Beethoven's sketch leaves are so numerous and at times so impenetrable that they are bound to throw up the occasional surprise when studied in detail. Around ten thousand such pages survive; they are generally difficult to decipher, and often more than one work appears on a single page. Nevertheless, few surprises will have been as unexpected as the identification of a fragment of a little-known Bach cantata amongst Beethoven's early sketches.

By the time Beethoven moved from Bonn to Vienna in November 1792 he had already filled many pages of loose sketch leaves and other manuscripts, and he took a large bundle of them with him. He gradually added to the collection up to 1798, but after that he made most of his sketches in actual manuscript sketchbooks, while any made on loose leaves were kept separate from the earlier collection (apart from three leaves from 1799–1800 added around that time).Footnote ¹ The collection remained more or less intact until his death, but thereafter some leaves or part-leaves were removed, and the remainder was eventually split into two batches. One batch ended up in the British Library and is known as the Kafka Miscellany; the other, known as the Fischhof Miscellany, went to what is now the Staatsbibliothek zu Berlin – Preussischer Kulturbesitz.Footnote ² The Kafka Miscellany was published in facsimile and transcription in 1970; the Fischhof Miscellany is less well known, though it is the subject of a detailed study by Douglas Johnson and has also been published in microfiche and more recently online.Footnote ³

The two batches contain a great variety of material, including autograph scores, sketches for works, keyboard exercises, ideas for works not completed and occasionally some material in another hand. Amongst this last group is folio 36 of the Fischhof Miscellany; Johnson comments that ‘the leaf was originally used for the violin part to an unidentified overture (not in Beethoven's hand)’.Footnote ⁴ The recto side is the title page for this overture and originally contained just blank staves plus the words ‘Violino 1mo’, while the verso repeats this heading and adds ‘Ouvertüre’, plus the actual violin part, which is in C major (Example 1). The part consists of twenty-one bars culminating in a perfect cadence and double bar, followed without a break or separate heading by a fresh time signature and a further thirty-four bars to the foot of the page, and it must clearly have continued on other leaves now lost. The scribe appears to have been a professional copyist, and his addition of the figure ‘21’ at the end of the first section suggests that payment was per bar. The hand does not match that in other Beethoven-related manuscripts from the period, such as that of Copyist F (Johann Baptist Paraquin), who copied several of Beethoven's Bonn works in the early 1790s;Footnote ⁵ but it is the hand of a known Bonn copyist, which ‘also appears prominently in orchestral parts for the 1793 production of Die Zauberflöte, as well as other locally copied music from the 1780s’.Footnote ⁶ The paper was described by Johnson as rather coarse, with no visible watermark element, and does not exactly match any other known leaves used by Beethoven in Bonn before his move to Vienna; but its stave ruling, ten staves per side, is very similar to that in several other of these leaves (notably folio 130 of the Kafka Miscellany), and it may even have been ruled with the same rastrum.Footnote ⁷

Example 1 Opening of an unascribed ‘Ouvertüre’, bars 1–4, transcribed from Staatsbibliothek zu Berlin – Preussischer Kulturbesitz, Beethoven Autograph 28 (Fischhof Miscellany), fol. 36v. Used by permission

Into the empty space on the recto of the leaf Beethoven has added numerous short sketches, mainly for his song Der freie Mann (WoO117). These must have been written in 1792, for the text of the song did not appear until that year (or the end of the previous one);Footnote ⁸ on the other hand, once Beethoven reached Vienna in November he began using Viennese paper. Moreover a complete score of Der freie Mann is also written on Bonn paper,Footnote ⁹ and the sketches clearly preceded this.

Since manuscript paper was typically used up quite quickly, and the paper type and stave ruling are similar to some of Beethoven's other Bonn papers, the violin part for the ‘Ouvertüre’ was almost certainly written out in Bonn, not long before 1792, although it could have been a slightly older leaf. It was clearly written out in preparation for an actual performance rather than merely for study, since it is an individual part copied professionally, and one must assume that the other parts were also written out. Thus this work was probably performed in Bonn, or at least tried out in rehearsal, some time in or before 1792. The parts must then have been discarded that year, at which time Beethoven, always short of manuscript paper, appropriated the first page of the first violin part so that he could use the empty staves on the recto for sketches. He may previously have played the violin part himself and acquired it that way, although he more often played viola in the court orchestra in Bonn.

This ‘Ouvertüre’ can now be identified as the beginning of Bach's cantata Der Herr denket an uns, bwv196, though whether Beethoven knew that the cantata was by Bach is uncertain, for the violin part gives no indication of composer. That the hitherto unidentified ‘overture’ turns out to be a Bach cantata rather than an instrumental work from Beethoven's own period is surely a great surprise. It is even more remarkable when viewed in the light of an inventory of 1794 (though begun earlier) that lists over three thousand works in manuscript or print available at the Bonn court. The inventory includes not a single work by J. S. Bach and only two by C. P. E. Bach, though quite a number by J. C. Bach. Among its subdivisions, such as ‘liturgical church music’ (which would all be in Latin), various types of opera, collections of lieder and several genres of instrumental music, no category would comfortably accommodate Lutheran church music, and no Lutheran cantatas are listed.Footnote ¹⁰

Moreover, Der Herr denket an uns is actually one of the least likely of Bach's cantatas to appear among Beethoven's manuscripts, for it is a very early work for which no autograph score survives, and there is only one other known source from the entire eighteenth century. This is a score copied by Johann Ludwig Dietel (1713–1773),Footnote ¹¹ one of Bach's main copyists. Andreas Glöckner has observed that Dietel's handwriting changed repeatedly while he was working with Bach, and that the score shows a hand later than a specimen of 27 August 1731 but earlier than one of 25 March 1732. The manuscript can therefore be dated to around the turn of the year 1731–1732, and is the earliest known complete score in Dietel's hand.Footnote ¹² No set of parts for the cantata is known, and the Beethoven leaf is the only known specimen of an instrumental part for it.

Dietel's score later entered the library of the composer and theorist Johann Philipp Kirnberger (1721–1783), a great admirer of Bach. From thence it passed to the library of Princess Anna Amalia of Prussia (1723–1787), youngest sister of Frederick the Great. She was also a composer and collector of music sources, especially of Bach works. Her library is now in the Staatsbibliothek zu Berlin – Preussischer Kulturbesitz, and the score of Der Herr denket an uns (Am.B. 103) is bound between those of two unrelated Bach cantatas, bwv16 and 64, in the composite volume Am.B. 102–104. Der Herr denket an uns appears on folios 9–16, which are numbered in pencil as pages 1–16. The title in the manuscript, ‘Der Herr denket an uns, a 4. Voci: 2 Violini, una Viola, Violoncello: Basso: e Continuo. di Sig: J. S. Bach’, is written on page 1, above the score. Two other scores of the work, both from the nineteenth century, also survive, but these are of little significance.Footnote ¹³

The cantata consists of five movements: an opening ‘Sinfonia’ (called ‘Ouvertüre’ in Beethoven's manuscript) of twenty-one bars, a four-part chorus (forty-two bars), a soprano aria (thirty-one bars including a da capo), a duet for tenor and bass (seventy-two bars) and a final chorus (fifty-nine bars). Beethoven's manuscript includes the whole Sinfonia plus the first thirty-four bars of the second movement, and so the Bonn performance that presumably took place was not just of the instrumental movement, but of at least one and probably all of the vocal movements. The text of the cantata is taken entirely from a single psalm – Psalm 115, verses 12–15. After the opening sinfonia, each movement corresponds to one verse of the psalm, as follows:

This cantata is itself something of a puzzle, for its origin and purpose are shrouded in mystery, although there is no reason to doubt the ascription to Bach. It is one of the few Bach cantatas that have yet to be dated precisely, but the style indicates that it was composed during his early period and probably dates from c1708–1714, when he was in Mühlhausen and subsequently (from July 1708) in Weimar, whereas most of his cantatas date from his first few years in Leipzig, in the 1720s. It is noteworthy that there are no recitatives, as there are in most of his later cantatas. Use of consecutive verses from a single psalm is another characteristic associated with his earliest cantatas, from Mühlhausen, although this cantata is the only one that relies exclusively on just one psalm.Footnote ¹⁴ Other features also indicate an early date, as suggested by Wilhelm Rust in 1864 in the first published edition of the work.Footnote ¹⁵ On the other hand, the appearance of a da capo aria with strongly Italianate features (the third movement) points to Bach's Weimar period rather than to Mühlhausen.Footnote ¹⁶ Why Dietel should make a copy in 1731–1732 is unknown, and it may not have been for Bach's use, since the paper type is different from any in other Bach manuscripts.Footnote ¹⁷

Rust published Der Herr denket an uns as the middle work in a group of three ‘Trauungs-Cantaten’ (wedding cantatas), sandwiching it between two others (Dem Gerechten muss das Licht, bwv195, and Gott ist unsre Zuversicht, bwv197). Ever since then it has been almost invariably described as a wedding cantata, but this description must be questioned. Rust made the claim on the basis of the text, especially its last two verses, which he claimed ‘point decisively to a wedding celebration’ (‘deuten entschieden auf eine Trauungsfeierlichkeit’).Footnote ¹⁸ But this passage from Psalm 115, like the psalm as a whole, has no known associations elsewhere with weddings. If a psalm text were wanted for a wedding cantata, several others would be more suitable, such as ‘Instead of thy fathers thou shalt have children’ (Psalm 45, verse 17) or ‘Thy wife shall be as the fruitful vine’ (Psalm 128, verse 3). Moreover, whereas Dem Gerechten muss das Licht is actually entitled ‘Copulations Cantata’ and Gott ist unsre Zuversicht is headed ‘Trauungs-Cantate’, and both have texts explicitly referring to a wedding, Der Herr denket an uns has no such heading, nor any reference to a couple or a wedding. Thus it is highly doubtful that it was ever intended for one. The nearest it comes to implying a wedding is its very general reference to ‘you and your children’ in verse 14, which is hardly convincing evidence. Even more speculative is the suggestion that the cantata was composed for the wedding of Pastor Johann Lorenz Staube and Regina Wedemann, the aunt of Bach's first wife Maria Barbara, on 5 June 1708. This suggestion, first put forward by Philipp Spitta in 1873 and since repeated by numerous other writers,Footnote ¹⁹ assumes not only that the reference to ‘your children’ implies a wedding, but that mention of the ‘House of Aaron’ implies one for a pastor, on the grounds that Aaron was a priest. There is no other evidence supporting this hypothesis. In sum, although the cantata would not be completely inappropriate at a wedding, it cannot legitimately be called a wedding cantata, any more than much other music that might occasionally be performed at weddings. It could have been composed for any joyful ceremony, such as a thanksgiving or a festival involving a blessing.

The appearance of such an obscure and early Bach cantata in Bonn around 1790 seems almost incredible, for Bach's cantatas were very little known at that date. Only one, Gott ist mein König, bwv71, was published during his lifetime, and there were exceptional circumstances for this. The first Bach cantata published after his death did not appear until 1821, when Ein feste Burg, bwv80, was issued by Breitkopf & Härtel.Footnote ²⁰ Manuscript copies were almost as rare, for the main Bach works in circulation in manuscript after his death were keyboard works, notably Das wohltemperirte Clavier, which Beethoven had already encountered as a boy. However, between 1760 and 1780 the publisher Breitkopf in Leipzig made available for purchase manuscript copies of numerous Bach compositions. These included no fewer than thirty-seven cantatas, but Der Herr denket an uns was not one of them.Footnote ²¹ Another publisher, Johann Christoph Westphal of Hamburg, also sold copies of about ten Bach cantatas between 1776 and 1792, but again not Der Herr denket an uns.Footnote ²² The appearance of part of this cantata in the Fischhof Miscellany therefore indicates that possibly several more of Bach's works were in circulation than previously thought, and that his cantatas were able to spread to previously unsuspected locations.

How the cantata arrived in Bonn is a matter of speculation, but one possibility is through Dietel himself. After moving from Leipzig to the nearby town of Falkenhain, he and his father supplemented their income there as cantors during the 1750s through extensive copying activities, and advertising ‘musicalia’ such as funeral motets and keyboard music.Footnote ²³ It may have been through this means that Kirnberger acquired the score of bwv196, if he did not obtain it in Leipzig. In addition, a second copy (see below) could have been obtained by someone who later took it to Bonn. The most likely figure here is Beethoven's teacher Christian Gottlob Neefe, who lived in Leipzig from 1769 to 1776 before moving to nearby Dresden and eventually to Bonn.

Also a matter of speculation is the reason why the cantata was apparently copied in Bonn. Copies of Bach cantatas made available in Leipzig and Hamburg could have been used locally for Lutheran services, since both cities were largely Lutheran; but the Bonn court was a Roman Catholic community, and could not be expected to use a Lutheran cantata in German in a liturgical context. One can only conjecture that some religious celebration, or even a concert that was not specifically religious, took place there on one occasion when a work with a German psalm text would have been considered appropriate, and that someone had access to Der Herr denket an uns. Since the work probably was performed in Bonn, it is more than likely that Beethoven took part, either on the violin, the viola or more probably the keyboard continuo. Thus this is the first evidence we have that he may have once played in a Bach cantata. There is, however, no obvious musical connection between Der Herr denket an uns and Beethoven's own two cantatas composed in 1790, on the death of Joseph II and on the accession of Leopold II. Beethoven clearly did not use the Bach cantata as a model.

The musical text of the violin part in the Beethoven leaf is similar to that in the extant score copied by Dietel, but there are a few variants and a few errors in common. Both sources have the last three notes of bar 10 a third too low and both omit a flat in bar 16, note 4. These errors also appear in the two nineteenth-century sources.Footnote ²⁴ In addition, the Beethoven leaf has the following variants from the Dietel score:

Thus the Beethoven leaf is less accurate than the Dietel score, but more accurate than the nineteenth-century copies, which have several additional errors in this section of the first-violin part.Footnote ²⁵ The leaf appears not to have been copied direct from the Dietel score, since the change of title and the added slur in bar 6 would be difficult to account for; but all other variants could be explained as simple misreadings or notational change, and so it was probably derived via an intermediate source. Such a source (the score supposedly taken from Falkenhain to Bonn) would probably have had a mark in bar 6 that was misread as a slur, and have lacked a title for the first movement, so that a fresh (and incorrect) one had to be added for the Beethoven leaf. Alternatively, both the leaf and Dietel's score could have derived from a slightly imperfect copy of the autograph, where three notes had been copied too low, as occurs also in the two nineteenth-century scores.

In later life Beethoven made copies of several Bach excerpts, but these were mainly keyboard works, including the Chromatic Fantasy, some movements from Das wohltemperirte Clavier and a passage from Die Kunst der Fuge.Footnote ²⁶ He also owned a copy of Kirnberger's treatise Die Kunst des reinen Satzes, which contains numerous passages from works by Bach .Footnote ²⁷ He even quoted one of these, the opening of the ‘Crucifixus’ in Bach's Mass in B minor, in a letter of 15 October 1810, in which he asked Breitkopf & Härtel for a complete score of the work (which unfortunately had not been published at the time).Footnote ²⁸ Kirnberger's only excerpt from a cantata, however, was a single cadence from the chorale at the end of Sehet, wir gehn hinauf gen Jerusalem, bwv159, quoted on page 159 (coincidentally) of volume 1 of his treatise.Footnote ²⁹ If this was the sum total of Beethoven's later encounters with Bach's cantatas, then the Fischhof leaf forms a considerable addition. At the very least, however, it indicates that he was familiar with a greater range of Bach's music, and at an earlier date, than previously suspected. It also shows that Bach's cantatas, or at least Der Herr denket an uns, circulated more widely in the late eighteenth century than was known hitherto; and the leaf provides a significant new source for the work itself.