlibet, Aria To some extent, then, they tend to alternate between contrapuntal and homophonic textures. In four cases, the character-pieces of Part I have a clear counterpart in Part II: the Aria, the trios (Var. 2 and 19), the alla breve pieces (Var. 10 and 22), and the cantabile movements (Var. 13 and 25). The Aria, of course, remains unchanged in its second statement, but the other movement types are each subjected to some form of enhancement on their return in Part II. The trio in Part I (Var. 2) is an imitative piece for two treble parts over a walking-quaver bass, a texture normally associated with the trio sonata; that of Part II (Var. 19), on the other hand, exchanges its four-bar figural theme between all three parts—the bass is no less thematic than the upper voices. The alla breve Fughetta in Part I (Var. 10) has a new entry of the subject every four bars; the

27 Regarding the style and authorship of this Aria, see Robert L. Marshall, The Music of Johann Sebastian Bach: The Sources, the Style, the Signiﬁcance (New York, 1989), pp. 54–8, and David Schulenberg, The Keyboard Music of J. S. Bach (London, 1993), p. 326. 28 See Werner Breig, ‘Bachs Goldberg-Variationen als zyklisches Werk’, Archiv fu¨r Musikwissenschaft, 32 (1975), pp. 243–65.

348 the we ll-temper ed clavier ii etc. Alla breve of Part II (Var. 22), on the other hand, treats its suspension-figure subject in close imitation throughout. Finally, the Cantabile from Part I (Var. 13), which recaptures something of the character of the original Aria, is a florid treble solo, to be played on one manual to the accompaniment of twin supporting parts on the other. This texture, a frequent resource in Bach’s mature keyboard slow movements, recurs in the cantabile Adagio from Part II (Var. 25) which, however, is a highly chromatic minore, introducing a new depth of feeling into the variations. Two successive character-pieces in Part I, Variations 4 and 7, are singular and nonrecurring: the Stretto and the ‘Al tempo di Giga’. The Stretto is a miniature counterfugue in four voices, making light-hearted play on the direct and inverted forms of its three-note subject. ‘Al tempo di Giga’, with its constant dotted rhythms in compoundduple time, belongs to the canarie type. Part II opens and closes with singular movement types, Ouverture (Var. 16) and Quodlibet (Var. 30). The French overture is deliberately chosen to mark the beginning of Part II of the variation set (it has a similar function in Clavieru¨bung I and II). Two of the three sections of the overture form (the third is absent here) are so disposed that the majestic dotted-rhythm section forms the first strain of the binary form, and the quick 3/8 fugato, the second strain. The Quodlibet provides a cheerful and witty conclusion. Phrases from two German folk songs, ‘Ich bin so lang nicht bei dir g’west’ and ‘Kraut und Ru¨ben haben mich vertrieben’, are ingeniously built into a four-part texture founded on the fundamental bass of the whole composition. It is often suggested that here Bach might have been recalling the quodlibets that, according to Forkel, used to be sung at the annual gatherings of the Bach family.29 The studies gradually increase in keyboard brilliance and virtuosity throughout. The first three studies exhibit a progression in manual requirements: one manual (Var. 1), one or two manuals (Var. 5), and two manuals (Var. 8). Thereafter two manuals are required in all the remaining studies (except that the last, Variation 29, is for one or two manuals). The five studies of Part I, Variations 1, 5, 8, 11, and 14, are all duets for the two hands and in every case, to an increasing degree, they demand handcrossing. Here, then, as in Part II, a keyboard technique that Bach had employed but sparingly in the past (the Giga from Partita No. 1 and the Praeludium in B♭ from WTC II are rare examples) becomes standard. We cannot rule out the possibility, despite the lack of evidence, that he had become acquainted with Domenico Scarlatti’s Essercizi not long after its publication in 1739. Scarlatti famously warned the player in his preface ‘not to expect any profound Learning, but rather an ingenious Jesting with Art’. In Bach, of course, one does indeed find ‘profound learning’, above all in the canons of this variation set. But in the brilliant studies one also finds a ‘jesting with art’ similar to that which Scarlatti might have had in mind. What better description could there be of Variation 23, for example, with its quixotic series of unexpected events?

BD VII, pp. 15–16; NBR, pp. 424–5.

goldber g variations

349

Whereas the first three studies of Part I (Var. 1, 5, and 8) are in 3/4 time with semiquavers against quavers, a clear rhythmic enhancement takes place in the last two: Variation 11 has ‘triplet’ semiquavers in 12/16 against unequal, ‘gigue’ rhythms (quaver plus semiquaver); and Variation 14, demisemiquaver groups and mordent figures divided between the hands. The first two studies of Part II, Variations 17 and 20, are hand-crossing duets of the type that has been established in Part I, albeit with novel figurations and modes of rhythmic movement. The last four, however, Variations 23, 26, 28, and 29, are largely freistimmig, which permits much variety of texture alongside diversity of rhythmic motion. Thus Variation 23 contains not only demisemiquaver and mordent figures but 3rds in alternating hands and contrary-motion scale figures in 3rds and 6ths. Variation 26, a trio, has one part in 18/16 semiquavers against two parts in 3/4 dotted rhythms. Variation 28 oscillates between two, three, and four voices, and its written-out trill figures later occur in both hands in 6ths or 10ths. Finally, the last study, Variation 29, provides a climax of brilliance, with its chords in quick alternation between the hands and its dazzling triplet figures, divided between two parts but to be played so as to sound like a single unbroken line. The last of each group of three variations is a canon at the unison, 2nd, 3rd, and so on, up to the 9th. If the original version contained only twenty-four variations, the last canon would have been at the octave (Var. 24). Why, one wonders, did Bach add two more groups of three variations (nos. 25–7 and 28–30) but only one more canon, that at the 9th (no. 27)? The answer might be connected with the special nature of the final variation, no. 30. As a Quodlibet it acts as a culmination of the series of character-pieces. But as a piece of ingenious counterpoint with much exact imitation, it might also be viewed as a light-hearted climax to the canonic series. This would also explain why the last group of three contains, in addition to the Quodlibet, two studies (Var. 28 and 29), for they too have their culmination at this point. An observation that lends support to the theory of an original group of eight canons only is connected with their structure: the eight canons from unison to octave are all trios with two upper canonic voices accompanied by a free bass; the Canon at the 9th (Var. 27), on the other hand, is for two voices on separate manuals, lacking a free part. This represents a remarkable new departure at such a late stage in the composition. The unison canon (Var. 3), with its two equal treble parts, replicates the trio-sonata texture of the immediately preceding Variation 2, though with the latter’s 2/4 metre replaced by a pastorale-like 12/8. The same texture is then applied to a light, flowing 3/8 in the Canon at the 2nd (Var. 6). The rather weightier Canon at the 3rd (Var. 9) is notable for its very distinct single-bar theme, which is of frequent recurrence. As in the non-canonic variations, there is a noticeable increase in intensity as the halfway mark approaches. Thus Variations 12 and 15 are both canons by inversion, at the 4th and 5th respectively; and Variation 15 is a minore—the first in the whole set—full of pathos-filled couplets, a figure that had long before become one of Bach’s hallmarks. This exceptionally rich, intricate movement forms a worthy conclusion to Part I of the set. The first canon of Part II, the Canon at the 6th (Var. 18), is written in a traditional

350 the we ll-temper ed clavier ii etc. alla breve style, foreshadowing the movement entitled ‘Alla breve’ (Var. 22). The Canon at the 7th (Var. 21), on the other hand, refers back to its counterpart in Part I, the 5th canon (Var. 15): it is another minore, and the chromatic 4th in the bass at the outset gives a cue for much affective chromatic writing thereafter. The last two canons are increasingly light in tone as a contrast to the weighty tonic-minor Adagio (Var. 25), and the climactic virtuosity of the last three studies (Vars. 26 and 28–9). The Canon at the Octave (Var. 24), is in a gigue-like 9/8, whose light touch conceals novel construction: whereas all previous canonic variations take the form of two canons, one for each strain of the binary form, this movement is constructed as four canons, one for each eight-bar period. The Canon at the 9th (Var. 27) is not only the last but the lightest, being in two-part texture only and preserving an even semiquaver ﬂow (in compound-duple time) almost without intermission. Hearing or playing the canons of this variation set, one is often struck by their naturalness. Unlike the Essercizi, they exhibit ‘profound learning’, but it is invariably unobtrusive, allowing them to be appreciated purely for their expressive beauty. Even awkward corners are redeemed by being made to serve a clear expressive purpose.

Verschiedene Canones (Fourteen Canons) When Bach’s Handexemplar (personal copy) of the Goldberg Variations came to light in 1975, it was found to contain, by way of appendix, the autograph of fourteen hitherto unknown canons by Bach, entitled ‘Verschiedene Canones u¨ber die ersteren acht Fundamental-Noten vorheriger Arie’ (‘Various Canons on the First Eight Notes of the Ground of the Preceding Aria’), BWV 1087. The manuscript dates from around 1747/8.30 Presumably reconsideration of the nine canons from the Goldberg Variations after their publication induced Bach to attempt further canonic work on the kernel of the same ground bass. Closest to a possible model among the variations is Variation 12, bars 1–8, where the ground is particularly clear. This is also one of only two cases in the set of canon by inversion (the other being Var. 15), which becomes standard in the Fourteen Canons. In general, the principle of canons on a ground is extracted from the variation set in a modiﬁed form: the 32-bar bass is represented by its ﬁrst eight bars only—the kernel out of which all four periods are built. Moreover, the bass is no longer elaborated but rather presented in a plain form throughout; in other words, it is reduced to absolute essentials. The concept of a given theme as the underlying subject of a series of ‘canonic variations’ recurs not only in the work of that name, the ‘Canonische Vera¨nderungen’ on the chorale Vom Himmel hoch, BWV 769, but also in the ten canons from the Musical Offering, BWV 1079, whose given subject is the ‘Thema Regium’ (Royal Theme). In each case we encounter two principal modes of treatment of the theme. 30 According to Kobayashi Chr, p. 60, and the same author’s NBA IX/2, p. 210. The canons might have been composed some years earlier, however (c. 1742/6), for two of them, BWV 1076 and 1077, survive in earlier sources.

four teen canons

351

First, it is treated in the manner of a cantus firmus alongside the canonic voices. This applies to all four trios from the Fourteen Canons (nos. 6–9) and also to the five-part double canon no. 11, which (in a variant version, BWV 1077) Bach dedicated to the theology student Johann Gottfried Fulde on 15 October 1747. The theme is in the bass in every case but one—no. 8, where it migrates to the middle part. The canonic voices frequently refer to the theme in some way, most noticeably in no. 7, where it is presented repeatedly in double diminution. A similar mode of treatment, albeit on a far greater scale, applies to the two trios that open the Vom Himmel hoch Variations. In principle, the four-part Variations III and IV from that set also belong to the same type, except that each of them includes a free voice—unthinkable in the strictly logical world of the Fourteen Canons. In the Musical Offering, treatment of the theme as cantus firmus alongside two canonic voices applies to five of the ten canons, the trios nos. 2–5 and 7. In the second principal mode of treatment, the theme itself is subjected to canonic treatment. The Fourteen Canons proceed from this method: nos. 1–4 contain nothing but the plain theme itself, shared by the two canonic voices. No. 3, a canon by inversion on the theme, then forms the basis of the double and triple canons nos. 5 and 13 respectively, in which one or two additional two-part canons are added to that on the theme itself. The six-part canon, no. 13, which may be construed as completely thematic, was (in a slightly different version, BWV 1076) reproduced in Haußmann’s famous Bach portrait in 1746, published as ‘Canon triplex a` 6 Voc:’ in 1747, and presented to Mizler’s ‘Sozieta¨t der musikalischen Wissenschaften’ in the same year. In the five- and four-part canons nos. 12 and 14, the theme is not only presented as a long-note cantus firmus in the bass but is also worked canonically in standard and diminished note-values. In the Vom Himmel hoch Variations, the theme as subject of the canonic voices comes into its own in Variation V, the finale of the printed version, where four canons—at the 6th, 3rd, 2nd, and 9th—are based exclusively on the plain chorale melody, free invention being restricted entirely to the accompanying part/s. In the Musical Offering, the Royal Theme is treated canonically in five of the ten canons, nos. 1, 6, and 8–10. Nos. 1 and 9 are in two parts only and thus wholly dominated by the theme. Nos. 6 and 8 are trios in which the canonically treated theme is accompanied by a freely invented bass (though with thematic references). Finally, no. 10 treats the theme in four-part canon according to a permutation scheme that embraces the subject plus three countersubjects. Of the nine canons in the Goldberg Variations, only two are canons by inversion (Vars. 12 and 15), but this becomes one of the leading principles of the Fourteen Canons, of which only one is answered directly rather than by inversion (no. 9). Bach was clearly fascinated by the more difficult inversion form at this time, so it is hardly surprising that it also features in the other sets of canonic variations of the 1740s. In Vom Himmel hoch, all four canons of the printed finale, Variation V (all those that treat the cantus itself) have the dux (leading voice) answered by inversion. The same is true of four of the Musical Offering canons, namely the trios nos. 3 and 4, in which the cantus is presented alongside two canonic voices, and nos. 8 and 9, in which variants of

352 the we ll-temper ed clavier ii etc. the Royal Theme are themselves answered by inversion. The inversion canon, however, is not the only type among the Fourteen Canons that resonates elsewhere among the works of Bach’s last decade. The set opens and closes with rather more abstruse types, the retrograde canon and the augmentation canon, both of which are also found in Bach’s published works of the period. No. 1 is a cancrizans or crab canon, in which the comes (trailing voice) presents the theme in reverse (no. 2 is the same but with both parts inverted in interval). Interestingly, the first canon from the Musical Offering is entitled ‘Canon a 2 cancrizans’ and proceeds in a similar fashion, though it is considerably more elaborate, introducing a quaver counterpoint to the Royal Theme. The last of the Fourteen Canons is an augmentation canon, a type that also occurs in the other three major canonic works of the period (Vom Himmel hoch, the Musical Offering, and The Art of Fugue). In Variation IV of Vom Himmel hoch, the florid treble voice (dux) is answered at the lower octave by the comes in augmented (doubled) note-values. In addition, there are two non-canonic voices, a freely invented part and the long-note cantus firmus in the bass. This last feature is shared by the last of the Fourteen Canons, though here the cantus firmus doubles as one of the canonic voices. Not only in this piece but in the augmentation canons from The Art of Fugue and the Musical Offering (no. 4), the comes is inverted as well as augmented. The Art of Fugue canon, however, is restricted to the two canonic voices. Canon no. 4 from the Musical Offering has three voices—the decorated Royal Theme occupies the middle ground between the two canonic voices. The last of the Fourteen Canons goes a step further: its four voices are all canonic and all present the principal theme, but in four different metrical forms—(from the bass upwards) minims, crotchets, quavers, and semiquavers. As Bach’s title indicates, it is not just an augmentation canon but a ‘Canon a` 4 per Augmentationem et Diminutionem’. Though lacking the breadth and emotional appeal of the other augmentation canons, it is technically the most remarkable of them and might have induced Bach to attempt the later canons of the same type. The simultaneous presentation of the theme in different note-values, both here and in the related no. 12, recurs in the printed finale to Vom Himmel hoch (Variation V), where it brings the set to a fine climax.

Die Kunst der Fuge The original version of The Art of Fugue as preserved in the autograph score dates largely from about 1742. It thus coincides roughly with the completion of the WTC II. Indeed, the use of manifold contrapuntal artiﬁces in the more complex fugues from that collection might have acted as an inducement to produce a more systematic survey of the principles of fugal writing. It is surely signiﬁcant, too, that the Goldberg Variations were published around the same time (late 1741). For the experience of working on the variation set and the WTC II simultaneously might have led to the idea of combining fugues and canons within an overall variation framework. This concept is already clear in the early version of the autograph score, which consists of

die kunst der fuge

353

twelve fugues (numbered but untitled) and two canons, one of which (the augmentation canon) is already given in two versions. All fourteen movements are based on the principal subject of the work, either direct or inverted, and not only in its original form but in six melodic or rhythmic variant forms. Other aspects of the later version are already in place too. The autograph collection bears the title ‘Die Kunst der Fuga’ in the hand of Johann Christoph Altnickol, Bach’s pupil and assistant from 1744 to 1748. And all the fugues are notated in four- (occasionally three-) stave open score in order to clarify the contrapuntal texture. Since they are presented in fair copies, it is likely that Bach had previously worked them out in separate manuscripts and in twostave keyboard notation. Preparations for a printed edition of the work seem to have begun in about 1747, but the publication process was still incomplete when Bach died in 1750. The work finally appeared posthumously (Leipzig, 1751) under the title ‘Die Kunst der Fuge durch Herrn Johann Sebastian Bach, ehemahligen Capellmeister und Musikdirector zu Leipzig’, and with a note on the verso of the title page explaining that a chorale prelude (BWV 668a) had been included ‘to compensate the friends of [Bach’s] muse’ for the unfinished state of the last fugue. A second impression appeared a few months later in which this note was replaced by a substantial preface by Friedrich Wilhelm Marpurg, recommending the work to students and identifying it as the inspiration of his forthcoming treatise Abhandlung von der Fuge (Berlin, 1753–4). In the printed edition, the fugues are entitled ‘Contrapunctus’ and their order is altered to reflect the systematic nature of the collection. Furthermore, three new pieces are added—Contrapunctus 4 and the canons at the 10th and 12th. And a thorough revision of the text is undertaken: in particular, four of the fugues, Contrapunctus 1–3 and 10, are significantly expanded. It appears that Contrapunctus 1–13 and the four canons were prepared for engraving under Bach’s own supervision, for the engraver’s copies were almost exclusively in his own hand. At a late stage in the publication process, however, Bach’s second-youngest son Johann Christoph Friedrich took over the preparation of the engraver’s copies, and there is considerable doubt as to whether the four pieces concerned really belong to the collection. Easiest to dispose of is the chorale Wenn wir in ho¨chsten No¨ten sein, BWV 668a, which we are told was added merely by way of compensation for the incomplete state of the last fugue. A second piece prepared by J. C. F. Bach, entitled ‘Contrap: a 4’, reproduces the early version of Contrapunctus 10, which Bach had since greatly expanded for inclusion in the printed edition. The early version, therefore, must have been included in error. J. C. F. Bach also includes both direct and inverted versions of the ‘Fuga a 2 Clav:’, his father’s arrangement for two keyboards of the three-part mirror fugue. This arrangement belongs with the early version of The Art of Fugue, dating from the same period as the autograph score, and it is doubtful whether Bach intended it to be included in the printed version. He cannot have meant it to replace the single-keyboard version, Contrapunctus 13, since the engraver’s copy of the inversus was in Bach’s own hand. As an alternative version for practical use, it lies outside the rigorous scheme of the work

354 the we ll-temper ed clavier ii etc. and could have been included only as an appendix. The most problematic of all the pieces prepared for publication by J. C. F. Bach is the great unfinished ‘Fuga a 3 Soggetti’. He copied it from the autograph fragment of 1748–9, making certain adjustments in the process: he added the title just quoted (Bach had left it untitled), altered the composer’s two-stave keyboard notation to four-stave open score, in line with the other fugues, and reduced the 239 bars of Bach’s fragment to 232 bars in order to end at a suitable cadence. There has long been considerable uncertainty as to whether this unfinished fugue genuinely belongs to The Art of Fugue. On the one hand, its first subject seems too similar to the principal theme of the collection but cannot be construed as a variation of it; its three subjects occur nowhere else in the work; and the Art of Fugue theme cannot be combined with them without adjustment. The engraver’s-copy character of Bach’s manuscript seems to rule out The Art of Fugue as its destination and point to some other use—a 1750 presentation to Mizler’s society has been suggested.31 On the other hand, would Bach have had the time or motivation to prepare a new fugue on an unprecedented scale for a quite different purpose when the engravers of The Art of Fugue were awaiting copy of the last two fugues and the composer was struggling to complete his expanded version of the collection before ill health intervened? Moreover, the unfinished piece is not only in the same key as The Art of Fugue (D minor) but in stylistic terms is fully in keeping with many of its constituent fugues. It is written on spare sheets of the paper that had been used for the engraver’s copy of the canons—paper that Bach might have used for the fugue simply because it was left over. It is perfectly possible that he intended the fugue, drafted without title and on two staves, to be given the title ‘Contrapunctus 14’ and to be copied into four-stave open score for the engravers. The well-known obituary account of the last stages in the composition of The Art of Fugue reads: His last illness prevented him from finishing his project of bringing the penultimate fugue to completion, and from working out the last one, which was to contain four themes and to have been afterwards inverted note-for-note in all four voices.32

This account is open to several different interpretations. One possibility is that ‘vorletzte’ and ‘letzte’ (‘penultimate’ and ‘last’) refer only to the unﬁnished third and unwritten fourth sections of the existing incomplete fugue.33 The obituary description of the last fugue, however, strongly suggests that it was to have been a mirror fugue a 4, based on four subjects and presented in two versions, of which the

31 By Gregory G. Butler, ‘Scribes, Engravers, and Notational Styles: The ﬁnal Disposition of Bach’s Art of Fugue’, in G. G. Butler, G. B. Stauffer, and M. D. Greer (eds.), About Bach (Urbana and Chicago, 2008), pp. 111–23 (esp. 117–18). 32 BD III, No. 666 (p. 86); NBR, No. 306 (p. 304). 33 This theory was originally put forward by Erich Bergel, Bachs letzte Fuge: Die ‘Kunst der Fuge’—ein zyklisches Werk (Bonn, 1985), p. 40.

die kunst der fuge

355

second would have been a total inversion of the first. This sounds very much like Contrapunctus 12 and 13 in all but the completely thematic texture, which would have rendered it the culmination of the group of mirror fugues. If this was to have been the last fugue, what was the unfinished penultimate one? It has recently been suggested that it might have been the previous mirror fugue, Contrapunctus 13, of which Bach prepared only the first version for the engraver.34 Whoever prepared the second version, however, would have found both versions complete in the autograph score. All he would have to do is copy out the second version while doubling the note-values. Contrapunctus 13, then, hardly counts as incomplete. The possibility remains that the ‘penultimate fugue’ really was the unfinished one that we know. There is a certain coherence between three crucial pieces of evidence: 1. Bach was working on it in 1749 till he became incapable of any further work;35 2. the obituary informs us that his last illness prevented him from completing the ‘penultimate fugue’; 3. C. P. E. Bach’s note in the score of the incomplete fugue reads, ‘While working on this fugue, in which the name “BACH” appears in the countersubject, the author died.’ Death, then, prevented the completion of the fugue, but it might have been Bach’s last illness that prevented its continuation. If the unfinished piece really does belong to The Art of Fugue, it would have belonged to the group of compound fugues (Contrapunctus 8–11). Just as Contrapunctus 8 is a triple fugue in which two new subjects are followed by and then combined with the Art of Fugue theme, so the unfinished fugue would have been a quadruple fugue (since we must postulate the return of the original theme) with three subjects followed by the Art of Fugue theme which, though absent from Bach’s autograph fragment, has long been known to combine with the other subjects.36 And just as all three subjects are inverted in the second triple fugue, Contrapunctus 11, so all four subjects would have been inverted in the completed version of the unfinished fugue. Thus, just as the ‘last fugue’ of the obituary account would have crowned the group of mirror fugues, so the ‘penultimate fugue’ would have crowned the group of compound fugues. The Art of Fugue adheres to a long-standing tradition of open-score notation for keyboard works that illustrate the principles of strict counterpoint, including Frescobaldi’s Fiori musicali (1635) and Scheidt’s Tabulatura nova (1624).37 Since it is primarily concerned with counterpoint rather than fugue, Bach entitles the pieces of the first four groups ‘Contrapunctus’, and the fifth group is entirely devoted to canon. It is not altogether clear, then, why the collection as a whole is entitled ‘Die Kunst der Fuge’; a 34

See Butler, ‘Scribes, Engravers, and Notational Styles’, p. 118. In Kobayashi Chr, p. 62, it is dated after Aug. 1748 to Oct. 1749. Ever since Gustav Nottebohm’s demonstration in ‘J. S. Bachs letzte Fuge’, Musik-Welt (1880–1), pp. 232–6 and 244–6. 37 See Friedrich W. Riedel, Quellenkundliche Beitra¨ge zur Geschichte der Musik fu¨r Tasteninstrumente in der zweiten Ha¨lfte des 17. Jahrhunderts (Kassel, 1960), pp. 82–7, and the same author’s ‘J. S. Bachs Kunst der Fuge und die Fugenbu¨cher der italienischen und o¨sterreichischen Organisten des 17. und 18. Jahrhunderts’, in F. Heidlberger, W. Osthoff, and R. Wiesend (eds.), Von Isaac bis Bach: Studien zur a¨lteren deutschen Musikgeschichte: Festschrift Martin Just (Kassel, 1991), pp. 327–33. 35

356 the we ll-temper ed clavier ii etc. more appropriate title might have been ‘Die Kunst des Kontrapunkts’. Nevertheless, the work is clearly divisible into five parts, of which four are concerned with various types of fugue and the fifth with canon: simple fugue (nos. 1–4), counter-fugue (nos. 5–7), compound fugue (nos. 8–11), mirror fugue (nos. 12–13), and canon (nos. 15–18). Within these categories, multiple forms of variation are in operation simultaneously—the work may be viewed as a series of contrapuntal variations upon a single theme. This theme is stated direct and inverted, diminished and augmented, and in eight variant forms. In addition, it is combined with itself in stretto and with other subjects in invertible counterpoint at the octave, 10th, and 12th. Thus all the main devices of strict counterpoint come into play at some point during the course of the work. The first two of the simple fugues, Contrapunctus 1 and 2, are concerned above all with the presentation of the principal theme in its direct and unvaried form. Thus there are no regular countersubjects or special contrapuntal devices of any kind. The accompanying parts, however, are often syncopated in No. 1 and in dotted rhythm throughout in No. 2, apparently as an afterthought,38 which highlights Bach’s treatment of free parts as another form of variation. Contrapunctus 3 and 4 are both based on the inverted subject—in No. 3, not only in its plain form but also (from b. 23) in an expressive, syncopated variant form. Both forms are combined with a partly chromatic regular countersubject (b. 5), which lends No. 3 exceptional expressive intensity. Contrapunctus 4, a late addition to The Art of Fugue, contains only the plain form of the inverted subject, but compensates by intervallic enhancement (at notes 4–5; see bb. 61, 65, 73, and 77), which gives it a ‘powerfully rhetorical modulating effect’,39 and later by presenting on- and off-beat entries of the subject in close stretto (bb. 107 and 111). The ostinato figure of the countersubject (bb. 5–6) and the cuckoo-like figure of the first episode (bb. 19–23) both recur throughout and lend the piece an insistent, forceful character in marked contrast to the expressive intensity of No. 3. Contrapunctus 5–7 form a group of counter-fugues or, to be more precise, strettoinversion fugues, since not only is the direct subject answered by its inversion, or vice versa, but rectus and inversus are frequently combined in stretto. All three fugues are based on variant 1 of the principal theme, the version with filled-in 3rds (here no longer syncopated) that was introduced in Contrapunctus 3. The smoothly flowing, largely stepwise counterpoint of No. 5 generates a limpid euphony. Bach demonstrates that the subject, whether direct or inverted, can be combined with itself in stretto at three different points: at the half-bar (b. 33), at one bar (b. 65), and at one-and-a-half bars (b. 47). The constant alternation of rectus and inversus is clinched by their simultaneous entry in the coda (b. 86). Contrapunctus 6 is written in the majestic style of the French ouverture, with its dotted rhythms and demisemiquaver upbeat

38 According to Wolfgang Wiemer, ‘Eine unbekannte Fru¨hfassung des Contrapunctus 2’, Die Musikforschung, 34 (1981), pp. 413–22. 39 In the words of Donald Francis Tovey, A Companion to ‘The Art of Fugue’ (London, 1931), p. 9.

die kunst der fuge

357

figures, hence the subtitle ‘In Stylo Francese’. It introduces the principle of metrical diminution. Typically, a subject entry in standard note-values is combined in stretto with two entries of the diminished subject. These same entries also alternate between rectus and inversus according to the principles of counter-fugue. In Contrapunctus 7 the subject is presented not only in metrical diminution but also in augmentation. A massive edifice is built around four entries of the augmented subject, ascending through the voices (bass, tenor, alto, soprano; bb. 5, 23, 35, 50), inverted and direct in alternation. Each augmented entry is accompanied by a myriad of stretto entries, in which the subject constantly alternates between rectus and inversus or between standard and diminished note-values. The next group, Contrapunctus 8–11, is made up of compound fugues. New subjects are first treated on their own and then combined with each other and with the Art of Fugue subject in invertible counterpoint. Contrapunctus 8 and 11 are both triple fugues; that is, they are based on three subjects, one of which is a new variant of the Art of Fugue theme. The two fugues share the same three subjects and are placed together in the autograph score. Their separation in the original edition might have two possible reasons: to avoid the direct juxtaposition of similar fugues or so that the simplest and most complex fugues, Nos. 8 and 11 respectively, should begin and end the compound-fugue group. Contrapunctus 8 is intimate and refined—a reflection of the character of its subjects and its restriction to three voices. Its four expositions, articulated by emphatic cadences (bb. 39, 93, and 123–4), clearly demarcate the subjects (S I–III) and their treatment. The last theme to enter is S III, a variant of the Art of Fugue theme in ostinato rhythm (b. 94). In the final exposition (b. 124), after a reprise of the first two subjects combined, we hear four permutations of the triple counterpoint of S I + II + III (bb. 147, 152, 158, and 170), of which the first recurs at the close (b. 183). In the four-voice Contrapunctus 11, the three subjects of No. 8 are inverted and enter in a different order: the ostinato variant of the Art of Fugue theme now comes first. All subjects are subsequently reinverted. The dense chromaticism developed out of the two subsidiary subjects renders this fugue one of the most intensely expressive in the whole collection. The principles of thematic combination and inversion that inform the whole fugue are brought to a climax in the last of the five expositions (bb. 129–84). Here, not only are all three subjects combined in triple counterpoint, but the Art of Fugue theme in its ostinato variant form is twice given direct and inverted simultaneously (bb. 158 and 164) in double counterpoint at the 10th. Contrapunctus 9 and 10 are double fugues in invertible counterpoint at the 12th and 10th respectively. Unlike the triple fugues, they contrast sharply: No. 9 is relatively light in tone, whereas No. 10 returns to the dark, passionate, intense tone of No. 3. The use of double counterpoint at the 12th in Contrapunctus 9 creates a different harmonic perspective for the combined subjects, varying their expressive character. The first subject is a lively, athletic theme in stepwise flowing quavers; the second, the Art of Fugue theme in its original form. The invertible counterpoint at the 10th of Contrapunctus 10 allows either or both subjects to be doubled in 3rds, 6ths, or 10ths. The first

358 the we ll-temper ed clavier ii etc. Ex. 2

a) 1st subject of Contrapunctus 10

b) 1st subject of Contrapunctus 8

c) 3rd subject of Contrapunctus 8 (bb. 94–8, middle part)

subject is a relative of the first and third subjects from No. 8, sharing its interval structure with the one and its rhythmic shape with the other (Ex. 2). The second subject is the first variant of the inverted Art of Fugue theme, the form used in the counter-fugues (Nos. 5–7) and ultimately derived from No. 3 (b. 23). Contrapunctus 12 and 13 form a pair of mirror fugues: they are totally invertible, for not only the melodic lines but also the entire contrapuntal texture are fully inverted in the ‘Contrapunctus inversus’ that follows each fugue and forms a mirror image of it. No. 12 is in a sedate, measured, sarabande-like 3/2 time. The subject is a new tripletime variant of the inverted Art of Fugue theme, which subsequently undergoes elaboration. The presence of plain and decorated forms of the original theme within the same fugue recalls Contrapunctus 3. No. 13, like No. 8, is for three voices only. The subject, a ‘brilliant comic variation’ of the direct Art of Fugue theme,40 is answered by inversion according to the counter-fugue principle of Nos. 5–7. If No. 12 is rhythmically akin to the sarabande, the lively, flowing triplet rhythms of No. 13 recall the gigue. The last group of pieces, Nos. 15–18, consists of four canons in two-part texture, each based on a different variant of the basic theme. Only the octave and augmentation canons were present in the early version of the autograph score; the canons at the 10th and 12th were added for the later, printed version. In the original edition, the octave, 10th, and 12th canons form a logical sequence. The augmentation canon precedes them, but evidence has been uncovered that it was originally intended to end the canonic group, in keeping with its character as the longest and most complex of the four canons.41

According to Tovey, A Companion to ‘The Art of Fugue’, p. 33. See Gregory G. Butler, ‘Ordering Problems in J. S. Bach’s Art of Fugue Resolved’, Musical Quarterly, 69 (1983), pp. 44–61. 41

die kunst der fuge

359

The ‘Canon alla Ottava’ is based on a lively, gigue-like variant of the inverted principal subject in compound triple time. There are four periods—the last being a reprise of the first (b. 77)—each led by the subject, which is presented in dux form in the first period, in comes form in the second (b. 25), and in inverted form in the third (b. 41). The ‘Canon alla Decima’ is based on a syncopated variant of the inverted principal subject in compound quadruple time. It falls into two halves, of which the second (b. 40) is a reprise of the first in double counterpoint at the 10th. The theme of the ‘Canon alla Duodecima’ is an elaborate variant of the direct principal subject, notable for its sextolet figures. The overall structure is virtually identical with that of the 10th Canon: it falls into two halves, the second being a reprise of the first but in double counterpoint at the 12th. This structure was presumably modelled on that of the ‘Canon per Augmentationem in Contrario Motu’, which was already present in the early version. This canon also falls into two parts, the second (b. 53) being a reprise of the first in double counterpoint at the octave. The theme, an elaborate variant of the direct principal subject, is answered in a form modified by inversion and augmentation. The great unfinished fugue was accurately named ‘Fuga a 3 soggetti’ in the original edition, for as it stands it is a triple fugue along the lines of Contrapunctus 8 and 11. Yet none of its three subjects is the principal theme of The Art of Fugue; and if it were destined for that work, the main theme would undoubtedly have been included in the finished version, as in all the other fugues and canons. Thus, in its complete form it would have been a quadruple fugue, treating four subjects and combining them in invertible counterpoint. As it stands, the fugue falls into three large sections or parts, of which the third is left incomplete. Part I is devoted to S[ubject] I, which has a superficial resemblance to the Art of Fugue theme, but is slow and massive in rhythm and palindromic in shape. Part II (b. 114) presents an exposition of S II, which moves in flowing quavers, followed by its combination with S I (b. 147). Part III (b. 193) gives a very thorough treatment of S III, a musical representation of Bach’s signature. Finally, at the climax of Part III (b. 233) all three themes are combined. Judging by Contrapunctus 8 and 11, Bach would not have been content with a single statement of this combination, but would have presented a number of different permutations of the triple counterpoint. If the piece was indeed intended for The Art of Fugue, a fourth and final part would have followed in which S IV—the Art of Fugue theme, either plain or decorated—would first have been presented alone (perhaps in direct and inverted forms and in stretto, as in the case of S I and S III), then combined with the other three subjects in quadruple counterpoint. The combination with the plain, direct principal subject was demonstrated in 1880;42 here it is also shown how it works with all subjects inverted (Ex. 3).

See n. 36.

360 t he w ell-tempe red clav ie r i i etc. Ex. 3

Author’s attempt at four-subject combination (all themes inverted) such as might have featured in lost or unwritten conclusion of incomplete fugue

Canonische Vera¨nderungen (Vom Himmel hoch) The Canonic Variations, Einige canonische Veraenderungen u¨ber das Weynachts-Lied: Vom Himmel hoch da komm ich her, BWV 769, are preserved in two versions: that of the original edition of 1747, and that of the autograph manuscript of 1746–7. Detailed textual differences between them are less significant than their different order of movements. Variation 5, the finale of the printed set, is situated at the midpoint in the autograph, giving rise to the order Variations 1–2–5–3–4. This major difference reflects the complex history of the work and its engraving, which has recently been reconstructed.43 Variations 1–3 were perhaps composed in late 1745 and engraved around Easter 1746. In these, the simplest of the five variations, two canonic voices are combined with the chorale cantus firmus. They were engraved in enigmatic notation, which suggests an initial abstract, theoretical conception of the work. Somewhat later Variation 5 was added. It is longer and more complex than nos. 1–3, containing four canons by inversion—at the 6th, 3rd, 2nd, and 9th—in which the chorale cantus firmus itself forms the subject of the canonic voices. Like the autograph fair copy of the whole set, it is notated in organ score, which suggests a shift from a theoretical to a more practical conception of the work. That it was originally intended to form the finale of the work, as it does in the printed version, is strongly suggested by the tour de force that forms its coda (bb. 52ff.), which is crowded with diminished and stretto entries—the latter involving all four lines of the chorale—and ends with Bach’s musical signature. Variation 4, the last to be composed, was added after a considerable 43 ¨ bung III: The Making of a Print; with a Companion Study of the By Gregory G. Butler, Bach’s Clavier-U Canonic Variations on ‘Vom Himmel hoch’, BWV 769 (Durham and London, 1990), pp. 91–116. See also the same author’s ‘J. S. Bachs Kanonische Vera¨nderungen u¨ber “Vom Himmel hoch”, BWV 769: Ein Schlußstrich unter die Debatte um die Frage der “Fassung letzter Hand” ’, BJ 86 (2000), pp. 9–34.

s ch u¨ b ler chor ales

361

interval: it was written no earlier than the end of 1746 and engraved in mid-1747. Like Variation 5, it is longer and more complex than the earlier variations and introduces a new type, the augmentation canon, which around the same time Bach included in The Art of Fugue. The printed version of Variation 4 is (like The Art of Fugue) notated in four-stave open score in order to clarify its contrapuntal structure. This has been connected with the next stage in the work’s history:44 on the occasion of his induction as a member of Mizler’s ‘Sozieta¨t der musikalischen Wissenschaften’ in June 1747, Bach submitted to the society a presentation copy of the Canonic Variations in which the canons were ‘fully worked out’; in other words, they were fully realized in score. This presumably means that the presentation copy, now lost, was notated in open score, just like the engraved version of Variation 4. Publication of the work followed in the latter half of 1747, perhaps at the Michaelmas Fair. Variations 1–4 fall into two pairs. Nos. 1 and 2 are in three-part texture with the chorale cantus firmus in the pedals. They accord most readily with Bach’s title-page specification: ‘vor die Orgel mit 2. Clavieren’. Nos. 3 and 4, on the other hand, are in four-part texture with a florid right-hand, cantabile voice in counterpoint with the chorale cantus firmus. In each case, the florid part is in Bach’s most ‘modern’ style, with considerable variety of note-values, including diminution and syncopated figures. This suggests that Variation 4 might have been conceived as a partner to No. 3. On the other hand, Variation 4, like No. 5, includes Bach’s musical signature towards the end (b. 39), which equips it to act as finale, as it does in the autograph version. Variation 5 might be viewed as occupying an intermediate position between Nos. 1–2 and 3–4: its first two canons (at the 6th and 3rd) share the three-part texture of Nos. 1– 2, whereas its second pair (at the 2nd and 9th) adopt the four-part texture of Nos. 3–4. Although these observations seem to support its central position in the autograph, its coda, which builds up progressively from three to six voices, seems an ideal ending for the work as a whole.

Sechs Chora¨le von verschiedener Art (Schu¨bler Chorales) Bach’s last publication of organ music, the six Schu¨bler Chorales, BWV 645–50, appeared in 1748/9 under the title ‘Sechs Chora¨le von verschiedener Art auf einer Orgel mit 2 Clavieren und Pedal vorzuspielen . . . In Verlegung Joh: Georg Schu¨blers zu Zella am Thu¨ringer Walde’ (‘6 Chorales of various kinds, to be performed on an Organ with two Manuals and Pedal . . . published by Johann Georg Schu¨bler at Zella in the Thuringian Forest’). These organ chorales may be viewed as retrospective in a manner comparable with the eighteen Chorales: those were revised versions of largeformat Weimar organ chorales; these were adapted from vocal-instrumental chorale arrangements drawn from Bach’s Leipzig cantatas of the period 1724–31. Whereas Bach’s other chorale-based organ works of the 1730s and 1740s create a somewhat

By Butler, Companion Study of the Canonic Variations, p. 109.

362 the we ll-temper ed clavier ii etc. austere impression—Clavieru¨bung III due to its use of the stile antico and the old church modes; the Canonic Variations due to their strictness and (in Nos. 1–3) enigmatic notation—the ‘Eighteen’ and Schu¨bler Chorales are more popular and accessible in style and more obviously geared to the practical needs of the church organist. Among the six Schu¨bler Chorales, this relatively ‘popular’ style applies in particular to Nos. 1, 5, and 6 (BWV 645, 649, and 650). All three are chorale trios, made up of cantus firmus (formerly sung), florid obbligato (formerly played on solo or unison strings), and bass (formerly continuo). The obbligato is in each case a highly attractive, melodious theme that first introduces the chorale, in the form of a substantial ritornello, then accompanies it and furnishes inter-line episodes and a conclusion. No. 2, Wo soll ich fliehen hin, BWV 646, is also a trio ‘a` 2 Clav. et Ped.’, but of a quite different kind: the pedal cantus firmus is combined with two manual parts, which are throughout imitative on the basis of an eight-note motive and its inversion. The remarkable similarity between this trio and the organ trio BWV 694, based on the same chorale, raises the possibility that Schu¨bler No. 2 might have been intended for the organ ab initio (no vocal version is known). Schu¨bler Chorales Nos. 3 and 4, Wer nur den lieben Gott la¨ßt walten, BWV 647, and Meine Seele erhebt den Herren, BWV 648, are both in four-part texture. Their vocal originals, BWV 93 no. 4 and 10 no. 5, were written for two successive Sundays in the chorale-cantata year (2 and 9 July 1724), and the type they represent occurs only in these two cases. The role of voices and instruments is reversed: the plain chorale cantus firmus is delivered instrumentally, and the chorale verse sung as an imitative duet to a freely invented theme. In No. 3 the free, imitative (indeed, fugal) theme is derived from the chorale melody; in No. 4 it is independent but descriptive, for the chromatic theme recalls the words ‘He remembers his mercy.’

III.3 The Musical Offering and other instrumental works

Title

Earliest source/s

Scribe/publisher, date

Sonata in E, BWV 1035 Sonata in G, BWV 1027 Sonata in D, BWV 1028 Concerto in A minor, BWV 1044 Musicalisches Opfer (Musical Offering), BWV 1079

Berlin, P 621 Berlin, P 226 Berlin, P 1057 Berlin, P 249 Berlin, St 134 Berlin, P 226 Original edition

Voß-Buch, post-1800 Autograph, c. 1742 C. F. Penzel, 1753 J. F. Agricola, c. 1750 Anon., c. 1750 Autograph, 1747 Leipzig: Breitkopf, 1747

Sonatas Of the four Bach sonatas that can reasonably be dated in the 1740s, two are continuo sonatas—the Flute Sonata in E, BWV 1035, and the trio sonata from the Musical Offering, BWV 1079 no. 3—and both were apparently occasioned by Bach’s visits to the Potsdam court of Frederick the Great in 1741 (BWV 1035) and 1747 (BWV 1079 no. 3). According to the sources, the E major Sonata was written for Michael Gabriel Fredersdorf, valet, private secretary, and fellow flautist of the king’s. Its immediately appealing, relatively undemanding style might have been a concession to the prevailing taste of the court,1 in which case it is more likely to have originated in the 1740s than to have been adapted from a work of Co¨then origin, as was formerly thought. Numerous details of style support the late origin of the sonata. The prevailing texture is homophonic—flute solo with continuo accompaniment. The phrase structure is regular throughout. The opening ‘Adagio ma non tanto’ changes its mode of rhythmic movement from melismatic demisemiquaver groups to triplet semiquavers or semiquaver couplets. The Allegro, no. 2, is in a modish 2/4 metre, increasingly common in Bach’s vocal and instrumental music from the late 1720s onwards. And in the finale, binary dance form (also employed in nos. 2 and 3) is allied with an unspecific dance rhythm. All these features are characteristic of the galant style that was prevalent throughout much of Europe in the mid-eighteenth century.

¨ ber J. S. Bachs Flo¨tensonaten mit Generalbaß’, BJ 58 (1972), pp. 12–23 According to Hans Eppstein, ‘U (esp. 14–15 and 18–19). 1

364

the m u si c a l o f fe r i n g a n d o t h e r w o r k s

For the most part, the continuo takes a supporting role, though it occasionally states the theme (no. 2, b. 73; no. 4, b. 4). In the opening Adagio, the continuo facilitates the lyrical freedom of the flute writing by virtue of its recurring dotted rhythm-cum-mordent figure (b. 2b), which acts like a very free basso quasi ostinato. The one movement in which the flute and continuo collaborate on equal terms is the Siciliana, no. 3. Both strains open with flute and continuo in canon at the octave, recalling the Sarabande from the Ouverture in B minor, BWV 1067. The second phrase (b. 5), with its play on direct and inverted forms of the initial dotted rhythm figure, is imitative; and during the second strain (b. 21) the same figure is treated in imitation by inversion. Such an elaborately contrapuntal treatment of a simple dance is characteristic of Bach:2 however much he accommodated his music to the latest galant fashions, he could never entirely divest himself of his polyphonic heritage. In the three gamba sonatas, BWV 1027–9, the solo instrument plays alongside obbligato harpsichord rather than continuo, which immediately links them with the violin and flute sonatas BWV 1014–19, 1030, and 1032. The Sonata in G minor, BWV 1029, belongs to a fundamentally different type from the other two gamba sonatas, namely the Sonate auf Concertenart (see Part II, Ch. 3). The gamba sonatas in G and D belong to the more traditional sonata da chiesa type, with its four movements disposed in two slow–fast pairs. Since Bach still employed this older type in the E major Flute Sonata and in the trio sonata from the Musical Offering, it can be no argument against a late dating of the gamba sonatas in G and D. Moreover, the autograph fair copy of the Sonata in G, BWV 1027, dates from about 1742. Its musical substance, however, is largely identical with that of the Sonata in G for two flutes and continuo (BWV 1039), which dates from before about 1726: the two works might have had a common source in a lost sonata in G for two violins and continuo.3 Leaving aside the many refinements that were made, the adaptation essentially involved transferring the flute or violin parts to gamba (which involved downward octave transposition) and harpsichord, right hand. The bass line was retained by harpsichord left hand but with the figuring removed. It is possible that not only the G major Sonata for gamba and harpsichord but also the D major Sonata for the same combination, BWV 1028, originated around 1742. For the autograph of the G major Sonata is written on paper of the same type as Bach used in new gamba parts for two movements from the St Matthew Passion (nos. 34–5); and both these parts and the D major Sonata require a larger, seven-string gamba. A possible protagonist for these late gamba parts might have been the virtuoso Carl

2 Features of this kind make it hard to doubt the authenticity of the work, even though such doubts are understandable in view of the adverse source situation; see Robert L. Marshall, ‘J. S. Bach’s Compositions for Solo Flute: A Reconsideration of Their Authenticity and Chronology’, in Marshall, The Music of Johann Sebastian Bach (New York, 1989), pp. 201–25 (see 220). 3 According to H. Eppstein, ‘J. S. Bachs Triosonate G-Dur (BWV 1039) und ihre Beziehungen zur Sonate fu¨r Gambe und Cembalo G-Dur (BWV 1027)’, Die Musikforschung, 18 (1965), pp. 126–37.

c o nc e rt o

365

Friedrich Abel (son of the court gambist at Co¨then when Bach was Capellmeister), who was resident in Leipzig in the late 1730s and early 1740s.4 Another argument for dating the D major Sonata in the early 1740s is the manner in which Bach incorporates elements of the galant style in the music. This hardly applies to the opening Adagio, where a walking-quaver bass accompanies gamba and harpsichord right hand parts that are imitative throughout—to the point of strict canon in bars 14–19. However, the galant element comes very much to the fore in the following Allegro, a non-fugal movement in binary dance form (like its equivalent in the E major Flute Sonata), whose syncopated theme in parallel 10ths over a decorative bass emanates a distinctly modish flavour. But this material is treated in Bach’s customarily strict motivic/thematic, contrapuntal manner. In the second strain, for example, a thematic variant is worked out in two pseudo-fugal expositions (bb. 33, 37, and 48, 51). The Andante, no. 3, like its equivalent in the E major Flute Sonata, is a siciliana (though not so-called in this case)—both themes open with the same dotted-rhythm headmotive. The texture, however, is pseudo-fugal: the theme functions like a fugue subject in the two upper parts over a merely supporting bass. The finale is a non-fugal Allegro—a through-composed concertante movement in 6/8 gigue rhythm. Despite the great charm of its double subject, the movement is problematic in its formal design—a variant of the concertante da capo form that Bach used in the violin and harpsichord sonatas and elsewhere. Section A being much longer than usual (69 bars), Bach compensates by renouncing a full da capo in favour of a severely abridged subdominant reprise (15 bars). This, however, leaves too little room for the full re-establishment of the original key and material. Moreover, only the first 30 bars of the main section A are tonally dynamic; the remaining 39 bars are static, being restricted to the tonic D. The middle section B (41 bars), by contrast, is marked by rapid modulation and includes a long, colourful, concerto-style episode for each of the two soloists. But the material of these episodes is largely free and not integrated into its surroundings, with the result that the movement is not only ill-proportioned but loosely constructed.

Concerto Bach’s set of six solo harpsichord concertos of around 1738 culminates in the Concerto in F, BWV 1057, in which the harpsichord leads a concertino of three instruments. The same is true of the Concerto in A minor, BWV 1044, though a closer analogy would be with the fifth Brandenburg Concerto, whose concertino is identical in make-up, consisting of harpsichord, flute, and violin. It is not impossible that Bach, finding the fifth Brandenburg Concerto a success at the Leipzig Collegium musicum concerts

4 The arguments about the dating of the gamba sonatas and the identity of their dedicatee are indebted to Laurence Dreyfus; see his Peters edn (Frankfurt and Leipzig, 1985), pp. 64–5.

366

the m u s i c a l o f f e r i n g a n d o t h e r w o r k s

in the 1730s (it seems inconceivable that he did not perform it there), deliberately chose to replicate its instrumentation in a new concerto, probably in the 1740s.5 Just as the solo harpsichord concertos are exclusively transcriptions, so too is the A minor Concerto—a fundamentally different type of transcription, however, since the original is not a solo concerto but, for the outer movements, the Praeludium et Fuga in A minor, BWV 894, and for the inner movement the ‘Adagio e dolce’ from the Organ Sonata in D minor, BWV 527 (no. 2). A far greater degree of intervention was required to make concerto movements out of pieces from other genres, whatever concertante elements they might contain. By far the most momentous change was the introduction of framing ritornellos, as well as internal ritornellos largely based on the same material, in the outer movements (no. 1, bb. 14, 35, 68; no. 3, bb. 37, 42, 120). In addition, new counterpoints for flute and violin are interpolated throughout. The keyboard essentially takes over the original harpsichord part, though in a greatly enriched and more cogent fashion. In the outer movements, the flute and violin take a subordinate role; in the slow movement, on the other hand, they are on equal terms with the harpsichord. Two main changes are made to the original organ-sonata movement. First, its trio texture is expanded to a quartet by the addition of a free part, with the result that flute, violin, and both hands of the harpsichordist are fully occupied throughout. Second, the repeats are written out in full, which allows the flute and violin parts to be interchanged.6 This concerto is primarily based on one of Bach’s most brilliant concertante harpsichord works of his earlier years (Weimar, c. 1715), which, however, had a serious flaw: both prelude and fugue were flooded with triplet-semiquaver movement, which created a monotonous impression and left insufficient room for variety. In the concerto version this defect is remedied by introducing a slow movement in a quite different rhythmic mode and by recasting the finale in the alla breve time of its newly composed ritornellos. The result is perhaps the most imaginative and far-reaching of all Bach’s concerto transcriptions. That it is not better known and more often played is perhaps to be attributed to a late survival of the old prejudice against arrangements, transcriptions, and parodies.

Musicalisches Opfer The greatest instrumental work of Bach’s late years is undoubtedly the Musicalisches Opfer (Musical Offering) of 1747. The circumstances of its origin are exceptionally well documented. Bach himself gave a brief account of his visit to the court of Frederick

5 Peter Wollny gives strong arguments for this late dating; see his ‘J. S. Bachs Tripelkonzert A-Moll BWV 1044’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke (Witten, 1997), pp. 283–91. 6 Ulrich Siegele points out that the varied reprise in binary movements, used as an occasional resource by Bach, was then taken up by his son C. P. E. Bach and much cultivated by him; see his Kompositionsweise und Bearbeitungstechnik in der Instrumentalmusik Johann Sebastian Bachs (Neuhausen and Stuttgart, 1975), pp. 45–6.

musicalisches opfer

367

the Great at Potsdam from 7 to 8 May 1747 in his dedication of the original edition to the king, dated 7 July 1747.7 A more detailed account was given in the Berlin press on 11 May, only a few days after the event itself.8 And further details are provided by Forkel, who took his information from Wilhelm Friedemann Bach, an eyewitness of the event.9 Bach himself recalls improvising a fugue on a theme provided by the king. The press report and Forkel both add that it was played on the fortepiano, one of several such instruments ‘made by Silbermann, which stood in several rooms of the palace’ (Forkel). Not mentioned by Bach is that ‘on the following evening, His Majesty charged him with the execution of a fugue in six parts’ (Berlin press report). Forkel also mentions this six-part improvised fugue but adds that it was based on a subject of Bach’s own choice. It is interesting to note that Bach’s intention to compose and publish a work based on the ‘Royal Theme’ is already mentioned in the Berlin press report, only three or four days after Bach played before the king. Bach acted swiftly: in little over four months the ‘Musical Offering’ to the king had been composed and published: it was announced in the Leipzig press on 30 September and copies were available immediately afterwards at the Michaelmas Fair. The Musical Offering was composed when the publication process for The Art of Fugue was about to begin, and up to a point it may be regarded as a by-product of the earlier composition. Both works are essentially monothematic, and in both cases the principal theme is repeatedly modified according to the variation principle. There are certain resemblances between the themes of the two works, particularly their triadic opening (Ex. 1); but the Royal Theme, unlike that of The Art of Fugue, introduces two motives of common occurrence in seventeenth- and early eighteenth-century fugal works, namely the falling diminished 7th and the chromatic descent. It seems likely that these motives originated with the music-loving king, even though Bach might have added refinements.10 The king’s input probably explains why the Royal Theme is not particularly susceptible to contrapuntal artifice, whereas the Art of Fugue theme was clearly designed with that in mind. In both works the genres of fugue and canon are represented, but only the Musical Offering includes a trio sonata, which takes it into the realm of instrumental chamber music. The Musical Offering in its final form consists of a sonata, two fugues, and ten canons. But it is not at all clear from the original edition what order Bach intended these constituent parts to be in. In a Leipzig press announcement of 30 September 1747, for which Bach himself was no doubt responsible, the contents of the work are described as follows:

BD I, No. 173; NBR, No. 245. BD II, No. 554; NBR, No. 239. 9 BD VII, pp. 22–3; NBR, pp. 429–30. 10 Christoph Wolff argues that the Royal Theme was a joint effort of this kind; see his ‘Apropos the Musical Offering: The Thema Regium and the Term Ricercar’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 324–31 (esp. 326–8). 8

368

the m u s ic a l o f f e r i n g a n d o t h e r w o r k s

Ex. 1

a) Thema Regium from Musical Offering transposed up a tone to D minor b) Principal theme of The Art of Fugue

Die Elaboration bestehet 1. in zweyen Fugen, eine mit 3, die andere mit 6 obligaten Stimmen; 2. in einer Sonata a Traversa, Violino e Continuo; 3. in verschiedenen Canonibus, wobey eine Fuga canonica beﬁndlich. (The Elaboration [of the Royal Theme] consists 1. in two Fugues, one with three, the other with six obbligato parts; 2. in a Sonata for Transverse Flute, Violin, and Continuo; 3. in Diverse Canons, among which is a Fuga canonica.)11

A similar ordering by genres is given in the Bach Obituary of 1754, where the contents of the publication are described as ‘Zwo Fugen, ein Trio, und etliche Canones, u¨ber das . . . von Seiner Majesta¨t dem Ko¨nige in Preussen aufgegebene Thema’ (‘Two Fugues, a Trio, and several Canons, on the . . . Theme given by His Majesty the King in Prussia’).12 There is plentiful evidence in the original edition that this order corresponds with Bach’s intentions and even with the order in which the constituent items were composed.13 An early version consisted primarily of the three- and six-part ricercars. At a later stage Bach added the ﬁve numbered canons (Nos. 1–5), the ‘Canones Diversi super Thema Regium’. Finally he added the ‘Sonata sopr’il Soggetto Reale’ (Sonata on the Royal Theme), which was probably the last item to be composed. Unanswered questions remain, however—not least why the other ﬁve canons are unnumbered and dispersed seemingly at random throughout the publication, forming appendices to the major items as shown:14

BD III, No. 558 a (p. 656); NBR, No. 248. BD III, No. 666 (p. 86); NBR, No. 306 (p. 304). 13 See Michael Marissen, ‘More Source-Critical Research on J. S. Bach’s Musical Offering’, Bach, 25/1 (1994), pp. 11–27; Gregory Butler, ‘Eine neue Interpretation der Druckgeschichte des Musikalischen Opfers’, in U. Leisinger (ed.), Bach in Leipzig—Bach und Leipzig (Hildesheim, 2002), pp. 309–20; and the same author’s ‘The Printing History of J. S. Bach’s Musical Offering: New Interpretations’, Journal of Musicology, 19 (2002), pp. 306–31. 14 Regarding the structure of the original edition in ﬁve printing units (A–E) and the editorial numbering of canons 6–10, see C. Wolff, ‘New Research on the Musical Offering’, in his Bach: Essays on His Life and Music, pp. 239–58 (esp. 240), and the same author’s Krit. Bericht, NBA VIII/1 (1976), pp. 47–9 and 125–6. 12

musicalisches opfer

Printing Unit

Contents

A B

Title and dedication Ricercar a 3 Canon perpetuus super Thema Regium [Canon 7] Sonata sopr’il Soggetto Reale Canon perpetuus [Canon 8] Canones Diversi super Thema Regium (Canons 1–5) Fuga canonica [Canon 6] Ricercar a 6 Canon a 2 [Canon 9] Canon a 4 [Canon 10]

C D E

369

In The Art of Fugue and the Musical Offering alike, the fugues lack preludes and are given special titles that, as far as we know, Bach used nowhere else, namely ‘contrapunctus’ and ‘ricercar’ respectively. Two distinct traditions of ricercar go back to the sixteenth century: the polyphonic ricercar with imitative texture, an instrumental counterpart to the vocal motet; and the improvisatory-style ricercar which originally had a preludial function, ‘seeking out’ the mode or key of the music that was to follow.15 The polyphonic type was much cultivated in the seventeenth century by Frescobaldi and Froberger, whose music was much studied by Bach, and the six-part Ricercar from the Musical Offering belongs to that tradition. The three-part Ricercar, on the other hand, has clear links with the second type: it has an unmistakable air of improvisation, and indeed may represent a revised version of the first fugue that Bach improvised before the king at Potsdam. The contrast between ‘ancient and modern’ styles, one of the most striking features of the Musical Offering, is heard at its clearest in the two ricercars. The Ricercar a 6 is written in a traditional polyphonic style, ultimately derived from the vocal polyphony of the Renaissance. Like The Art of Fugue, it is notated in open score in the original edition, the traditional notation for keyboard music written in strict counterpoint. The normal instrument for the rendition of such music was the harpsichord or organ—in this case, harpsichord. The Ricercar a 3, on the other hand, is printed in two-stave keyboard notation and is couched in a much freer, more ‘modern’ style. Particularly in its episodes it has a pronounced improvisatory air that conjures up the image of Bach extemporizing at one of the king’s Silbermann fortepianos. At times it refers clearly to the style galant or to that special branch of it known as the empfindsamer Stil, which was cultivated by Bach’s sons, to whose generation the king belonged. Thus the fortepiano, at that time a modern invention, is perhaps the most appropriate instrument for the three-part Ricercar. According to his pupil Johann Friedrich Agricola, Bach admired the tone of Silbermann’s first fortepianos

These two ricercar traditions are discussed by Wolff, ‘Apropos the Musical Offering’, pp. 329–30.

370

the m u s ic a l o f f e r i n g a n d o t h e r w o r k s

but pointed out certain faults that the maker corrected on later instruments, which then met with Bach’s complete approval.16 In his last years Bach is known to have acted as a sales agent for Silbermann’s fortepianos in Leipzig. In the absence of the special artifices of strict counterpoint, which would be inappropriate in the relatively informal context of the three-part Ricercar, two things are of overriding importance: the presentation of the Royal Theme and the episodes between thematic entries, which take on greater significance than in any contrapunctus from The Art of Fugue. The ricercar is divisible into four large paragraphs: A exposition, B counter-exposition, C development, and D reprise. Within this scheme a profusion of colourful subsidiary material is presented—more than Bach would have permitted himself in a tighter structure. The countersubject opens with staccato crotchets (bb. 9 and 24), which are taken up in the first episode (b. 18) and in later derivatives of it (bb. 109 and 161). Paragraph A ends with episodic treatment of the chromatic descent from the subject (b. 31). Then at the end of B the subject unexpectedly modulates up a tone by sharpening the tonic (bb. 79ff.), after which the chromatic descent returns. The same subject modulation returns in C (bb. 102ff.), now greatly extended, leading to a huge, discursive episode based primarily on the chromatic figure, which is presented in both diminished and inverted forms. Paragraphs B and C both begin with triplet and syncopated figures in turn (bb. 38 and 87), the former being combined with the headmotive of the subject (bb. 46 and 95). The next significant motive in B consists of smoothly flowing quavers (b. 52), which at the next subject entry (b. 59) contribute to a regular triple-counterpoint combination. All three subject entries in the concluding exposition D are graced by different permutations of this elegant triple counterpoint. Although the six-part Ricercar is notated in open score in the original edition (though not in the autograph) and written in a style traditionally associated with strict counterpoint, it ignores the contrapuntal devices employed in much of The Art of Fugue. There are two possible explanations. While the Art of Fugue theme was designed with contrapuntal manipulation in mind, the Royal Theme of the Musical Offering, whose kernel at least came from the king, is simply not susceptible to such devices, as we have already seen. Secondly, since it was a ‘royal’ theme and its workingout dedicated to the king, Bach might have wished it to remain clearly audible rather than buried in clever fugal artifices. Accordingly, our interest focuses on the entries of the Royal Theme and on the episodes that connect them. There are three overall paragraphs (A, B, and C; bb. 1, 40, and 83), marked off from each other by prominent cadences in the mediant E♭ and submediant A♭. They are interrelated on account of recurring episodes. Thus A and B both culminate in a massive, sustained ‘tutti’ built on a rising hexachord bass (bb. 29, 33, and 79). Within this ‘tutti’ texture an inner part has a rising decorated, chromatic motive, derived by inversion from the chromatic

BD III, No. 743; NBR, No. 358d.

m u s i c a l i s ch e s o p f e r

371

descent of the subject. In the first episode of B (b. 40) this chromatic motive is inverted and imitated in a swift build-up of parts from one to six. The episode continues (b. 45) with imitation of the triadic headmotive of the subject, combined with a new counterpoint—a combination that returns in varied form at the beginning of C (b. 83). In the later stages of B (bb. 62 and 70) the headmotive is inverted, diminished, and combined with a new quaver counterpoint. In the most intricate and highly organized episode of B (b. 52) a chromatic motive, derived from the subject, is combined with three recurring figures in four-part invertible counterpoint. A varied reprise of this episode occurs in the concluding paragraph C (b. 90), now with a new counterpoint in crotchets, derived from the original countersubject (bb. 5, 9, and 13). This motive proceeds to dominate not only the remainder of the episode but the accompanying parts of the concluding bass entry (b. 99), creating a clear aural link with the opening exposition. The ‘Sonata sopr’il Soggetto Reale a Traversa, Violino e Continuo’ (Sonata on the Royal Theme for flute, violin, and continuo) was clearly included in the Musical Offering as a direct tribute to Frederick the Great, who was a proficient flautist. At Potsdam the king at that time held regular chamber concerts every evening, at which sonatas and concertos were performed featuring, in many cases, Frederick himself as soloist. Bach would have gained inside knowledge of these concerts from his son Carl Philipp Emanuel, who had been employed as Frederick’s harpsichordist since 1738. Johann Sebastian’s trio sonata was clearly written with these evening concerts in mind. Like the ricercars and canons, it is constructed around the Royal Theme. Although only hidden, questionable allusions to it occur in the slow movements, it plays a prominent part in both of the fast movements. It is curious that, despite the relatively ‘modern’, galant quality of much of the sonata, Bach reverts to the older trio-sonata texture as opposed to the more forward-looking obbligato-harpsichord texture that he himself had pioneered. Perhaps he wished to avoid giving the impression that through his own music he was advancing the claims of his harpsichord-playing son to a position that in musical terms competed with the king. The opening Largo, cast in rounded binary form with varied subdominant reprise (b. 33) and rhyming close (bb. 11 and 43), is notable for the strictly motivic figure-work of the flute and violin parts (the continuo has a largely supporting role). The dottedrhythm main theme is repeatedly stated in canonic imitation between the leading parts (bb. 1, 9, 17, 25, 33). Moreover, it is not only inverted after the double bar (b. 17) but enhanced in interval by expanding its first figure from rising 6th to 7th (bb. 1 and 9) and in the second strain from 6th to octave (bb. 17 and 25). A motive in Bach’s characteristic paired semiquavers is not only combined in counterpoint with the theme but treated independently. At the end of the first strain (b. 13) a new motive emerges in broken semiquaver figures—imitated by inversion—which then proceeds to replace the paired semiquavers in the middle period (bb. 17–32). The Allegro no. 2 is a fugal da capo movement in which, after the initial exposition, the fugue subject is combined in counterpoint with a plain, cantus firmus-like version

372 the musical offer ing a n d o t he r w o r k s of the Royal Theme and a regular countersubject with ‘sigh’ figures to form triple counterpoint. The great events of this movement are the six statements of this triple counterpoint (bb. 46, 67, 117, 160, 207, 228), which is heard in four different permutations. The fugue subject is particularly rich in memorable figures, two of which form the chief material of the many long, strictly motivic episodes. In the second Allegro, no. 4, the Royal Theme no longer acts as a countersubject but rather as the fugue subject itself—not in plain form, however, but in a decorative variant form in 6/8 gigue rhythm. Again, the main events are the subject entries, which vary in key according to the exposition: 1. tonic and dominant (bb. 1, 9, and 23); 2. a key sequence by rising 5ths (E♭, B♭, f; bb. 44, 53, 69); 3. tonic only (bb. 89 and 95). As in the ricercars, Bach shows himself to be a consummate master of old and new styles: the two fugal movements are divided by an Andante, no. 3, which perhaps represents his furthest excursion into the progressive galant style of the mideighteenth century. In particular, Bach alludes here to the rather more localized empfindsamer Stil cultivated by his sons and by the resident composers at the Potsdam court, including Frederick the Great himself. Two characteristics of this style are prominent: frequent appoggiaturas, or ‘sigh’ figures, and subtle dynamic nuances. Both occur in this Andante: the chief motive of the main theme consists of a figure made up of repeated note plus appoggiatura—a formula that also occurs in the lengthy chromatic episode from the three-part Ricercar—and this motive is repeatedly subjected to echo effects, being stated forte and then immediately repeated piano at a different pitch. The ten canons of the Musical Offering form two groups of five, of which the first group is entitled ‘Canones diversi super Thema Regium’. Accordingly, its contents are numbered 1–5 and placed together as a discrete set. The second group, on the other hand, are untitled, unnumbered, and distributed seemingly at random among the folios of the original edition. This might indicate that Bach had no particular order for them in mind, but it might also reflect the haste with which the work was composed and published. In terms of the manner in which the Royal Theme is treated, no single technique is wholly concentrated within one of the two groups. The theme occurs in its original plain form in nos. 1–3 and 6–7 but in a decorated form, as in the finale of the sonata, in nos. 4–5 and 8–10. It is treated as a cantus firmus, lying outside the canonic voices, in nos. 2–5 and 7, but forms the subject of the canon in nos. 1, 6, and 8–10. It is interesting to note, however, that each of these different modes of treatment of the Royal Theme occurs five times and thus applies to exactly half of the ten canons. Among the canons of the first group there appears to be a clear progression from simple to complex. No. 1, the ‘cancrizans’ (crabwise) canon, is in two voices only; all the others in three. The Royal Theme as cantus firmus is in the bass, treble, and middle part in turn (nos. 2, 3, and 4); and it is first presented in plain form (nos. 2 and 3), then decorated (nos. 4 and 5). The canonic dux (leading voice) is first answered direct (no. 2), then by inversion (no. 3), and finally by inversion and augmentation simultaneously (no. 4). The modulating ‘Canon a 2 per Tonos’, no. 5, may be

musicalisches opfer

373

regarded as the climax of the first group of canons. If all six potential modulations are carried out (c–d–e–f♯–g♯–b♭–c) a structure of 48 bars results. It is striking to observe that the same modulation up a tone by sharpening the tonic at the end of the Royal Theme occurs twice in the Ricercar a 3 (bb. 79–80 and 102–3). It seems likely that the idea originated in the ricercar and was later transferred to the canon, rather than the other way round. Two of the second group of canons are relatively small, circumscribed pieces that have clear counterparts in the first group. The two-voice canon, no. 9, corresponds in texture with the ‘Canon a 2 cancrizans’, no. 1 of the first group; and the three-voice ‘Canon perpetuus super Thema Regium’, no. 7, with its largely plain cantus firmus and two florid canonic voices, corresponds closely with nos. 2 and 3. However, the remaining canons of the second group, nos. 6, 8, and 10, are larger and more complex than anything in the first group (except perhaps the Canon per Tonos, no. 5). This possibly indicates that they were composed at a later date. Indeed, we cannot exclude the possibility that, according to Bach’s original conception, the work contained only the canons numbered 1–5. He might have added the other five canons hastily during the publication process, which might explain why they are unnumbered and scattered in various different parts of the edition. The Fuga canonica no. 6 and the Canon perpetuus no. 8 are very substantial pieces in trio-sonata texture, which does not occur in the first group—two canonic treble parts over a largely free continuo part. The Canon perpetuus, which follows the sonata in the original edition, expressly requires flute, violin, and continuo; and the Fuga canonica may be performed on the same instruments. The Fuga canonica and the Canon a 4, no. 10, both employ the permutation principle: each canonic voice opens with the Royal Theme and then continues with several counter-themes, with the result that the texture repeatedly combines all this thematic material. The Fuga canonica owes its name to the answer at the 5th, fundamental to fugal technique; in the Canon a 4, on the other hand, the canonic voices enter at the unison or octave. In the Fuga canonica, due to its largely free bass, the fugal-canonic writing is restricted to the two treble parts, whereas the Canon a 4 is not only uniquely in four-part texture but completely canonic throughout in all four parts. Moreover, since the canonic dux is made up of a variant of the Royal Theme followed by three regular counter-themes, the entire texture is thematic from beginning to end. From this point of view, the Canon a 4 may be regarded as the culmination of all ten canons. In the original edition it follows the Ricercar a 6 and is coupled with the Canon a 2—smallest and largest together—under the rubric ‘Quaerendo invenietis’ (‘Seek and you shall find’).

III.4 The B minor Mass and other vocal works

Title

Earliest source/s

Scribe, date

Cantate burlesque (Peasant Cantata), BWV 212 Tilge, Ho¨chster, meine Su¨nden, BWV 1083 Mass in B minor, BWV 232

Berlin, P 167

Autograph, 1742

Berlin, Mus. ms. 30199, Mus. ms. 17155/16 Berlin, P 180

Autograph, 1746/7 part-autograph, 1746/7 Autograph, 1748–9

Cantate burlesque (Peasant Cantata) The ‘Cantate [en] burlesque’, known today as the Peasant Cantata (BWV 212), was performed on 30 August 1742 as an act of homage to Carl Heinrich von Dieskau, who had inherited the manor of Klein-Zschocher, south-west of Leipzig, from his recently deceased mother. The burlesque element lies in the comic treatment of the two peasant characters, Mieke (soprano) and her young lover (bass), in the banter between them, in the use of the Upper Saxon dialect (mainly in the duet no. 2), and in the many humorous references to taxation—Dieskau and the librettist Picander were both tax officials in the Leipzig area. Bach enters with great gusto into the rustic setting of the plot. The sinfonia, the first recitative, and ten of the twelve arias (all but nos. 14 and 20) are scored for a rustic trio of violin, viola, and continuo, to which a hunting horn is added in nos. 16 and 18. Many of the arias are recognizable as specific dance types: bourreé (nos. 2 and 24), polonaise (nos. 4 and 6), sarabande (no. 8), mazurka (no. 12), minuet (no. 14), passepied (no. 20), and paysanne (no. 22). And throughout the work most of the melodies have the character of folk songs or folk dances. Many of them have been recognized as existing tunes that Bach borrowed to add local colour and to help characterize the peasant milieu.1 Others might be Bach’s own inventions, but it is more likely that they are existing tunes whose origin has not yet been established. It cannot be mere coincidence that in the previous year, 1741, Bach had published the

1 See Werner Neumann, ‘J. S. Bachs “Rittergutskantaten” BWV 30a und 212’, BJ 58 (1972), pp. 76–90; HansJoachim Schulze, ‘Melodiezitate und Mehrtextigkeit in der Bauernkantate und in den Goldbergvariationen’, BJ 62 (1976), pp. 58–72; and Tim Crawford, ‘Peasant Cantata: Borrowed Material’, in M. Boyd (ed.), Oxford Composer Companions: J. S. Bach (Oxford, 1999), pp. 362–3.

cantate burle sque

375

Goldberg Variations (BWV 988), whose Quodlibet-finale includes two German folk songs.2 Nor is it without significance that many of the ‘peasant’ tunes come from Poland (no. 1, bb. 8 and 39; nos. 2, 4, 6, 12, and 24), for native Polish music had been popular in Germany ever since the Elector of Saxony became King of Poland in 1697. The rustic character of the work is immediately established in the introductory sinfonia, a patchwork or potpourri in seven short, contrasting sections, each in the style of a popular song or dance. Bach gives it shape by returning to the opening Presto, albeit greatly extended, at the close. It has been suggested that this sinfonia might be a reminiscence of the quodlibets of Bach-family days in the composer’s youth, of which Forkel gives us an entertaining glimpse.3 The vocal solos that follow are framed by two bourreé-like duets, nos. 2 and 24. The intervening solos take the form of ten recitative–aria pairs, mostly very short, though each of the two soloists also sings a much more substantial and elaborate aria in da capo form (nos. 14 and 20). The fundamental theme of the cantata is, of course, praise of Dieskau, the new lord of the manor. Alongside this theme, however, three sub-themes run their course: the amorous relationship between Mieke and her lover (nos. 2–12), their singing contest (nos. 13–20), and finally their repairing to the local tavern for drinking and dancing (nos. 21–4). By far the most interesting of these musically is the singing contest, for whereas the other sub-themes are confined to the rustic style, the contest places the rustic and ‘town’ styles side by side in a deliberate juxtaposition of opposites. Mieke introduces the singing contest in the words ‘Sollst du . . . ein neues Liedchen von mir ho¨ren’ (‘You shall hear from me . . . a new little song’)—new, presumably, in the sense that it (no. 14) is of recent origin and does not adhere to the old, familiar folk idiom: it is parodied from one of Bach’s own compositions (the lost cantata BWV Anh. I 11 of 1732),4 hence its da capo form and sophisticated style. By the same token it is the first movement in the Peasant Cantata to depart from the rustic trio of violin, viola, and continuo, which is here replaced by the delicate, refined tones of the transverse flute, accompanied by a standard string ensemble. Mieke’s lover comments afterwards, in no. 15, that the piece is ‘zu klug vor dich und nach der sta¨dter Weise’ (‘too clever for you and after the town manner’). Nevertheless, it exhibits clear links with the songs she has sung previously: it is in dance style—that of a minuet5—hence the binary dance form of its ritornello and A-section. Moreover, each strain of its binary form is subjected to a varied repeat, a procedure that recurs frequently among the dance movements of this work.

See Schulze, ‘Melodiezitate und Mehrtextigkeit’, pp. 65–7. BD VII, pp. 15–16; NBR, pp. 424–5. 4 See Neumann, ‘J. S. Bachs Rittergutskantaten’, pp. 88–9, and the same author’s Krit. Bericht, NBA I/39 (1977), pp. 126–9. 5 Regarding the possible origin of this melody, see the literature cited under n. 4. 3

376 the b m i n or m as s and o the r v oc al w o rk s Mieke’s lover then brings her down to earth in the words, ‘Wir Bauern singen nicht so leise. Das Stu¨ckchen, ho¨re nur, das schicket sich vor mich!’ (‘We peasants don’t sing so delicately. Listen to the little piece that suits me!’). He then sings an old French hunting song6 (hence the addition of ‘corne de chasse’ to the rustic trio) which sounds coarse after the refined beauty of Mieke’s da capo aria. She says as much—‘Das klingt zu liederlich’ (‘That sounds too uncouth’)—but then offers to match it (retaining the horn) in what must have been a familiar folk song, judging by her description of it as ‘die alte Weise’ (‘the old tune’). Finally, Mieke’s lover attempts to emulate her in ‘singing something in the town style’ (‘was Sta¨dtisches zu singen’). The aria he sings, ‘Dein Wachstum sei feste’ (no. 20) corresponds with Mieke’s town-style piece (no. 14) in numerous ways. Both are in the original key of A; both are structured in ABA da capo form; both diverge from the prevailing rustic trio, in this case by employing obbligato violin and continuo; both are in a 3/8 dance rhythm, here that of the passepied; and both are parodied from earlier compositions of Bach’s. The parody model of ‘Dein Wachstum sei feste’—Pan’s aria ‘Zu Tanze, zu Sprunge’ from Der Streit zwischen Phoebus und Pan (BWV 201 no. 7, 1729)—is highly significant, since Phoebus and Pan are also engaged in a singing contest. In view of this borrowing, Bach can hardly have remained unaware of the analogy between the two compositions. In the earlier work, however, Phoebus’ aria ‘Mit Verlangen’ (no. 5) represents Bach’s own style—florid, refined, and serious—whereas Pan’s ‘Zu Tanze, zu Sprunge’ stands for the light, facile, and superficial style of those who saw fit to criticize him, a style that was becoming increasingly widespread at that time. In the Peasant Cantata, on the other hand, Pan’s aria (no. 20), together with Mieke’s in the same key (no. 14), represents the ‘town’ style as opposed to the rustic style of the hunting-horn arias (nos. 16 and 18). The implication is surely that Mieke has won the contest: she has nothing to be ashamed of in her aria ‘after the town manner’, with its lovely melody and decorative flute obbligato. Her lover, on the other hand, invites ridicule in his failed attempt at ‘singing something in the town style’ (Pan’s aria), with its ‘laughing’ motive in the ritornello (bb. 5–8) and its staccato quavers on ‘lache’ (‘laugh’, bb. 41–3 etc.). This is surely to be regarded as one of the comic highlights of a cantata in which Bach shows that, though of ‘serious temperament . . . he could also, when the occasion demanded, adjust himself . . . to a lighter and more humorous way of thought’.7

Tilge, Ho¨chster, meine Su¨nden Pergolesi’s Stabat Mater, composed towards the end of his short life (c. 1736), enjoyed immense posthumous fame and became one of the central documents of the ‘modern’ style galant. Bach’s adaptation of it to German words, Tilge, Ho¨chster, meine Su¨nden, BWV 1083, dates from 1746/7, only about ten years after the Latin original. The text of 6 See Schulze, ‘Melodiezitate und Mehrtextigkeit’, pp. 62–5, and Crawford, ‘Peasant Cantata: Borrowed Material’, p. 363. 7 According to the Bach obituary, BD III, No. 666 (p. 87); NBR, No. 306 (p. 305).

t i l ge , h o¨ chster, meine su¨ nde n

377

the Stabat Mater belonged to the Roman Catholic liturgy and had no place in the Lutheran Church, hence Bach’s use of German words, which enabled the music to be heard in the Leipzig churches. The text, a metrical paraphrase of Psalm 51, was clearly written for Pergolesi’s music, since it follows the verse structure of the original Latin text. As a penitential text (for an unknown occasion) Psalm 51 on the whole fits the music well: the anguish and sorrow of Mary at the foot of the Cross become that of the sinner in his or her awareness of guilt and consequent estrangement from God. Only at the end (Versus 12–13) did Bach have to alter the order of Pergolesi’s movements in order to give appropriate major-mode expression to the more positive mood that takes over at the end of the psalm. For the same reason Bach repeats Pergolesi’s minormode Amen in the major. Bach makes significant interventions in the make-up of the instrumental ensemble. Violin I and II are now both divided into solo and ripieno—in general, the solo violinists play during the vocal solos and the ripieno violinists between them. Where violins I and II were originally unisono and the viola doubled the continuo at the upper octave, which happened frequently, Bach gives all three upper strings their own independent part. He also makes many alterations to the two vocal parts (soprano and alto). These are often connected with the task of adapting the music to the new text, but elsewhere Bach seems to be imposing his own personal style and aesthetic principles upon the music (Ex. 1). Thus repeats are often varied, rests filled in, and short phrases joined together to create longer ones—lyrical continuity takes precedence for Bach over rhetorical breaks in the flow. Plain melodic lines are often elaborated, as one might expect of the great master of ‘florid expression’,8 and this often involves diversifying plain repeated notes, breaking up long notes into smaller note-values, and decorating the main themes when they recur later on in a movement. Finally, an entirely new vocal counterpoint is, on occasion, combined with one of Pergolesi’s themes. To a considerable extent, then, Bach brings his own personal style to bear on the Italian composer’s music. Many of the most galant features of the Stabat Mater, however, he leaves untouched. Moreover, his decision to perform the work might in itself be regarded as a tribute to Pergolesi and to the galant style of which he was the most celebrated exponent. Certainly the work seems to have made a strong impression on him, judging by the reminiscences of it that can be heard in some of his late works, notably the three-part Ricercar from the Musical Offering and ‘Et incarnatus est’ from the B minor Mass.9 It would be mistaken, however, to assume that the Stabat Mater is entirely galant in its stylistic orientation. The main theme of Versus 1 is

As J. A. Scheibe justly observes in an appreciative comment; NBR, No. 332. Compare the end of BWV 1083 no. 1 with the 3-part Ricercar, bb. 109–12 and 183–5; see Werner Breig, ‘J. S. Bachs Leipziger Klaviermusik und das Prinzip “Empﬁndsamkeit”’, in S. Schmalzriedt (ed.), Aspekte der Musik des Barock: Auffu¨hrungspraxis und Stil (Laaber, 2006), pp. 295–315 (esp. 311–12). The link between Pergolesi and the ‘Et incarnatus’ was established by Christoph Wolff, ‘“Et incarnatus” and “Cruciﬁxus”: The Earliest and Latest Settings of Bach’s B-minor Mass’, in M. A. Parker (ed.), Eighteenth-Century Music in Theory and Practice: Essays in Honour of Alfred Mann (Stuyvesant, NY, 1994), pp. 1–17 (esp. 12–13). 9

378

the b minor mass and other vocal works

Ex. 1

dum

sit

spi

ri - tum

a) Pergolesi, Stabat Mater, 6th movement, bb. 36–8, soprano and continuo only (strings omitted)

willst die

Wahr - heit

selbst

ha - ben,

mir of

die ge - hei - men Weis - heits - ga - ben hast du

- fen - (bart.)

b) Same passage in Bach’s embellished version, BWV 1083, Versus 7, bb. 36–8, soprano and continuo only (strings omitted) written in the style of a slow movement from a Corellian trio sonata; only later does it revert to the galant manner. Furthermore, Versus 9 and 14 are both alla breve fugues in a traditional ecclesiastical style, though Versus 9 is diversiﬁed by incorporating galant episodes. This broad stylistic reach might have struck a chord for Bach, who himself took pains to reconcile diverse styles in his later years. In particular, the historically opposite extremes of stile antico and style galant are at times encountered side by side in Clavieru¨bung III, The Well-Tempered Clavier II, the Musical Offering, and the B minor Mass.

Mass in B minor The majority of the Mass settings in Bach’s music library,10 as well as the ﬁve Masses that he himself composed in the 1730s, consisted of Kyrie and Gloria only in accordance with widespread Lutheran and, to some extent, Catholic usage. As early as the Weimar period, however, he and his cousin J. G. Walther had copied out a ‘Missa tota’—all ﬁve divisions of the Mass Ordinary—by Johann Baal. Further engagement with the complete Mass on Bach’s part is known to have taken place in the mid-to-late Leipzig years. Around 1735 he and an assistant copied out the Acroama missale by 10 By Peranda, Baal, Pez, J. L. Bach, Durante, Wilderer, Lotti, Bassani, Palestrina, and anon.; see the VBN in Kirsten Beißwenger, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992), pp. 226–400.

mass in b m inor

379

Giovanni Battista Bassani, a collection whose six Masses lack only the Agnus Dei. And around 1742 Bach and a copyist wrote out the performing parts for all five movements—Kyrie, Gloria, Credo, Sanctus, and Agnus Dei—from Palestrina’s Missa sine nomine.11 The experience of copying and performing these complete Mass settings might have encouraged Bach to attempt his own Missa tota by adding the Credo, Sanctus, and Agnus Dei to the Dresden Missa of 1733. This seems to have been carried out during the last two years of Bach’s life—between summer 1748 and winter 1749— and it has until recently been considered his last major creative project altogether, though The Art of Fugue probably has a better right to this description.12 What was the purpose of this great undertaking? It is often assumed that it was done for its own sake—as a supreme monument to Bach’s art, for his own personal satisfaction, and for the benefit of posterity. It has been written, for example, that ‘Bach probably wished to compose in a field that represented the highest achievements [in sacred music] ever since the time of Josquin and Palestrina, who elevated the Mass to an independent work of art. Bach took it outside the realm of liturgy as an expression of his personal mastery.’13 It is unlikely, however, that the Missa tota was conceived as a purely abstract work without performance in mind.14 The notion of Bach in his late years as a lone genius, more and more divorced from external reality and immersed in his own inner world, is probably a romantic illusion. To the end of his days Bach remained an extremely practical musician whose works invariably had a clear external purpose, and it seems most unlikely that the B minor Mass is an exception. This view receives some confirmation from the character of the autograph score, which is divided into four folders, each with a title page on the front listing the vocal and instrumental resources required. Moreover, numerous details in the autograph have been observed that are clearly aimed at performance.15 In this connection it has recently been observed that Bach’s second-youngest son Johann Christoph Friedrich made certain marks in the autograph that imply he was to prepare a set of performing parts before his departure for Bu¨ckeburg in December 1749.16 This is the clearest evidence to have emerged so far that Bach had a concrete performance in mind for late 1749 or early 1750.

Beißwenger (see n. 10), Baal: VBN I/B/1; Bassani: VBN I/B/48; Palestrina: VBN I/P/2. The view of the Mass as Bach’s last work was the product of research by Dadelsen, Wolff, and Kobayashi. See the last-named writer’s ‘Die Universalita¨t in Bachs h-Moll-Messe’, Musik und Kirche, 57 (1987), pp. 9–24; Eng. trans. in Understanding Bach’s B-minor Mass (Belfast, 2007), vol. ii, pp. 387–403 (esp. 387–8). According to Anatoly P. Milka, however, Bach’s last work on The Art of Fugue took place in the first three months of 1750, which makes it Bach’s opus ultimum, as it is described in the obituary (BD III, No. 666, p. 86; NBR, No. 306, p. 304). See A. P. Milka, ‘Zur Datierung der h-Moll-Messe und der Kunst der Fuge’, BJ 96 (2010), pp. 53–68. 13 ¨ ber Bach und anderes: Aufsa¨tze und Vortra¨ge Georg von Dadelsen, ‘Bachs h-Moll Messe’, in Dadelsen, U 1957–1982 (Laaber, 1983), pp. 139–43. 14 In his afterword to the facsimile of the autograph score (Kassel, 1965; 2nd edn 1984), Alfred Du¨rr pointed out that it is characteristic of composers of that era to write with a definite performance in view. Thus, he added, there are grounds for assuming that Bach saw a definite possibility for performance when he wrote the ‘Symbolum Nicenum’ (Credo) and the ‘Osanna’ to ‘Dona nobis pacem’. 15 As pointed out by Kobayashi, ‘Die Universalita¨t in Bachs h-Moll-Messe’, Eng. trans. (n. 12), p. 402. 16 See Peter Wollny, ‘Beobachtungen am Autograph der h-Moll-Messe’, BJ 95 (2009), pp. 135–51. 12

380

the b m inor mas s and o ther vocal w orks

If the B minor Mass was to have been performed, then, where might the intended performance have been and upon what occasion? Since a full concerted setting of the complete Mass was excluded from the Lutheran liturgy, it can only have been intended for a Catholic service, which in Bach’s case points to Dresden as the most likely destination of the work. However, since the B minor Mass as a whole is far larger than any of the complete Mass settings within the Dresden repertoire, it is unlikely that it was intended for an ordinary service. Instead, it might well have been written for a special festive occasion—one that has been suggested is the dedication of the magnificent new Dresden Hofkirche (court church), which in the event took place after Bach’s death, in 1751.17 Another possibility, a commission from Count Johann Adam von Questenberg in the spring of 1749 for a performance at his palace in Jaromerice, has recently been rejected. But Questenberg was a member of a musical ‘Congregation’ in Vienna, founded in 1725, which gave splendid, lavish musical performances every year on St Cecilia’s Day, 22 November, at first in St Michael’s Church but later in St Stephen’s Cathedral. Masses of huge dimensions and performing forces were given at these events, including vast works by Ferdinand Schmidt, Georg Reutter the Younger, F. L. Gassmann, and Joseph Haydn (the Missa cellensis, Hob. XXII: 5). It is not beyond the bounds of possibility that Questenberg contacted Bach in March 1749 to ask whether he would compose a Mass for the Congregation’s annual concert on 22 November 1749.18 The division of the autograph score into four separate folders suggests that Bach might have had an additional purpose in mind. This fourfold division, in place of the fivefold division of the liturgy, clearly reflects the manner in which the work was compiled. The Kyrie and Gloria (folder I) had been composed as a discrete Missa in 1733; the Credo (folder II) was added in 1748–9; and Bach used a Sanctus setting (folder III) that he had composed in 1724 for his second Christmas in Leipzig. However, the Lutheran liturgy employed an abbreviated form of the Sanctus that lacked the Osanna and Benedictus, so these movements were added in folder IV (1748–9) alongside the Agnus Dei and its concluding prayer ‘Dona nobis pacem’. Bach’s fourfold division of the Mass might be more than simply a reflection of the compilation process, however. The Dresden Masses were also customarily stored in separate folders so that parts of them could be extracted for independent performance as and when necessary.19 Bach’s four folders in all probability had a similar purpose. The Kyrie–Gloria Missa (part I) and the opening portion of the Sanctus (part III) could be extracted for use within the Lutheran liturgy; and any of the sections could be used independently within the Catholic liturgy. In this way Bach might have

17 See Wolfgang Osthoff, ‘Das Credo der h-moll-Messe: Italienische Vorbilder und Anregungen’, in W. Osthoff and R. Wiesend, Bach und die italienische Musik (Venice, 1987), pp. 109–40 (esp. 134). 18 See Michael Maul, ‘“Die große catholische Messe”: Bach, Graf Questenberg und die Musicalische Congregation in Wien’, BJ 95 (2009), pp. 153–75. 19 See George B. Stauffer, Bach: The Mass in B minor (New York, 1997), p. 22.

mass in b m inor

381

foreseen greater practical usefulness for the work than if it had been restricted to complete performance on special festive occasions, for there were very few such events at that time that would warrant the performance of a Mass on such a massive scale. Several of Bach’s activities in the sphere of sacred music in the mid-to-late 1740s are significant in the run-up to the composition of the B minor Mass. Around 1745 he extracted three movements from the Gloria of the 1733 Missa, ‘Gloria in excelsis Deo’, ‘Domine Deus’, and ‘Cum Sancto Spiritu’, to form a Latin Christmas cantata (BWV 191). This work might have been performed at a special thanksgiving service in the university church at Leipzig on Christmas Day 1745 to mark the signing of the Dresden peace treaty following the Prussian invasion of Saxony. It is not unlikely that Bach’s experience of arranging and performing this cantata, especially in conjunction with the revival of the six-part Sanctus of 1724, probably on the same occasion, acted as a catalyst towards his expansion of the 1733 Missa.20 A more immediate catalyst seems to have been Bach’s revival of Bassani’s Acroama missale in 1747/8. In Missa V from this collection he interpolated his own setting of ‘Credo in unum Deum’ (BWV 1081), which looks very much like a sketch for the setting of the same words in the Credo of the B minor Mass.21 Both movements are scored for four voices and continuo (plus two violins in Bach’s own Mass) and they exhibit the same conjunction of Renaissance-style polyphony and Baroque-style walking bass. A recently discovered manuscript in the hand of Bach’s pupil Johann Friedrich Agricola contains an early version of the ‘Credo in unum Deum’ from the B minor Mass, which is a tone lower than the definitive version.22 Though more complex than the Bassani interpolation, it is similar in style, as already indicated, and might have been composed not long afterwards. The Credo, Sanctus, and Agnus Dei were built up using the same process of compilation as the Kyrie and Gloria of the 1733 Missa, involving much use of parody (the re-texting of existing music). Of the nine movements of the Credo, three are known to be parodies—‘Patrem omnipotentem’, ‘Crucifixus’, and ‘Et expecto resurrectionem mortuorum’—and there may well be parodies whose models no longer survive among the remaining six movements. There is only one known parody in the Sanctus, namely the ‘Osanna’, though the first movement was, of course, borrowed from a much older, independent composition. Both movements of the Agnus Dei are parodies, the second, ‘Dona nobis pacem’, being a re-texted version of the ‘Gratias agimus tibi’ from the Gloria of the 1733 Missa. One forms the impression that Bach came to associate the Latin words of the Mass with permanence and perfection, and

See Gregory G. Butler, ‘J. S. Bachs Gloria in excelsis Deo BWV 191: Musik fu¨r ein Leipziger Dankfest’, BJ 78 (1992), pp. 65–71. 21 See G. von Dadelsen, ‘Eine unbekannte Messenbearbeitung Bachs’, in H. Hu¨schen (ed.), Festschrift Karl Gustav Fellerer (Regensburg, 1962), pp. 88–94. 22 See Peter Wollny, ‘Ein Quellenfund zur Entstehungsgeschichte der h-Moll-Messe’, BJ 80 (1994), pp. 163–9. The early version is published in U. Wolf (ed.), Fru¨hfassungen zur h-Moll-Messe, NBA II/1 a (2005), pp. 135–9.

382

t h e b m in o r m a s s a n d o t h e r vo c a l w o r k s

therefore sought by means of parody to elevate some of the ﬁnest of his Germantexted vocal works to that exalted state. It is also worth pointing out that parody played a signiﬁcant role in the Dresden Mass repertoire.23 Thus in all Bach’s Mass settings he was falling in line with standard practice in the Saxon capital. The ‘Symbolum Nicenum’ (Credo), like the Gloria, is divided into nine movements as shown:

1 2 3 2 4 45 6 7 8 9

Credo in unum Deum Patrem omnipotentem Et in unum Dominum Et incarnatus est Cruciﬁxus Et resurrexit Et in Spiritum Sanctum Conﬁteor Et expecto resurrectionem

Chor. Chor. Aria (duet) Chor. Chor. Chor. Aria Chor. Chor.

A D G b e–G D A f♯ D

Father Son Incarnation Cruciﬁxion Resurrection Holy Spirit Baptism Resurrection of the Dead

The overall design is thus clearly symmetrical: three large complexes at the beginning, middle, and end, linked by two arias. The opening and closing complexes correspond with each other: both are made up of two choruses, the first in stile antico and the second in stile moderno; and in both cases the two choruses are linked together by tonal transitions and overlapping texts. The two arias form interludes, concerned with the Second and Third Persons of the Trinity respectively. At the centre lies a triptych of choruses that deal with the main events in the life of Christ: the Incarnation, Crucifixion, and Resurrection. The ‘Crucifixus’, as the turning point between Christ’s earthly and heavenly life, lies at the centre of the whole Credo. The opening complex of two choruses, ‘Credo in unum Deum’ and ‘Patrem omnipotentem’, is concerned with God the Father. It may have been partly the great antiquity of the Credo—formulated by the Council of Nicaea in AD 325—that led Bach to set the opening statement, ‘Credo in unum Deum’, in the stile antico—the Baroque understanding of Palestrina’s style.24 The theme is not one of his own composition but a German variant of the Gregorian plainchant intonation for the Credo. Not only is this chant in keeping with the antiquity of the text, but it also gives Bach’s setting an air of objectivity. On the basis of this chant, stretched out into a long-note cantus firmus, Bach builds a massive stretto fugue in no fewer than seven real parts—five for voices and two for violins. This entire Renaissance-style texture is supported by a continuo bass made up of treading crotchets in the style of Bach’s own period, so that past and present are musically interlinked. The movement falls into three large stretto

See Stauffer, Bach: The Mass in B minor, pp. 20–2. The stile antico is dealt with comprehensively by Christoph Wolff in Der Stile antico in der Musik Johann Sebastian Bachs: Studien zu Bachs Spa¨twerk (Wiesbaden, 1968). For the Zarlino–Theile–Bach line of development see Paul Walker, ‘Bach’s Use of Fugue in the stile antico Vocal Writing of the B-minor Mass’, in Tomita et al. (eds.), Understanding Bach’s B-minor Mass (n. 12), vol. ii, pp. 368–86. 24

m ass i n b m in o r

383

expositions (bb. 1, 17, and 33) in which intensity is progressively increased by gradually diminishing the time and pitch intervals between the stretto entries. The close entries of the third exposition are underpinned by a massive bass subject entry in augmented note-values, creating an impression of overwhelming, elemental power well suited to a movement concerned with the Deity Himself. In keeping with its deliberately antiquated character, the opening movement of the Credo is strictly speaking modal (A-Mixolydian, hence the two-sharp signature) rather than tonal. Therefore, in order to create a smooth join with the following ‘Patrem omnipotentem’ fugue, Bach starts this movement with a subject entry in the dominant A rather than the tonic D. Furthermore, the first three subject entries, though set to the words ‘Patrem omnipotentem’, are accompanied by the words of the first movement, ‘Credo in unum Deum’, as a choral refrain in the other voice parts. This has the effect of identifying the second movement as an amplification of the first, which is what it amounts to in textual terms. ‘Patrem omnipotentem’ is parodied from the opening chorus of Cantata No. 171, Gott, wie dein name, so ist auch dein Ruhm (for New Year’s Day 1729) or its model, but both the opening dominant-key subject entry and the ‘Credo’ refrain were added when Bach revised the movement for inclusion in the Mass. The refrain introduces a festive, homophonic component into this otherwise polyphonic movement. The text of the cantata movement, drawn from Psalm 48: 10, is very close to the Credo text: both are concerned with the worldembracing almighty power of God. In the fugue subject, slightly adapted for the Latin words, a bold 7th leap on ‘omnipotentem’ stands for the almighty power of the Creator, and the vast 11th descent on ‘factorem caeli et terrae’ for his creativity that spans both heaven and earth. As in ‘Gratias agimus tibi’, much of the exalted character of the music stems from the sparing use of obbligato trumpets, which in Bach’s day were associated with kingship. In the first aria interlude, ‘Et in unum Dominum’, a substantial text relating to God the Son is set as a soprano–alto duet with two oboes d’amore, strings and continuo. Bach’s first version also included the following words that deal with the Incarnation, but on second thoughts he decided that that crucial doctrine merited a movement of its own. The words of the duet dwell on the relationship between Father and Son. Since that relationship is one of love, it is not altogether inappropriate that Bach at one time considered using the music of the love duet ‘Ich bin deine, du bist meine’ from the Hercules Cantata (BWV 213, 1733). The musical symbolism Bach employs in ‘Et in unum Dominum’ to represent the loving relationship between Father and Son recalls ‘Domine Deus’ from the Gloria. He is clearly struck by the mystical paradox that while Christ, as ‘the only begotten Son of God’, is separate from the Father, he is also ‘of one substance with the Father’. Consequently he gives a musical representation of duality within unity. The ritornello theme, taken up by the voices at the start of each paragraph (except the last), opens with imitation at the unison or octave, representing unity, but then proceeds with imitation at the 4th or 5th, representing duality. Moreover, the imitating instruments give the theme in two different modes of

384

t h e b m i n o r ma s s a n d o t h e r v o c a l w o r ks

articulation, first staccato and then legato. The love relationship is brought most clearly into focus in the homophonic coda (bb. 56–62) to the middle paragraph (bb. 34–62) of this ABA1 reprise structure, where both voices and violins move in amorous parallel 3rds and 6ths. The central triptych of the Credo is concerned with the three main events of Christ’s earthly life, the Incarnation, Crucifixion, and Resurrection. The key relationship between the three movements is symbolic: a descending 5th in minor mode (b–e) from ‘Et incarnatus’ to ‘Crucifixus’, followed by an ascending 5th in major mode (G–D) from ‘Crucifixus’ to ‘Et resurrexit’. The change from the first key sequence to the second (e–G) occurs during the central ‘Crucifixus’, which represents the turning point between Christ’s temporal and eternal life. There is also a clear stylistic distinction between the three choruses. The ‘Crucifixus’ is among the most retrospective movements in the entire Mass, whereas the surrounding movements are considerably more ‘modern’ in their stylistic orientation. This corresponds with the relative date of the music: ‘Crucifixus’ is the earliest movement in the Mass, parodied from Cantata No. 12, Weinen, Klagen, Sorgen, Zagen, of 1714, whereas ‘Et incarnatus est’ seems to have been newly composed in 1748/9 and might have been Bach’s last vocal composition.25 It was apparently inspired by Pergolesi’s ultra-modern, galant Stabat Mater of 1736, which Bach had arranged as a German-text cantata, Tilge, Ho¨chster, meine Su¨nden, BWV 1083, only a short time before the composition of the Credo (Ex. 2).26 The doctrine of the Incarnation is set to a chorus which ‘for simplicity, depth, and mystery cannot be surpassed’.27 The two paragraphs (bb. 1 and 23) each open with imitation of the subject in all five voices but then continue in homophonic texture. The subject, a broken-chordal descent, may be understood as signifying Christ’s descent from heaven. In the coda (b. 39) it is inverted at the words ‘et homo factus est’ (bb. 45–6), presumably to signify Christ’s ascent into manhood. The unison violins’ figure, perhaps borrowed from Pergolesi, is a variant of the vocal theme, decorated with expressive appoggiaturas; presumably it represents the movement of the Holy Spirit, for it ‘hovers like the Spirit of God, moving on the face of the waters’.28 The German text of the parody model for the ‘Crucifixus’ reads (in English translation): ‘Weeping, lamenting, grieving, trembling, anguish, and distress’; and the Latin text is concerned with crucifixion, suffering, death, and burial. Thus there is no conflict between the two texts and they are equally suited to the musical treatment they receive: a lament in chaconne form over the traditional lamento ground bass with its chromatic-4th descent. The thirteen statements of the ground bass are woven into an overall tripartite scheme (ABA1) with coda. The introductory instrumental

See Wolff, ‘“Et incarnatus” and “Cruciﬁxus”’, pp. 1–17. For the link between the two compositions, see Wolff, ‘“Et incarnatus” and “Cruciﬁxus”’, pp. 12–13. 27 According to Donald Francis Tovey, ‘J. S. Bach: Mass in B minor’, in his Essays in Musical Analysis, vol. v (London, 1937), pp. 20–49 (see p. 40). 28 Tovey, ‘J. S. Bach: Mass in B minor’, p. 40. 26

m ass i n b m in o r

385

Ex. 2

a) Pergolesi, Stabat Mater, 5th movement, bb. 13–15, repeated accompaniment ﬁgure for unison violins

b) ‘Et incarnatus est’ from B minor Mass, bb. 1–2, repeated accompaniment figure for unison violins statement is new to the Latin version, as is the piano coda for voices and continuo only (b. 49) to the words ‘sepultus est’ (‘He was buried’). Here Bach was ‘inspired with one of the greatest of all his strokes of genius in the unexpected modulation to G major, with a cadence of immeasurable depth’.29 The impression formed by the coda is of Christ reaching the deep peace of rest in death with his mission accomplished. The full festive vocal and instrumental ensemble, including three trumpets and drums, now bursts out at the words ‘Et resurrexit’ (‘And he rose again’). The movement is structured in ritornello form within an overall ABA1 reprise structure. In the opening ritornello and in its reprise at b. 86, the choir briefly sing the principal theme with its triplet ‘joy’ motive as a motto, whereas the other ritornellos are purely instrumental. The three components of the ritornello, a (b. 1), b (b. 24), and c (b. 40), do not all come together till the final ritornello—all previous ones are abridged. And since the final ritornello opens with c, which is designed as a continuation of a, it seems likely that a definitive ritornello would take the form a c a b. This suggests that the movement was parodied from a lost original that began with a full ritornello in the form just described. On this hypothesis the opening ritornello would have been greatly abbreviated when attached to the ‘Et resurrexit’ text in the interests of immediacy and swift continuity. The second aria-interlude, ‘Et in Spiritum Sanctum’, an aria for bass, two oboes d’amore, and continuo, is concerned primarily with the Holy Spirit. A decidedly pastoral tone is conveyed by the lilting 6/8 rhythms, the choice of obbligato instruments, the parallel 3rds of the ritornello theme, and the folksong-like internal repetitions. It has

Tovey, ‘J. S. Bach: Mass in B minor’, p. 41.

386

t h e b mi n o r m a s s a n d o t h e r vo c a l w o r k s

been suggested that Bach might have derived this mode of treatment from Martin Luther’s account in the catechism of the Holy Spirit shepherding the Christian flock. The concluding complex of the Credo is concerned with the sacrament of baptism and the doctrine of the resurrection of the dead. These matters are dealt with in two choruses which are not only sung without a break but overlap in text—the line ‘Et expecto resurrectionem mortuorum’ both concludes the ‘Confiteor’ and initiates the ‘Et expecto’ chorus. The structure of this concluding complex of the Credo is similar to that of the opening complex. Both begin with a motet-style, alla breve chorus in stile antico for five-part a cappella choir and continuo (though with the addition of two obbligato violin parts in the Credo intonation). In both cases, the first chorus is in a secondary key (A in the first case, f♯ in the second), employs the appropriate Gregorian plainchant melody, and is closely linked in text and music with the second chorus, which forms the greatest possible contrast with the first, being written in the stile moderno and scored for the entire vocal and instrumental ensemble. Like the Credo intonation, the ‘Confiteor’ is a five-part double stretto fugue, structured as shown: Paragraph

Content

Key

Exp. I Exp. II Exp. III Exp. IV Chant exp. I Chant exp. II Conclusion

SI S II S I + II S I, S II Canon at 5th Augmented C.F. Homophonic

f♯ b D c♯ f♯ g (half-close) d (half-close)

The two subjects, ‘Confiteor’ and ‘in remissionem peccatorum’, receive separate expositions (bb. 1 and 16), are combined in the third exposition (b. 31), then separate again in the fourth (b. 54). The plainchant then enters (b. 73) in bass–alto canon at one bar and the upper 5th while the other voices continue freely with whole or partial entries of the two subjects. After a tonic cadence, the plainchant returns (b. 92) in augmented note-values as a tenor cantus firmus. One is immediately reminded of the augmented entry of the plainchant in the concluding exposition of the Credo intonation. These two movements are among the few pieces that appear to have been composed specifically for the Credo in the late 1740s rather than extracted from earlier sources. They are alike in so many ways that it seems reasonable to assume that Bach had the Credo intonation in mind when he wrote the ‘Confiteor’. Like ‘Et incarnatus est’, ‘Confiteor’ must have been one of the last movements Bach wrote for the Mass; and, comparing the two, it is fascinating to observe him even at this very late stage of his life drawing inspiration from opposite ends of the historical spectrum. The largely homophonic conclusion of the ‘Confiteor’ (b. 123b) gives the first of two musical responses to the words ‘Et expecto resurrectionem mortuorum’ (‘And I look for the resurrection of the dead’)—one of the most profound mystery. This is conveyed in an essentially homophonic texture (though with some canonic writing between the outer

m as s in b mi n o r

387

voices), chromatic harmony, and unexpected modulations to unrelated keys. In its extraordinary depth of insight this passage, which acts as a transition to the following fast movement, is equalled only by the setting of ‘sepultus est’ at the end of the ‘Crucifixus’. The second musical response to the words ‘Et expecto resurrectionem mortuorum’ is one of great joy, expressed in a D major ‘Vivace e allegro’ chorus with trumpets and drums. This movement was parodied from a composition that Bach had used on at least three occasions between 1729 and 1742 (in BWV 120 and variants). The words of the council election version read: ‘Sing for joy, you glad voices, climb up to heaven’; thus the theme of jubilation was already attached to the music before its use in the Credo. It was, however, more drastically altered than any other parodied movement in the Mass, with the exception of the Agnus Dei. The German original opened and closed with an instrumental ritornello which, in the context of the Credo, would have detracted from the sudden outburst of joy after the ‘Confiteor’ and from the sense of purposeful continuity that Bach endeavoured to establish within each division of the Mass. For these reasons the concluding ritornello is dropped altogether and the opening one substantially abridged (the missing bars are present in the instrumental parts of bb. 77–87). In addition, the opening ritornello is now furnished with vocal parts, with the result that the feeling of excited anticipation is present from the outset. As in the ‘Crucifixus’, Bach borrows only the A-section of the original da capo chorus. The Sanctus is an ancient hymn of adoration, probably dating from the second century, which unites two quite distinct elements, both of biblical origin. The first is the vision of God in Isaiah 6: ‘I saw the Lord sitting on a throne, high and lofty; and the hem of his robe filled the temple. Seraphs were in attendance above him. Each had six wings; with two they covered their faces, and with two they covered their feet, and with two they flew. And one called to another and said, “Holy, holy, holy is the Lord of hosts; the whole earth is full of his glory.”’ The second part of the Sanctus is drawn from Matthew’s account of Christ’s entry into Jerusalem on a donkey, the event celebrated on Palm Sunday. The relevant passage in Matthew 21: 8–9 reads: ‘A very large crowd spread their cloaks on the road, and others cut branches from the trees and spread them on the road. The crowds that went ahead of him and that followed were shouting, “Hosanna to the Son of David! Blessed is the one who comes in the name of the Lord! Hosanna in the highest heaven!”’ Only the Old Testament part of the Sanctus was sung regularly in the Lutheran liturgy, so for this opening part—‘Sanctus’ and ‘Pleni sunt coeli’—Bach could simply borrow a magnificent six-part setting that he had composed for Christmas 1724. This Old Testament complex, which is concerned with God the Father, is bipartite—a vocal adaptation of the instrumental form of prelude and fugue, common in Bach’s German sacred music and already encountered in ‘Et in terra pax’ from the Gloria. The Sanctus ‘prelude’ undoubtedly counts as one of the most sublime of all Bach’s creations. The seraphim’s hymn of adoration is represented by the antiphony of various vocal and instrumental choirs. Numerical symbolism plays a significant part: the six voices stand

388 t h e b m i n o r m a s s a n d o t h e r v o c a l w o r k s for the six wings of the seraphs, and their arrangement in groups of two leads to the division of the choir into two sopranos, two altos, and two lower voices (tenor and bass). The invocation is the threefold Trisagion ‘Holy, holy, holy’, hence the grouping of the instrumental choirs in threes—three trumpets, three oboes, three upper strings. The glory of God that fills both heaven and earth is now celebrated in a five-part fugue, ‘Pleni sunt coeli’, in a dance-like 3/8 time. One of the chief properties of the subject is its capability of being doubled in 3rds, 6ths, or 10ths (bb. 66 and 72). This leads to an effect of great power in the second exposition (b. 113), where the heights of the heavens and the depths of the earth are rendered by the simultaneous entry of the subject at the bottom of the texture in bass and continuo and at the top in the highest register of the first trumpet (b. 131). In the first of the two ‘developments’ (bb. 78 and 137) that alternate with the two expositions, an important subsidiary theme enters (b. 78) that looks back, as it were, in its obvious derivation from the fugue subject but also forward in its near-identity with the theme of the following ‘Osanna’ (Ex. 3).30 Immediately before the final subject entry this subsidiary theme is hammered out unisono by vocal bass, trumpets, and drums in order to prepare for the following ‘Osanna’. The New Testament part of the Sanctus, which is concerned with God the Son, had to be parodied and/or newly composed. This complex is tripartite with da capo— Osanna, Benedictus, Osanna—a scheme dictated by the liturgical text and indeed by its biblical source (Matthew 21). The Osanna survives in an earlier German version, the opening chorus of the secular cantata Preise dein Glu¨cke, BWV 215, of 1734. The two versions had a common source in a lost secular cantata Es lebe der Ko¨nig, BWV Anh. I 11, of 1732. Both German versions were written in praise of a king. In the Latin version, praise of an earthly king becomes praise of the heavenly King. This was an easy transition to make in an age when kings were widely regarded as God’s

Ex. 3 Sanctus from B minor Mass:

Ple

ni sunt coe - li

ter

a) Subsidiary theme from ‘Pleni sunt coeli’ (bass, bb. 78–82)

O - san - na,

o - san - na

b) Incipit of ‘Osanna’ (SATB + SATB in unison)

30 This near-identity is difﬁcult to account for in view of the origin of the movements concerned eight years apart (1724 and 1732).

mas s in b mi n o r

389

representatives on earth, a view to which Bach no doubt subscribed. The choir, already expanded to six voices in the first Sanctus complex, is now expanded still further to eight voices—two four-part choirs used antiphonally in polychoral style. In the Latin version Bach drops the opening ritornello, as often elsewhere in the Mass, for the sake of continuity and immediacy—the voices now enter at the outset with a great unison cry of ‘Osanna’. It is only towards the end that we hear the whole magnificent ritornello in its entirety, first with inbuilt antiphonal voices and then played by the instrumental ensemble alone. For the Benedictus there is a change not only from major to minor mode (D to b) but from the largest vocal-instrumental forces in the whole Mass to the smallest—solo tenor, obbligato flute or violin, and continuo. This seems to signify that the focus of our attention now turns away from the ecstatic praises of the crowd and towards the single figure of Christ, riding alone into Jerusalem to face the Passion and Crucifixion. The obbligato instrument is not specified in Bach’s score, but the part is undoubtedly more idiomatic to transverse flute than to violin, especially in view of the absence of any notes below d1.31 The Agnus Dei is a prayer addressed to Christ as the ‘Lamb of God’, said or sung at the breaking of the bread shortly before Communion. Like the Sanctus, it is biblical in origin. Isaiah’s description of the ‘suffering servant’, later identified with Christ, reads: ‘He was oppressed, and he was afflicted, yet he did not open his mouth; like a lamb that is led to the slaughter, and like a sheep that is silent before its shearers, he did not open his mouth’ (Isaiah 53: 7). John the Baptist echoes this passage when he sees Jesus coming towards him and declares, ‘Here is the Lamb of God, who takes away the sin of the world’ (John 1: 29). These words were introduced into church liturgy in the form of a prayer in the late seventh century. By the eleventh century this prayer took the threefold form that is still in regular use today: three petitions, each addressed to the ‘Agnus Dei’ (Lamb of God). Bach condenses this prayer, omitting the opening of the third petition and setting it in two movements rather than three: an alto aria in the subdominant minor (g) for ‘Agnus Dei’ (first two petitions) and a great chorus in D, the overall tonic of the Mass (not B minor, as the nickname suggests), for the key words of the third petition, ‘Dona nobis pacem’. An older, German version of the ‘Agnus Dei’ aria survives as the alto aria ‘Ach, bleibe doch, mein liebstes Leben’ from the Ascension Oratorio (c. 1735). However, the oratorio and Mass versions are derived from a common original that no longer survives, the aria ‘Entfernet euch, ihr kalten Herzen’ from the secular wedding cantata Auf! Su¨ß entzu¨ckende Gewalt, BWV Anh. I 196, of 1725. The oratorio and Mass texts are both prayers addressed to Jesus Christ. The oratorio aria expresses grief at his departure from this earth; in the Mass this grief becomes the human response to the doctrine of the Atonement, Christ’s expiation of the sins of the world. The imploring

31 It is often thought that the Benedictus might have been parodied from a lost aria, but Kobayashi has discovered an ink draft underneath the fair copy. See his ‘Die Universalita¨t in Bachs h-Moll-Messe’, Eng. trans. (n. 12), pp. 397–8.

390 t h e b mi n o r m a s s a n d o t h e r v o c a l w o r k s gestures of the opening theme are no less convincing when applied to the plea for mercy of the Mass (‘miserere nobis’) than to the plea for Jesus not to depart from this world in the oratorio. Comparison of the two movements reveals very extensive alterations.32 Of the original ABA1 reprise structure, the A-sections are to a considerable extent recomposed and the central B-section is altogether omitted. Most striking is Bach’s introduction of an entirely new vocal theme for the words ‘Agnus Dei’ (b. 9), broader than the ritornello theme but no less expressive, and in canon at the lower 5th between alto voice and unison violins. In setting the words ‘Dona nobis pacem’ Bach reused the music of ‘Gratias agimus tibi’ from the Gloria. This has the great advantage that it establishes a clear link between the earlier and later portions of the Mass, binding the whole work together. In this context it is worth recalling that ‘Dona nobis pacem’ is more than just a prayer for peace; it is also the grand finale of the entire Mass. For this purpose a return to earlier music, and specifically to music associated with thanksgiving, makes perfect sense. A practical problem arises, however. ‘Gratias agimus tibi’ was constructed on the basis of two quite distinct subjects: a relatively plain theme for the first words, ‘Gratias agimus tibi’; and a more florid one for those that follow, ‘propter magnam gloriam tuam’. The text of ‘Dona nobis pacem’, on the other hand, consists of nothing more than those three words, which consequently have to serve for both subjects. The plain simplicity of the first subject is eminently well suited to these words, but not so the lively, florid character of the second subject. Bach attempts to adapt words and themes to each other by reversing the word-order and by replacing the repeated notes of the second subject with a suspension figure in long notes. But the problem remains that the following melisma in quavers, formerly associated with the glory of God, is less meaningful in relation to the text ‘Dona nobis pacem’. Despite these problems, the movement, as one of the most sublime in the entire Mass, forms a fitting conclusion to the great work as a whole. It is well known that in 1818 the Swiss publisher Hans Georg Na¨geli, one of the leaders of the Bach renaissance, described the B minor Mass as ‘the greatest musical artwork of all times and all people’.33 This is an extraordinary claim but perhaps not an inordinate one when considered in the light of the work’s meaning and significance in all its various aspects. In the context of Bach’s stated aim to produce a well-regulated church music for the glory of God,34 the Mass may be regarded as the culmination of his life’s work. The proprium—in Bach’s case, music written to German words earlier in his career for specific occasions in the church year—here, by means of the parody process, becomes the ordinarium, music to Latin words suitable for any occasion. What was merely

32 As shown in Christoph Wolff, ‘The Agnus Dei of Bach’s B-minor Mass: Parody and New Composition Reconciled’, in P. Brainard and R. Robinson (eds.), A Bach Tribute: Essays in Honour of William H. Scheide (Kassel and Chapel Hill, NC, 1993), pp. 233–40; repr. in C. Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 332–9. 33 In the Leipziger Allgemeine musikalische Zeitung (1818); facs. repr. in Friedrich Smend, Krit. Bericht, NBA II/1 (1956), p. 215. 34 Mu¨hlhausen, 25 June 1708; BD I, No. 1; NBR, No. 32.

m ass i n b m in o r

391

transient now becomes permanent. And in carrying out this transformation Bach draws on music from every stage of his maturity—from the late Weimar years, when he first began the regular composition of church music, right up to the end of his life. In terms of its stylistic reference, Bach’s Mass covers the entire history of music as it was then known, from the traditional Renaissance style of the sixteenth century right up to the most progressive pre-Classical style of the mid-eighteenth century. From Bach’s own point of view it may be assumed that the Mass had both an inner and an outer purpose. Externally, the Kyrie and Gloria were a means to preferment and a homage to the Electoral House of Saxony. However, the work was a homage not only to an outer monarch of the physical world but also to the inner Monarch of the spiritual world. And this brings us to the question of Bach’s own Christian faith. Various attempts were made in the twentieth century to undermine the view of Bach as a committed Christian whose church works reflect genuine conviction in relation to their texts and liturgical purposes.35 This scepticism was finally laid to rest, however, by the discovery of Bach’s copy of the three-volume Die deutsche Bibel (Wittenberg, 1681–2) with annotations by Abraham Calov. The many marginal notes in Bach’s own hand in this Bible demonstrate beyond doubt how profound was his personal commitment to the Christian faith.36 His comment on 2 Chronicles 5: 13 reads: ‘Where there is a devotional music God is always present with His grace’, a clear sign that for Bach sacred music and the divine power that calls it forth were inextricably linked.37 Such works as the St Matthew Passion or the Christmas Oratorio give only partial expression to this profound faith because they deal with particular events in the Christian calendar. The B minor Mass, on the other hand, gives complete and comprehensive expression to it because it deals with Christian doctrine in its entirety. The choice of the word ‘Christian’ rather than ‘Lutheran’ here is deliberate. Orthodox Lutherans in Bach’s day were brought up to feel and even express considerable animosity towards the Catholic Church.38 It is possible, however, that by the 1730s and 1740s Bach had arrived at a rather more tolerant view of Catholicism. The Catholic persuasion of the Elector of Saxony might have played a crucial role here. For Bach’s frequent use of the same music to honour secular and divine kingship suggests that he shared the widely held view of kings as God’s earthly representatives. 35 Most notoriously by Friedrich Blume; see his ‘Outlines of a new Picture of Bach’, Music and Letters, 44 (1963), pp. 214–27. 36 See C. Trautmann, ‘“Calovii Schrifften. 3. Ba¨nde” aus J. S. Bachs Nachlass und ihre Bedeutung fu¨r das Bild des lutherischen Kantors Bach’, Musik und Kirche, 39 (1969), pp. 145–60. 37 ‘NB Bey einer anda¨chtigen Music ist allezeit Gott mit seiner Gnaden Gegenwart’; see H. H. Cox (ed.), The Calov Bible of J. S. Bach (Ann Arbor, 1985), p. 419. 38 Thus Martin Luther’s Litany of 1528, quoted in the third movement of Gleichwie der Regen und Schnee vom Himmel fa¨llt, BWV 18 (Weimar, c. 1715), includes the prayer ‘Und uns fu¨r des Tu¨rken und des Papsts grausamen Mord und La¨sterungen, Wu¨ten und Toben va¨terlich behu¨ten. Erho¨r uns, lieber Herre Gott!’ (‘And from the Turk’s and the Pope’s cruel murder and blasphemies, rages and storms, preserve us like a father. Hear us, dear Lord God!’). Similarly, a well-known hymn of Luther’s from 1542—the basis of Bach’s chorale cantata BWV 126 of 1725—opens with the words ‘Erhalt uns, Herr, bei deinem Wort und steur des Papsts und Tu¨rken Mord’ (‘Uphold us, Lord, by Your Word and ward off the murderousness of Pope and Turk’).

392

the b m i n or m as s and o the r v oc al w o rk s

If, then, Bach accepted Friedrich Augustus II as God’s representative, he must surely have reconciled himself up to a point with the Catholic faith. The very dedication of the Kyrie–Gloria Missa to Friedrich Augustus in 1733 appears to be a sign of such a reconciliation; indeed, it has been viewed as a reflection of Bach’s desire to foster unity between Protestants and Catholics.39 Similarly, his decision to extend the Missa to cover all five divisions of the Mass Ordinary could reasonably be taken as an ecumenical statement. Only in the Catholic Church were the Kyrie, Gloria, Credo, Sanctus, and Agnus Dei regularly sung in Latin each Sunday in elaborate polyphonic or concerted settings. In the Lutheran Church, according to the Saxon Agenda established in 1539, all five divisions could be employed in the main Sunday service, the Hauptgottesdienst. In practice, however, monophonic settings and German hymn paraphrases were common; concerted settings were largely restricted to the Kyrie and Gloria, and sung only on special feast days. The Sanctus was regularly sung without its New Testament accretions, Osanna and Benedictus. The Agnus Dei was sung sub communione, most often in Luther’s German version, Christe, du Lamm Gottes. The Ordinary of the Latin Mass was nonetheless shared by the Lutheran and Catholic Churches, and it may be due to this universality that the complete Mass of 1748/9 is described in C. P. E. Bach’s estate catalogue of 1790 as ‘Die große Catholische Messe’, ‘the great Catholic Mass’.40 By the end of his life Bach must have attained a view of Christianity that transcended the divisions between Lutheranism and Catholicism. And it is this all-embracing attitude to the spiritual life that is enshrined within the work itself. This attitude is also relevant to our response. Because the B minor Mass covers the whole gamut of states of the soul, from the most profound sorrow to the most ecstatic joy, it acts as an expression of the wholeness of the psyche and consequently can have a restorative effect on its listeners, so often torn and divided by doubt, despair, depression, and other debilitating states of mind. The B minor Mass can do this supremely well because it is music of such immense depth and power. The spiritual states involved are not contingent but archetypal—they are not called forth by external events but exist in a pure, detached form as permanent constituents of our inner life. In representing these archetypal states of the soul with such awe-inspiring power, the B minor Mass puts us back in touch with the very source of our spiritual life itself. The feelings of uplift and exaltation that we experience upon entering fully into this sublime work are a measure of its power to reunite us with that which is of ultimate value.

Mary Dalton Greer takes this view; see her ‘Bach’s Calov Bible and His Quest for the Title of Royal Court Composer’, in Tomita et al. (eds.), Understanding Bach’s B-minor Mass (n. 12), vol. i, pp. 150–76 (esp. 175). 40 As Robin Leaver argues; see his ‘How “Catholic” is Bach’s “Lutheran” Mass?’, in Tomita et al. (eds.), Understanding Bach’s B-minor Mass (n. 12), vol. i, pp. 177–206 (esp. 198–206). Yoshitake Kobayashi is also of the view that ‘the B-minor Mass is not only Protestant-Lutheran but also Catholic; in other words, ecumenical’. See his ‘Die Universalita¨t in Bachs h-Moll-Messe’, Eng. trans. (n. 12), vol. ii, pp. 387–403 (esp. 392).

III.5 Conclusion

It is clear from the Entwurff of 17301 that Bach then considered himself to be a ‘modern’ composer. Accordingly, during the 1720s and 1730s his published works were advertised in the local press, the Leipziger Post-Zeitungen, alongside new-style works by his contemporaries, such as Georg Friedrich Kauffmann’s Harmonische Seelen-Lust (1733–6) and Conrad Friedrich Hurlebusch’s Compositioni musicali (c. 1734), both of which seem to have exerted a considerable influence upon him.2 Towards the end of the 1730s many references are found in the Leipzig press to the ‘neue Gusto’ (‘new taste’),3 and the question arises how Bach stood in relation to this novel style, both in his own eyes and in those of the musical public. After the Goldberg Variations of 1741, whose canonic technique is remarkably unobtrusive, he appears in his major printed works—the Canonic Variations, the Musical Offering, and The Art of Fugue—as a great master of the artifices of strict counterpoint. By the 1740s, then, he might well have been viewed by the musical public as a conservative, retrospective composer, out of touch with the latest developments of the day. This would accord with the Prussian king’s desire to test his legendary fugal skills at Potsdam in May 1747, for significantly Frederick the Great presented him with a fugue subject couched in the old style—the opposite pole to the galant or empfindsam style cultivated at the court. As for Bach and his associates, they repeatedly referred to the Musical Offering as the ‘Prussian Fugue’,4 thereby ignoring the ‘Sonata sopr’il Soggetto Reale’ that contains some of its most ‘modern’ music. The wider reality is rather different, however. The first phase of the new style was represented by the ‘galant homme’ of Johann Mattheson’s Das Neu-Ero¨ffnete Orchestre (1713), who stood for the modern and rational as opposed to tradition. Later, in Das Forschende Orchestre of 1721, Mattheson referred to B. Marcello, Vivaldi, Caldara, A. Scarlatti, Lotti, Keiser, Handel, and Telemann as ‘die allerberu¨hmtesten und galantesten Componisten in Europa’ (‘the most celebrated and most galant composers 1

BD I, No. 22; NBR, No. 151. See John Butt, ‘Bach and G. F. Kauffmann: Reﬂections on Bach’s Later Style’, in D. R. Melamed (ed.), ¨ bung III: The Making of a Print; Bach Studies 2 (Cambridge, 1995), pp. 47–61; Gregory Butler, Bach’s Clavier-U with a Companion Study of the Canonic Variations on Vom Himmel hoch, BWV 769 (Durham and London, 1990), pp. 4–16; and Peter Williams, The Organ Music of J. S. Bach (rev. 2nd edn, Cambridge, 2003), p. 388. 3 See Christian Ahrens, ‘Joh. Seb. Bach und der “neue Gusto” in der Musik um 1740’, BJ 72 (1986), pp. 69–79. 4 BD II, No. 557; III, No. 558 a (p. 656); I, No. 49. NBR, Nos. 247–8 and 257. 2

394 p art i i i in Europe’). Bach knew and performed the music of all these composers, and four of them—Caldara, Keiser, Handel, and Telemann—he particularly admired, according to his son C. P. E. Bach.5 Moreover, we have found plentiful evidence that in the 1720s and 1730s Bach himself embraced certain key elements of the galant style, while never shedding his fundamentally contrapuntal mode of musical thinking (see Part I Ch. 5 and Part II Ch. 5). In the late 1730s and 1740s Bach clearly showed an interest in the second phase of the galant movement, as represented by the music of his son Carl Philipp Emanuel, by Hasse and the Graun brothers, whose music we are told he admired in his later years,6 and by Quantz, Vinci, and Pergolesi.7 According to Quantz, music written in this style was expected to be clear, natural, pleasant, and agreeable. These qualities would be manifest in clearly periodic melody, simple harmony, light texture, and formulaic cadences. One of the clearest signs of Bach’s interest in this style is his Tilge, Ho¨chster, meine Su¨nden, BWV 1083 (1746/7), a parody of the celebrated Stabat Mater by Giovanni Battista Pergolesi, who was and is widely regarded as the arch-exponent of the later phase of the style galant. To a certain extent Bach imposes elements of his own style upon Pergolesi’s music, and yet he does nothing to alter the galant character of the melodic writing. And reminiscences of the Italian composer’s work in the threepart Ricercar from the Musical Offering and in the ‘Et incarnatus est’ from the Credo of the B minor Mass demonstrate what a lasting impression it made upon him.8 In his last decade, as in the 1730s, Bach showed a simultaneous preoccupation with old and new styles—typically, the stile antico and the style galant—and a tendency to unite them or place them in direct juxtaposition. In two sonatas from the 1740s—the Flute Sonata in E, BWV 1035, and the ‘Sonata sopr’il Soggetto Reale’, BWV 1079 no. 3—he reverted to old forms and textures that he had long since largely repudiated in his chamber music, namely the sonata da chiesa and the basso continuo texture. Yet the E major Sonata—like the Gamba Sonata in D, BWV 1028 (particularly its second movement), which might have been written around the same time—counts as one of Bach’s most galant compositions; and the Andante third movement of the Sonata in C minor from the Musical Offering represents one of his most thoroughgoing essays in the North German subspecies of the galant known as the empﬁndsamer Stil, a style associated above all with the music of his son Carl Philipp Emanuel. Empﬁndsamkeit may be translated as ‘sensibility’ and denotes the search for a subjective, keenly felt mode of expression that would have an immediate effect on the listener. The style 5

Letter to Forkel of 13 Jan. 1775; BD III, No. 803; NBR, No. 395. C. P. E. Bach, letter to Forkel of 13 Jan. 1775; see under n. 5. The two-phase view of the galant presented here is indebted to David A. Sheldon, ‘The galant Style Revisited and Re-evaluated’, Acta musicologica, 47 (1975), pp. 240–69. 8 See Werner Breig, ‘J. S. Bachs Leipziger Klaviermusik und das Prinzip “Empﬁndsamkeit”’, in S. Schmalzriedt (ed.), Aspekte der Musik des Barock: Auffu¨hrungspraxis und Stil (Laaber, 2006), pp. 295–315; and Christoph Wolff, ‘“Et incarnatus” and “Cruciﬁxus”: The Earliest and Latest Settings of Bach’s B-minor Mass’, in M. A. Parker (ed.), Eighteenth-Century Music in Theory and Practice: Essays in Honour of Alfred Mann (Stuyvesant, NY, 1994), pp. 1–17. 6 7

c on c l usi on

395

shares much with its galant parent—specific characteristics are sighing appoggiatura figures, detailed dynamic nuances, and subtle use of chromatic inflections. As we shall see, all of these features are found in certain compositions by Bach of the period 1739–50. Despite the progressive traits noted here, there is a strong retrospective element in Bach’s music of the 1740s. The ‘Eighteen’ and Schu¨bler Chorales both refer back to bygone periods: the one to large-scale organ chorales of the Weimar years; the other to chorale arrangements from Leipzig cantatas of the period 1724–31. The Concerto in A minor, BWV 1044, is based primarily on a massive keyboard work of the Weimar period (BWV 894); and with its dual identity as solo harpsichord concerto and concerto grosso it refers back to Brandenburg Concerto No. 5, whose concertino is identical in composition (flute, violin, and harpsichord). Nevertheless, the sophistication of the ripieno passages, with their detailed articulation, pizzicato–arco contrasts, and finely graded dynamics, seems like an attempt to ape the orchestral style of the Berlin school as cultivated by C. P. E. Bach in the 1740s and 1750s. It is possible that the work was associated with one of Bach’s Berlin visits, either in 1741 or in 1747.9 The Well-Tempered Clavier II likewise looks backwards and forwards simultaneously. It is clearly conceived as a successor to the original set of twenty-four preludes and fugues, completed at Co¨then in 1722. In terms of style and technique, however, it is no mere duplicate of the earlier set but reflects Bach’s priorities and preoccupations twenty years later. There are fewer arpeggiated or pseudo-improvisatory preludes than in Part I, and in general the preludes tend to be more thoroughly worked out and designed on a considerably larger scale. Significantly, no fewer than ten preludes are cast in binary dance form with repeats, a structure that had undergone considerable expansion in the keyboard Partitas of Clavieru¨bung I and had long since been emancipated from the dance. The binary preludes in D and B♭, like the Fantasia in C minor, BWV 906, resemble Scarlattian sonatas in their exceptional brilliance and highly idiomatic writing for the harpsichord. As in the partitas, Bach’s handling of the binary structure often seems to foreshadow sonata form. A clear example is the Praeludium in E, whose first strain includes a distinct thematic group in the dominant key, recapitulated in the tonic at the end of the second strain. In two of the binary preludes, those in the minor keys of F and G♯, Bach seeks to capture the empfindsam style described earlier—the repeated-note-plus-appoggiatura figure of their themes is a hallmark of his essays in this style throughout the 1730s and 1740s. Other aspects of the preludes (whether binary or not) that distinguish them clearly from those of Part I are an expansive use of ritornello form (F♯, A♭), a cantabile style already developed in Clavieru¨bung I–II (c♯, f♯), and strict counterpoint (a, b♭), which has a clear precedent in the Inventions and Sinfonias. The chromatic ‘twelve-note’ double theme of the remarkable Praeludium in A minor may be intended as a microcosm 9 See Peter Wollny, ‘J. S. Bachs Tripelkonzert a-Moll BWV 1044’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke (Witten, 1997), pp. 283–91.

396

par t i i i

of the totally chromatic key scheme of the work as a whole. Empfindsam chromatic writing is here allied with strict counterpoint, rendering the piece a worthy successor to the Sinfonia in F minor, BWV 795. Ten of the fugues are free of the artifices of strict counterpoint (apart from the invertible combination of subject and countersubject) and might be described as character-fugues, since their individual quality and expressiveness rely greatly on the specific character of their themes. They include two of the most ‘modern’ fugues in the collection: the cantabile-style Fuga in B♭, with its regular phrase structure, clear cadencing, and rhyming-close; and the Fuga in F♯, with its prominent empfindsam characteristics. Another character-fugue, that in A minor, makes highly original use of traditional material, namely the falling diminished 7th that belongs to the stock figures of Baroque fugue writers. At the opposite end of the spectrum are three successive major-mode stretto fugues, those in D, E♭, and E, composed in a traditional style derived from vocal polyphony. Of these the most extreme is the Fuga in E, a study in the stile antico and a salute to J. C. F. Fischer (quoting one of his themes), whose Ariadne musica had inspired Part I of Bach’s magnum opus some twenty years previously. A rare resource in this fugue (also encountered in the Fuga in D♯ minor from Part I) is subject variation, which would later play a significant part in The Art of Fugue. Metrical augmentation of the subject was at all times less common in Bach than diminution. However, it occurs in two fugues from Part II of The Well-Tempered Clavier (c, C♯), as well as in one from Part I (d♯). The strict fugues of Part II thoroughly explore the possibilities not only of augmentation and diminution but of double and triple counterpoint (g, A♭), double and triple fugue (c♯, g♯, B, f♯), and stretto and inversion (C♯, c, d, d♯, b♭)—techniques that would soon play a fundamental role in The Art of Fugue. The highly organized stretto-inversion Fuga in B♭ minor is the equal of any ‘contrapunctus’ from the later collection in its combination of systematic structure and depth and beauty of content. A subtle, new feature of certain double fugues (c♯, g♯) is the adumbration of the second subject during the first half of the fugue, so that its entry midway is gradually prepared rather than abrupt. It has been suggested10 that the chapters on counterpoint from Johann Mattheson’s Der vollkommene Capellmeister, which was published in Leipzig at the Easter Fair of 1739, might have influenced the character of Bach’s fugal writing in the 1740s. Mattheson makes a direct reference to Bach: ‘Of double fugues with three subjects . . . I would much rather see something of the same sort published by the famed Herr Bach of Leipzig, who is a great master of fugue.’ This remark might have prompted Bach’s composition of the concluding Fuga in E♭ from the Clavieru¨bung III, which belongs to the type specified by Mattheson. In a broader sense, Mattheson’s reference to multiple fugue might have suggested the contrapuntal form of the fugues in c♯, g♯, B, and f♯ 10 By Gregory Butler, ‘Der vollkommene Capellmeister as a Stimulus to J. S. Bach’s Late Fugal Writing’, in G. J. Buelow and H. J. Marx (eds.), New Mattheson Studies (Cambridge, 1983), pp. 293–305.

c on c lus io n

397

from The Well-Tempered Clavier II. Moreover, his discussion and examples of contrapuntal oblighi (self-imposed technical constraints within a composition, cultivated by Frescobaldi and others), borrowed from Angelo Berardi’s Documenti armonici, Part I (1687), appear to have served as a model for Bach’s use of such devices in the same collection: for example, the countersubject built out of a repeated, syncopated figure (Fuga in B, bb. 5–7), or out of a repeated motive (Fuga in f♯, bb. 9–11); or the use of an ostinato figure as countersubject (Fuga in b♭, bb. 67–8). That the Goldberg Variations were published in 1741, around the time when Bach was completing Part II of The Well-Tempered Clavier, cannot be entirely without significance. The fugal element of the latter collection would shortly be combined with the variation and canonic elements of the former to become the defining features of The Art of Fugue and the Musical Offering. The variation technique of the Goldbergs, however, embraces more than canons: the canonic movements alternate with studies and character-pieces that often recall the recently composed preludes of The WellTempered Clavier II. The cantabile variations, for example, with their highly florid treble and twin supporting parts, nos. 13 and 25—the latter, with its subtle chromatic writing, among the most empfindsam of Bach’s late harpsichord pieces—recall the Praeludium in F♯ minor and, to a lesser extent, that in C♯ minor. Similarly, the Scarlattian hand-crossing studies not only refer back to the Giga from Partita No. 1 (BWV 825) and the Fantasia in C minor (BWV 906) but have a clear counterpart in the Praeludium in B♭. The use of genuine folk songs in the Quodlibet has a quite different orientation, referring back to the Bach-family quodlibets of the composer’s youth as described by Forkel,11 and forward to the more extensive use of such folk material in the Cantate burlesque (Peasant Cantata), BWV 212, of the following year (1742).12 Further canonic work on the fundamental bass of the Goldberg set led to the Verschiedene Canones, BWV 1087, which are tantamount to a set of canonic variations, despite their numbering as fourteen separate canons. Everything is here reduced to essentials: the 32-bar Goldberg bass is restricted to its kernel, the first eight notes; and it is no longer paraphrased, as in the Goldberg set, but presented in its plain, underlying form throughout. The relatively difficult form of canon by inversion, represented by only two variations in the Goldbergs (nos. 12 and 15), here becomes standard; and two rarer types also occur, namely retrograde canon and augmentation canon. There are two basic modes of procedure: either the theme (extracted from the Goldberg bass) is treated as a cantus firmus in combination with the canonic voices, or else the Goldberg theme becomes the subject of the canonic voices. Both types recur in Bach’s two other sets of canonic variations from this period: those based on the chorale Vom Himmel hoch da komm ich her (BWV 769) and the ten canons on the

BD VII, pp. 15–16; NBR, p. 424. See Hans-Joachim Schulze, ‘Melodiezitate und Mehrtextigkeit in der Bauernkantate und in den Goldbergvariationen’, BJ 62 (1976), pp. 58–72. 12

398

par t i ii

‘Thema Regium’ from the Musical Offering. The canonic types employed in the Vom Himmel hoch Variations include canons by inversion and augmentation, as in the Fourteen Canons; but the chorale-based set is less austere, partly because two of the variations (nos. 3 and 4) present a florid cantabile part in Bach’s most ‘modern’ style as a free counterpoint to the chorale cantus firmus. The ten canons of the Musical Offering also distinguish between theme as cantus firmus and theme as canonic subject; in addition, however, they are governed by another basic distinction— between canons based on the original plain version of the ‘Thema Regium’ and those based on various decorated versions. This distinction even applies to the ‘Sonata sopr’il Soggetto Reale’: the Allegro second and fourth movements employ the Royal Theme in opposite ways: as a plain cantus firmus in counterpoint with the Allegro theme (no. 2); and in a highly varied form in 6/8 time as the subject of the second Allegro (no. 4). In this plain/decorated antithesis the Musical Offering clearly follows The Art of Fugue, in which it acts as one of the chief modes of variation. The theme of the Goldberg Variations and its offshoot, the Verschiedene Canones, is of course the bass rather than the melody of the florid, sarabande-like Aria with which it begins. This bass has a purely functional role as determinant of the harmonic structure that underlies every variation. The theme of The Art of Fugue, on the other hand, is melodic—just like the ‘given’ (externally sourced) themes of two contemporaneous collections that also rely on the variation principle, the choralebased Canonic Variations (BWV 769) and the Musical Offering, which is of course based throughout on the Royal Theme. Bach worked on The Art of Fugue, at least intermittently, from about 1742—shortly after the completion of The Well-Tempered Clavier II and the Goldberg Variations—until early 1750, not long before his death. And it represents the most significant product of that in-depth absorption in the techniques of strict counterpoint which lasted throughout most of his last decade. As in The Well-Tempered Clavier II, certain aspects of the fugal writing appear to be indebted to the chapters on counterpoint in Mattheson’s Der vollkommene Capellmeister,13 particularly the multiple-fugue design of Contrapunctus 8–11 and the employment of contrapuntal oblighi in the countersubjects of Contrapunctus 1–5: for example, the persistent dotted rhythm in No. 2, the constant stepwise motion in No. 3 (bb. 5–8), and the use of ostinato in No. 4 (bb. 107–9), combined with the subject in close canonic stretto. The Art of Fugue theme is subject to multiple forms of variation: rectus and inversus, metrical diminution and augmentation, melodic variation, and combination with other subjects or countersubjects. Nevertheless, the work lacks the stylistic diversity and melodic inventiveness of The Well-Tempered Clavier II or the Goldberg Variations. As in the Verschiedene Canones, one forms the impression that everything has been reduced to bare essentials. Preludes, variety of key, the invention of ‘character-

According to Gregory Butler, ‘Der vollkommene Capellmeister’.

c on c lus io n

399

subjects’: all are renounced in favour of a single principal subject of neutral character, moulded with its contrapuntal possibilities in mind. Bach goes a long way towards compensating for these limitations by virtue of the varied character of his secondary subjects and the inventiveness of the many decorated forms of the principal subject. The diverse forms of strict counterpoint are nonetheless the main argument here. With the exception of the mirror fugues (Nos. 12–13), they can be more or less matched with contrapuntal modes of treatment found in The Well-Tempered Clavier II: simple fugue (Nos. 1–4), stretto-inversion fugue (Nos. 5–7), double fugue (Nos. 9–10), and triple fugue (Nos. 8 and 11). In the earlier work, however, they contribute to the characterization of a fugue and are subordinate to that aim. In The Art of Fugue, on the other hand, the entire raison d’eˆtre is the systematic exploration of these techniques of strict counterpoint. In view of the textbook character of the collection, with its logical ordering according to contrapuntal type, it is a wonder that Bach managed to create music of such beauty and profundity. There are obvious continuities with the 1730s in the compositions of Bach’s last decade. His first attempt at secular comedy, the Coffee Cantata of 1734, was followed up by another in 1742, the Cantate burlesque (Peasant Cantata). Above all, his expansion of the 1733 Missa (Kyrie and Gloria) to form the B minor Mass by adding the Credo, Sanctus, and Agnus Dei in 1748–9 may be viewed with hindsight as the completion of an unfinished project. Whether this Missa tota was inspired by works of that kind by Baal, Bassani, Palestrina, and others that he studied and performed, or whether it was the result of a commission, we do not know. What has become clear in recent years is that, alongside The Art of Fugue, it was Bach’s last major composition,14 occupying him during the year 1748–9. The greatest effort must have gone into the composition of the massive Credo, or ‘Symbolum Nicenum’, which in some respects seems to have been modelled on the Gloria: both were built up by a process of compilation that involved much use of parody; and both are divided into nine movements according to symmetrical schemes. For the Old Testament part of the Sanctus (‘Sanctus, sanctus, sanctus’ and ‘Pleni sunt coeli’), sung regularly in the Lutheran Church, Bach was able to borrow his finest setting, the six-part Sanctus of 1724. For the New Testament part (‘Osanna’, ‘Benedictus’, ‘Osanna’) he had to resort to parody (‘Osanna’) and possibly new composition (‘Benedictus’). Bach goes out of his way to unite the Old and New Testament parts of the Sanctus by introducing a powerful anticipation of the ‘Osanna’ theme at the end of ‘Pleni sunt coeli’ (b. 159). Yet the two choruses are perhaps too similar: both are in a dance-like 3/8 time and in D major with trumpets and drums. This is a relatively small point, however, by comparison with the structuring of the Agnus Dei. The threefold form of this prayer, no less crucial than that of the Kyrie (which Bach observes), is completely ignored by him. Instead, he sets the Agnus Dei as a dual structure of aria and chorus. The 14 Work on The Art of Fugue continued after the completion of the Mass, according to A. P. Milka, ‘Zur Datierung der h-Moll-Messe und der Kunst der Fuge’, BJ 96 (2010), pp. 53–68.

400 par t i ii bipartite aria may be construed as a setting of the first two petitions, but the chorus sets only the last words of the third petition, ‘Dona nobis pacem’. A possible explanation is that Bach was already feeling hindered by ill health and even becoming conscious of his approaching end by the time he reached the fifth and last division of the Mass. It might be for the same reason that he reused the music of the ‘Gratias agimus tibi’ from the Gloria for the ‘Dona nobis pacem’. This has advantages—it knits the 1733 and 1748/9 portions of the Mass together and allows the work to end on a sublime note—but there is no concealing the fact that the music fits less well to the ‘Dona nobis pacem’ than to the ‘Gratias agimus tibi’ text. In the B minor Mass and the Musical Offering, the historic extremes of ‘ancient’ and ‘modern’ styles, such as stile antico and style galant, are often placed side by side. Whereas the Ricercar a 6 from the Musical Offering is written in a traditional style derived from Renaissance polyphony, the Ricercar a 3 refers to the empfindsam style current at the court of the dedicatee Frederick the Great. Similarly, the opening and closing complexes of the Credo from the B minor Mass each consist of two choruses, one in stile antico and the other in what might be termed stile moderno. The two antico movements, ‘Credo in unum Deum’ and ‘Confiteor’, are both five-part stretto fugues in alla breve metre and both make use of the appropriate Gregorian plainchant—as fugue subject in the Credo intonation and as canonic (later augmented) cantus firmus in the ‘Confiteor’. The retrospective elements of plainchant, cantus firmus technique, and (in the Credo movement) modality all recall another Mass movement—the Kyrie–Christe–Kyrie sequence from the organ Missa of Clavieru¨bung III (1739). The moderno movement that follows the Credo intonation, ‘Patrem omnipotentem’, is also fugal but couched in a contemporary Baroque style, hence the repercussion headmotive of the subject, its illustrative features, and the homophonic choral refrain that accompanies the subject entries. The moderno chorus that follows the ‘Confiteor’, ‘Et expecto resurrectionem mortuorum’, is a ritornello movement (albeit truncated for the sake of economy) in Bach’s most mature concertante style. The juxtaposition of movements at opposite ends of the historical spectrum is taken to extremes in the central complex of the Credo, which deals with the Incarnation, Crucifixion, and Resurrection. The ‘Crucifixus’ is not only the earliest composition in the entire Mass—parodied from the opening chorus of Weinen, Klagen, Sorgen, Zagen, BWV 12, of 1714—but it is also relatively antiquated in compositional make-up, its chaconne form and chromatic lamento bass reaching far back into the seventeenth century. ‘Et resurrexit’ (like ‘Et expecto’), on the other hand, is a highly mature concertante piece in ritornello form. As for ‘Et incarnatus est’, not only was it perhaps Bach’s last vocal piece but it appears to have been inspired by Pergolesi’s galant-style Stabat Mater, which Bach had parodied only a few years previously.15 Bach thus ended his composing career with a Mass so comprehensive in its stylistic reference that it covered the

According to Christoph Wolff (see n. 8).

c o nc l usi on

401

entire known history of music at the time. Old and new styles, which must have seemed mutually exclusive to many of their exponents in the early eighteenth century, were like grist to the mill for Bach and could be combined or juxtaposed at will. Like Shakespeare, Bach stands as a colossus astride all periods and developments. Only thus can his music be fully understood and appreciated.

Bibliography Due to the vast extent of the Bach literature, this bibliography is necessarily selective. It is restricted to books and articles that have proved particularly useful in the preparation of this study. Ahrens, Christian, ‘Joh. Seb. Bach und der “neue Gusto” in der Musik um 1740’, BJ 72 (1986), pp. 69–79. Ambrose, Z. Philip, ‘Klassische und neue Mythen in Bachs weltlichen Kantaten’, in C. Wolff (ed.), Die Welt der Bach Kantaten, vol. ii. J. S. Bachs weltliche Kantaten (Stuttgart, 1997), pp. 139–55. Apel, Willi, Geschichte der Orgel- und Klaviermusik bis 1700 (Kassel, 1967); trans. and rev. by H. Tischler as The History of Keyboard Music to 1700 (Bloomington and Indianapolis, 1972). Axmacher, Elke, ‘Ein Quellenfund zum Text der Mattha¨us-Passion’, BJ 64 (1978), pp. 181–91. ——‘Aus Liebe will mein Heyland sterben’: Untersuchungen zum Wandel des Passionsversta¨ndnisses im fru¨hen 18. Jahrhundert (Neuhausen and Stuttgart, 1984). Baron, Carol K. (ed.), Bach’s Changing World: Voices in the Community (Rochester, NY and Woodbridge, 2006). Basso, Alberto, ‘Oper und “dramma per musica” ’, in C. Wolff (ed.), Die Welt der Bach Kantaten, vol. ii. J. S. Bachs weltliche Kantaten (Stuttgart, 1997), pp. 49–63. Beißwenger, Kirsten, Johann Sebastian Bachs Notenbibliothek (Kassel, 1992). ——Kritischer Bericht, NBA II/9 (Kassel and Leipzig, 2000). Berger, Christian, ‘J. S. Bachs Cembalokonzerte: Ein Beitrag zur Gattungsgeschichte des Klavierkonzerts im 18. Jahrhundert’, Archiv fu¨r Musikwissenschaft, 47 (1990), pp. 207–16. Bertling, Rebekka, ‘Das Arioso und das ariose Accompagnato im Vokalwerk J. S. Bachs’, Musik und Kirche, 62 (1992), pp. 327–34. Blankenburg, Walter, ‘Eine neue Textquelle zu sieben Kantaten Johann Sebastian Bachs und achtzehn Kantaten Johann Ludwig Bachs’, BJ 63 (1977), pp. 7–25. ——Das Weihnachts-Oratorium von J. S. Bach (Kassel, 1982). Boyd, Malcolm, Bach: The Brandenburg Concertos (Cambridge, 1993). ——(ed.), Oxford Composer Companions: J. S. Bach (Oxford, 1999). ——Bach (London and Melbourne, 1983; 3rd edn Oxford, 2000). Brainard, Paul, ‘Bach’s Parody Procedure and the St. Matthew Passion’, Journal of the American Musicological Society, 22 (1969), pp. 241–60. ——(ed.), NBA II/7: Oster-Oratorium (Kassel and Leipzig, 1977); Kritischer Bericht (1981). ——‘The Aria and its Ritornello: The Question of “Dominance” in Bach’, in W. Rehm (ed.), Bachiana et Alia Musicologica: Festschrift Alfred Du¨rr (Kassel, 1983), pp. 39–51. ——‘Aria and Ritornello: New Aspects of the Comparison Handel/Bach’, in P. Williams (ed.), Bach, Handel, Scarlatti: Tercentenary Essays (Cambridge, 1985), pp. 21–33. ——‘The Regulative and Generative Roles of Verse in Bach’s “Thematic” Invention’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 54–74.

b i b l i o gr a p h y

403

——‘The “Non-quoting” Ritornello in Bach’s Arias’, in P. Brainard and R. Robinson (eds.), A Bach Tribute: Essays in Honour of William H. Scheide (Kassel and Chapel Hill, NC, 1993), pp. 27–44. Breig, Werner, ‘Bachs Goldberg-Variationen als zyklisches Werk’, Archiv fu¨r Musikwissenschaft, 32 (1975), pp. 243–65. ——‘Bachs Violinkonzert d-Moll: Studien zu seiner Gestalt und seiner Entstehungsgeschichte’, BJ 62 (1976), pp. 7–34. ——‘Zu Bachs Umarbeitungsverfahren in den “Achtzehn Chora¨len” ’, in T. Kohlhase and V. Scherliess (eds.), Festschrift Georg von Dadelsen (Neuhausen and Stuttgart, 1978), pp. 33ff.; rev. and enlarged version as ‘The “Great Eighteen” Chorales: Bach’s Revisional Process and the Genesis of the Work’, in G. Stauffer and E. May (eds.), J. S. Bach as Organist (London, 1986), pp. 102–20. ——‘Bachs Cembalokonzert-Fragment in d-Moll (BWV 1059)’, BJ 65 (1979), pp. 29–36. ——‘J. S. Bach und die Entstehung des Klavierkonzerts’, Archiv fu¨r Musikwissenschaft, 36 (1979), pp. 21–48. ——‘Zur Chronologie von J. S. Bachs Konzertschaffen: Versuch eines neuen Zugangs’, Archiv fu¨r Musikwissenschaft, 40 (1983), p. 77–101. ——‘Bachs freie Orgelmusik unter dem Einﬂuß der italienischen Konzertform’, in R. Szeskus and J. Asmus (eds.), Johann Sebastian Bachs Traditionsraum, Bach-Studien 9 (Leipzig, 1986), pp. 29–43. ——‘Grunzu¨ge einer Geschichte von Bachs vierstimmigem Choralsatz’, Archiv fu¨r Musikwissenschaft, 45 (1988), pp. 165–85, 300–19. ——‘Zum Kompositionsprozeß in Bachs Cembalokonzerten’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld: 61. Bachfest der Neuen Bachgesellschaft, Duisburg 1986 (Kassel, 1988), pp. 32–47. ——‘Das Ostinato-Prinzip in J. S. Bachs langsamen Konzertsa¨tzen’, in W. Osthoff and R. Wiesend (eds.), Von Isaac bis Bach: Studien zur a¨lteren deutschen Musikgeschichte: Festschrift Martin Just zum 60. Geburtstag (Kassel, 1991), pp. 287–300. ——‘Versuch einer Theorie der Bachschen Orgelfuge’, Musikforschung, 48 (1995), pp. 14–52. ——‘Composition as Arrangement and Adaptation’, in J. Butt (ed.), The Cambridge Companion to Bach (Cambridge, 1997), pp. 154–70. ——‘Zur Werkgeschichte von Bachs Cembalokonzert BWV 1056’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke [conference report, Dortmund, 1996] (Witten, 1997), pp. 265–82. ——‘Freie Orgelwerke’, in K. Ku¨ster (ed.), Bach-Handbuch (Kassel and Stuttgart, 1999), pp. 614–712. ——Kritischer Bericht, NBA VII/4: Konzerte fu¨r Cembalo (Kassel and Leipzig, 2001). ——‘Beobachtungen an einigen spa¨ten Klavierfugen Bachs’, in U. Leisinger (ed.), Bach in Leipzig—Bach und Leipzig [conference report, Leipzig, 2000] (Hildesheim, 2002), pp. 207–18. ——‘Zur Vorgeschichte von Bachs Ouvertu¨re h-Moll BWV 1067’, BJ 90 (2004), pp. 41–63. ——‘J. S. Bachs Leipziger Klaviermusik und das Prinzip “Empﬁndsamkeit” ’, in S. Schmalzriedt (ed.), Aspekte der Musik des Barock: Auffu¨hrungspraxis und Stil [conference report, Karlsruhe, 2001–4] (Laaber, 2006), pp. 295–315. Brokaw II, James A., ‘The Genesis of the Prelude in C major, BWV 870’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 225–39.

404

bi b l io g rap hy

Buelow, George J., ‘In Defence of J. A. Scheibe against J. S. Bach’, Proceedings of the Royal Musical Association, 101 (1974–5), pp. 85–100. ——‘Expressivity in the Accompanied Recitatives of Bach’s Cantatas’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 18–35. Butler, Gregory G., ‘Der vollkommene Capellmeister as a Stimulus to J. S. Bach’s Late Fugal Writing’, in G. J. Buelow and H. J. Marx (eds.), New Mattheson Studies (Cambridge, 1983), pp. 293–305. ——‘Ordering Problems in J. S. Bach’s Art of Fugue Resolved’, Musical Quarterly, 69 (1983), pp. 44–61. ——‘Neues zur Datierung der Goldberg-Variationen’, BJ 74 (1988), pp. 219–23. ¨ bung III: The Making of a Print; with a Companion Study of the Canonic ——Bach’s Clavier-U Variations on ‘Vom Himmel hoch’, BWV 769 (Durham and London, 1990). ——‘J. S. Bachs Gloria in excelsis Deo BWV 191: Musik fu¨r ein Leipziger Dankfest’, BJ 78 (1992), pp. 65–71. ——‘J. S. Bach’s Reception of Tomaso Albinoni’s Mature Concertos’, in D. R. Melamed (ed.), Bach Studies 2 (Cambridge, 1995), pp. 20–46. ——‘Towards a More Precise Chronology of Bach’s Concerto for Three Violins and Strings, BWV 1064a: The Case for Formal Analysis’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke [conference report, Dortmund, 1996] (Witten, 1997), pp. 235–47. ——‘J. S. Bachs Kanonische Vera¨nderungen u¨ber “Vom Himmel hoch” BWV 769: Ein Schlußstrich unter die Debatte um die Frage der “Fassung letzter Hand” ’, BJ 86 (2000), pp. 9–34. ——‘Eine neue Interpretation der Druckgeschichte des Musikalischen Opfers’, in U. Leisinger (ed.), Bach in Leipzig—Bach und Leipzig [conference report, Leipzig, 2000] (Hildesheim, 2002), pp. 309–20. ——‘The Printing History of J. S. Bach’s Musical Offering: New Interpretations’, Journal of Musicology, 19 (2002), pp. 306–31. ——‘The Prelude to the Third English Suite, BWV 808: An Allegro Concerto Movement in Ritornello Form’, in A. Leahy and Y. Tomita (eds.), Bach Studies from Dublin, Irish Musical Studies 8 [conference report, Dublin, 2000] (Dublin, 2004), pp. 93–101. ——‘Scribes, Engravers, and Notational Styles: The Final Disposition of Bach’s Art of Fugue’, in G. G. Butler, G. B. Stauffer, and M. D. Greer (eds.), About Bach (Urbana and Chicago, 2008), pp. 111–23. Butt, John, Bach: Mass in B Minor (Cambridge, 1991). ——‘Bach and G. F. Kauffmann: Reﬂections on Bach’s Later Style’, in D. R. Melamed (ed.), Bach Studies 2 (Cambridge, 1995), pp. 47–61. ——(ed.), The Cambridge Companion to Bach (Cambridge, 1997). Chafe, Eric, ‘Key Structure and Tonal Allegory in the Passions of J. S. Bach: An Introduction’, Current Musicology, 31 (1981), pp. 39–54. ——‘J. S. Bach’s St. Matthew Passion: Aspects of Planning, Structure, and Chronology’, Journal of the American Musicological Society, 35 (1982), pp. 49–114. ——‘Allegorical Music: The Symbolism of Tonal Language in the Bach Canons’, Journal of Musicology, 3 (1984), pp. 340–62. ——‘The St. John Passion: Theology and Musical Structure’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 75–112. ——Tonal Allegory in the Vocal Music of J. S. Bach (Berkeley, 1991).

b i b l i o gr a p h y

405

——Analyzing Bach Cantatas (Oxford, 2000). Clement, Albert, Der dritte Teil der Clavieru¨bung von J. S. Bach: Musik, Text, Theologie (Middelburg, 1999). Cox, Howard H., The Calov Bible of J. S. Bach (Ann Arbor, 1985). ——‘Bach’s Knowledge of the Bible’, in P. Brainard and R. Robinson (eds.), A Bach Tribute: Essays in Honour of William H. Scheide (Kassel and Chapel Hill, NC, 1993), pp. 87–99. Crist, Stephen A., ‘Aria Forms in the Cantatas from Bach’s ﬁrst Leipzig Jahrgang’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 36–53. ——‘The Question of Parody in Bach’s Cantata “Preise dein Glu¨cke, gesegnetes Sachsen”, BWV 215’, in R. Stinson (ed.), Bach Perspectives [I] (Lincoln, Nebr., 1995), pp. 135–62. ——‘J. S. Bach and the Conventions of the Da Capo Aria, or How Original was Bach?’, in P. F. Devine and H. White (eds.), The Maynooth International Musicological Conference 1995: Selected Proceedings, Part I, Irish Musical Studies 4 (Dublin, 1996), pp. 71–85. ——‘Bach, Theology, and Harmony: A New Look at the Arias’, Bach, 27/1 (1996), pp. 1–30; also pub. in J. Knowles (ed.), Critica Musica: Essays in Honour of Paul Brainard (Amsterdam and New York, 1996), pp. 65–95. Dadelsen, Georg von, Beitra¨ge zur Chronologie der Werke Johann Sebastian Bachs, Tu¨binger Bach Studien 4/5 (Trossingen, 1958). ——‘Anmerkungen zu Bachs Parodieverfahren’, in W. Rehm (ed.), Bachiana et Alia Musicologica: Festschrift Alfred Du¨rr (Kassel, 1983), pp. 52–7. ——‘Friedrich Smends Ausgabe der h-moll-Messe von J. S. Bach’, Die Musikforschung, 12 (1959), ¨ ber Bach und anderes: Aufsa¨tze und Vortra¨ge 1957–1982 (Laaber, pp. 315–34; repr. in Dadelsen, U 1983), pp. 18–40; Eng. trans. by J. A. Brokaw in Bach, 20/2 (1989), pp. 49–74. ——‘Eine unbekannte Messenbearbeitung Bachs’, in H. Hu¨schen (ed.), Festschrift Karl Gustav ¨ ber Bach und anderes (Laaber, Fellerer (Regensburg, 1962), pp. 88–94; repr. in Dadelsen, U 1983), pp. 68–74. ¨ ber Bach und anderes ——‘Bachs h-moll-Messe’, notes to recording, 1977; repr. in Dadelsen, U (Laaber, 1983), pp. 139–43. ——‘Bachs Kantate 77’, Almanach der Sommerakademie J. S. Bach, Stuttgart 1980, pp. 91–8; repr. ¨ ber Bach und anderes (Laaber, 1983), pp. 185–93. in Dadelsen, U ——‘Bach und die katholische Kirchenmusik’, in H. Dechant and W. Sieber (eds.), Gedenk¨ ber Bach und anderes schrift Hermann Beck (Laaber, 1982), pp. 81–5; repr. in Dadelsen, U (Laaber, 1983), pp. 200–3. ——‘Herkules an der Elbe: Mythologie und Allegorie in Bachs weltlichen Kantaten’, Berliner ¨ ber Bach und anderes (Laaber, 1983), pp. 212–21. Bachtagen 1982; repr. in Dadelsen, U ——‘Bemerkungen zu Bachs Cembalokonzerten’, in W. Hoffmann and A. Schneiderheinze (eds.), Bericht u¨ber die wissenschaftliche Konferenz zum V. Internationalen Bachfest der DDR, Leipzig 1985 (Leipzig, 1988), pp. 237–40. ——‘Wenn Bach Opern geschrieben ha¨tte’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld [conference report, Duisburg, 1986] (Kassel, 1988), pp. 177–83. ——‘Bach der Violinist: Anmerkungen zu den Soli fu¨r Violine und fu¨r Violoncello’, in K. Lehmann (ed.), J. S. Bach: Schaffenskonzeption, Werkidee, Textbezug [conference report, Leipzig 1989]; pub. in Beitra¨ge zur Bach-Forschung, 9/10 (Leipzig, 1991), pp. 70–6. ——‘Exkurs u¨ber die h-moll-Messe’, in Dadelsen, Beitra¨ge zur Chronologie, pp. 143–56.

406 b i b l i o g r a p h y David, Hans T., and Mendel, A., The New Bach Reader: A Life of J. S. Bach in Letters and Documents, rev. and enlarged by C. Wolff (New York and London, 1998); orig. pub. as The Bach Reader (1945; rev. edn 1966). ¨ berlegungen zu Bachs Suite f-moll BWV 823’, in M. Geck (ed.), Bachs Musik Dirksen, Pieter, ‘U fu¨r Tasteninstrumente [conference report, Dortmund, 2002] (Dortmund, 2003), pp. 119–31. ——‘Ein verschollenes Weimarer Kammermusikwerk J. S. Bachs? Zur Vorgeschichte der Sonate e-Moll fu¨r Orgel (BWV 528)’, BJ 89 (2003), pp. 7–36. ——(ed.) Johann Sebastian Bach: Sa¨mtliche Orgelwerke, vol. v. Sonatas, Trios, Concertos (Wiesbaden, 2010). Dreyfus, Laurence (ed.), Joh. Seb. Bach: Drei Sonaten fu¨r Viola da gamba und Cembalo BWV 1027–1029 (Frankfurt and Leipzig, 1985). ——‘Matters of Kind: Genre and Subgenre in Bach’s Well-Tempered Clavier, Book I’, in P. Brainard and R. Robinson (eds.), A Bach Tribute: Essays in Honour of William H. Scheide (Kassel, 1993), pp. 101–19. ——Bach and the Patterns of Invention (Cambridge, Mass. and London, 1996). Du¨rr, Alfred, ‘Zu den verschollenen Passionen Bachs’, BJ 38 (1949–50), pp. 81–99. ——‘Zur Chronologie der Leipziger Vokalwerke J. S. Bachs’, BJ 44 (1957), pp. 5–162; 2nd rev. edn pub. separately under same title (Kassel, 1976). ——‘Wieviele Kantatenjahrga¨nge hat Bach komponiert?’, Die Musikforschung, 14 (1961), pp. 192–5. ——‘Bachs Trauer-Ode und Markus-Passion’, Neue Zeitschrift fu¨r Musik, 74 (1963), pp. 460–6; repr. in Du¨rr, Im Mittelpunkt Bach (Kassel, 1988), pp. 115–25. ——(ed.), J. S. Bach: Messe in h-Moll, BWV 232: Faksimile der autographen Partitur (Kassel, 1965; 2nd edn 1984). ——‘Bach’s Chorale Cantatas’, in J. Riedel (ed.), Cantors at the Crossroads: Essays in Church Music in Honour of Walter E. Buszin (St Louis, Mo., 1967), pp. 111–20. ——J. S. Bach: Weihnachts-Oratorium BWV 248 (Munich, 1967). ——‘Neues u¨ber Bachs Pergolesi-Bearbeitung’, BJ 54 (1968), pp. 89–100. ——‘Zur Chronologie der Handschrift Johann Christoph Altnickols und Johann Friedrich Agricolas’, BJ 56 (1970), pp. 44–65. ——Die Kantaten von J. S. Bach (Kassel, 1971; 6th edn 1995); rev. Eng. trans. by R. D. P. Jones as The Cantatas of J. S. Bach (Oxford, 2005). ——‘Gibt es einen Spa¨tstil im Kantatenschaffen J. S. Bachs?’, in Bach-Tage, Vortra¨ge 1973 (Berlin, 1974), pp. 1–9; repr. in Du¨rr, Im Mittelpunkt Bach (Kassel, 1988), pp. 146–57. ——Supplement to NBA VII/2 (Kassel and Leipzig, 1975), pp. 243–84. ——‘Zur Entstehungsgeschichte des 5. Brandenburgischen Konzerts’, BJ 61 (1975), pp. 63–9. ——‘Heinrich Nicolaus Gerber als Schu¨ler Bachs’, BJ 64 (1978), pp. 7–18. ——‘Tastenumfang und Chronologie in Bachs Klavierwerken’, in T. Kohlhase and V. Scherliess (eds.), Festschrift Georg von Dadelsen (Neuhausen and Stuttgart, 1978), pp. 73–88; repr. in Du¨rr, Im Mittelpunkt Bach (Kassel, 1988), pp. 220–31. ——‘Zur Form der Pra¨ludien in Bachs Englischen Suiten’, in P. Ansehl et al. (eds.), Beitra¨ge zum Konzertschaffen J. S. Bachs, Bach-Studien 6 (Leipzig, 1981), pp. 101–8; repr. in Du¨rr, Im Mittelpunkt Bach (Kassel, 1988), pp. 232–8.

bibliography

407

——‘Zur Fru¨hgeschichte des Wohltemperierten Klaviers I von J. S. Bach’, in Nachrichten der Akademie der Wissenschaften in Go¨ttingen, i. Philologisch-historische Klasse (Go¨ttingen, 1984), pp. 1–67. ——‘The Historical Background of the Composition of J. S. Bach’s Clavier Suites’, Bach, 16/1 (1985), pp. 53–68. ——‘Melodievarianten in J. S. Bachs Kirchenliedbearbeitungen’, in A. Du¨rr and W. Killy (eds.), Das protestantische Kirchenlied im 16. und 17. Jahrhundert (Wiesbaden, 1986), pp. 149–63. ——‘Noch einmal: Wo blieb Bachs fu¨nfter Kantatenjahrgang?’, BJ 72 (1986), pp. 121–2. ——‘Gedanken zu den spa¨ten Kantaten Bachs’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld [conference report, Duisburg, 1986] (Kassel, 1988), pp. 58–63. ¨ berlieferung, Werkeinfu¨hrung (Munich ——Die Johannes-Passion von J. S. Bach: Entstehung, U and Kassel, 1988); Eng. trans. by Alfred Clayton as Johann Sebastian Bach’s St. John Passion: Genesis, Transmission and Meaning (Oxford, 2000). ——‘Zur Textbeziehung der Arienritornelle bei Bach’, in K. Lehmann (ed.), J. S. Bach: Schaffenskonzeption, Werkidee, Textbezug [conference report, Leipzig, 1989]; pub. in Beitra¨ge zur Bach-Forschung, 9/10 (Leipzig, 1991), pp. 34–43. ——‘Die Entstehungsgeschichte der Mattha¨us-Passion’, in U. Prinz (ed.), J. S. Bach: Mattha¨usPassion BWV 244; Vortra¨ge der Sommerakademie J. S. Bach 1985 (Kassel, 1990), pp. 76–93. ——‘Der Passionbericht des Johannes in Bachs Deutung—aus der Sicht des Musikwissenschaftlers’, in U. Prinz (ed.), J. S. Bach: Johannes-Passion BWV 245; Vortra¨ge des Meisterkurses 1986 und der Sommerakademie J. S. Bach 1990 (Kassel, 1993). ——Johann Sebastian Bach: Das Wohltemperierte Klavier (Kassel, 1998). ——and Kobayashi, Y. (eds.), Bach-Werke-Verzeichnis: Kleine Ausgabe (Wiesbaden, 1998). Eller, Rudolf and Heller, K., Kritischer Bericht, NBA VII/6: Konzerte fu¨r drei und vier Cembali (Kassel and Leipzig, 1976). ¨ berlegungen zu den Konzert- und Instrumentalsa¨tzen in J. S. Bachs KantaEmans, Reinmar, ‘U ten’, in R. Emans and M. Wendt (eds.), Beitra¨ge zur Geschichte des Konzerts: Festschrift Siegfried Kross (Bonn, 1990), pp. 41–55. ——‘Gibt es eine Entwicklung im Wort-Ton Verha¨ltnis bei Bach?’, in K. Lehmann (ed.), J. S. Bach: Schaffenskonzeption, Werkidee, Textbezug [conference report, Leipzig, 1989], pub. as Beitra¨ge zur Bach-Forschung 9/10 (Leipzig, 1991). ¨ berlegungen zur Genese der Kantate “Du Hirte Israel, ho¨re” (BWV 104)’, in Acht kleine ——‘U Pra¨ludien und Studien u¨ber BACH; Georg von Dadelsen zum 70. Geburtstag (Wiesbaden, 1992), pp. 44–50. Eppstein, Hans, ‘Zur Probematik von J. S. Bachs Sonate fu¨r Violine und Cembalo G-dur (BWV 1019)’, Archiv fu¨r Musikwissenschaft, 21 (1964), pp. 217ff. ——‘J. S. Bachs Triosonate G-dur (BWV 1039) und ihre Beziehungen zur Sonate fu¨r Gambe und Cembalo G-Dur (BWV 1027)’, Die Musikforschung, 18 (1965), pp. 126–37. ——Studien u¨ber J. S. Bachs Sonaten fu¨r ein Melodieinstrument und obligates Cembalo (Uppsala, 1966). ——‘Grundzu¨ge in J. S. Bachs Sonatenschaffen’, BJ 55 (1969), pp. 5–30. ¨ ber J. S. Bachs Flo¨tensonaten mit Generalbaß’, BJ 58 (1972), pp. 12–23. ——‘U ——‘Chronologieprobleme in Johann Sebastian Bachs Suiten fu¨r Soloinstrument’, BJ 62 (1976), pp. 35–57.

408 b i b l i o g r a p h y ¨ ber die Beziehungen zwischen Konzert und Sonate bei J. S. Bach’, in P. Ansehl Eppstein, Hans, ‘U et al. (eds.), Beitra¨ge zum Konzertschaffen J. S. Bachs, Bach-Studien 6 (Leipzig, 1981), pp. 87–93. ——‘Zur Problematik von J. S. Bachs Flo¨tensonaten’, BJ 67 (1981), pp. 77–90. ——‘Konzert und Sonate’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke [conference report, Dortmund, 1996] (Witten, 1997), pp. 145–9. Finke-Hecklinger, Doris, Tanzcharaktere in J. S. Bachs Vokalmusik, Tu¨binger Bach-Studien 6 (Trossingen, 1970). Finscher, Ludwig, ‘Zum Parodieproblem bei Bach’, in M. Geck (ed.), Bach-Interpretationen (Go¨ttingen, 1969), pp. 94–105, 217–18. Fischer, Wilfried, NBA VII/7: Verschollene Solokonzerte in Rekonstruktionen (Kassel and Leipzig, 1970), Kritischer Bericht (1971). Forkel, Johann Nicolaus, Ueber Johann Sebastian Bachs Leben, Kunst und Kunstwerke (Leipzig, 1802); repr. in BD VII, pp. 9–89. Franklin, Don O., ‘Reconstructing the Urpartitur for WTC II: A Study of the “London Autograph” (BL Add. MS 35021)’, in Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 240–78. Freeman, Daniel E., ‘J. S. Bach’s “Concerto” Arias: A Study in the Amalgamation of EighteenthCentury Genres’, Studi musicali, 27 (1998), pp. 123–62. Geck, Martin, ‘Ko¨then oder Leipzig? Zur Datierung der nur in Leipziger Quellen erhaltenen Orchesterwerke J. S. Bachs’, Die Musikforschung, 47 (1994), pp. 17–24. ——(ed.), Bachs Orchesterwerke [conference report, Dortmund 1996] (Witten, 1997). ——(ed.), Bach und die Stile [conference report, Dortmund 1998] (Dortmund, 1999). ——(ed.), Bachs 1. Leipziger Kantatenjahrgang [conference report, Dortmund 2000] (Dortmund, 2002). ——(ed.), Bachs Musik fu¨r Tasteninstrumente [conference report, Dortmund 2002] (Dortmund, 2003). Gerber, Christian, Historie der Kirchen-Ceremonien in Sachsen (Dresden and Leipzig, 1732). Germann, Sheridan, ‘The Mietkes, the Margrave and Bach’, in P. Williams (ed.), Bach, Handel, Scarlatti: Tercentenary Essays (Cambridge, 1985), pp. 119–48. Glo¨ckner, Andreas, ‘Bach and the Passion Music of His Contemporaries’, Musical Times, 116 (1975), pp. 613–16. ——‘J. S. Bachs Auffu¨hrungen zeitgeno¨ssischer Passionsmusiken’, BJ 63 (1977), pp. 75–119. ——‘Neuerkenntnisse zu J. S. Bachs Auffu¨hrungskalender zwischen 1729 und 1735’, BJ 67 (1981), pp. 43–75. ——‘Anmerkungen zu J. S. Bachs Ko¨thener Kantatenschaffen’, Co¨thener Bach-Hefte 4 (Ko¨then, 1986), pp. 89–95. ¨ berlegungen zu J. S. Bachs Kantatenschaffen nach 1730’, Beitra¨ge zur Bach-Forschung 6 ——‘U (1987), pp. 54–64; repr. in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld (Kassel, 1988), pp. 64–73. ——‘Bemerkungen zu den Leipziger Kantatenauffu¨hrungen vom 3. bis 6. Sonntag nach Trinitatis 1725’, BJ 78 (1992), pp. 73–6. ¨ berlieferung und Datierung der Kantate “Amore ——‘ “Das kleine italienische Ding”—zu U traditore” (BWV 203)’, BJ 82 (1996), pp. 133–7. ——‘Bachs Leipziger Collegium musicum und seine Vorgeschichte’, in C. Wolff (ed.), Die Welt der Bach Kantaten, vol. ii. J. S. Bachs weltliche Kantaten (Stuttgart, 1997), pp. 105–17.

bibliography

409

——‘Zur Vorgeschichte des “Bachischen” Collegium musicum’, in M. Geck (ed.), Bachs Orchesterwerke (Witten, 1997), pp. 293–303. ——‘Eine Michaeliskantate als Parodievorlage fu¨r den sechsten Teil des Bachschen WeihnachtsOratorium?’, BJ 86 (2000), pp. 317–26. ——‘Bachs Es-Dur Magniﬁcat BWV 243a—eine genuine Weihnachtsmusik?’, BJ 89 (2003), pp. 37–45. ——‘Ein weiterer Kantatenjahrgang Gottfried Heinrich Sto¨lzels in Bachs Auffu¨hrungsrepertoire?’, BJ 95 (2009), pp. 95–115. Go¨llner, T., ‘J. S. Bach and the Tradition of Keyboard Transcriptions’, in H. C. Robbins Landon and R. E. Chapman (eds.), Studies in Eighteenth-Century Music: A Tribute to Karl Geiringer on His 70th Birthday (London, 1970), pp. 253–60. Greer, Mary Dalton, ‘Embracing Faith: The Duet as Metaphor in Selected Sacred Cantatas by J. S. Bach’, Bach, 34/1 (2003), pp. 1–71. ——‘Bach’s Calov Bible and His Quest for the Title of Royal Court Composer’, in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass (Belfast, 2007), vol. i, pp. 150–76. Grubbs, John W., ‘Ein Passions-Pasticcio des 18. Jahrhunderts’, BJ 51 (1965), pp. 10–42. Heller, Karl, ‘Zur Stellung des Concerto C-Dur fu¨r zwei Cembali BWV 1061 in Bachs KonzertOeuvre’, in W. Hoffmann and A. Schneiderheinze (eds.), Bericht u¨ber die Wissenschaftliche Konferenz zum V. Internationalen Bachfest der DDR, Leipzig 1985 (Leipzig, 1988), pp. 241–52. ——‘Eine Leipziger Werkfassung und deren unbekannte Vorlage: Thesen zur Urform des Konzerts BWV 1063’, in U. Leisinger (ed.), Bach in Leipzig—Bach und Leipzig [conference report, Leipzig 2000] (Hildesheim, 2002), pp. 89–108. Heller, Karl and Schulze, Hans-Joachim (eds.), NBA VII/5: Konzerte fu¨r zwei Cembali (Kassel and Leipzig, 1985), Kritischer Bericht (1990). Helms, Marianne, ‘Zur Chronologie der Handschrift des Anonymus 5’, Kritischer Bericht, NBA V/7 (Kassel and Leipzig, 1981), pp. 183–95. Hermelink, Siegfried, ‘Das Pra¨ludium in Bachs Klaviermusik’, Jahrbuch des Staatlichen Instituts fu¨r Musikforschung Preußischer Kulturbesitz, 1976 (Berlin, 1977), pp. 7–80. Herz, Gerhard, ‘Toward a new Image of Bach’, Bach, 1/4 (1970), pp. 9–27, and 2/1 (1971), pp. 7–28; repr. in Herz, Essays on J. S. Bach (Ann Arbor, 1985), pp. 149–84. ——‘Der lombardische Rhythmus im “Domine Deus” der h-Moll-Messe J. S. Bachs’, BJ 60 (1974), pp. 90–7; Eng. trans. in Bach, 8/1 (1977), pp. 3–11; repr. in Herz, Essays on J. S. Bach (Ann Arbor, 1985), pp. 221–9. ——‘Thoughts on the First Movement of J. S. Bach’s Cantata No. 77’, programme notes at annual meeting of American Musicological Society, 1974; repr. in Herz, Essays on J. S. Bach (Ann Arbor, 1985), pp. 205–17. ——‘Der lombardische Rhythmus in Bachs Vokalschaffen’, BJ 64 (1978), pp. 148–80; rev. Eng. version in Herz, Essays on J. S. Bach (Ann Arbor, 1985), pp. 233–68. Hobohm, Wolf, ‘Neue “Texte zur Leipziger Kirchen-Music” ’, BJ 59 (1973), pp. 5–32. ¨ ber ein unbeachtetes Sammelwerk Hofmann, Klaus, ‘ “Fu¨nf Pra¨ludien und fu¨nf Fugen”: U J. S. Bachs’, in W. Hoffmann and A. Schneiderheinze (eds.), Bericht u¨ber die Wissenschaftliche Konferenz zum V. Internationalen Bachfest der DDR, Leipzig 1985 (Leipzig, 1988), pp. 227–35. ¨ ber Themenbildung und thematische Arbeit in einigen zweiteiligen Pra¨ludien des ——‘U Wohltemperierten Klaviers II’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld (Kassel, 1988), pp. 48–57.

410 b i b l i o g r a p h y ¨ berlegungen zu Hofmann, Klaus, ‘J. S. Bachs Kantate “Jauchzet Gott in allen Landen” BWV 51: U Entstehung und urspru¨nglicher Bestimmung’, BJ 75 (1989), pp. 43–54. ¨ berlegungen zu der Kantate “Non sa che sia dolore” BWV 209’, BJ 76 ——‘Alte und neue U (1990), pp. 7–25. ——(ed.), NBA VIII/2.1–2, Die Kunst der Fuge, BWV 1080 (Kassel and Leipzig, 1995); Kritischer Bericht (1996). ——‘Zur Fassungsgeschichte des zweiten Brandenburgischen Konzerts’, in M. Geck (ed.), Bachs Orchesterwerke [conference report, Dortmund 1996] (Witten, 1997), pp. 185–92. ——‘Auf der Suche nach der verlorenen Urfassung: Diskurs zur Vorgeschichte der Sonate in h-Moll fu¨r Querflo¨te und obligates Cembalo von J. S. Bach’, BJ 84 (1998), pp. 31–59. ——‘Perfidia-Techniken und -Figuren bei Bach’, in R. Steiger (ed.), Die Quellen J. S. Bachs: Bachs Musik im Gottesdienst [conference report, Stuttgart 1995] (Heidelberg, 1998), pp. 201–99. ——‘Ein verschollenes Kammermusikwerk J. S. Bachs: Zur Fassungsgeschichte der Orgelsonate Es-Dur (BWV 525) und der Sonate A-Dur fu¨r Flo¨te und Cembalo (BWV 1032)’, BJ 85 (1999), pp. 67–79. ¨ berlegungen zur Fru¨h——‘Die Klangfla¨chenpra¨ludien des “Wohltemperierten Klaviers” I: U geschichte der Sammlung’, in M. Staehelin (ed.), Die Zeit, die Tag und Jahre macht: Zur Chronologie des Schaffens von J. S. Bach [conference report, Go¨ttingen 1998] (Go¨ttingen, 2001), pp. 157–68. ——Johann Sebastian Bach: Die Motetten (Kassel, 2003). Horn, Victoria, ‘French Influence in Bach’s Organ Works’, in G. Stauffer and E. May (eds.), J. S. Bach as Organist (London, 1986), pp. 256–73. Horn, Wolfgang, Die Dresdner Hofkirchenmusik 1720–1745: Studien zu ihren Voraussetzungen und ihrem Repertoire (Kassel, 1987). Hutchings, Arthur, The Baroque Concerto (London, 1959; 3rd edn 1973). Irwin, Joyce L., Neither Voice Nor Heart Alone: German Lutheran Theology of Music in the Age of the Baroque (New York, 1993). Jacob, Andreas, Studien zu Kompositionsart und Kompositionsbegriff in Bachs Klavieru¨bungen (Stuttgart, 1997). Jones, Richard D. P., NBA V/1: Erster Teil der Klavieru¨bung (Kassel and Leipzig, 1976); Kritischer Bericht (Kassel and Leipzig, 1978; Suppl. 1997). ——(ed.), J. S. Bach: The Well-Tempered Clavier, Parts I and II, with commentaries by Donald Francis Tovey (London, 1994). ——‘The Keyboard Works: Bach as Teacher and Virtuoso’, in J. Butt (ed.), The Cambridge Companion to Bach (Cambridge, 1997), pp. 136–53. ——‘The History and Text of Bach’s Clavieru¨bung I’, dissertation, University of Oxford, 1988. Kan, Rebecca, ‘Vivaldi, Bach and Their Concerto Slow Movements’, in A. Leahy and Y. Tomita (eds.), Bach Studies from Dublin, Irish Musical Studies 8 [conference report, Dublin 2000] (Dublin, 2004), pp. 65–91. Kast, Paul, Die Bach-Handschriften der Berliner Staatsbibliothek, Tu¨binger Bach-Studien 2/3 (Trossingen, 1958). Kauffmann, Georg Friedrich, Harmonische Seelen Lust (Leipzig, 1733–6). Kevorkian, Tanya, Baroque Piety: Religious Practices, Society, and Music in Leipzig, 1650–1750 (Aldershot, 2007).

bibliography

411

Kilian, Dietrich, ‘Dreisa¨tzige Fassungen Bachscher Orgelwerke’, in M. Geck (ed.), Bach-Interpretationen (Go¨ttingen, 1969), pp. 12–21. ——Kritischer Bericht, NBA IV/5–6 (Kassel and Leipzig, 1978–9). ——Kritischer Bericht, NBA IV/7 (Kassel and Leipzig, 1988). Klein, Hans-Gu¨nter, Der Einﬂuß der Vivaldischen Konzertform im Instrumentalwerk J. S. Bachs (Strasbourg, 1970). Kobayashi, Yoshitake, ‘Die Universalita¨t in Bachs h-Moll-Messe’, Musik und Kirche, 57 (1987), pp. 9–24; Eng. trans. in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass [conference report] (Belfast, 2007), vol. ii, pp. 387–403. ——‘Zur Chronologie der Spa¨twerke J. S. Bachs: Kompositions- und Auffu¨hrungsta¨tigkeit von 1736 bis 1750’, BJ 74 (1988), pp. 7–72. ——Die Notenschrift J. S. Bachs: Dokumentation ihrer Entwicklung, NBA IX/2 (Kassel and Leipzig, 1989). ——‘Noch einmal zu J. S. Bachs “Solo pour la ﬂuˆte traversie`re” BWV 1013’, Tibia, 16 (1991), pp. 379–82. ——and Beißwenger, Kirsten, Die Kopisten Johann Sebastian Bachs: Katalog und Dokumentation, NBA IX/3 (Kassel, 2007). Koch, Ernst, ‘Tro¨stendes Echo: Zur theologischen Deutung der Echo-Arie im IV. Teil des Weihnachts-Oratoriums von J. S. Bach’, BJ 75 (1989), pp. 203–11. Krausse, Helmut K., ‘Eine neue Quelle zu drei Kantatentexten J. S. Bachs’, BJ 67 (1981), pp. 7–22. ¨ ber Bachs geistliche Dialoge’, in Krummacher, Friedhelm, ‘Gespra¨ch und Struktur: U K. Lehmann (ed.), J. S. Bach: Schaffenskonzeption, Werkidee, Textbezug [conference report, Leipzig 1989], pub. as Beitra¨ge zur Bach-Forschung, 9/10 (Leipzig, 1991), pp. 45–59. ——Bachs Zyklus der Choralkantaten: Aufgaben und Lo¨sungen (Go¨ttingen, 1995). ——‘Chronologie und Entwicklung: Zu Bachs Jahrgang der Choralkantaten’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke [conference report, Dortmund 1996] (Witten, 1997), pp. 151–8. ——‘Chronologie und Interpretation: Chorsa¨tze in Bachs erstem Kantatenjahrgang’, in M. Staehelin (ed.), Die Zeit, die Tag und Jahre macht: Zur Chronologie des Schaffens von J. S. Bach [conference report, Go¨ttingen 1998] (Go¨ttingen, 2001), pp. 61–88. Ku¨ster, Konrad, ‘Meininger Kantatentexte um Johann Ludwig Bach’, BJ 73 (1987), pp. 159–64. ¨ berlieferung der Kantaten Johann Ludwig Bachs’, BJ 75 ——‘Die Frankfurter und Leipziger U (1989), pp. 65–106. ——‘Konzertvorlage oder Originalkomposition? Zu den obligaten Orgelanteilen in Bachs Kantaten aus dem Jahr 1726’, in U. Leisinger (ed.), Bach in Leipzig—Bach und Leipzig [conference report, Leipzig 2000] (Hildesheim, 2002), pp. 45–58. Landmann, Ortrun, ‘The Dresden Hofkapelle during the Lifetime of J. S. Bach’, Early Music, 17 (1989), pp. 17–30. Leaver, Robin A., ‘Bach and the German Agnus Dei’, in P. Brainard and R. Robinson (eds.), A Bach Tribute: Essays in Honour of William H. Scheide (Kassel and Chapel Hill, NC, 1993), pp. 163–71. ——‘Music and Lutheranism’, in J. Butt (ed.), The Cambridge Companion to Bach (Cambridge, 1997), pp. 35–45. ——‘How “Catholic” is Bach’s “Lutheran” Mass?’, in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass [conference report] (Belfast, 2007), vol. i, pp. 177–206.

412

bi b l io g rap hy

Ledbetter, David, Bach’s Well-Tempered Clavier (New Haven and London, 2002). ——‘Les Gouˆts reúnis and the Music of J. S. Bach’, Basler Jahrbuch fu¨r historische Musikpraxis, 28 (2004), pp. 63–80. ——‘A Question of Genre: J. S. Bach and the “Mixed Style” ’, in T. Donahue (ed.), Music and Its Questions: Essays in Honour of Peter Williams (Richmond, Va., 2007), pp. 205–24. ——Unaccompanied Bach (New Haven and London, 2009). Leisinger, Ulrich, ‘Forms and Functions of the Choral Movements in J. S. Bach’s St. Matthew Passion’, in D. R. Melamed (ed.), Bach Studies 2 (Cambridge, 1995), pp. 70–84. ——‘Musikalische Textauslegung: Kontinuita¨t und Wandel’, in C. Wolff (ed.), Die Welt der Bach-Kantaten, vol. iii. J. S. Bachs Leipziger Kirchenkantaten (Stuttgart, 1999), pp. 223–31. ——(ed.), J. S. Bach: Chromatische Fantasie und Fuge BWV 903 (Vienna, 1999). ——‘Die zweite Fassung der Johannes-Passion von 1725. Nur ein Notbehelf?’, in Leisinger (ed.), Bach in Leipzig—Bach und Leipzig [conference report, Leipzig 2000] (Hildesheim, 2002), pp. 29–44. Lester, Joel, Between Modes and Keys: German Theory, 1592–1802 (Stuyvesant, NY, 1989). ——Compositional Theory in the Eighteenth Century (Cambridge, Mass., 1992). Mann, Alfred, ‘Bach’s Parody Technique and its Frontiers’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 115–24. Marissen, Michael, ‘A Trio in C major for Recorder, Violin and Continuo by J. S. Bach?’, Early Music, 13 (1985), pp. 384–90. ——‘A Critical Reappraisal of J. S. Bach’s A-major Flute Sonata’, Journal of Musicology, 6 (1988), pp. 367–86. ——‘On Linking Bach’s F-major Sinfonia and His Hunt Cantata, BWV 208’, Bach, 23/2 (1992), pp. 31–46. ——‘More Source-Critical Research on J. S. Bach’s Musical Offering’, Bach, 25/1 (1994), pp. 11–27. ——The Social and Religious Designs of J. S. Bach’s Brandenburg Concertos (Princeton, 1995). ——Bach’s Oratorios: The Parallel German-English Texts with Annotations (Oxford, 2008). Ma¨rker, Michael, ‘Die Tradition des Jesus-Seele-Dialoges und ihr Einfluß auf das Werk Bachs’, in K. Lehmann (ed.), J. S. Bach: Schaffenskonzeption, Werkidee, Textbezug [conference report, Leipzig 1989], pub. as Beitra¨ge zur Bach-Forschung, 9/10 (Leipzig, 1991), pp. 235–41. Marpurg, Friedrich Wilhelm, Abhandlung von der Fuge (Berlin, 1753–4). Marshall, Robert L., Introduction to J. S. Bach: Fantasia per il Cembalo, BWV 906 (facs. edn, Leipzig, 1976); repr. as ‘The Autograph Fair Copies of the Fantasia per il Cembalo, BWV 906’, in Marshall, The Music of J. S. Bach: The Sources, the Style, the Significance (New York, 1989), pp. 193–200. ——‘Bach the Progressive: Observations on His Later Works’, Musical Quarterly, 62 (1976), pp. 313–57; repr. in Marshall, The Music of J. S. Bach (New York, 1989), pp. 23–58. ——‘J. S. Bach’s Compositions for Solo Flute: A Reconsideration of Their Authenticity and Chronology’, Journal of the American Musicological Society, 32 (1979), pp. 463–98; repr. in Marshall, The Music of J. S. Bach (New York, 1989), pp. 201–25. ——‘Beobachtungen am Autograph der h-moll-Messe: Zum Kompositionsprozess J. S. Bachs’, Musik und Kirche, 50 (1980), pp. 230–9; Eng. version pub. as ‘The Mass in B minor: The Autograph Scores and the Compositional Process’, in Marshall, The Music of J. S. Bach (New York, 1989), pp. 175–89.

bibliography

413

——‘On the Origin of Bach’s Magniﬁcat: A Lutheran Composer’s Challenge’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 3–17. Marx, Hans Joachim, ‘Bach und der Theatralische Stil’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld [conference report, Duisburg 1986] (Kassel, 1988), pp. 148–54. Mattheson, Johann, Der vollkommene Capellmeister (Hamburg, 1739); Eng. trans. by E. C. Harriss (Ann Arbor, 1981). Maul, Michael, ‘ “Die große catholische Messe”: Bach, Graf Questenberg und die “Musicalische Congregation” in Wien’, BJ 95 (2009), pp. 153–75. ——‘Johann Adolph Scheibes Bach-Kritik: Hintergru¨nde und Schaupla¨tze einer musikalischen Kontroverse’, BJ 96 (2010), pp. 153–98. Max, Hermann, ‘Verwandtes im Werk Bachs, seiner Schu¨ler und So¨hne’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld [conference report, Duisburg 1986] (Kassel, 1988), pp. 117–47. Melamed, Daniel R., J. S. Bach and the German Motet (Cambridge, 1995). ——‘The Double Chorus in J. S. Bach’s St. Matthew Passion, BWV 244’, Journal of the American Musicological Society, 57 (2004), pp. 3–50. ——Hearing Bach’s Passions (Oxford, 2005). Mendel, Arthur, ‘Traces of the Pre-history of Bach’s St. John and St. Matthew Passions’, in Festschrift Otto Erich Deutsch zum 80. Geburtstag (Kassel, 1963), pp. 31–48. ——‘More on the Weimar Origin of Bach’s “O Mensch, bewein” (BWV 244/35)’, Journal of the American Musicological Society, 17 (1964), pp. 203–6. ——Kritischer Bericht, NBA II/4: Johannes-Passion (Kassel and Leipzig, 1974). Mies, Paul, Das instrumentale Rezitativ: Von seiner Geschichte und seinen Formen (Bonn, 1968). Milka, Anatoly P., ‘Zur Datierung der h-Moll-Messe und der Kunst der Fuge’, BJ 96 (2010), pp. 53–68. Neumann, Werner, ‘Das “Bachische Collegium Musicum” ’, BJ 47 (1960), pp. 5–27. ¨ ber Ausmaß und Wesen des Bachschen Parodieverfahrens’, BJ 51 (1965), pp. 63–85. ——‘U ——‘J. S. Bachs “Rittergutskantaten” BWV 30a und 212’, BJ 58 (1972), pp. 76–90. ——Kritischer Bericht, NBA I/39 (Kassel and Leipzig, 1977). ——and Schulze, Hans-Joachim (eds.), BD I: Schriftstu¨cke von der Hand Johann Sebastian Bachs (Leipzig, 1963). ——and Schulze, Hans-Joachim (eds.), BD II: Fremdschriftliche und gedruckte Dokumente zur Lebensgeschichte Johann Sebastian Bachs 1685–1750 (Leipzig, 1969). Newman, William S., ‘Concerning the Accompanied Clavier Sonata’, Musical Quarterly, 33 (1947), pp. 327–49. Nott, Kenneth, ‘ “Tilge, Ho¨chster, meine Su¨nden”: Observations on Bach and the style galant’, Bach, 23/1 (1992), pp. 3–30. O’Donnell, John, ‘Mattheson, Bach, Krebs and the Fantasia & Fugue in C minor, BWV 537’, Organ Yearbook, 20 (1989), pp. 88–95. Osthoff, Wolfgang, ‘Das Credo der h-moll-Messe: Italienische Vorbilder und Anregungen’, in W. Osthoff and R. Wiesend (eds.), Bach und die italienische Musik (Venice, 1987), pp. 109–40. Paczkowski, Szymon, ‘On the Role and Meaning of the Polonaise in the Mass in B minor by J. S. Bach, as Exempliﬁed by the Aria “Quoniam tu solus sanctus” ’, in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass [conference report] (Belfast, 2007), vol. i, pp. 33–51.

414

b i bl i og r aphy

Pankratz, Herbert R., ‘J. S. Bach and His Leipzig Collegium musicum’, Musical Quarterly, 69 (1983), pp. 323–53. Petzoldt, Martin, ‘Vertonte Dogmatik: Zu theologischen Aspekten der h-Moll-Messe Bachs’, in Bach fu¨r Kenner und Liebhaber: Festschrift zum 70. Geburtstag von Diethard Hellmann (Stuttgart, 1998), pp. 67–75; Eng. trans. in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass (Belfast, 2007), pp. 425–34. Pfau, Marc-Roderich, ‘Ein unbekanntes Leipziger Kantatentextheft aus dem Jahr 1735—Neues zum Thema Bach und Sto¨lzel’, BJ 94 (2008), pp. 99–122. Platen, Emil, ‘Eine Pergolesi-Bearbeitung Bachs’, BJ 48 (1961), pp. 35–51. ——Die Mattha¨us-Passion von J. S. Bach: Entstehung, Werkbeschreibung, Rezeption (Munich and Kassel, 1991). ——and Helms, Marianne, Kritischer Bericht, NBA II/2: Lutherische Messen etc. (Kassel and Leipzig, 1982). Plath, Wolfgang, Kritischer Bericht, NBAV/5: Clavierbu¨chlein fu¨r W. F. Bach (Kassel and Leipzig, 1963). Powell, Ardal and Lasocki, David, ‘Bach and the Flute: The Players, the Instruments, the Music’, Early Music, 23 (1995), pp. 9–29. Prinz, Ulrich, ‘Zur Entstehungsgeschichte der Johannes-Passion und ihrer Fassungen’, in Prinz (ed.), J. S. Bach: Johannes-Passion BWV 245 [conference report, Internationale Bachakademie Stuttgart] (Kassel, 1993), pp. 100–11. Quantz, Johann Joachim, Versuch einer Anweisung die Flo¨te traversiere zu spielen (Berlin, 1752); Eng. trans. by E. R. Reilly (London, 1966; 2nd edn 1985). Rampe, Siegbert (ed.), Bach: Das Wohltemperierte Klavier I: Tradition, Entstehung, Funktion, Analyse (Munich and Salzburg, 2002). Rampe, Siegbert, and Sackmann, Dominik, Bachs Orchestermusik: Entstehung, Klangwelt, Interpretation (Kassel, 2000). Rasch, Rudolf, ‘Does Well-Tempered Mean Equal-Tempered’, in P. Williams (ed.), Bach, Handel, Scarlatti: Tercentenary Essays (Cambridge, 1985), pp. 293–308. Rathey, Markus, ‘Weimar, Gotha oder Leipzig: Zur Chronologie der Arie “Himmel, reiße” in der zweiten Fassung der Johannes-Passion (BWV 245/11+)’, BJ 91 (2005), pp. 291–300. ¨ berlegungen zur Chronologie der drei Fassungen der Sonate G-Dur fu¨r Rempp, Frieder, ‘U Violine und konzertierenden Cembalo (BWV 1019)’, in M. Staehelin (ed.), Die Zeit, die Tag und Jahre macht: Zur Chronologie des Schaffens von J. S. Bach [conference report, Go¨ttingen 1998] (Go¨ttingen, 2001), pp. 169–83. Riedel, Friedrich Wilhelm, Quellenkundliche Beitra¨ge zur Geschichte der Musik fu¨r Tasteninstrumente in der zweiten Ha¨lfte des 17. Jahrhunderts (Kassel, 1960). ——‘J. S. Bachs Kunst der Fuge und die Fugenbu¨cher der italienischen und o¨sterreichischen Organisten des 17. und 18. Jahrhunderts’, in F. Heidlberger, W. Osthoff, and R. Wiesend (eds.), Von Isaac bis Bach: Studien zur a¨lteren deutschen Musikgeschichte: Festschrift Martin Just (Kassel, 1991), pp. 327–33. Rifkin, Joshua, ‘The Chronology of Bach’s St. Matthew Passion’, Musical Quarterly, 61 (1975), pp. 360–87. ——‘Ein langsamer Konzertsatz J. S. Bachs’, BJ 64 (1978), pp. 140–7. ——‘ “O Mensch, bewein dein Su¨nde groß” and the History of Bach’s Passions’ (unpublished paper, c.1990).

bibliography

415

——‘Verlorene Quellen, verlorene Werke: Miszellen zu Bachs Instrumentalkomposition’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke [conference report, Dortmund 1996] (Witten, 1997), pp. 59–75. ¨ berlieferung: Bemerkungen zur Ouvertu¨re BWV 1068’, BJ 83 ——‘Besetzung, Entstehung, U (1997), pp. 169–76. ——‘Klangpracht und Stilauffassung: Zu den Trompeten der Ouvertu¨re BWV 1069’, in M. Geck and K. Hofmann (eds.), Bach und die Stile [conference report, Dortmund 1998] (Dortmund, 1999), pp. 255–89. ——‘The “B-minor Flute Suite” Deconstructed: New Light on Bach’s Ouverture BWV 1067’, in G. G. Butler (ed.), Bach Perspectives 6 (Urbana and Chicago, 2007), pp. 1–98. Rubin, Norman, ‘ “Fugue” as a delimiting Concept in Bach’s Choruses’, in R. L. Marshall (ed.), Studies in Renaissance and Baroque Music in Honour of Arthur Mendel (Kassel, 1974), pp. 195–208. Sackmann, Dominik, ‘Bachs langsame Konzertsa¨tze unter dem Einﬂuß von Arcangelo Corelli: vom Ostinatoprinzip zum Primat der expressive Solostimme’, in M. Geck and K. Hofmann (eds.), Bach und die Stile [conference report, Dortmund 1998] (Dortmund 1999), pp. 303–26. ——Bach und Corelli: Studien zu Bachs Rezeption von Corellis Violinsonaten Op. 5 (Munich and Salzburg, 2000). ——‘ “A` la recherche du Pre´lude perdu”: Die Pra¨ludien im Wohltemperierten Klavier I und ihre Stellung in der Geschichte der Gattung’, in S. Rampe (ed.), Bach: Das Wohltemperierte Klavier I: Tradition, Entstehung, Funktion, Analyse (Munich and Salzburg, 2002), pp. 161–80. Scheibel, Gottfried Ephraim, Zufa¨llige Gedancken von der Kirchen-Music, wie sie heutiges Tages beschaffen ist (Frankfurt and Leipzig, 1721); Eng. trans. as ‘Random Thoughts about Church Music in Our Day’, introd. and trans. by Joyce Irwin, in Carol K. Baron (ed.), Bach’s Changing World: Voices in the Community (Rochester, NY and Woodbridge, 2006), pp. 227–49. Scheide, William H., ‘J. S. Bachs Sammlung von Kantaten seines Vetters Johann Ludwig Bach’, BJ 46 (1959), pp. 52–94; 48 (1961), pp. 5–24; and 49 (1962), pp. 5–32. ——‘Ist Mizlers Bericht u¨ber Bachs Kantaten korrekt?’, Die Musikforschung, 14 (1961), pp. 60–3. ——‘Some Miscellaneous Chorale Forms in J. S. Bach’s Vocal Works’, in R. L. Marshall (ed.), Studies in Renaissance and Baroque Music in Honour of Arthur Mendel (Kassel, 1974), pp. 209–27. ——‘Bach und der Picander-Jahrgang: eine Erwiderung’, BJ 66 (1980), pp. 47–51. ——‘Eindeutigkeit und Mehrdeutigkeit in Picanders Kantatenjahrgangs-Vorbemerkung und im Werkverzeichnis des Nekrologs auf J. S. Bach’, BJ 69 (1983), pp. 109–13. Schlage, Thomas, ‘Form-Modelle in den Chorsa¨tzen des 1. Leipziger Kantatenjahrgangs’, in M. Geck (ed.), Bachs 1. Leipziger Kantatenjahrgang [conference report, Dortmund 2000] (Dortmund, 2002), pp. 177–94. Schulenberg, David, ‘Composition as Variation: Inquiries into the Compositional Procedures of the Bach Circle of Composers’, Current Musicology, 33 (1982), pp. 57–87. ——‘Expression and Authenticity in the Harpsichord Music of J. S. Bach’, Journal of Musicology, 8 (1990), pp. 449–76. ——‘Musical Expression and Musical Rhetoric in the Keyboard Works of J. S. Bach’, in S. L. Benstock (ed.), Johann Sebastian Bach: A Tercentenary Celebration (Westport, Conn. and London, 1992), pp. 95–109. ——The Keyboard Music of J. S. Bach (London, 1993).

416

bi b l io g rap hy

Schulenberg, David, ‘Composition and Improvisation in the School of J. S. Bach’, in R. Stinson (ed.), Bach Perspectives [I] (Lincoln, Nebr., 1995), pp. 1–42. Schulze, Hans-Joachim (ed.), BD III: Dokumente zum Nachwirken Johann Sebastian Bachs 1750–1800 (Leipzig, 1972). ——‘Melodiezitate und Mehrtextigkeit in der Bauernkantate und in den Goldbergvariationen’, BJ 62 (1976), pp. 58–72. ——(ed.) J. S. Bach: Konzert c-Moll fu¨r zwei Cembali und Streichorchester BWV 1062; Sonate A-Dur fu¨r Flo¨te und Cembalo BWV 1032, facsimile of autograph score (Leipzig, 1979). ¨ bersicht’, in Zu Fragen der Impro——‘Cembaloimprovisation bei J. S. Bach: Versuch einer U visation in der Instrumentalmusik der ersten Ha¨lfte des 18. Jahrhunderts [conference report, Blankenburg 1979] (Blankenburg, 1980), pp. 50–7. ¨ berlieferung und Chronologie’, in P. Ahnsehl et al. (eds.), ——‘J. S. Bachs Konzerte: Fragen der U Beitra¨ge zum Konzertschaffen J. S. Bachs, Bach-Studien 6 (Leipzig, 1981), pp. 9–26. ——‘Der franzo¨sische Einfluß im Instrumentalwerk J. S. Bachs’, in Studien zur Auffu¨hrungspraxis und Interpretation von Instrumentalmusik im 18. Jahrhundert (Blankenburg, 1981), pp. 57–63; Eng. trans. as ‘The French Influence in Bach’s Instrumental Music’, Early Music, 13 (1985), pp. 180–4. ——(ed.), J. S. Bach, Missa h-Moll BWV 232I: Faksimile nach dem Originalstimmensatz der Sa¨chsischen Landesbibliothek Dresden (Neuhausen and Stuttgart, 1983). ¨ berlieferung im 18. Jahrhundert (Leipzig and Dresden, 1984). ——Studien zur Bach-U ——‘The B-minor Mass: Perpetual Touchstone for Bach Research’, in P. Williams (ed.), Bach, Handel, Scarlatti: Tercentenary Essays (Cambridge, 1985), pp. 311–20. ——‘ “Wer der alte Bach gewesen, weiß ich wohl”: Anmerkungen zum Thema Kunstwerk und Biographie’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld [conference report, Duisburg, 1986] (Kassel, 1988), pp. 23–31. ——‘The Parody Process in Bach’s Music: An Old Problem Reconsidered’, Bach, 20 (1989), pp. 7–21; paper read to the American Bach Society, 1988. ——‘Bach’s Secular Cantatas: A New Look at the Sources’, Bach, 21/1 (1990), pp. 26–41. ——‘J. S. Bachs Passionsvertonungen’, in U. Prinz (ed.), J. S. Bach: Matthaüs-Passion BWV 244 [conference report 1985, Internationale Bachakademie Stuttgart] (Kassel, 1990), pp. 24–49. ——‘Florilegium—Pasticcio—Parodie—Verma¨chtnis: Beobachtungen an ausgewa¨hlten Vokalwerken J. S. Bachs’, in K. Lehmann (ed.), Schaffenskonzeption, Werkidee, Textbezug [conference report, Leipzig 1989], pub. in Beitra¨ge zur Bach-Forschung, 9/10 (Leipzig, 1991), pp. 199–204. ——‘Bemerkungen zum zeit- und gattungsgeschichtlichen Kontext von J. S. Bachs Passionen’, in R. Szeskus (ed.), J. S. Bachs Historischer Ort, Bach-Studien 10 (Wiesbaden and Leipzig, 1991), pp. 202–15. ——‘J. S. Bachs Johannes-Passion: die Spa¨tfassung von 1749’, in U. Prinz (ed.), J. S. Bach: Johannes-Passion BWV 245 [conference report 1986, 1990, Internationale Bachakademie Stuttgart] (Kassel, 1993), pp. 112–27. ——‘ “O Jesu Christ, meins Lebens Licht”: On the Transmission of a Bach Source and the Riddle of Its Origin’, in P. Brainard and R. Robinson (eds.), A Bach Tribute: Essays in Honour of William H. Scheide (Kassel and Chapel Hill, NC, 1993), pp. 209–20. ——et al. (eds.), Passionmusiken im Umfeld J. S. Bachs [conference report, Leipzig 1994], pub. as Leipziger Beitra¨ge zur Bach-Forschung I (Hildesheim, 1995).

b ib l i o g r a p h y

417

——‘Bachs Parodieverfahren’, in C. Wolff (ed.), Die Welt der Bach-Kantaten, vol. ii. J. S. Bachs Weltliche Kantaten (Stuttgart, 1997), pp. 167–87. ——‘Kantatenformen und Kantatentypen’, in C. Wolff (ed.), Die Welt der Bach-Kantaten, vol. ii. J. S. Bachs Weltliche Kantaten (Stuttgart, 1997), pp. 157–65. ——‘Texte und Textdichter’, in C. Wolff (ed.), Die Welt der Bach-Kantaten, vol. iii. J. S. Bachs Leipziger Kirchenkantaten (Stuttgart, 1999), pp. 109–25. ——‘J. S. Bachs dritter Leipziger Kantatenjahrgang und die Meininger “Sonntags- und FestAndachten” von 1719’, BJ 88 (2002), pp. 193–9. ——(ed.), BD V: Dokumente zu Leben, Werk und Nachwirken Johann Sebastian Bachs 1685–1800 (Kassel, 2007). ——‘Johann Sebastian Bachs Himmelfahrts-Oratorium und Picanders Geburtstagkantate fu¨r “Herrn J. W. C. D.” ’, BJ 95 (2009), pp. 191–9. Shabalina, Tatiana, ‘ “Texte zur Music” in Sankt Petersburg: Neue Quellen zur Leipziger Musikgeschichte sowie zur Kompositions- und Auffu¨hrungsta¨tigkeit J. S. Bachs’, BJ 94 (2008), pp. 33–98. ——‘ “Texte zur Music” in Sankt Petersburg—Weitere Funde’, BJ 95 (2009), pp. 11–48. ——‘Neue Erkenntnisse zur Entstehungsgeschichte der Kantaten BWV 34 und 34a’, BJ 96 (2010), pp. 95–109. Sheldon, David A., ‘The galant Style Revisited and Re-evaluated’, Acta musicologica, 47 (1975), pp. 240–69. Siegele, Ulrich, Kompositionsweise und Bearbeitungstechnik in der Instrumentalmusik J. S. Bachs, dissertation, Tu¨bingen 1956 (Neuhausen and Stuttgart, 1975). ——‘Bachs vermischter Geschmack’, in M. Geck and K. Hofmann (eds.), Bach und die Stile [conference report, Dortmund 1998] (Dortmund, 1999), pp. 9–17. Smallman, Basil, The Background of Passion Music: J. S. Bach and His Predecessors (London, 1957; 2nd edn 1970). Smend, Friedrich, Bach in Ko¨then (Berlin, 1951); Eng. trans. by J. Page, ed. and rev. by S. Daw (St Louis, Mo., 1985). ——‘Bachs Markus-Passion’, BJ 37 (1940–8), pp. 1–35; repr. in F. Smend, Bach-Studien: Gesammelte Reden und Aufsa¨tze, ed. C. Wolff (Kassel, 1969), pp. 110–36. Spitta, Philipp, Johann Sebastian Bach, 2 vols. (Leipzig, 1873 and 1880); Eng. trans. by C. Bell and J. A. Fuller-Maitland, 3 vols. (London, 1884–5; repr. 1952). Stauffer, George, The Organ Preludes of J. S. Bach (Ann Arbor, 1980). ——‘Bach’s Pastorale in F: A Closer Look at a Maligned Work’, Organ Yearbook, 14 (1983), pp. 44–60. ——‘Fugue Types in Bach’s Free Organ Works’, in G. Stauffer and E. May (eds.), J. S. Bach as Organist (London, 1986), pp. 133–56. ——‘ “This Fantasia . . . never had its like”: On the Enigma and Chronology of Bach’s Chromatic Fantasia and Fugue in D minor, BWV 903’, in D. O. Franklin (ed.), Bach Studies [I] (Cambridge, 1989), pp. 160–82. ——‘Boyvin, Grigny, d’Anglebert, and Bach’s Assimilation of French Classical Organ Music’, Early Music, 21 (1993), pp. 83–96. ——Bach: The Mass in B minor (New York, 1997).

418 b i b l i o gr a p h y Stiller, Gu¨nther, J. S. Bach und das Leipziger gottesdienstliche Leben seiner Zeit, dissertation, Leipzig 1966 (Berlin and Kassel, 1970); Eng. version, ed. R. A. Leaver, as J. S. Bach and Liturgical Life in Leipzig (St Louis, Mo., 1984). Stinson, Russell, ‘Towards a Chronology of Bach’s Instrumental Music: Observations on Three Keyboard Works’, Journal of Musicology, 7 (1989), pp. 440–70. ——The Bach Manuscripts of Johann Peter Kellner and His Circle (Durham and London, 1989). ——‘ “Ein Sammelband aus Johann Peter Kellners Besitz”: Neue Forschungen zur Berliner Bach-Handschrift P 804’, BJ 78 (1992), pp. 45–64. ——J. S. Bach’s Great Eighteen Organ Chorales (Oxford, 2001). ——‘Neues u¨ber Bachs Leipziger Orgelchora¨le’, in U. Leisinger (ed.), Bach in Leipzig—Bach und Leipzig [conference report, Leipzig 2000] (Hildesheim, 2002), pp. 109–29. Stockigt, Janice B., ‘Consideration of Bach’s Kyrie e Gloria BWV 232I within the Context of Dresden Catholic Mass Settings, 1729–1733’, in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass [conference report] (Belfast, 2007), vol. i, pp. 52–92. Strohm, Reinhard, ‘Johann Adolph Hasses Oper “Cleoﬁde” und ihr Vorgeschichte’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld [conference report, Duisburg 1986] (Kassel, 1988), pp. 170–6. ——‘Die Epochenkrise der deutschen Opernpﬂege’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld (Kassel, 1988), pp. 155–66. Swack, Jeanne R., ‘On the Origins of the Sonate auf Concertenart’, Journal of the American Musicological Society, 46 (1993), pp. 369–414. ——‘J. S. Bach’s A major Flute Sonata BWV 1032 Revisited’, in D. R. Melamed (ed.), Bach Studies 2 (Cambridge, 1995), pp. 154–74. Synofzik, Thomas, ‘ “Fili Ariadnaei”: Entwicklungslinien zum Wohltemperierten Klavier’, in S. Rampe (ed.), Bach: Das Wohltemperierte Klavier I (Munich and Salzburg, 2002), pp. 109–46. Talbot, Michael, ‘The Concerto Allegro in the Early Eighteenth Century’, Music and Letters, 52 (1971), pp. 8–18, 159–72. ——‘Purpose and Peculiarities of the Brandenburg Concertos’, in M. Geck and K. Hofmann (eds.), Bach und die Stile [conference report, Dortmund 1998] (Dortmund, 1999), pp. 255–89. ¨ berlieferung in der Talle, Andrew, ‘Nu¨rnberg, Darmstadt, Ko¨then—Neuerkenntnisse zur Bach-U ersten Ha¨lfte des 18. Jahrhunderts’, BJ 89 (2003), pp. 143–72. Tomita, Yo, et al. (eds.), Understanding Bach’s B-minor Mass [conference report] (Belfast, 2007), 2 vols. Tovey, Donald Francis, A Companion to ‘The Art of Fugue’ (London, 1931). ——Essays in Musical Analysis, 6 vols. (London, 1935–9). Trautmann, Christoph, ‘ “Calovii Schriften, 3. Ba¨nde” aus J. S. Bachs Nachlass und ihre Bedeutung fu¨r das Bild des lutheranischen Kantors Bach’, Musik und Kirche, 39 (1969), pp. 145–60; Eng. trans. as ‘J. S. Bach: New Light on His Faith’, Concordia Theological Monthly, 42 (1971), pp. 88–99. Walker, Paul, ‘Bach’s Use of Fugue in the stile antico Vocal Writing of the B-minor Mass’, in Y. Tomita et al. (eds.), Understanding Bach’s B-minor Mass [conference report] (Belfast, 2007), vol. ii, pp. 368–86; rev. and expanded version of paper read at biennial meeting of the American Bach Society, Leipzig, 2006.

b i bl i og r aphy

419

Walther, Johann Gottfried, Musicalisches Lexicon oder musicalische Bibliothec (Leipzig, 1732); facs. edn, ed. R. Schaal (Kassel, 1953). Whaples, Miriam K., ‘Bach’s Recapitulation Forms’, Journal of Musicology, 14 (1996), pp. 475–513. Wiemer, Wolfgang, ‘Eine unbekannte Fru¨hfassung des Contrapunctus 2’, Die Musikforschung, 34 (1981), pp. 413–22. Williams, Peter, The Organ Music of J. S. Bach, 3 vols. (Cambridge, 1980–4); rev. 2nd edn in one vol. (Cambridge, 2003). ——Bach: The Goldberg Variations (Cambridge, 2001). ¨ berlegungen anhand eines Quellenfundes’, Wolf, Uwe, ‘J. S. Bach und der Weißenfelser Hof: U BJ 83 (1997), pp. 145–50. ——Kritischer Bericht, NBA V/9.2 (Kassel and Leipzig, 2000). ¨ berlieferung der chromatischen Fantasie BWV 903/1’, in ——‘Fassungsgeschichte und U M. Geck (ed.), Bachs Musik fu¨r Tasteninstrumente [conference report, Dortmund 2002] (Dortmund, 2003), pp. 145–58. ——NBA II/1a: Fru¨hfassungen zur h-Moll-Messe (Kassel and Leipzig, 2005); Kritischer Bericht (2005). Wolff, Christoph, ‘Der Terminus “Ricercar” in Bachs Musikalischem Opfer’, BJ 53 (1967), pp. 70–81; rev. Eng. version as Part II of ‘Apropos the Musical Offering: The Thema Regium and the Term Ricercar’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 324–31. ——‘Zur musikalischen Vorgeschichte des Kyrie aus J. S. Bachs Messe in h-Moll’, in M. Ruhnke (ed.), Festschrift fu¨r Bruno Sta¨blein zum 70. Geburtstag (Kassel, 1967), pp. 316–26; Eng. trans. in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 141–51. ——Der Stile antico in der Musik Johann Sebastian Bachs: Studien zu Bachs Spa¨twerk (Wiesbaden, 1968). ——‘Ordnungsprinzipien in den Originaldrucken Bachscher Werke’, in M. Geck (ed.), BachInterpretationen (Go¨ttingen, 1969), pp. 144–67, 223–5; Eng. version as ‘Principles of Design and Order in Bach’s Original Editions’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 340–58. ——‘New Research on Bach’s Musical Offering’, Musical Quarterly, 57 (1971), pp. 379–408; Eng. trans. in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 239–58. ¨ berlegungen zum “Thema Regium” ’, BJ 59 (1973), pp. 33–8; Eng. trans. as Part I of ——‘U ‘Apropos the Musical Offering: The Thema Regium and the Term Ricercar’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 324–31. ——‘Bachs Sterbechoral: Kritische Fragen zu einem Mythos’, in R. L. Marshall (ed.), Essays in Renaissance and Baroque Music in Honour of Arthur Mendel (Kassel, 1974), pp. 283–97; Eng. trans. as ‘The Deathbed Chorale: Exposing a Myth’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 282–94. ——NBA VIII/1: Kanons, Musikalisches Opfer (Kassel and Leipzig, 1974); Kritischer Bericht (1976). ——‘The Last Fugue: Unﬁnished?’, Current Musicology, 19 (1975), pp. 71–7; repr. in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 259–64.

420 b i b l i o g r a p h y Wolff, Christoph, ‘Bach’s Handexemplar of the Goldberg Variations: A New Source’, Journal of the American Musicological Society, 29 (1976), pp. 224–41; repr. in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 162–77. ——‘Bachs Leipziger Kantoratsprobe und die Auffu¨hrungsgeschichte der Kantate “Du wahrer Gott und Davids Sohn” BWV 23’, BJ 64 (1978), pp. 78–94; Eng. trans. as ‘Bach’s Audition for the St. Thomas Cantorate: The Cantata “Du wahrer Gott und Davids Sohn” ’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 128–40. ——‘Bach’s Cantata Ein feste Burg: History and Performance Practice’, American Choral Review 24 (1982), pp. 27–38; repr. as ‘The Reformation Cantata “Ein feste Burg” ’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 152–61. ——‘Wo blieb Bachs fu¨nfter Kantatenjahrgang?’, BJ 68 (1982), pp. 151–2. ——‘Zur Chronologie und Kompositionsgeschichte von Bachs “Kunst der Fuge” ’, Beitra¨ge zur Musikwissenschaft, 25 (1983), pp. 130–42; Eng. trans. as ‘The Compositional History of the Art of Fugue’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 265–81. ¨ bung (facs. edn Leipzig, 1984); commentary repr. as ——(ed.), Johann Sebastian Bach’s Clavier-U ¨ ‘The Clavier-Ubung Series’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 189–213. ——‘Bach’s Leipzig Chamber Music’, Early Music, 13 (1985), pp. 165–75; repr. in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 223–38. ——‘Bach’s Personal Copy of the Schu¨bler Chorales’, in G. Stauffer and E. May (eds.), J. S. Bach as Organist (London, 1986), pp. 121–32; repr. in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 178–86. ——‘Anmerkungen zu Bach und “Cleofide” ’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld [conference report, Duisburg 1986] (Kassel, 1988), pp. 167–9. ——‘J. S. Bachs Spa¨twerk: Versuch einer Definition’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld (Kassel, 1988), pp. 15–22. ——‘Musikalische Formen und dramatische Gestaltung in Bachs Matthaüs-Passion’, in U. Prinz (ed.), J. S. Bach: Matthaüs-Passion BWV 244 [conference report 1985, Internationale Bachakademie Stuttgart] (Kassel, 1990), pp. 94–109. ——‘ “Intricate Kirchen-Stu¨cke” und “Dresdener Liederchen”: Bach und die Instrumentalisierung der Vokalmusik’, in C. Wolff and H.-J. Schulze (eds.), J. S. Bach und der su¨ddeutsche Raum [conference report, Munich 1990] (Regensburg, 1991), pp. 19–23. ——‘Die musikalischen Formen der Johannes-Passion’, in U. Prinz, J. S. Bach: Johannes-Passion BWV 245 [conference report, 1986 and 1990, Internationale Bachakademie Stuttgart] (Kassel, 1993), pp. 128–41. ——‘Origins of the Kyrie of the B minor Mass’, in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass., and London, 1991), pp. 141–51. ——‘The Agnus Dei of Bach’s B-minor Mass: Parody and New Composition Reconciled’, in P. Brainard and R. Robinson (eds.), A Bach Tribute: Essays on Honour of William H. Scheide (Kassel and Chapel Hill, NC, 1993), pp. 233–40; also pub. in Wolff, Bach: Essays on His Life and Music (Cambridge, Mass. and London, 1991), pp. 332–9. ——‘ “Et incarnatus” and “Crucifixus”: The Earliest and the Latest Settings of Bach’s B-minor Mass’, in M. A. Parker (ed.), Eighteenth-Century Music in Theory and Practice: Essays in Honour of Alfred Mann (Stuyvesant, NY, 1994), pp. 1–17.

bibliography

421

——(ed.), Die Welt der Bach-Kantaten, vol. ii. J. S. Bachs Weltliche Kantaten (Stuttgart, 1997). ——(ed.), Die Welt der Bach-Kantaten, vol. iii. J. S. Bachs Leipziger Kirchenkantaten (Stuttgart, 1999). ——Johann Sebastian Bach: The Learned Musician (Oxford and New York, 2000). ——(ed. with M. Maul), BD VII: Johann Nikolaus Forkel, Ueber Johann Sebastian Bachs Leben, Kunst und Kunstwerke (Kassel, 2008). ——‘J. S. Bachs Oratorien-Trilogie und die große Kirchenmusik der 1730 er Jahre’, BJ 97 (2011), pp. 11–25. Wollny, Peter, ‘Ein Quellenfund zur Entstehungsgeschichte der h-Moll-Messe’, BJ 80 (1994), pp. 163–9. ——‘Ein “musikalischer Veteran Berlins”: Der Schreiber Anonymous 300 und seine Bedeutung ¨ berlieferung’, in G. Wagner (ed.), Jahrbuch des Staatlichen Instituts fu¨rs fu¨r die Berliner Bach-U Musikforschung Preußischer Kulturbesitz 1995 (Stuttgart and Weimar, 1996), pp. 80–133. ——‘Neue Bach-Funde’, BJ 83 (1997), pp. 7–50. ——‘J. S. Bachs Tripelkonzert a-Moll BWV 1044’, in M. Geck and W. Breig (eds.), Bachs Orchesterwerke [conference report, Dortmund 1996] (Witten, 1997), pp. 283–91. ——(ed.), Facsimile edn of the Eighteen Chorales and the Canonic Variations on Vom Himmel hoch, BWV 769 (Laaber, 1999). ——and Wolff, Christoph, ‘Allgemeine Strategien in Bachs 1. Leipziger Kantatenjahrgang’, in M. Geck (ed.), Bachs 1. Leipziger Kantatenjahrgang [conference report, Dortmund 2000] (Dortmund, 2002), pp. 23–40. ¨ berlieferung in Mit——‘Tennstedt, Leipzig, Naumburg, Halle—Neuerkenntnisse zur Bach-U teldeutschland’, BJ 88 (2002), pp. 29–60. ——‘ “Bekennen will ich seinen Namen”—Authentizita¨t, Bestimmung und Kontext der Arie BWV 200: Anmerkungen zu J. S. Bachs Rezeption von Werken Gottfried Heinrich Sto¨lzels’, BJ 94 (2008), pp. 123–58. ——‘Beobachtungen am Autograph der h-Moll-Messe’, BJ 95 (2009), pp. 135–51. ——‘Zwei Bach-Funde in Mu¨geln: C. P. E. Bach, Picander und die Leipziger Kirchenmusik in den 1730er Jahren’, BJ 96 (2010), pp. 111–51. Zenck, Martin, ‘1740–1750 und das a¨sthetische Bewußtsein einer Epochenschwelle? Zum Text und Kontext von Bachs Spa¨twerk’, in C. Wolff (ed.), J. S. Bachs Spa¨twerk und dessen Umfeld [conference report, Duisburg 1986] (Kassel, 1988), pp. 109–16. Zohn, Steven, ‘Bach and the Concert en ouverture’, in G. G. Butler (ed.), Bach Perspectives 6 (Urbana and Chicago, 2007), pp. 137–56.

This page intentionally left blank

Index of Bach’s Works The reference to the main discussion of each work is given in bold type. Aria variata in A minor, BWV 989: 347 Art of Fugue, see Die Kunst der Fuge Ascension Oratorio, see Himmelfahrts-Oratorium Aufrichtige Anleitung (Inventions & Sinfonias), BWV 772–801: 4, 5, 10, 15, 16, 31–5, 39, 206, 210–11, 395 Inventions 11, 39, 237, 242, 337–8 Sinfonias 45, 48, 54, 99, 170, 263, 304, 337–8, 340, 396 ‘Bekennen will ich seinen Namen’ (aria after Sto¨lzel), BWV 200: 328 Brandenburg Concertos, BWV 1046–51: 4, 10, 38, 65–81, 82–3, 85, 86, 87, 206, 208, 211–12, 255, 268 Concerto No. 1 in F, BWV 1046: 66–8, 111 Concerto No. 2 in F, BWV 1047: 70–2, 255 Concerto No. 3 in G, BWV 1048: 72, 195, 255, 267 Concerto No. 4 in G, BWV 1049: 73, 225, 252, 257, 261, 317 Concerto No. 5 in D, BWV 1050: 30, 39–40, 73, 98, 208, 252, 268, 317, 365, 395 Concerto No. 6 in B ﬂat, BWV 1051: 72, 255 Canon triplex a 6, BWV 1076: 331, 351 Canon for J. G. Fulde, BWV 1077: 351 14 Canons, BWV 1087, see Verschiedene Canones Canonic Variations, see Canonische Vera¨nderungen Canonische Vera¨nderungen u¨ber das Weihnachtslied: Vom himmel hoch da komm ich her, BWV 769: 331, 333, 350–2, 360–1, 362, 393, 397–8 cantatas, sacred: Ach Gott, vom Himmel sieh darein, BWV 2: 142, 147, 153, 176, 215 Ach Gott, wie manches Herzeleid, BWV 3: 144, 147, 148, 149–50, 215 Ach Gott, wie manches Herzeleid, BWV 58: 169, 171–2, 173, 180, 186, 196, 216, 218 Ach Herr, mich armen Su¨nder, BWV 135: 142, 147, 176, 215, 287 Ach! ich sehe, itzt, da ich zur Hochzeit gehe, BWV 162: 118 n. 20

Ach, lieben Christen, seid getrost, BWV 114: 143, 150, 153, 154, 216 Ach wie ﬂu¨chtig, ach wie nichtig, BWV 26: 143, 149, 153, 215 Allein zu dir, Herr Jesu Christ, BWV 33: 143, 149 Alles nur nach Gottes Willen, BWV 72: 168, 305 Alles, was von Gott geboren, BWV 80a: 121, 279, 280, 283 Also hat Gott die Welt geliebt, BWV 68: 162, 163–6, 215 Am Abend aber desselbigen Sabbats, BWV 42: 107, 162, 163, 164, 165, 167, 174 ¨ rgre dich, o Seele, nicht, BWV 186: 116, 118, A 125, 126 ¨ rgre dich, o Seele, nicht, BWV 186a: 118, 121, 146 A Auf Christi Himmelfahrt allein, BWV 128: 162, 164–5 Aus tiefer Not schrei ich zu dir, BWV 38: 143, 147, 151, 153, 176, 215 Barmherziges Herze der ewigen Liebe, BWV 185: 118, 121 Bisher habt ihr nichts gebeten in meinem Namen, BWV 87: 162, 164, 166, 167, 179 Bleib bei uns, denn es will Abend werden, BWV 6: 162, 164, 165–6, 176 Brich dem Hungrigen dein Brot, BWV 39: 8, 168, 169, 177, 178, 179 Bringet dem Herrn Ehre seines Namens, BWV 148: 168, 177 Christ lag in Todes Banden, BWV 4: 118, 119, 145, 162, 176, 215, 217, 282 Christ unser Herr zum Jordan kam, BWV 7: 142, 147, 153, 215 Christen, a¨tzet diesen Tag, BWV 63: 80, 107, 121 Christum wir sollen loben schon, BWV 121: 143, 147, 176, 215 Christus, der ist mein Leben, BWV 95: 116, 125, 126, 127–8, 146, 150, 151, 176, 215 Darzu ist erschienen der Sohn Gottes, BWV 40: 116, 124, 129, 130, 177, 214, 307 Das neugeborne Kindelein, BWV 122: 143, 151, 152 Dem gerechten muß das Licht, BWV 195: 322, 327

424 i n de x of bac h ’ s wo r ks cantatas, sacred: (cont.) Der Friede sei mit dir, BWV 158: 168, 180, 196, 197, 216 Der Herr ist mein getreuer Hirt, BWV 112: 279, 282, 322 Der Himmel lacht! Die Erde jubilieret, BWV 31: 83, 118, 119, 121, 125 Die Elenden sollen essen, BWV 75: 6, 113, 115, 117, 118, 119, 123, 125, 126, 129, 177, 214 Die Himmel erza¨hlen die Ehre Gottes, BWV 76: 113, 116, 118, 119, 123, 125, 126, 177, 214, 263, 265 Du Friedefu¨rst, Herr Jesu Christ, BWV 116: 143, 153 Du Hirte Israel, ho¨re, BWV 104: 117, 124, 129, 177, 214 Du sollt Gott, deinen Herren, lieben, BWV 77: 116, 119, 126, 129, 215, 284 Du wahrer Gott und Davids Sohn, BWV 23: 6, 115, 117, 118, 119, 126, 129, 131, 138, 146, 151, 156, 160, 197, 214, 306 Ehre sei Gott in der Ho¨he, BWV 197a: 193, 194 Ein feste Burg ist unser Gott, BWV 80: 279, 282, 283–4, 316 Ein feste Burg ist unser Gott, BWV 80b: 279, 283 Ein Herz, das seinen Jesum lebend weiß, BWV 134: 117, 119 Ein ungefa¨rbt Gemu¨te, BWV 24: 116, 118, 119, 123, 125, 177 Er rufet seinen Schafen mit Namen, BWV 175: 162, 163, 164, 167 Erforsche mich, Gott, und erfahre mein Herz, BWV 136: 116, 124, 125, 177, 214, 303 Erfreut euch, ihr Herzen, BWV 66: 107, 117, 119, 163, 171 Erfreute Zeit im neuen Bunde, BWV 83: 117, 128, 129, 216 Erhalt uns, Herr, bei deinem Wort, BWV 126: 144, 151 Erho¨htes Fleisch und Blut, BWV 173: 117, 119, 163, 193 Erschallet, ihr Lieder, erklinget, ihr Saiten, BWV 172: 107, 118, 119, 120, 121, 125, 171, 218 Erwu¨nschtes Freudenlicht, BWV 184: 107, 117, 119, 193–4 Es erhub sich ein Streit, BWV 19: 169, 173, 272–3 Es ist das Heil uns kommen her, BWV 9: 279, 282, 284 Es ist dir gesagt, Mensch, was gut ist, BWV 45: 8, 168, 169, 177, 179

Es ist ein trotzig und verzagt Ding, BWV 176: 162, 164 Es ist euch gut, daß ich hingehe, BWV 108: 162, 164, 166, 167 Es ist nichts Gesundes an meinem Leibe, BWV 25: 116, 126–7, 129, 215 Es reißet euch ein schrecklich Ende, BWV 90: 116 Es wartet alles auf dich, BWV 187: 8, 168, 169, 177, 179, 305, 318 Falsche Welt, dir trau ich nicht, BWV 52: 169, 170, 174, 175, 195, 217, 218 Freue dich, erlo¨ste Schar, BWV 30: 223, 279, 281, 322 Geist und Seele wird verwirret, BWV 35: 168, 170, 174, 175, 195, 217, 218 Gelobet sei der Herr, mein Gott, BWV 129: 193–4, 196, 215, 282, 316 Gelobet seist du, Jesu Christ, BWV 91: 143, 151 Gleichwie der Regen und Schnee vom Himmel fa¨llt, BWV 18: 83, 118, 120–1, 151 Gott der Herr ist Sonn und Schild, BWV 79: 168, 177–8, 179, 180, 301, 303, 304, 318 Gott fa¨hret auf mit Jauchzen, BWV 43: 8, 168, 169, 177, 179 Gott ist unsre Zuversicht, BWV 197: 279, 281, 320, 322 Gott, man lobet dich in der Stille, BWV 120: 279, 327, 387 Gott, man lobet dich in der Stille, BWV 120b: 279 Gott soll allein mein Herze haben, BWV 169: 169, 170, 174, 175, 179, 180, 195, 217, 218, 272–3 Gott, wie dein Name, so ist auch dein Ruhm, BWV 171: 193, 194, 383 Gottes Zeit ist die allerbeste Zeit, BWV 106: 171, 199, 218 Gottlob! nun geht das Jahr zu Ende, BWV 28: 168, 176 Halt im Geda¨chtnis Jesum Christ, BWV 67: 117, 122–3, 124, 132, 177, 214, 302, 318 Herr Christ, der einge Gottessohn, BWV 96: 143, 147, 153 Herr, deine Augen sehen nach dem Glauben, BWV 102: 8, 168, 169, 177, 178, 179, 301, 305, 306, 318 Herr, gehe nicht ins Gericht, BWV 105: 116, 123, 125, 130, 131, 139–40, 177, 214 Herr Gott, Beherrscher aller Dinge, BWV 120a: 279, 316 Herr Gott, dich loben alle wir, BWV 130: 143, 153 Herr Gott, dich loben wir, BWV 16: 168, 176

index o f b ach ’s w or k s Herr Jesu Christ, du ho¨chstes Gut, BWV 113: 143, 148, 149, 150, 151, 153, 216 Herr Jesu Christ, wahr’ Mensch und Gott, BWV 127: 144, 153–4, 156–7, 197, 306, 329 Herr, wie du willt, so schicks mit mir, BWV 73: 117, 127, 128, 132, 146, 176, 215 Herz und Mund und Tat und Leben, BWV 147: 116, 118, 125, 126, 130 Herz und Mund und Tat und Leben, BWV 147a: 118, 121, 146 Himmelsko¨nig, sei willkommen, BWV 182: 118, 120, 121 Ho¨chsterwu¨nschtes Freudenfest, BWV 194: 107, 108, 116, 118, 119, 125, 147, 173, 216 Ich armer Mensch, ich Su¨ndenknecht, BWV 55: 169, 170, 179, 217, 273 Ich bin ein guter Hirt, BWV 85: 162, 164, 165, 167 Ich bin ein Pilgrim auf der Welt, BWV Anh. I 190: 195 Ich bin vergnu¨gt mit meinem Glu¨cke, BWV 84: 169, 170, 217 Ich elender Mensch, wer wird mich erlo¨sen, BWV 48: 116, 126, 129, 215, 284 Ich freue mich in dir, BWV 133: 143 Ich geh und suche mit Verlangen, BWV 49: 169, 171, 172–3, 174, 175, 179–80, 186, 195, 196, 216, 218 Ich glaube, lieber Herr, hilf meinem Unglauben, BWV 109: 116, 125–6, 129, 215 Ich hab in Gottes Herz und Sinn, BWV 92: 144, 148, 150, 151, 153, 216 Ich habe genung, BWV 82: 114, 169, 170, 171, 179, 217, 218, 272–3 Ich habe meine Zuversicht, BWV 188: 193, 194, 195, 322 Ich hatte viel Beku¨mmernis, BWV 21: 68, 118, 119, 120, 121, 123, 171, 199, 218 Ich lasse dich nicht, du segnest mich denn, BWV 157: 192, 194 Ich lebe, mein Herze, zu deinem Ergo¨tzen, BWV 145: 193, 194, 195, 197 Ich liebe den Ho¨chsten von ganzem Gemu¨te, BWV 174: 193, 194, 195, 322 Ich ruf zu dir, Herr Jesu Christ, BWV 177: 279, 282 Ich steh mit einem Fuß im Grabe, BWV 156: 193, 194, 195, 196–7, 216, 260 Ich will den Kreuzstab gerne tragen, BWV 56: 169, 170, 171, 172, 179, 180, 217 Ihr, die ihr euch von Christo nennet, BWV 164: 168, 180

425

Ihr Menschen, ru¨hmet Gottes Liebe, BWV 167: 116, 125 Ihr Tore zu Zion, BWV 193: 109, 193, 194 Ihr werdet weinen und heulen, BWV 103: 162, 164, 166 In allen meinen Taten, BWV 97: 279, 282, 322 Jauchzet Gott in allen Landen, BWV 51: 279, 280, 316, 321 Jesu, der du meine Seele, BWV 78: 143, 149–50, 153, 215 Jesu, nun sei gepreiset, BWV 41: 143, 148 Jesus nahm zu sich die Zwo¨lfe, BWV 22: 6, 115, 117, 118, 119, 121–2, 125, 129, 146, 167, 197, 214, 215 Jesus schla¨ft, was soll ich hoffen, BWV 81: 117, 122, 130 Komm, du su¨ße Todesstunde, BWV 161: 121, 125 Kommt, gehet und eilet, BWV 249: 162–3, 174, 217, 314 Leichtgesinnte Flattergeister, BWV 181: 117, 118 Liebster Gott, wenn werd ich sterben, BWV 8: 143, 148, 149, 153, 154, 215, 273 Liebster Immanuel, Herzog der Frommen, BWV 123: 144, 148 Liebster Jesu, mein Verlangen, BWV 32: 168, 171, 172, 218 Lobe den Herren, den ma¨chtigen Ko¨nig der Ehren, BWV 137: 168, 175–6, 196, 215, 282, 316 Lobe den Herrn, meine Seele, BWV 69: 327 Lobe den Herrn, meine Seele, BWV 69a: 116, 119, 123–4, 129, 165, 177, 214, 327 Lobet den Herrn, alle seine Heerscharen, BWV Anh. I 5: 3, 213 Mache dich, mein Geist, bereit, BWV 115: 143, 153, 154 Man singet mit Freuden vom Sieg, BWV 149: 193, 194, 322 Mein Herze schwimmt im Blut, BWV 199: 3, 113, 118, 121, 170, 217 Mein liebster Jesus ist verloren, BWV 154: 116, 122, 129 Meine Seel erhebt den Herren, BWV 10: 143, 150–1, 153, 362 Meine Seufzer, meine Tra¨nen, BWV 13: 168 Meinen Jesum laß ich nicht, BWV 124: 144, 153 Mit Fried und Freud ich fahr dahin, BWV 125: 144, 151, 153–4, 155–6 Nimm von uns, Herr, du treuer Gott, BWV 101: 143, 148, 151–2, 153, 155, 176, 215, 216–17, 287

426

i n dex of b ac h ’s w o rk s

cantatas, sacred: (cont.) Nimm, was dein ist, und gehe hin, BWV 144: 117, 124, 214 Nun danket alle Gott, BWV 192: 279, 282 Nun komm, der Heiden Heiland, BWV 61: 147, 148, 120, 121 Nun komm, der Heiden Heiland, BWV 62: 143, 148, 153 Nur jedem das Seine, BWV 163: 121 O ewiges Feuer, o Ursprung der Liebe, BWV 34: 192, 193 O ewiges Feuer, o Ursprung der Liebe, BWV 34a: 169, 193 O Ewigkeit, du Donnerwort, BWV 20: 142, 147, 153, 215 O Ewigkeit, du Donnerwort, BWV 60: 116, 127, 130, 131–2, 159, 171, 173, 215, 218 O heilges Geist- und Wasserbad, BWV 165: 118 Preise, Jerusalem, den Herrn, BWV 119: 116, 125, 139, 147, 173 Schau, lieber Gott, wie meine Feind, BWV 153: 116, 122, 129 Schauet doch und sehet, ob irgend ein Schmerz sei, BWV 46: 116, 123, 125, 130, 131, 177, 214, 296–7, 318 Schmu¨cke dich, o liebe Seele, BWV 180: 143, 151, 153, 154 Schwingt freudig euch empor, BWV 36: 109, 192, 194, 223, 279–80, 281, 282, 283, 284, 316 Sehet, welch eine Liebe hat uns der Vater erzeiget, BWV 64: 116, 119, 124, 129–30, 214 Sehet, wir gehn hinauf gen Jerusalem, BWV 159: 193, 194, 196–7, 216, 322 Sei Lob und Ehr dem ho¨chsten Gut, BWV 117: 279, 282 Selig ist der Mann, BWV 57: 168, 171, 172, 218 Sie werden aus Saba alle kommen, BWV 65: 116, 124, 129, 177, 214 Sie werden euch in den Bann tun, BWV 44: 117, 122, 125, 128, 129, 150, 166, 215 Sie werden euch in den Bann tun, BWV 183: 162, 164 Siehe, eine Jungfrau ist schwanger, BWV Anh. I 199: 118 Siehe, ich will viel Fischer aussenden, BWV 88: 8, 168, 169, 177, 178–9 Siehe zu, daß deine Gottesfurcht nicht Heuchelei sei, BWV 179: 116, 118, 124, 214, 301, 303, 304, 318

Singet dem Herrn ein neues Lied, BWV 190: 116, 126, 127, 129, 151, 176, 215 Su¨ßer Trost, mein Jesus ko¨mmt, BWV 151: 168 Tritt auf die Glaubensbahn, BWV 152: 121, 171, 218, 265 Tue Rechnung! Donnerwort, BWV 168: 168 Unser Mund sei voll Lachens, BWV 110: 84, 168, 173–4, 175, 177, 180, 217 Vergnu¨gte Ruh, beliebte Seelenlust, BWV 170: 168, 170–1, 175, 179, 217, 218 Wachet auf, ruft uns die Stimme, BWV 140: 273, 279, 280–1, 282–3, 284, 316 Wachet, betet, betet, wachet BWV 70: 116, 118, 126, 151 Wachet, betet, betet, wachet BWV 70a: 85, 118, 121, 125, 146 Wahrlich, wahrlich, ich sage euch, BWV 86: 117, 122, 128, 150, 166, 179, 215 Wa¨r Gott nicht mit uns diese Zeit, BWV 14: 279, 282, 283, 284, 316 Warum betru¨bst du dich, mein Herz, BWV 138: 116, 125–6, 127, 128, 129, 146, 151, 176, 215, 303 Was frag ich nach der Welt, BWV 94: 143, 149, 151, 153 Was Gott tut, das ist wohlgetan, BWV 98: 169 Was Gott tut, das ist wohlgetan, BWV 99: 143, 154 Was Gott tut, das ist wohlgetan, BWV 100: 279, 282, 322 Was mein Gott will, das gscheh allzeit, BWV 111: 144 Was soll ich aus dir machen, Ephraim, BWV 89: 116, 122, 129 Was willst du dich betru¨ben, BWV 107: 143, 146, 152, 153, 165, 176, 215, 282, 316 Weinen, Klagen, Sorgen, Zagen, BWV 12: 121, 125, 149, 384, 400 Wer da gla¨ubet und getauft wird, BWV 37: 117, 125, 128, 150, 215 Wer Dank opfert, der preiset mich, BWV 17: 8, 168, 169, 177, 178, 179, 304 Wer mich liebet, der wird mein Wort halten, BWV 59: 117, 118, 119, 164, 166 Wer mich liebet, der wird mein Wort halten, BWV 74: 162, 164, 166 Wer nur den lieben Gott la¨ßt walten, BWV 93: 143, 148, 150–1, 152, 153, 362 Wer sich selbst erho¨het, der soll erniedriget werden, BWV 47: 61, 169, 175, 177, 178

i n d e x of ba ch’ s w or ks Wer weiß, wie nahe mir mein Ende, BWV 27: 169, 175, 176 Widerstehe doch der Su¨nde, BWV 54: 113, 121, 170, 217, 286 Wie scho¨n leuchtet der Morgenstern, BWV 1: 144, 153, 154, 157, 160 Wir danken dir, Gott, wir danken dir, BWV 29: 279, 281, 295, 316, 318 Wir mu¨ssen durch viel Tru¨bsal, BWV 146: 168, 193, 194, 195–6, 217 Wo Gott der Herr nicht bei uns ha¨lt, BWV 178: 143, 148–9, 150, 151, 216 Wo gehest du hin, BWV 166: 122, 128, 150, 177, 215 Wo soll ich ﬂiehen hin, BWV 5: 143, 151 Wohl dem, der sich auf seinen Gott, BWV 139: 143, 153 cantatas, secular: Aeolus Placated, see Der zufriedengestellte Aeolus Amore traditore, BWV 203: 170, 217, 277 Angenehmes Wiederau, BWV 30a: 271, 277, 278, 281, 322 Auf, schmetternde To¨ne der muntern Trompeten, BWV 207a: 109, 274 Auf, su¨ß entzu¨ckende Gewalt, BWV Anh. I 196: 108, 109, 223, 313, 389 Blast La¨rmen, ihr Feinde! versta¨rket die Macht, BWV 205a: 109, 274 Cantate burlesque: Mer hahn en neue Oberkeet, BWV 212: 329, 374–6, 397, 399 Coffee Cantata, see Schweigt stille, plaudert nicht Der Himmel dacht auf Anhalts Ruhm und Glu¨ck, BWV 66a: 106, 107, 163, 213 Der Streit zwischen Phoebus und Pan: Geschwinde, geschwinde, ihr wirbelnden Winde, BWV 201: 115, 271–4, 275, 281, 294, 316, 322, 329, 376 Der zufriedengestellte Aeolus: Zerreißet, zersprenget, zertru¨mmert die Gruft, BWV 205: 10, 106, 109, 110, 111, 112, 213, 271 Die Feier des Genius: Verjaget, zerstreuet, zerru¨ttet, ihr Sterne, BWV 249b: 10, 109, 110, 271 Die Freude reget sich, BWV 36b: 109, 277 Die Zeit, die Tag und Jahre macht, BWV 134a: 106, 107, 213 Durchlauchtster Leopold, BWV 173a: 106, 107–8, 163, 213 Entfernet euch, ihr heitern Sterne, BWV Anh. I 9: 109, 274

427

Entfliehet, verschwindet, entweichet, ihr Sorgen, BWV 249a: 10, 67, 106, 108, 109, 110–11, 136, 162, 174, 213, 217, 222, 271, 314, 318 Es lebe der Ko¨nig, der Vater im Lande, BWV Anh. I 11: 274, 275, 276, 375, 388 Froher Tag, verlangte Stunden, Anh. I 18: 223, 277, 313, 322 Frohes Volk, vergnu¨gte Sachsen, Anh. I 12: 274 Funeral Music for Prince Leopold, see Klagt, Kinder, klagt es aller Welt Geschwinde, geschwinde, see Der Streit zwischen Phoebus und Pan Hercules auf dem Scheidewege: Laßt uns sorgen, laßt uns wachen, BWV 213: 223, 271, 274, 275, 309, 310, 311, 312, 319, 383 Hercules Cantata, see Hercules auf dem Scheidewege Hunt Cantata, see Was mir behagt, ist nur die muntre Jagd Ich bin in mir vergnu¨gt, see Von der Vergnu¨gsamkeit Ihr Haüser des Himmels, ihr scheinenden Lichter, BWV 193a: 109, 110, 194, 271, 274 Klagt, Kinder, klagt es aller Welt, BWV 244a: 10, 114, 213, 223, 285 Laß, Fu¨rstin, laß noch einen Strahl, see Trauer Music Laßt uns sorgen, laßt uns wachen, see Hercules auf dem Scheidewege Mer hahn en neue Oberkeet, see Cantate burlesque Murmelt nur, ihr heitern Ba¨che, Anh. I 195: 109 Non sa che sia dolore, BWV 209: 271, 274, 277–8, 316 O! angenehme Melodei, BWV 210a: 106, 109, 110, 112, 114–15, 213, 278, 329 O holder Tag, erwu¨nschte Zeit, BWV 210: 110, 114, 277, 329 Pastoral Cantata, see Entfliehet, verschwindet, entweichet, ihr Sorgen Peasant Cantata, see Cantate burlesque Phoebus and Pan, see Der Streit zwischen Phoebus und Pan Polish Cantata, see Preise dein Glu¨cke, gesegnetes Sachsen Preise dein Glu¨cke, gesegnetes Sachsen, BWV 215: 198, 223, 271, 274, 275–6, 294, 309, 312, 322, 388 Queen’s Cantata, see To¨net, ihr Pauken! Erschallet, Trompeten

428

i nd ex o f b ac h ’s w o rk s

cantatas, secular: (cont.) Rivers Cantata, see Schleicht, spielende Wellen Schleicht, spielende Wellen, und murmelt gelinde, BWV 206: 271, 274, 275–6, 294, 322 Schweigt stille, plaudert nicht, BWV 211: 271, 277, 316, 329, 399 Schwingt freudig euch empor, BWV 36c: 106, 109, 112, 194, 213, 223, 277, 279 So ka¨mpfet nur, ihr muntern To¨ne, BWV Anh. I 10: 277 Steigt freudig in die Luft, BWV 36a: 10, 108, 109, 223 Thomana saß annoch betru¨bt, BWV Anh. I 19: 277 To¨net, ihr Pauken! Erschallet, Trompeten, BWV 214: 223, 271, 274, 275, 285, 294, 309, 310, 322 Trauer Music: Laß, Fu¨rstin, laß noch einen Strahl, BWV 198: 10, 106, 109, 112, 113–14, 115, 193, 213, 223, 285–6 Trauer-Ode, see Trauer Music Vereinigte Zwietracht der wechselnden Saiten, BWV 207: 10, 68, 106, 109, 110, 111–12, 213, 271, 274 Vergnu¨gte Pleißenstadt, BWV 216: 10, 108 Verjaget, zerstreuet, zerru¨ttet, ihr Sterne, see Die Feier des Genius Von der Vergnu¨gsamkeit: Ich bin in mir vergnu¨gt, BWV 204: 10, 106, 109, 112, 113, 114, 170, 213, 217, 277 Was mir behagt, ist nur die muntre Jagd, BWV 208: 67, 107, 110, 163–4, 222, 329 Was mir behagt, ist nur die muntre Jagd, BWV 208a: 329 Weichet nur, betru¨bte Schatten, BWV 202: 113, 170, 172, 217, 277 Willkommen! Ihr herrschenden Go¨tter der Erden, BWV Anh. I 13: 227, 274, 321 Wo sind meine Wunderwerke, BWV Anh. I 210: 277 Zerreißet, zersprenget, zertru¨mmert die Gruft, see Der zufriedengestellte Aeolus [title unknown], BWV 184a: 107–8, 213, 275 [title unknown], BWV 194a: 107, 108, 119, 213, 216 [title unknown], BWV Anh. I 20: 109 Cello Suites, BWV 1007–12: 4, 36–7, 51, 55, 86, 91, 93–6, 97, 206, 207, 209, 231, 243, 244 6 Chora¨le von verschiedener Art (Schu¨bler Chorales), BWV 645–50: 151, 165, 176, 333, 361–2, 395 Christe in G minor, BWV 242: 292

Christmas Oratorio, see Weihnachts-Oratorium Chromatic Fantasia, see Fantaisie chromatique Clavierbu¨chlein: for Wilhelm Friedemann Bach, 1720: 4, 15, 16–18, 23, 31–2, 36, 45, 54, 95, 224, 241 for Anna Magdalena Bach, 1722: 4, 15, 36, 41, 42, 49, 242 for Anna Magdalena Bach, 1725: 11, 15, 36, 41, 42, 49, 51, 55, 56, 57, 102, 210, 224 Clavieru¨bung: Part I, 6 Partitas, BWV 825–30: 11, 36, 41, 49–58, 91, 206, 209, 211, 212, 225, 228, 229–30, 231, 239, 245, 319, 337, 395 Partita No. 1 in B flat, BWV 825: 10, 53, 239, 339, 348, 397 Partita No. 2 in C minor, BWV 826: 44, 53–4, 96, 97, 113, 231, 240, 322 Partita No. 3 in A minor, BWV 827: 11, 54–5, 113, 205, 210, 240, 264 Partita No. 4 in D, BWV 828: 44, 55–6, 95, 113, 224, 228, 230, 231, 299–300 Partita No. 5 in G, BWV 829: 56–7, 239, 299–300 Partita No. 6 in E minor, BWV 830: 11, 26, 57–8, 60, 63, 102, 213, 239, 245 Part II, Italian Concerto and French Overture 50, 55, 91, 92, 225, 228–31, 232, 237–8, 319, 395 Concerto nach italienischen Gusto, BWV 971: 50, 228, 229–30, 255, 262, 317, 319, 321 Ouverture nach franzo¨sischer Art, BWV 831: 228, 230–1, 319, 322 Part III, organ Missa, catechism chorales, etc. 225, 228, 231–8, 322–3, 330, 333, 335, 361, 378 Praeludium et Fuga in E flat, BWV 552: 237–8, 320–1, 396 4 Duetti, BWV 802–5: 237, 240 Missa: Kyrie-Christe-Kyrie, Allein Gott in der Ho¨h sei Ehr, BWV 669–77: 234–5, 245, 400 Catechism chorales, BWV 678–89: 235–6 [Part IV], Goldberg Variations: Aria mit verschiedenen Vera¨nderungen, BWV 988: 55, 229, 331, 333, 346–50, 351, 352, 375, 393, 397–8 Concerto in A minor for flute, violin, and harpsichord, BWV 1044: 66, 208, 263–4, 363, 365–6, 395 concerto transcriptions, BWV 972–87 and 592–6: 13, 74, 77, 81, 83, 86, 90, 207, 229, 253, 255, 257, 319 Credo in unum Deum in F, BWV 1081: 381

i n d e x of ba ch’ s w or k s Easter Oratorio, see Oster-Oratorium Eighteen Chorales, BWV 651–68: 262, 330, 333–5, 361–2, 395 Vor deinen Thron tret ich, BWV 668: 334–5 Wenn wir in ho¨chsten No¨ten sein, BWV 668 a: 334–5, 353 English Suites, BWV 806–11: 4, 10, 11, 36–41, 43, 45–6, 48–9, 51, 55, 57, 68–9, 91, 93, 94–6, 188, 206, 208–9, 211–12, 231, 240, 244 Fantaisie chromatique in D minor, BWV 903: 12–14, 15, 208 Fantaisie sur un rondeau in C minor, BWV 918: 238, 240–1, 245, 322 Fantasia in G minor, BWV 542 no. 1: 12, 14–15, 208 Fantasia in C minor, BWV 562 no. 1: 58–9, 61–2, 239 Fantasia et Fuga in C minor, BWV 537: 58, 62, 63–4, 210, 239 Fantasia et Fuga in A minor, BWV 904: 238–9, 243, 320 Fantasia et Fuga in C minor, BWV 906: 238, 239–40, 244, 320–1, 395, 397 Fantasia [et Fuga] in A minor, BWV 944: 86, 229 Flute Solo in A minor, BWV 1013: 86, 96–7 ﬂute sonatas: with obbligato harpsichord: Sonata in B minor, BWV 1030: 208, 227, 249–50, 261, 266–7, 268–9, 270, 316, 364 Sonata in A, BWV 1032: 208, 227, 261, 266–7, 269–70, 316, 364 with continuo: Sonata in E minor, BWV 1034: 97, 103, 104 Sonata in E, BWV 1035: 330, 363–4, 365, 394 Sonata in G for two ﬂutes, BWV 1039: 97, 103, 104, 364 French Overture, BWV 831, see Clavieru¨bung, Part II French Suites, BWV 812–17: 4, 10, 36, 41–7, 48–9, 53–6, 93, 97, 206, 209, 211, 231, 339 Fuga in F, BWV 540 no. 2: 58, 63, 210, 239, 320 Fuga in C minor on a Theme of Legrenzi, BWV 574b: 63 Fuga in G minor, BWV 1026: 87, 207 Gloria in excelsis, BWV 191: 133, 293, 327, 381 Goldberg Variations, see Clavieru¨bung [Part IV] harpsichord concertos 10, 224, 317 for solo harpsichord 225, 252, 257–61, 317, 365

429

Six Concertos, BWV 1052–7: 208, 257–61, 317 Concerto No. 1 in D minor, BWV 1052: 69, 83, 87, 93, 257, 258–9, 261, 267, 317 Concerto No. 2 in E, BWV 1053: 175, 195, 257, 259, 263, 274 Concerto No. 3 in D, BWV 1054: 224, 257, 259 Concerto No. 4 in A, BWV 1055: 257, 259–60 Concerto No. 5 in F minor, BWV 1056: 257, 260–1 Concerto No. 6 in F, BWV 1057: 225, 257, 261, 317, 365 Concerto in G minor, BWV 1058: 224, 257 Concerto in D minor, BWV 1059: 175, 257 for two harpsichords 225, 254–7, 317 Concerto in C minor, BWV 1060: 254, 256–7, 260 Concerto in C, BWV 1061: 225, 229, 254–6, 259, 262, 317, 319 Concerto in C minor, BWV 1062: 224, 254, 256 for three harpsichords 225, 252–4, 317 Concerto in D minor, BWV 1063: 252, 253, 259 Concerto in C, BWV 1064: 252, 253, 254, 323 for four harpsichords 225, 252–4, 317 Concerto in A minor (after Vivaldi), BWV 1065: 225, 252, 253 Inventions and Sinfonias, see Aufrichtige Anleitung Italian Concerto, BWV 971, see Clavieru¨bung, Part II Die Kunst der Fuge, BWV 1080: 26, 331, 333, 334, 352–60, 361, 367–70, 379, 393, 396, 397, 398–9 Kyrie eleison-Christe, du Lamm Gottes, BWV 233a: 287, 290, 305–6, 317 Magnificat in E flat, BWV 243a: 43–4, 132–6, 198, 287, 290, 317 Magnificat in D, BWV 243: 198, 287–8, 318 Mass in B minor, BWV 232: 133, 136, 281, 286, 288, 290, 315, 328, 331–2, 374, 378–92, 399–401 Missa (Kyrie and Gloria), BWV 232I: 133, 198, 222–3, 227, 281, 288–300, 301, 303–7, 317–18, 320, 323, 327–8, 399–400 Symbolum Nicenum (Credo), BWV 232II: 377, 382–7, 394, 399, 400 Sanctus, BWV 232III: 136, 290, 315, 317, 327–8, 387–9, 399 Osanna, Benedictus, Agnus Dei, Dona nobis pacem, BWV 232IV: 198, 276, 294, 314, 388–90, 399–400

430 i n de x o f b ac h ’s w or k s Ouverture in D, BWV 1068: 84, 248, 249, 251, 282, 316, 319 Ouverture in D, BWV 1069: 10, 65, 67, 84, 85–6, 174, 248

4 Missae, BWV 233–6: 223, 225, 233, 287, 289, 299–307, 318, 378 Missa in F, BWV 233: 287, 305–7, 323 Missa in A, BWV 234: 287, 301–3, 313 Missa in G minor, BWV 235: 287, 305 Missa in G, BWV 236: 287, 303–5, 323 motets 198–205, 284–7 Der Geist hilft unser Schwachheit auf, BWV 226: 198, 200, 201–5 Fu¨rchte dich nicht, BWV 228: 198, 199 Ich lasse dich nicht, du segnest mich denn, BWV Anh. III 159: 198 n. 94 Jauchzet dem Herrn, alle Welt, BWV Anh. III 160: 198 n. 94 Jesu, meine Freude, BWV 227: 133, 198–9, 201, 202–5 Komm, Jesu, komm, BWV 229: 198–9, 200–1, 202, 205 Lobet den Herrn, alle Heiden, BWV 230: 198 n. 94 O Jesu Christ, meins Lebens Licht, BWV 118: 284, 286–7 Singet dem Herrn ein neues Lied, BWV 225: 198, 201–2, 203 Musical Offering, see Musicalisches Opfer Musicalisches Opfer, BWV 1079: 330–1, 338–9, 350–2, 363–4, 366–73, 377–8, 393–4, 397–8, 400

partitas, see Clavieru¨bung, Part I; violin solos Pastorella in F, BWV 590: 244–5, 321 5 Praeludien und 5 Fugen, BWV 870a, 899–902: 241–2, 335 Praeludium in C minor, BWV 546 no. 1: 58–9, 60–1, 211, 239 Praeludium et Fuga in G, BWV 541: 24, 263 Praeludium et Fuga in B minor, BWV 544: 58, 59, 61, 62–3, 210, 211, 320 Praeludium et Fuga in C, BWV 545: 263 Praeludium et Fuga in C, BWV 547: 244, 245–6, 320 Praeludium et Fuga in E minor, BWV 548: 58–9, 60, 63, 210, 211, 239, 240, 244, 320 Praeludium et Fuga in A minor, BWV 894: 14, 86, 229, 366, 395 Prelude [Fugue and Allegro] in E ﬂat, BWV 998: 239, 240, 243–4, 320–1 6 Preludes ‘fu¨r Anfa¨nger auf dem Clavier’, BWV 933–8: 238, 241–3 4 Preludes, BWV 939–42: 241–2

oratorios 307–15 Himmelfahrts-Oratorium, BWV 11: 196, 223, 307, 313–14, 318–19, 322, 389 Oster-Oratorium, BWV 249: 67, 109, 136, 162, 217, 223, 307, 314–15, 318–19 Weihnachts-Oratorium, BWV 248: 196, 223, 274–6, 284–6, 288, 294, 301–2, 307, 308–13, 314, 318–20, 322, 391 orchestral suites, see ouvertures organ chorales: Jesu, meine Freude, BWV 713: 203–4 Valet will ich dir geben, BWV 735a: 246 Valet will ich dir geben, BWV 736: 244, 246–7 Wo soll ich ﬂiehen hin, BWV 694: 362 Organ Sonatas, BWV 525–30: 59, 60, 227, 261, 262–6, 267, 273, 317, 321–2, 366 Orgelbu¨chlein, BWV 599–644: 4–5, 158, 206, 207, 236, 334 Ouverture in C, BWV 1066: 10, 39–40, 65, 67, 84, 85, 248 Ouverture in B minor, BWV 1067: 84, 227, 240, 248, 249–51, 316, 319, 322–3, 364

St John Passion, BWV 248: 9, 132, 136–42, 157–61, 181, 183–6, 188–9, 191, 216, 218, 224, 285–6, 306, 327–8 St Mark Passion, BWV 247: 9, 284–6, 114, 213, 223–4, 288, 307, 312, 318, 328 St Matthew Passion, BWV 244: 9, 113–14, 142, 153, 157–8, 176, 181–92, 197–8, 213, 218, 223–4, 244, 272–3, 285–6, 318, 328, 364, 391 Sanctus in D, BWV 232III, see Mass in B minor Sanctus in C, BWV 237: 290, 317 Sanctus in D, BWV 238: 290, 317 Schu¨bler Chorales, see 6 Chora¨le von verschiedener Art Sinfonia in F, BWV 1046a: 66–8, 175 Sonata in G, BWV 1021: 97, 103, 104–5 Sonata in F, BWV 1022: 104 Sonata in E minor, BWV 1023: 103 n. 59 Sonata in G, BWV 1038: 104 Sonatas and Partitas for unaccompanied violin, see violin solos Suite in A minor, BWV 818: 36, 42, 47–9, 53, 56–7, 209

i nd ex o f b a c h’s w o rk s Suite in E ﬂat, BWV 819: 36, 42, 47–9, 54–6 Suite in F minor, BWV 823: 188, 243 Suite in C minor, BWV 997: 188, 239, 243–4, 320 Tilge, Ho¨chster, meine Su¨nden (after Pergolesi), BWV 1083: 328, 374, 376–8, 384, 394 Toccata et Fuga in F, BWV 540: 25, 59, 86, 239 Verschiedene Canones (14 Canons), BWV 1087: 331, 333, 350–2, 397–8 viola da gamba sonatas: Sonata in G, BWV 1027: 208, 363, 364 Sonata in D, BWV 1028: 363, 364–5, 394 Sonata in G minor, BWV 1029: 69, 227, 261, 266, 267–8, 270, 316, 323, 364 violin concertos: Concerto in A minor, BWV 1041: 82, 251–2, 257, 320–1 Concerto in E, BWV 1042: 10, 65, 69, 82–3, 93, 173, 251, 257, 259 Concerto in D minor for two violins, BWV 1043: 82, 251, 252, 256, 320–1 Concerto in D minor (lost original of BWV 1052): 83, 87, 93, 195

431

violin solos: Sei Solo a violino senza basso accompagnato, BWV 1001–6: 37, 86–93, 206, 207 Sonatas 88–91, 98 Partitas 46, 50, 82–3, 91–3, 95, 96–7, 102, 209, 279, 281, 319 violin sonatas with obbligato harpsichord: Sei Sonate a cembalo certato e violino solo, BWV 1014–19: 4, 38, 97–103, 104, 206, 207, 208, 211–12, 270, 272–3, 364 Sonata No. 6 in G, BWV 1019: 51, 101–3, 262, 265, 267–8 Well-Tempered Clavier I, II, see Das Wohltemperierte Clavier I, II Das Wohltemperierte Clavier I, BWV 846–69: 4, 10–11, 15, 17–31, 33–5, 206, 208, 212–13 Preludes 17–18, 22–6, 47, 95, 210, 211, 213, 238, 242 Fugues 26–31, 90, 210, 237 Das Wohltemperierte Clavier II, BWV 870–93: 19, 242, 330–1, 333, 335–46, 395, 397 Preludes 247, 336–40, 348, 395–6, 397 Fugues 245, 341–6, 352, 396–7, 398, 399

General Index Citations of modern writers are invariably located in the footnotes. Abel, Carl Friedrich 364–5 Abendmusik 107 Agricola, Johann Friedrich 5, 8, 12, 102, 243, 248, 336, 369, 381 Ahle, Johann Rudolf 171, 218 Ahrens, Christian 393 Albinoni, Tomaso Giovanni 74, 253, 267 alla breve style 27, 29, 78, 294, 350 Altnickol, Johann Christoph 36, 42, 47, 97, 168, 254, 279, 287, 301, 330, 334–5, 342, 353 Apel, Willi 21 Axmacher, Elke 183 Baal, Johann 289, 317, 378, 399 Bach, Anna Magdalena 4–5, 10–11, 15, 41–2, 49, 51, 93, 102, 104–5, 210, 224, 230, 255, 336 Bach, Carl Philipp Emanuel 5, 8, 97–8, 104, 193, 195, 197, 211, 225, 241, 248, 252–3, 328–30, 355, 371, 392, 394–5 Bach, Johann Bernhard 21, 222, 248–50, 268 Bach, Johann Christian 37 Bach, Johann Christoph 32 Bach, Johann Christoph (Eisenach) 198, 205 Bach, Johann Christoph Friedrich 287, 353–4, 379 Bach, Johann Gottfried Bernhard 253 Bach, Johann Heinrich 98, 101 Bach, Johann Ludwig 8, 169, 177, 289, 317 Bach, Johann Michael 198, 205 Bach, Maria Barbara 5 Bach, Wilhelm Friedemann 4, 8, 15–18, 22–3, 31, 45, 54, 174, 224–5, 242, 252–3, 262, 367 Bassani, Giovanni Battista 299, 379, 381, 399 basso quasi ostinato 105, 254, 280, 364 Beißwenger, Kirsten 32, 37, 55, 84, 222, 223, 233, 248, 289, 299, 328, 329, 333, 378, 379 Benda, Franz 330 Berardi, Angelo 397 Bergel, Erich 354 Berlin 3, 71, 94, 226, 329–30, 367, 395 Besseler, Heinrich 85 Biber, Heinrich Ignaz Franz von 87, 92, 207

Biedermann, Johann Gottlieb 329 binary form: rounded binary 41, 97, 239, 244, 264, 266, 371 with varied reprise 81, 201, 253, 366, 375 Birnbaum, Johann Abraham 226, 233 n. 11 Blankenburg, Walter 8, 169 Blume, Friedrich 391 Bo¨hm, Georg 92, 95 Bonporti, Francesco Antonio 32, 55 Bordoni, Faustina 227, 295, 323 Boyd, Malcolm 68, 70, 71 Boyvin, Jacques 262 Brainard, Paul 110 Breig, Werner 58, 59, 62, 63, 203, 249, 257, 260, 334, 338, 347, 377, 394 Briegel, Wolfgang Carl 171, 218, 224 Brockes, Barthold Heinrich 8, 136, 139, 141–2, 183–4, 218, 224, 328 Brokaw, James A. 336 Bru¨hl, Count Heinrich von, 329 Buelow, George J. 226 Burney, Charles 295 Butler, Gregory G. 38, 232, 234, 254, 259, 327, 346, 354, 355, 358, 360, 361, 368, 381, 393, 396, 398 Butt, John 234, 292, 293, 294, 393 Buttstedt, Johann Heinrich 21–2 Buxtehude, Dieterich 26, 92, 144, 171, 218 Caldara, Antonio 213, 234, 289, 330, 343, 393–4 Calov, Abraham 391 Calvinism 3, 106 canonic variations 331, 333, 350–1, 360–2, 393, 397–8 cantabile style 18, 23–4, 32–5, 44, 46, 49, 53, 209–11, 395–6 cantata: concertante cantata 170, 173–5, 217 Cycle I 8–9, 115–32 Cycle II 8–9, 142–57, 162–7 Cycle III 8, 168–80 Cycle IV 8, 193 Cycle V 8

g e n er a l i n d e x dialogue cantata 127, 131–2, 170–3, 175, 179, 186, 197, 216, 218, 280–1, 283, 316 Picander Cycle 8, 192–7, 216–17, 321 solo cantata 10, 108, 170–1, 213, 217–18, 277–8, 280, 316, 321 chaconne 88, 92, 149, 215, 243, 332, 384, 400 chorale: chorale-aria 157–9, 161, 172–3, 176, 180, 186, 196, 201, 216, 218, 284 chorale motet 147–8, 176, 283–4, 286–7 chorale-trope 127–8, 146, 151–2, 155, 176, 187, 196, 202, 215, 309, 311 chorale cantata 142–57, 284 per omnes versus 145, 175, 196, 215, 217, 282, 284, 316 semi-chorale cantata 282–4, 316 Choreinbau 124–5, 177–8, 196, 312 Christian Ludwig, Margrave of Brandenburg 65, 71 Christian, Duke of Saxe-Weißenfels 67, 109–10, 114, 164, 217, 222, 271, 280, 314–15, 329 Christiane Eberhardine, Electress of Saxony and Queen of Poland 113, 193, 285 Chromatic 4th 63, 149–50, 239, 260, 265–6, 292, 338, 344, 350, 384 see also lamento bass Clement, Albert 231, 346 Collegium musicum 9, 13, 109, 115, 174, 195, 221, 223, 225, 248–52, 258, 267–8, 271, 274, 316–17, 365 concert en ouverture 84, 227, 248–9, 316, 319 concerto: concerto da camera (chamber concerto) 71–2, 86 concerto grosso 71–3, 86, 208, 261, 317, 395 concerto ripieno 70, 208 concerto senza ripieno 72, 86, 208, 253, 255, 317, 319 ensemble concerto 70–1 Corelli, Archangelo 25, 52, 80–1, 88, 91, 98–9, 103, 105, 209–10, 230, 282, 378 Cox, Howard H. 391 Crawford, Tim 374, 376 Crist, Stephen A. 276 Cru¨ger, Johann 202 D’Anglebert, Jean-Henri 16 Dadelsen, Georg von 41, 66, 126, 193, 300, 379, 381 Decius, Nicolaus 186, 191 Dieskau, Carl Heinrich von 374–5 Dieupart, Charles 16, 37–8, 51, 94 Dirksen, Pieter 243, 263

433

dramma per musica 110, 112, 163, 213, 217, 223, 271, 274, 277–8, 314, 318–19 Dresden 6, 10, 70, 209, 222, 226, 243, 258, 287–90, 292, 295, 299, 316–18, 329–32, 346, 380–2 Dreyfus, Laurence 365 Du Mage, Pierre 262 Durante, Francesco 289–90, 292, 317 Du¨rr, Alfred 17, 23, 38, 66, 107, 108, 133, 160, 193, 204, 207, 241, 277, 310, 379 ecclesiastical/church/sacred style 152–3, 162, 170, 172, 197, 213–14, 216, 218, 378 Eller, Rudolf 76, 252, 253, 255 Emery, Walter 229 Empfindsamkeit, empfindsamer Stil (empfindsam style) 321, 330, 338, 342, 369, 372, 393–7, 400 Eppstein, Hans 36, 94, 98, 100, 103, 104, 212, 363, 364 Erdmann, Georg 5, 221 Ernesti, Johann Heinrich 198 Ernst Ludwig, Duke of Saxe-Meiningen 170, 177 Fasch, Johann Friedrich 227 Finke-Hecklinger, Doris 53, 107, 113, 129, 153, 322 Fischer, Johann Caspar Ferdinand 21, 28–9, 84, 92, 212–13, 228, 343, 396 Fischer, Wilfried 83, 254 Fitzpatrick, Horace 67 Flemming, Count Joachim Friedrich von 109–10, 277 Forkel, Johann Nikolaus 8, 13, 37, 43, 97–8, 208, 242–3, 253, 262, 329–30, 346, 348, 367, 375, 397 fortepiano 367, 369–70 Franck, Johann 202 Franck, Salomo 120–1, 136, 183 Franklin, Don O. 31, 336 Fredersdorf, Michael Gabriel 330, 363 Freeman, Daniel E. 128 French style 43, 51–5, 57, 68, 85, 87–8, 91–6, 107, 209–10, 228, 231, 236, 243–4 Frescobaldi, Girolamo 13, 50, 54, 233–4, 245, 355, 369, 397 Friederica Henrietta of Anhalt-Bernburg 5 Friedrich II, King of Prussia (‘Frederick the Great’) 329–31, 363, 366–7, 371–2, 393, 400 Friedrich Augustus I (‘Augustus the Strong’), Elector of Saxony; as Augustus II, King of Poland 109, 271, 274, 276, 287 Friedrich Augustus II, Elector of Saxony; as Augustus III, King of Poland 222, 288–9, 299, 312, 392

434 g e ne r al i n dex Froberger, Johann Jacob 43, 54, 57, 369 Fro¨de, Christine 281 fugue: counter-fugue 236, 283, 303, 335, 356–8 free fugue 27, 95 strict fugue 26–7, 396 Fulde, Johann Gottfried 351 Fux, Johann Joseph 213, 234, 289, 330, 343 galant style 22, 35, 52–3, 56–7, 107, 112, 147, 208–9, 211, 226, 230, 233, 242, 245, 260, 264, 282, 292, 321–3, 328, 330, 347, 363–5, 369, 371–2, 376–8, 384, 393–5, 400 Gassmann, Florian Leopold 380 Geck, Martin 35, 211 Gerber, Christian 9 Gerber, Ernst Ludwig 10 Gerber, Heinrich Nicolaus 10, 11, 15, 33, 41–2, 46, 47 Gerhardt, Paul 191, 197, 199, 285 Gerlach, Carl Gotthelf 226 Germann, Sheridan 65 Girdlestone, Cuthbert M. 57 Glo¨ckner, Andreas 9, 109, 133, 169, 223, 224, 249, 308, 321, 328 Goldberg, Johann Gottlieb 346 Go¨rner, Johann Gottlieb 226 Gottsched, Johann Christoph 109, 113, 193, 213, 285, 313 Graaf, Jan Jacob de 229 Gramann, Johann 202 Graun, Carl Heinrich 226–7, 328, 330–1, 394 Graun, Johann Gottlieb 330, 394 Graupner, Christoph 6 Greer, Mary Dalton 392 Griepenkerl, Friedrich Conrad 252–3 Grigny, Nicolas de 61–2, 233, 262 Grubbs, John W. 329 Gru¨ss, Hans 85 Hamburg 5, 22, 88, 90, 98, 136, 183, 226, 255 Hammerschmidt, Andreas 171, 218 Handel, George Frideric 27, 123, 222, 224, 243, 328, 330–1, 393–4 Harrer, Gottlob 329 Hasse, Johann Adolf 222, 226–7, 289, 295, 330, 394 Hauser, Frank 285 Haydn, Joseph 123, 380 Heder, Samuel Gottlieb 12 Heermann, Johann 191

Heinichen, Johann David 22, 289 Heller, Karl 252, 253, 255 Helms, Marianne 301 Hennicke, Johann Christian, Count von 278 Herz, Gerhard 126 Heyden, Sebald 161, 191 Hildebrandt, Zacharias 243 Hobohm, Wolf 8, 169 Hofmann, Klaus 18, 71, 200, 263, 267, 270, 277, 280, 335, 337 Horn, Victoria 61 Horn, Wolfgang 290 Hunold, Christian Friedrich 107, 113, 183, 218, 224 Hurlebusch, Conrad Friedrich 55, 226, 234, 393 Hutchings, Arthur 46, 266 improvisatory style 13–14, 336 integration 77, 212 inventio 5, 32, 33, 55, 210, 251 Italian style 38, 51–4, 56–7, 68–9, 85, 87–8, 91–3, 95, 107, 209–10, 228, 244–5, 319 Jacob, Andreas 53, 54 Janovka, Toma´sˇ Baltazar 21 Johann Ernst, Prince of Saxe-Weimar 88, 229 Jones, Richard D. P. 31, 49, 51–2, 73, 108, 207, 277, 340–1 Josquin Desprez 379 Kan, Rebecca 83 Kast, Paul 333 Kauffmann, Georg Friedrich 234, 246, 393 Kayser, Bernhard Christian 10, 17–19, 33, 36, 41, 47, 96 Keiser, Reinhard 8, 224, 328, 330, 393–4 Kellner, Johann Peter 59, 103, 239, 244 Kevorkian, Tanya 7 Keyserlingk, Heinrich Christian von 346 Kilian, Dietrich 14, 263 Kindermann, Johann Erasmus 21 n. 23 Kirchhoff, Gottfried 22 Kirnberger, Johann Philipp 20 Kittel, Johann Christian 244 Kittel, Johann Heinrich 20–1 Knauer, Johann 119 Knu¨pfer, Sebastian 144 Kobayashi, Yoshitake 59, 96, 287, 328, 333, 336, 350, 355, 379, 389, 392 Koch, Ernst 311

g e n er a l i n d e x Kortte, Gottlieb 111 Krausse, Helmut K. 119 Kra¨uter, Philipp David 84 n. 31 Krebs, Johann Ludwig 59, 64, 225, 253 Krebs, Johann Tobias 12, 59, 64 Krieger, Johann 21 n. 23 Krieger, Johann Philipp 120, 144 Kropffgans, Johann 243 Krummacher, Friedhelm 144, 145, 160, 171 Kuhnau, Johann 6, 7, 9, 19, 50, 95, 120, 144, 212, 229, 329 Kusser, Johann Sigismund 84 Ku¨ster, Konrad 169, 173, 177 lamento bass 140, 149, 260, 332, 384, 400 Lange, Gottfried 120 Lautenwerk (lute-harpsichord) 243–4 Le Roux, Gaspard 37 Leaver, Robin A. 145, 289, 392 Ledbetter, David 18, 20, 21, 25, 29, 87, 90, 92, 94, 210 Lehms, Georg Christian 170–2, 217 Leisinger, Ulrich 13, 160–1 Leopold, Prince of Anhalt-Co¨then 3, 5, 10, 22, 106, 114, 118, 126, 163, 213, 223, 225, 235, 297 Linigke, Christian Bernhard 94 Locatelli, Pietro Antonio 222 Lombard rhythm 235, 278, 313 Lotti, Antonio 233, 289, 317, 393 Lully, Jean-Baptiste 27, 41, 42, 46, 67, 84, 85, 88, 92, 96, 107, 149, 163, 209, 215, 228 Luther, Martin 127, 133, 145, 147, 148, 155, 162, 191, 217, 283, 291, 306, 308, 309, 386, 391 n. 38, 392 Marais, Marin 84 Marcello, Alessandro 81, 90, 229 Marcello, Benedetto 229, 393 Marchand, Louis 209, 316 Marissen, Michael 67, 68, 159, 184, 269, 368 Ma¨rker, Michael 171 Marpurg, Friedrich Wilhelm 20, 26–7, 42–3, 353 Marshall, Robert L. 134 Mattheson, Johann 20–2, 31, 35, 47, 52–4, 63–4, 88–90, 171, 208–9, 211, 224, 393, 396, 398 Maul, Michael 226, 380 Meiningen 8, 169–70, 177–9 Melamed, Daniel R. 184, 190, 199, 286 Mendel, Arthur 160, 161, 295, 327 Meyer, Ulrich 246 Michel, Johann Heinrich 343

435

Mietke, Michael 65 Milka, Anatoly P. 379, 399 mixed style, see vermischte Geschmack Mizler, Lorenz Christoph 226–7, 233, 321, 331, 351, 354, 361 monodic style 179 Monteverdi, Claudio 134 motet style 114, 124, 133, 147–8, 166, 176, 178, 187, 214, 281, 292, 295, 296, 303, 305, 306, 310 Muffat, Georg 84, 92, 245 Mu¨ller, August Friedrich 111 Mu¨ller, Heinrich 183 Na¨geli, Hans Georg 390 Neidhardt, Johann Georg 20 Neumann, Werner 276, 330, 374, 375 Neumeister, Erdmann 7, 119, 120, 146, 214 Newman, William S. 98 Nicolai, Philipp 173, 280, 283 Niedt, Friedrich Erhard 31, 53 Nottebohm, Gustav 355 n. 36 Obituary, Bach 5, 8–9, 248, 354–5, 368, 376, 379 n. 12 O’Donnell, John 64 Oley, Johann Christoph 229 n. 2 operatic style 9, 107, 120, 146–7, 152, 162, 167, 172, 214, 292, 294–6, 314 Osthoff, Wolfgang 380 Ovid (Publius Ovidius Naso) 271 Pachelbel, Johann 21, 32, 144, 236, 335 Paczkowski, Szymon 298 Palestrina, Giovanni Pierluigi da 233–4, 289, 343, 379, 382, 399 Passion: oratorio-Passion 9, 136, 218, 224, 285, 307, 314, 318, 328 Passion-oratorio 183–4, 218, 224, 319, 328 pasticcio Passion 328–9 Penzel, Christian Friedrich 36 Peranda, Marco Giuseppe 289, 317 perﬁdia 76, 78, 93, 103, 252, 254, 256, 258, 265 Pergolesi, Giovanni Battista 328, 332, 376–8, 384–5, 394, 400 Pez, Johann Christoph 289, 317 Picander (Christian Friedrich Henrici) 8–9, 109–10, 181, 183–8, 191, 192, 194–7, 216–18, 224, 271–2, 277, 285–6, 307, 321, 329, 374 Pisendel, Johann Georg 70, 87, 207, 209

436

g en e r a l in d ex

Platen, Emil 301 Plath, Wolfgang 16, 18, 31 Postel, Christian Heinrich 136 Potsdam 329–31, 363, 367, 369, 371–2, 393 Praetorius, Michael 198 Quantz, Johann Joachim 209, 322, 330, 394 Questenberg, Count Johann Adam von 380 Raison, Andre´ 262 Rameau, Jean-Philippe 22, 44, 56–7 Rampe Siegbert 257, 314 Rasch, Rudolf 20 Rathey, Markus 161 Reincken, Jan Adams 255–6 Rempp, Frieder 101, 102 Reusner, Adam 285 Reutter, Georg 380 Riedel, Friedrich Wilhelm 355 Rifkin, Joshua 84, 85, 107, 160, 161, 163, 174, 181, 196, 248, 249, 250, 256, 260, 278, 281, 291, 293, 294, 295, 314 Ringk, Johannes 103 Rist, Johann 127, 311 Roger, Estienne 84 Rost, Johann Christoph 9 Sackmann, Dominik 16, 88, 257, 314 Saxony, Electoral House of 109, 113, 213, 222–3, 271, 274, 276, 288–9, 299, 301, 309, 318–19, 329, 375, 391 Scarlatti, Alessandro 75, 170, 213, 259, 280, 289, 290, 393 Scarlatti, Domenico 53, 239, 266, 321, 338, 348, 395, 397 Schalling, Martin 139 Scheibe, Johann Adolph 84, 226, 229, 233, 238, 248, 262, 316, 321 Scheibel, Gottfried Ephraim 120, 129, 174, 216–17 Scheide, William H. 8, 169 Scheidt, Samuel 233–4, 355 Schein, Johann Hermann 144, 202 Schelle, Johann 7, 144, 200 Schmelzer, Johann Heinrich 87, 207 Schmidt, Ferdinand 380 Schneider, Johann 226 Schott, Georg Balthasar 109 Schu¨bler, Johann Georg 361 Schulenberg, David 35, 43, 45, 48, 55, 57, 347

Schulze, Hans-Joachim 76, 110, 120, 145, 199, 227, 255, 256, 258, 259, 269, 288, 289, 313, 374, 375, 376, 397 Schu¨tz, Heinrich 6, 171, 179, 198, 218 Schwanenberger, Georg Heinrich Ludwig 93 Selle, Thomas 136 serenata 107–8, 110–12, 119–20, 129, 163, 171–2, 194, 197, 213–14, 275 Shabalina, Tatiana 163, 193, 194, 224, 286, 328 Sheldon, David A. 394 Sicul, Christoph Ernst 133 n. 36 Siegele, Ulrich 104, 209, 253, 316, 366 Silbermann, Gottfried 367, 369–70 Smend, Friedrich 107, 197, 202, 286, 390 sonata: sonata da camera 25, 50, 88, 91, 93, 95–6, 209 sonata da chiesa 25, 52, 87, 88, 98–100, 103, 210, 262, 364, 394 Sonate auf Concertenart (sonata in concerto style) 101, 227, 261–70, 316, 364 Sorge, Georg Andreas 20 Spitta, Philipp 12, 194 Stauffer, George B. 12, 61, 245, 290, 292, 293, 295, 380, 382 Steffani, Agostino 84, 222 Steger, Adrian 7 stile antico 234, 236, 238, 292, 322–3, 343, 362, 378, 382, 386, 394, 396, 400 Stinson, Russell 59, 103, 244, 245, 334 Stockigt, Janice B. 290 Stockmann, Paul 138, 158, 285 Sto¨lzel, Gottfried Heinrich 119, 223–4, 226, 319, 328, 331 Stradella, Alessandro 213 Strohm, Reinhard 222 Stu¨bel, Andreas 145 style luthe´ 37, 43, 44, 46, 339 Suppig, Friedrich 22 Swack, Jeanne R. 262 Sweelinck, Jan Pieterszoon 54 Synofzik, Thomas 21, 241 Talbot, Michael 68, 69, 70 Talle, Andrew 10, 17 Telemann, Georg Philipp 6, 8, 9, 45, 70, 74, 84–5, 88, 109, 120, 169, 221, 223–4, 226–7, 229, 248–9, 316, 319, 329–31, 393–4 temperament 20, 208 Theile, Johann 322

gener al i nd ex Thymich, Paul 200 Torelli, Giuseppe 56, 70, 74, 229, 253, 267 Tovey, Donald Francis 31, 259, 272, 299, 340, 342, 356, 358, 384, 385 Trautmann, Christoph 391 Venturini, Francesco 70 vermischte Geschmack (mixed style) 39, 43, 52, 55, 85, 87, 92, 94, 209, 316, 319, 342 Vinci, Leonardo 394 Virgil (Publius Vergilius Maro) 110, 111 Vivaldi, Antonio 13, 70–2, 74, 77, 79, 81, 83, 90, 208, 225, 229, 253–4, 257–8, 262, 268, 289, 317, 323, 393 Vogler, Johann Caspar 10, 41–2, 46, 47, 263 Vokaleinbau 82, 122 Volumier, Jean Baptiste 209 vox Christi 120, 122, 130, 132, 153–4, 156, 165–7, 172, 177, 179, 180, 197, 214, 218, 281, 305 Wagner, Georg Gottfried 20 Walker, Paul 292, 322, 382 Walther, Johann 145 Walther, Johann Gottfried 18, 21, 37, 43, 112, 226, 234, 319, 378 Walther, Johann Jakob 87, 207 Weise, Christian 136

437

Weiß, Silvius Leopold 243 Weiss, Wisso 287 Weiße, Michael 138 Weißenfels 108, 162 Werckmeister, Andreas 18, 20 Westhoff, Johann Paul von 87, 207 Whaples, Miriam K. 128 Wilderer, Johann Hugo von 289, 291, 317 Wiemer, Wolfgang 356 Williams, Peter 12, 62, 63, 65, 203, 233, 234, 245, 246, 335, 393 Wolf, Uwe 13, 238, 280 Wolff, Christoph 120, 222, 229, 232, 234, 291, 292, 319, 322, 334, 343, 346, 367, 368, 369, 377, 379, 382, 384, 390, 394, 400 Wollny, Peter 66, 82, 108, 120, 195, 197, 223, 328, 329, 334, 366, 379, 381, 395 Zachow, Friedrich Wilhelm 95, 144 Zarlino, Gioseffo 292 n. 29, 322 Zehnder, Jean-Claude 56 Zelenka, Jan Dismas 234, 289–90, 330, 343 Ziegler, Mariane von 164, 166–7 Zipoli, Domenico 245 Zimmermann, Gottfried 221 Zohn, Steven 84, 248

The Creative Development of Johann Sebastian Bach, Volume II.pdf - PDFCOFFEE.COM (2025)