Number of Cantatas
Discussions - Part 2

Continue from Part 1

Missing Bach Cantatas--possible locations and slightly off topics

William Rowland (Ludwig) wrote (December 16, 2015):
I have found indications that Bach's Cantatas in the original manuscript forms long ago reached other parts of Europe--such as Monasteries and Palaces of Austria among other locations. It is said that Bach wrote 400 Cantatas of which we have approximately only 200 currently known to exist. 'New' Organ works are constantly being found and published by Peters.

Recently I was going through my collection of the complete Symphonies of Joseph Haydn (whose brother was Michael--and who worked with religious groups most of his career as a performing musician as well as composer) and came across an astounding symphony (of the earlier London series) that bears an astounding similarity to parts of BWV 8. Most notably in the use of the repeated flute notes of BWV 8 and the use of material that seems to be from the spurious BWV 15--both Funeral Cantatas.

As most of Haydn scholars and fans know--he worked most of his life for the Prince Esterhazy family and whose library contains (or did) a great amount of music scores. Based on the fact that Haydn was probably very familiar with BWV 8---there are possibilities that other scores of the Cantatas exist in the Esterhazy collections--yet undiscovered as well in the Monastic and Church libraries where Michael Haydn also worked.

A diligent search through the libraries and private collections throughout Europe and through the auspices of such auction and publishing houses as Peters, Southeby's and Park Benet could turn these up but because cause of Wars and social unrest since the time of Bach--we also must consider that these may have been lost forever and if folks like Haydn's wife got a hold of any of them--well they ended up in the fireplace or as her curlers.

Question--what really happened to Mozart's body after famous thunderstorm? According to German and Austrian Customs--most probably Mozart may have been temporarily buried but when no one came to pay the grave digger or the church--it was dug up and thrown away. Even if it was given a proper burial--failure to pay the upkeep of the grave results in the body being dug up and remains dumped. Supposedly this is because land space is very limited. One must remember that Mozart was heavily in debt at the time of his death.

At the time of Mozart's death--Medical students needed bodies for Anatomical studies. Grave Robbers in cahoots with Churches and Grave diggers often were only too happy to supply the schools with bodies. It was this very same thing that descrated Haydn's tomb--when some pseudo-scientists dug him up and beheaded him because they thought that criminality etc could be determind by the shape of ones skull and face--something that was once used in Police work as late as the 1900s in the United Stats.

Claudio Di Veroli wrote (December 16, 2015):
Ludwig wrote:
< 'New' Organ works are constantly being found and published by Peters. >
Indeed, the list of JSB works rediscovered within the last 40 years or so is fascinating, though perhaps the frequency, inevitably, is dwindling:

(I abridge from Wikipedia and other websites)

BWV 1087 – 14 "Goldberg" Canons. rediscovered in 1974.
BWV 1088 – Arioso for Bass
BWV 1089 – Da Jesus an dem Kreutze stund (organ chorale)
BWV 1090-1120 – 31 "Neumeister Chorales" for organ red. in 1985
BWV 1121 – Fantasie in C minor (organ)
BWV 1122 – 1125 Four organ Chorales in Dietel copy.
BWV 1126 – Lobet Gott, unsern Herrn (organ chorale)
BWV 1127 – Aria "Alles mit Gott" for soprano and strings red. in 2005.
BWV 1128 – Wo Gott der Herr (organ chorale) red. in 2008.

Googling BWV 1129 and higher produces webpages unrelated to new Bach works.

Luke Dahn wrote (December 16, 2015):
[To Claudio Di Veroli] Variations of this statement are frequently made: "It is said that Bach wrote 400 Cantatas of which we have approximately only 200 currently known to exist." In most cases, the statement is made matter-of-factly, with the assumption that it is common knowledge.

I am interested in knowing and studying the evidence for this 400 number. Is it simply assumed that the Nekrolog is trustworthy when it states that there were 5 complete cantata cycles? To what extent does the entire weight of the 400 figure rest on the reliability of the Nekrolog? In other words, what other secondary sources give evidence to the missing cantatas? Are there any scholars who question such a high number of total cantatas?

If anyone has input here or can point me to other scholarly sources, I would be grateful.

Evan Cortens wrote (December 16, 2015):
[To Luke Dahn] Agreed. The only detailed scholarly discussion of this issue I'm aware of was a brief colloquy between William Scheide and Alfred Dürr, published in Die Musikforschung in 1961, with Scheide suggesting that the "5 cycles" number ought not to be taken at face value.

Here are the citations:

Scheide, William H. "Ist Mizlers Bericht über Bachs Kantaten korrekt?" Die Musikforschung 14, no. 1 (Jan–Mar 1961): 60–63.
Dürr, Alfred. "Wieviele Kantatenjahrgänge hat Bach komponiert?" Die Musikforschung 14, no. 2 (Apr–Jun 1961): 192–95.
Scheide, William H. "Nochmals Mizlers Kantatenbericht—Eine Erwiderung." Die Musikforschung 14, no.4 (Oct–Dec 1961): 423–27.

I certainly question the figure--it seems to me rather unlikely that such a vast quantity of music would go missing with virtually no trace. Five cycles gives us roughly 300 sacred cantatas, of which we have roughly 200. Could a handful, even a dozen, have gone missing? Possibly--but 100? It seems a bit of a stretch.

William L. Hoffman wrote (December 16, 2015):
The myth began with Forkel quoting "five complete sets of church music"; the original Obituary simply says "five sets of church pieces." Chorale cantata cycle 2 distributed only about 45 pieces (none for Easter season), third cycle is incomplete and includes the Easter Season pieces, mostly von Ziegler.

Stephen Clarke wrote (December 16, 2015):
[To Luke Dahn] Yes, interested here, too. My first wish would be to have those missing 200 found. My second wish would be for to not be necessary.

sa'a naghai bikhe' hozho

Claudio Di Veroli wrote (December 16, 2015):
[To Evan Cortens] I also agree. It is not surprising that they have mostly found organ chorales, which easily end up in lost copies in church and other archives. Cantata scores are larger things and, should they exist, they would be easier to find, not more difficult.

Evan Cortens wrote (December 16, 2015):
[To William L. Hoffman] A quick bit of additional context here.

=====

The original statement in the obituary (Nekrolog), printed in Mizler's Musikalische Bibliothek (1754, p. 168) is as follows:
"Die ungedruckten Werke des seligen Bachs sind ungefehr folgende:
1) Fünf Jahrgänge von Kirchenstücken, auf alle Sonn- und Festtage."

The New Bach Reader (p. 304) translates this as follows:
"The unpublished works of the late Bach are approximately as follows:
(1) Five full annual cycles [Jahrgänge] of church pieces, for all the Sundays and holidays;"

Forkel, in 1802, writes as follows (p. 61):
"V. Sing-Compositionen:
1) Fünf vollständige Jahrgänge von Kirchenstücken auf alle Sonn- und Festtage"

The New Bach Reader (p. 472) translates as follows:
"v. Vocal Compositions
(1) Five complete sets of Church Music [cantatas] for all the Sundays and Holidays of the year."
=====

All this to say, I don't think we can blame Forkel here for adding "complete" ("vollständige"), I think it's fair to read the obituary's "annual cycles" ("Jahrgänge") as implying that they are complete.

Luke Dahn wrote (December 16, 2015):
[To Evan Cortens] Thank you, Evan and all. I too question the high figure. My skepticism is primarily based on my study of the chorales. I personally seriously question, even seriously doubt, that all or even most of the isolated chorales BWV 253-438 which came to us by way of CPE's compilation of chorales, originally came from cantatas now lost. I am sure there are a few chorales among these that originally come from cantatas, but I believe that number to be quite low. The patterns that emerge among these isolated chorales just don't lend support, in my opinion, to the idea that they originally came from chorales -- that is, unless Bach drastically changed his approach to chorale tune selection for the later cantata cycles. (I could go into describing these "patterns" further if it interests anyone.) Rather, I believe that most of these chorales either came from didactic sources, or from a now-lost compilation of chorales that JS Bach himself had made (a hypothetical source which the NBA editors discuss and label as source [Y] if memory serves me correct) from which he possibly drew chorale settings when composing his cantatas. (This would explain why several of the chorales in CPE Bach's edition that do come from extant cantatas are in keys different than their appearances in those cantatas. Bach, drawing from his chorale compilation, would have to transpose the chorale to the key of the cantata, while CPE Bach drew from the original compilation.)

For what it's worth...

Douglas Cowling wrote (December 16, 2015):
Chorales for Weekday masses and offices, and Sunday mass for Choirs III & IV

Luke Dahn wrote:
< Rather, I believe that most of these chorales either came from didactic sources, or from a now-lost compilation of chorales that JS Bach himself had made (a hypothetical source which the NBA editors discuss and label as source >
I think a good case can be made that the chorale movements contain specific liturgical items like the Te Deum which could very well have been used at weekday offices. A complete German Mass can be reconstructed as well, and this could have been used at weekday liturgies on feast days when no cantata was mandated. This "German Mass" would have been useful for Bach's Choir III and IV which had no concerted or polyphonic repertoire. I've never seen a detailed study of the music of these choirs, or of the weekday services which must have been substantial.

Luke Dahn wrote (December 16, 2015):
[To Douglas Cowling] Thanks for this. One of the reasons the individual chorales "don't fit" his cantata chorales as a group is that a very high percentage of the tunes in these individual chorales do not appear anywhere else in the extant choral works of Bach. They are "one-offs", so to speak. Thus, the chorale tunes among these individual chorales were apparently less common or from less well known sources. (One quick example: Six chorale tunes by Matthäus Apelles von Löwenstern appear among these individual chorales while not one of Löwenstern's tunes appear in the chorales from the extant choral works.)

Contrast that with the chorales from the extant cantatas and passions where there are very few "one-offs". Bach returned to the same tunes again and again for his cantatas and passions, presumably tunes that the congregation would be well familiar with.

Could there be a reason why Bach might turn to lesser known tunes for the specific liturgical occasions that Doug mentions? These individual chorales which include numerous "one-offs" seem to be of a different nature, which might give plausibility to the idea of a liturgical occasion that is similarly of a different nature.

Douglas Cowling wrote (December 16, 2015):
Chorales for Weekday masses and offices, and Sunday mass for Choirs III & IV

These individual chorales which include numerous "one-offs" seem to be of a different nature, which might give plausibility to the idea of a liturgical occasion that is similarly of a different nature.

The best examples of these "Liturgical Chorales" are the Te Deum and Credo which are unlike any other of the "Cantata Chorales". They are extended multi-sectioned works that are quite challenging to sing. I could see Choir I or II singing them on a weekday celebration of a saint's day. We forget that St. Thomas operated like a modern English cathedral with daily choral services. With nothing to back up my opinion, I'm convinced that the chorale collections need to be placed in the context of a major ecclesiastical operation.

William L. Hoffman wrote (December 17, 2015):
Adding to Luke Dahn's comments, the so-called "free-standing" or "unattached" chorales increasingly are seen as part of Bach's "well-ordered church music to the glory of God" as well as a teaching instrument. For the latter, there are recent discoveries of chorale collections of Johann Ludwig Dietel, Christoph Friedrich Penzel, and the so-called Bach-Choral Buch of 1740. As to the former, there are some fine harmonizations of Michael Weisse and his Bohemian Brethren and in my recent studies of the late Trinity Time chorales, there are several free-standing ones while Trinity 27 and 26 rarely occurred anywhere, Bach, having composed Cantata 140 (Tr. 27) and J.L. Bach 3 (2/10/1726) to satisfy his needs.

There is only one complete Bach cantata cycle, No. 1. No. 2, chorale cantatas, has only some 46 actual chorale cantatas, while Cycle 3 (1725-27) is an enigma that involves 18 J.L. Cantatas. Bach also possessed and performed in the middle 1730s, two cycles of Gottfried Heinreich Stoezel. As to the so-called Picander cycle (No. 4, 1728-29), there are some nine works extant. As for the the "fifth" Christoph Wolff suggested in 1982 Bachjahrbuch short article, "Wo bleib Bachs Fuenfer Kantatenjahrgang?, that Bach's first cycle includes a "double-annual cycle (mini-cycle) of two part (BWV 75, 76, 21, 147, 186, 70, 63) cantatas and double-performances (BWV 24+185, 179-199, 181-18, 22-23, A.199-182, 31-4, 172-59, and 194-165).

IMVHO the fifth cycle is a christological cycle of vocal works, mostly oratorios, composed 1725-35.

Most telling, Wolf says (p. 151), that to what extent every one of the five cycles collectively at any time had been completely altered cannot be determined. It is quite plausible that smaller and larger gaps endure in all the cycles (the interruption of the chorale cantatas at Easter 1725).

Thus there are three ways of looking at the chorale cantata cycle: The original some 40 composed 1724-25; the addition of the undesignated and filler pieces (1725-35), and the some 46 actually distributed to Friedemann (scores) and Anna Magdalena (parts) in 1750.

Julian Mincham wrote (December 17, 2015):
"Missing" Bach Cantatas

[T Evan Cort]Separating what we know from what we can surmise, it is certain that the idea of the 'five full cycles' originated in the Obituary, published by CPE Bach and Johann Agricola in 1750. Forkel simply rode on the back of this.

HOWEVER it is also generally accepted that not all aspects of this document are fully accurate. There are two possible reasons for this:

1 CEP was putting his father in the best possible light. He omitted aspects of Bach's wrangling with authorities and other aspects of his life which might not enhance his reputation. This is only natural. This was an obituary written by a son fully aware of his father's talents, not a critical review.

2 With regard to the cantata cycles, CPE was a boy around 8 years old when JS settled in Leipzig. He is now writing over 30 years later, years in which his own life had its tensions and demands. It is more than possible that a) he might not have noted certain things accurately as a child and b) time may have edited certain memories over the years.

Thus we should not simply swallow what he says uncritically and decloine to subject it to scrutiny because of who he was.

I think it very likely that he was not fully aware that so many of the early cantatas were reformed for use in the first Leipzig cycle. He may well have seen bundles of scores and parts labelled misleadingly in his father's study. Furthermore he may not have been fully aware that his father was making use of other composer's works after the chorale cycle, as he (JS) appeared to lose interest in churning out a new cantata every week.

Put these together and you have a compelling scenario in which CPE might have counted the Weimer cantatas as cycle 1, the first two years at Leipzig produing cycles 2 and 3 and two further cycles as 4 and 5 (as set out by Wolff as cycles 3 4 on pages 281-285, these latter being filled out with cantatas by other composers (we know of the 15 by Johann Ludwig Bach used in the 1725-6 cycle).

Now we cannot be sure of this although the theory strikes me as being much more compelling than, for example, the Andreas Stübel theory of expiring luricists! It would suggest that there is nothing like 100-200 missing cantatas (which I personally find hard to believe) and it fits the facts, few though they be.

William L. Hoffman wrote (December 17, 2015):
[To Julian Mincham] The Obituary, while rather vague about specific compositions and numbers has one interesting fact: All of the types of works listed have been found to exist, especially "No. 2, Many oratorios etc." No type of work listed has not been found. The possible exception is the Weimar 1713 birthday ode, "Alles mit Gott," although Obituary No. 2 includes "music for birthdays." Also, the works accounting of Agricola and Emmanuel, begun in the fall of 1750 with Friedemann in Leipzig, probably following his father's directions (Wolff), went to the shelves in the music room and started sorting the manuscript bundles for distribution to family members (Gerhard Herz). The topical outline is found in Emmanuel's estate catalogue of 1790.

Douglas Cowling wrote (December 16, 2015):
William L. Hoffman wrote:
< Also, the works accounting of Agricola and Emmanuel, begun in the fall of 1750 with Friedemann in Leipzig, probably following his father's directions (Wolff), went to the shelves in the music room and started sorting the manuscript bundles for distribution to family members (Gerhard Herz). >
Wolff is pretty convincing that Sebastian had already made the decisions about legacies, even to the point of commissioning copies which he thought both CPE and WF should have. What is not clear is what source of income he envisioned for his widow.

Julian Mincham wrote (December 17, 2015):
[To William L. Hoffman] Its a good point although I wonder if CPE was always quite clear about the numbers of works in each category--i.e. the numbers of cantatas or, indeed the oratorios and masses about which he is (perhaps deliberately) vague.

Luke Dahn wrote (December 17, 2015):
[To Douglas Cowling] Does Wolff specify (i.e. did Johann specify) precisely which pieces he thought CPE and WF should have? The copying of manuscripts for this purpose might account for some of the chorale setting duplicates found in the Breitkopf chorale collection. (There are a couple other possible explanations, but this might be a very logical one.)

Also, the patterns that emerge in the Breitkopf collection itself coincides perfectly with the idea of sorted "manuscript bundles". The collection of 371 chorales can be divided into three sections (first section: 1-125; second section: 126-247; third section: 248-371), with the middle section being so distinctly different from the other two in this regard: a very high percentage of the chorales from this middle section are individual chorales belonging to no extant larger work (e.g. cantata, passion), chorales which were eventually assigned an individual BWV number (BWV 253-438).

Here are the statistics:
Section 1: Chorales #1-125 (125 chorales) -- Only 32 (25.6%) are isolated BWV 253-438 chorales
Section 2: Chorales #126-247 (122 chorales) -- 101 (82.8%!) are isolated BWV 253-438 chorales
Section 3: Chorales #248-371 (124 chorales) -- Only 59 (47.6%) are isolated BWV 253-438 chorales

One can easily picture the following scenario: The original compiler(s) of the Breitkopf chorale edition, whoever they may be (e.g. CPE, Marpurg), mined a "bundle" of large choral works for the Section 1 chorales. Then, they turned to another source, perhaps a didactic source or separate compilation of chorales that Johann himself had made (the hypothetical source I mentioned earlier), for Section 2 chorales. Finally, they turned to yet another "bundle" of large choral works for Section 3 chorales.

The patterns in the Breitkopf chorale compilation (and in others like the Dietel collection, a section of which seems to feature chorales in sequential order of the liturgical year) are too significant to ignore, in my opinion. And I think there is still much that we can learn through further investigation into such patterns.

William L. Hoffman wrote (December 17, 2015):
[To Douglas Cowling] By donating the chorale cantata parts sets to the Thomas School, Anna Magdalena was able to stay in the family residence for another six months before vacating. Beyond this, she and her eldest daughter had no provisions and they lived in virtual poverty. I believe Emmanual did provide some financial assistance, besides possibly foreswearing his portion of the chorale cantatas to his step mother.

Kim Patrick Clow wrote (December 17, 2015):
Julian Mincham wrote:
<
Separating what we know from what we can surmise, it is certain that the idea of the 'five full cycles' originated in the Obituary, published by CPE Bach and Johann Agricola in 1750. Forkel simply rode on the back of this. 2 With regard to the cantata cycles, CPE was a boy around 8 years old when JS settled in Leipzig. He is now writing over 30 years later, years in which his own life had its tensions and demands. It is more than possible that a) he might not have noted certain things accurately as a child and b) time may have edited certain memories over the years. >
But CPE didn't have to rely on his childhood memory. Johann Sebastian Bach had only died the year previous to the obituary, so there had been plenty of opportunities for Bach to discuss with his son(s) the nature of his musical legacy. Also, the notion it seems unlikely there could be so much missing music from Bach has come up several times, in other words, there couldn't be a large number of missing Bach cantatas.

But there aren't many Baroque composers who have had most of their musical legacies (or subsets of that legacy) survive intact. In terms of baroque German cantata composers, Christoph Graupner (due to his unique circumstances in Darmstadt) is the only exception I can think of. Johann Friedrich Fasch, in Zerbst, composed nearly 12 cantata cycles, but only about 120 survive (it's theorized his son, Carl Friedrich Christian Fasch, the founder of the Sing-Akademie zu Berlin, may have destroyed most of the scores, and it's not clear what happened to the performance parts in Zerbst). Gottfried Heinrich Stoelzel composed 1.358 cantatas but only 605 have survived (most of them destroyed by his successor in Gotha, Georg Anton Benda, who by the way, who himself wrote a lage number of cantatas that haven't survived). Telemann suffered significant losses as well, including several cantata cycles before 1740 (his grandson Georg Michael Telemann, the music director of Riga Cathedral, purchased a large part of Georg Philipp's estate from his relatives (was there family drama?) and he apparently threw out large batches of the cantatas because they were, in his view, out of date). The kapellmeister in Weimar while Bach was employed there, Johann Wilhem Drese: not a single cantata survives ( there was a massive fire in the Weimar castle in the 1780s, and it destroyed the music archives). So within the historical context of Bach's life, a potential loss of 140- 200 Bach cantatas wouldn't be that unusual. Christoph Wolff suggests there are at least 20 missing Bach cantatas from Weimar, not an insignificant number. And I've been always struck by how close we came to losing some of the cantatas, based on reading Thomas Braatz's write-ups on the cantatas' provenance, it's fascinating to see how many of them survived by the slenderest of threads due to a single copy.

Fascinating thread.

Evan Cortens wrote (December 17, 2015):
[To William L. Hoffman] This raises an interesting point about inheritance, ownership and (to be a bit anachronistic) intellectual property. On the one hand, clearly Bach thought of his cantatas (by which I primarily mean the physical musical materials, rather than the more abstract notion of an 'intellectual work') as his property, and dispensed of them as he wished. This despite the fact that he had beenin the employ of the Thomasschule for some 27 years, and of course cantatas in Weimar and Mühlhausen (etc) before that.

On the other hand, we have Christoph Graupner. He was nearly an exact contemporary (born two years earlier) and with a similar kind of training and employment, and died just ten years after Bach. Yet there was no sense by his employer that the cantatas he wrote in Darmstadt (over 1,400 across five decades) belonged to him--rather the court argued that since he was paid to write them for the court, they belonged to the court. Graupner's heirs, perhaps familiar with Bach's situation, but certainly familiar with the general practice (they wrote to G. A. Benda for support, for instance), clearly expected that they would inherit the music and were surprised they didn't, and appealed to the court to be paid for the value of the music.z

Kim's message came in while I was typing this. Indeed, it is the very fact that the heirs were prevented from inheriting the music that means it survives almost entirely intact in one place (a couple stolen works excepted--one at Yale, another in Paris). What was unfortunate for the heirs is something that is fortunate for us today.

Evan Cortens wrote (December 17, 2015):
Kim Patrick Clow wrote::
< But CPE didn't have to rely on his childhood memory. Johann Sebastian Bach had only died the year previous to the obituary, so there had been plenty of opportunities for Bach to discuss with his son(s) the nature of his musical legacy. Also, the notion it seems unlikely there could be so much missing music from Bach has come up several times, in other words, there couldn't be a large number of missing Bach cantatas. >
Actually even closer than that! Bach died in July of 1750, and the obituary must have been written by the end of 1750 in order for Mizler to have received it when he said he did (March 1751, in Warsaw; see NBR, p. 297n28).

Setting aside his cataloguing abilities ca. 1750, by the end of his life (late 1780s), C. P. E. Bach was exemplary in this regard. The Nachlassverzeichnis (Estate Catalogue) of 1790, already mentioned by William L. Hoffman, is a fabulous and generally very accurate catalogue of the music in CPEB's possession at the time of his death, including many works by JSB.

Another thing alluded to by Kim is: what exactly is the value of this music anyway?

For Endler in Darmstadt (Graupner's successor), Benda in Gotha (Stölzel's successor) and G. M. Telemann in Riga (not G. P. Telemann's successor, but in possession of many of his church works--see the letters between him and CPEB), this was not some abstract question. The value of the music was in its ability to be reused or reperformed. If you couldn't do that (e.g., because it was old-fashioned), then it wasn't particularly valuable at all--so why not save some shelf space and just junk it?

Of course, it is over the late eighteenth and early nineteenth centuries that this attitude begins to shift. CPEB's own large 1786 concert is often cited as one of the watershed moments. It featured some of the best of the old (excerpts from Handel's Messiah and JSB's B minor mass, both some four decades old) and the new (CPEB's double-choir Heilig, clearly one of his favourite works).,

Ed Myskowski wrote (December 17, 2015):
Evan Cortens wrote:
< Kim's message came in while I was typing this. Indeed, it is the very fact that the heirs were prevented from inheriting the music that means it survives almost entirely intact in one place (a couple stolen works excepted--one at Yale, another in Paris). What was unfortunate for the heirs is something that is fortunate for us today. >
A neat point, thank you for making it!

Peter Smaill wrote (December 18, 2015):
Of some interest in this thread is the discovery by Michael Maul (Bach-Archiv, Leipzig) of an account of a funeral in Leipzig which makes specific reference to a chorale harmonised by Bach, namely BWV 1122, Denket doch, ihr Menschenkinder". It's one of a small group of chorale settings whose attribution was doubted but is now certain. A most beautiful harmonisation, that and other lately catalogued/discovered chorales can be heard at: https://www.youtube.com/watch?v=7QC-t4AJCY4

It follows, as Luke I think suggests, that the existence of many chorale settings with no surviving related Cantata does not suggest the absence of Cantatas, but rather occasional use by Bach. Some settings are of hymns associated to weddings, for example.

One might expect the settings for Cantatas to be much more sophisticated than these unattached pieces. If anyone has pondered that question, I'd like to find out if any conclusion is possible!

Julian Mincham wrote (December 18, 2015):
[To Evan Cortens] Given that CPE was best placed to list his father's music, the fact remains that he is imprecise about much of it in the obituary----'many oratorios, various concertos, a mass of other instrumental pieces, some double-chorus motets' etc etc. Whether he knew exactly what JS had produced is a matter of debate and this allows us to raise the question, in the light of no other corroborating evidence, is his assertion about five complete cantata cycles correct and upon what is it based? Whilst we know of a number of individual works that disappeared or for which only the text survives (see Dürr) where is the evidence of one or two complete missing cycles? When would they have been written? How might they have fitted in with Bach's diminishing interest in weekly cantata production particularly after 1727?

With reference to the Weimar cantatas ( and of course my argument does not depend on CPE's childhood memory ------- he was only an infant at that time) he may, at the time of Bach's death, have been unaware that his father reused so many of them in the first Leipzig cycle. However, these were not, for the most part, just taken off the peg and reperformed. Bach altered the instrumentation, added movements and often transposed the keys. This would have produced two sets of parts which the cantor, no doubt would have meticulously labelled and set aside. It would be easy to assume that this entailed two cycles when the second is, to a large extent a reuse of the first.

I guess many have the romantic hope that another cantata (let alone a full cycle) may yet turn up in some Eastern European catacomb, especially since the end of the cold war. But for me the evidence falls on the side of an error in the obituary---which, in one sense is good news as it would indicate that far fewer great works are missing than was once thought.

But I am happy to look at any corrobotating evidence for CPE's statement. Bring it forward!

Douglas Cowling wrote (December 18, 2015):
Peter Smaill wrote:
< Of some interest in this thread is the discovery by Michael Maul (Bach-Archiv, Leipzig) of an account of a funeral in Leipzig which makes specific reference to a chorale harmonised by Bach, namely BWV 1122, Denket doch, ihr Menschenkinder". It's one of a small group of chorale settings whose attribution was doubted but is now certain. A most beautiful harmonisation, that and other lately catalogued/discovered chorales can be heard at: >
That suggests that a concerted effort to look for liturgical context rather than sexier 'lost cantatas" would be more fruitful. For instance, the three free-standing chorales with obligato horns are settings of the three traditional wedding chorales. This is a dissertation topic!

Julian Mincham wrote (December 18, 2015):
[To Douglas Cowling] As Peter indicated, it is not a useful exercise to attempt to trace lost cantata though unattached or individual chorales.

Luke Dahn wrote (December 18, 2015):
[To Julian Mincham] Right. But, as per my earlier comments about the distinct nature of these isolated chorales taken as a whole, I think we can say more than simply "it's not useful." I have argued that we can say with some confidence that it's likely that most of them did NOT come from cantatas that are now lost. That's different than simply saying it's not a useful exercise. In other words, my suggestion is that the nature of the cis counter-evidence to the idea that all or most of these chorales once belonged to cantatas. (To be clear, this does not necessarily mean that additional cantata cycles never existed. It only removes any consideration that these individual chorales provides evidence to that fact.)

I'm quite interested in Doug's mention of Maul's reference to a specific occasion for BWV 1122, which is indeed a beautiful setting of a tune possibly by Bach himself. The tune is certainly not a congregational chorale with it's large leaps (two leaps of a seventh and one octave) and very wide range. It reminds me of the "chorale" (which Bach labels "aria") that ends the motet "Komm, Jesu Komm!", BWV 229.2. Most of the isolated four-part chorale settings, however, are more congregational in nature (i.e. more singable by the non-professional). I am very interested in any information regarding the possible purpose, occasion and origin of these individual chorale settings.

(Doug, I believe you may have forgotten to include a URL in your last note.)

Julian Mincham wrote (December 18, 2015):
Luke Dahn wrote:
< Right. But, as per my earlier comments about the distinct nature of these isolated chorales taken as a whole, I think we can say more than simply "it's not useful." >
Just to set the record straight, I said that is was 'not a useful exercise to attempt to trace lost cantata though unattached or individual chorales'. It is, of course valuable to track and study them for other reasons.

Ed Myskowski wrote (December 18, 2015):
Julian Mincham wrote:
< I guess many have the romantic hope that another cantata (let alone a full cycle) may yet turn up in some Eastern European catacomb, especially since the end of the cold war. But for me the evidence falls on the side of an error in the obituary---which, in one sense is good news as it would indicate that far fewer great works are missing than was once thought. >
As is most often the case, good scholarship coincides with (not so) common sense, and Julian expresses it as well as anyone.

Douglas Cowling wrote (December 18, 2015):
Luke Dahn wrote:
< Right. But, as per my earlier comments about the distinct nature of these isolated chorales taken as a whole, >
Here is a quick outline of Bach's "German Mass" according to Luther's ordo, assembled from the Riemenschnieder collection:

Introit Hymn: (seasonal)
Kyrie: 132. "Kyrie,Gott Vater in Ewigkeit" (extended choir setting)
Gloria: 125, 249, 313, 326, "Allelin Gott in der Höh"
Hymn de Tempore: (seasonal)
51, 160, 288."Gelobet seist du" (Christmas)
15, 184, 261, 371."Christ Lag in Todesbnden" (Easter)
197. "Christ ist Erstanden" (choir setting)
Credo: 133. "Wir Glauben" (choir setting)
Chancel Hymn: (seasonal)
Sanctus: 319. "Sanctus, Sanctus, Sanctus"
Agnus Dei: 165, "O Lamm Gottes" (choir setting)
Communion Hymns: (seasonal)
Dismissal: 32, 330. "Nun Danket Alle Gott"

The next step would be to look at Bach's hymn book and identify which settings can be identified for the variable hymns of the church year. I suspect that this woud account for more than half of the chorales in the collection. The above outline offers an impressive musical repertoire for Choirs 2 – 4 on Sunday and for all weekday masses that were sung. The items marked "Choir setting" are quite challenging music.

Luke Dahn wrote (December 18, 2015):
Julian Mincham wrote:
< Just to set the record straight, I said that is was 'not a useful exercise to attempt to trace lost cantata though unattached or individual chorales'. It is, of course valuable to track and study them for other reasons. >
Right, Julian. I think I just misunderstood you, or at least read something into your note that was unintended. Didn't mean to come off as overly defensive.

Luke Dahn wrote (December 19, 2015):
[To Douglas Cowling] Below is your list with a few things added: BWV numbers are provided for each of the Riemenschneider chorales, additional chorale settings of these chorale tunes are listed in bold, an equal sign is inserted when two Riemenschneider settings are duplicates, and finally the tune locations in the Neu Leipziger Gesangbuch. (It should be noted that some of these Riemenschneider chorales come from cantatas, as can be seen from the Schmieder numbers.)

Introit Hymn: (seasonal)
Kyrie: 132 (BWV 371). "Kyrie,Gott Vater in Ewigkeit" (extended choir setting) (NLGB p.423)
Gloria: 125=326 (BWV 104.6), 249 (BWV 260), 313=353 (BWV 112.5) "Allein Gott in der Höh" (NLGB p.425)
Hymn de Tempore: (seasonal)
51 (BWV 91.6), 160 (BWV 64.2), 288 (BWV 314). Dietel 123 (BWV 248.28) "Gelobet seist du" (Christmas) (NLGB p.30)
15 (BWV 277), 184 (BWV 4.8), 261 (BWV279), 371 (BWV 278), Dietel 24 (BWV 158.4)."Christ Lag in Todesbanden" (Easter) (NLGB p.272)
197 (BWV 276), Dietel 79 (BWV 66.6). "Christ ist Erstanden" (choir setting) (Dietel 79 is much shorter, less elaborate) (NLGB p.290)
Credo: 133 (BWV 437). "Wir Glauben" (choir setting) (NLGB p.501)
Chancel Hymn: (seasonal)
Sanctus: 235=319 (BWV 325). "Sanctus, Sanctus, Sanctus" (NLGB: not present)
Agnus Dei: 165 (BWV 401), "O Lamm Gottes" (choir setting) (NLGB p.173)
Communion Hymns: (seasonal)
Dismissal: 32 (BWV 386), 330 (BWV 252), BWV 79.3 (not in Riemenschneider or Dietel). "Nun Danket Alle Gott" (NLGB p.638)

A table giving the location of chorales in the NLGB can be found at the following link, a table which can be sorted by Riemenschneider, BWV number, and a variety of other criteria.
http://www.lukedahn.net/ChoralesInVopelius.htm

I'm not sure how one goes about determining whether a setting is considered a "variable setting". Numerous hymns in the NLGB contain the words "Ein anders", but I presume that only indicates that the hymn is an alternate setting for the particular liturgical occasion under which it is listed. By "variable setting" you mean a tune that can serve multiple liturgical occasions, correct?

Julian Mincham wrote (December 19, 2015):
Luke Dahn wrote:
< Right, Julian. I think I just misunderstood you, or at least read something into your note that was unintended. Didn't mean to come off as overly defensive. >
No problem. Just did not want to be misunderstood.

Douglas Cowling wrote (December 19, 2015):
Luke Dahn wrote:
< I'm not sure how one goes about determining whether a setting is considered a "variable setting". Numerous hymns in the NLGB contain the words "Ein anders", but I presume that only indicates that the hymn is an alternate setting for the particular liturgical occasion under which it is listed. By "variable setting" you mean a tune that can serve multiple liturgical occasions, correct? >
The project is daunting. Bach had a seven-day a week job in Leipzig. (Wolff outlines Bach's amazing first Christmas on p.265)

You would need to set up a 365 day grid for every year Bach was in Leipzig, enter the mandated services for each day, enter the chorales from the Leipzig hymn book, and then the dated choral works. Will Hoffman has done great for the Sundays, but there is still slog work to be done for the 8 or 9 services for which Bach has responsibility each (I think I produced a weekly schedule for this group a few years ago.)

William Rowland (Ludwig) wrote (December 24, 2015):
Luke Dahn wrote to Evan Cortens:
< Thank you, Evan and all. I too question the high figure. My skepticism is primarily based on my study of the chorales. I personally seriously question, even seriously doubt, that all or even most of the isolated chorales BWV 253-438 which came to us by way of CPE's compilation of chorales, originally came from cantatas now lost. I am sure there are a few chorales among these that did originally come from cantatas, but I believe that number to be quite low. The patterns that emerge among these isolated chorales just don't lend support, in my opinion, to the idea that they originally came from chorales -- that is, unless Bach drastically changed his approach to chorale tune selection for the later cantata cycles. (I could go into describing these "patterns" further if it interests anyone.) Rather, I believe that most of these chorales either came from didactic so, or from a now-lost compilation of chorales that JS Bach himself had made (a hypothetical source which the NBA editors discuss and label as source [Y] if memory serves me correct) from which he possibly drew chorale settings when composing his cantatas. (This would explain why several of the chorales in CPE Bach's edition that do come from extant cantatas are in keys different than their appearances in those cantatas. Bach, drawing from his chorale compilation, would have to transpose the chorale to the key of the cantata, while CPE Bach drew from the original compilation.)
For what it's worth... >
We Bach lovers just need a more through search of the world's libraries,auction houses, palaces and monasteries to turn up more 'lost' works as well as private collections.

Kim Patrick Clow wrote (December 24, 2015):
[To Ludwig] There's been a massive effort to do just that, under the lead of Christoph Wolff, e.g. advertisements have been placed in German newspapers in most cities and towns, encouraging citizens to inspect their homes, attics, etc for any possible unknown music manuscripts and report their discoveries. RISM (Répertoire International des Sources Musicales) has been active for decades (since 1952) in creating a massive database of music manuscripts in state, church and monastic libraries.

The biggest hope of finding things seems to rest with misidentified or unidentified scores and or parts, or things that weren't cataloged properly. That was the case for the long missing Missa sopra Ecco sì beato giorno composed by Florentine Renaissance composer Alessandro Striggio (c. 1536/1537 – February 29, 1592). The mass was composed for a monumental 72 individual parts. The mass vanished for 400 years; and Irish musicologist Davitt Moroney only found the music after years of looking in the Paris Bibliothèque Nationale. The parts were mislabeled with the wrong composer's name (Alessandro Strusco), the wrong number of voices (4), and it also was mis-shelved. You can watch a mini documentary about the rediscovery of the Striggio mass @: https://www.youtube.com/watch?v=4ls_9id5ba4

The rediscovery of music by other composers has occurred largely in a similar manner, so I remain hopeful that some missing Bach could turn up in a similar manner.

Kim Patrick Clow wrote (December 24, 2015):
The mass was written for 60 voices. Not 72. My apologies.

Continue on Part 3

Number of Cantatas: Part 1 | Part 2 | Part 3 | Lost Bach Cantatas: Part 1 | Part 2