Why aren't the sets of Bach's parts corrected?

Bradley Lehman wrote (September 16, 2006):
Hypothesis: the sets of parts (usually only one of each exist, more rarely two) were exemplars. Whether they were actually used in the performance by one or more players, we don't know. But, the thrust of my hypothesis: the student(s) assigned to play/sing each movement would have--as part of the assignment--the task of making HIS OWN copy of the whole thing, by hand, as part of the learning process. And then keep it afterward, not file it back away with the master set of parts and score, in Bach's folders.

That's how Bach taught the WTC and other keyboard music to his keyboard students: they started by making their own hand-copy from the master score. Why would it be any different for the cantatas, given sufficient lead time for the students to do this?

This was also Bach's OWN practice in educating himself about other people's music: start by making his own copy of it by hand.

And yes, we must first discard the over-romanticized notion of the cantatas being written only as late as a couple days before the gig.

Note that this is independent of the number of singers or players assigned to perform any given part. Within this hypothesis, we can't use the number of extant parts (master copies) to extrapolate the number of singers/players, either to many or few; it's an unknown.

Corrections to the master parts, if any, would tend not to get written down: that particular part might not be on anyone's music stand at all, if the players were playing from their own copies (where any corrections would get written in -- the copies they kept for themselves afterward). The master part itself might be back in the library during the performances, having served its purpose as exemplar....

Bradley Lehman wrote (September 16, 2006):
Bradley Lehman wrote:
< Hypothesis: the sets of parts (usually only one of each exist, more rarely two) were exemplars. Whether they were actually used in the performance by one or more players, we don't know. But, the thrust of my hypothesis: the student(s) assigned to play/sing each movement would have--as part of the assignment--the task of making HIS OWN copy of the whole thing, by hand, as part of the learning process. And then keep it afterward, not file it back away with the master set of parts and score, in Bach's folders.
That's how Bach taught the WTC and other keyboard music to his keyboard students: they started by making their own hand-copy from the master score. (...) >
p.s. Note that this hypothesis also offers an explanation why some of the blisteringly difficult continuo parts remain unfigured in the official part, for the archive files with the score in the library.

Within this model, there's no compelling reason for them be figured: the guy playing them would be working from his own copy anyway, where he could write out the thing on one or two staves (whatever he needed, at his skill level) and work out what he was going to play, across the week (or more) of rehearsals with his colleagues. The unfigured exemplar part would already be back in the library by that time, not needed for further reference. If the player really got stuck, during the process of working out his own private playing part, he could go look at the score for further information about the necessary or implied harmonies at any spot.

An example, the one that got me thinking along these lines: my own private printed copy of Richard Mix's realization for the BWV 211 aria already has a set of my own penciled changes. It's sitting on my harpsichord at the moment, and I might eventually file it away somewhere or I might lose it. If I have opportunity sometime to perform this piece, and if I remember that I have this sketch available, I'll consult it...and probably change it further at the time. My own changes and suggestions wouldn't necessarily get back into Richard's master score, either today or years from now; why would they need to? The surviving performance materials on paper diverge every time a piece is seriously worked on by intelligent musicians. That's a natural dynamic process for learning music. It blows away the concept of "Urtext", but so what? That was always a chimera anyway: the notion of fixing some single "best" version that is to become the standard for all time forward, for all players, for all acoustical situations!

What I would actually play in performance depends on many things: the quality of the instrument given, the skills of the singer, the choices of tempo, the acoustics of the space, the presence or absence of a bowed bass-line player, and any inspiration of the moment. The more ideas that come up during rehearsals, the better to craft an effective performance. Every time I play through the thing, I improvise different details anyway, and this is normal. If I have to deal with somebody else's less-palatable tuning, or a sticking note on the instrument, or a too-quiet instrument, or any other hazards beyond immediate control, I improvise around them with more or fewer notes...always with an ear to making the particular situation sound as well as can be done.

Douglas Cowling wrote (September 16, 2006):
Bradley Lehman wrote:
< Hypothesis: the sets of parts (usually only one of each exist, more rarely two) were exemplars. Whether they were actually used in the performance by one or more players, we don't know. But, the thrust of my hypothesis: the student(s) assigned to play/sing each movement would have--as part of the assignment--the task of making HIS OWN copy of the whole thing, by hand, as part of the learning process. And then keep it afterward, not file it back away with the master set of parts and score, in Bach's folders. >
This is an attractive hypothesis to account for mutliple copies but I can't imagine that the extra copies would be tossed -- the cost of paper was pretty stiff. It is possible that the extra copies were stored away from the exemplars and disposed of over the centuries. Still it's odd that not even one such external copy surivives when throwaway items like the cantata libretti do.

I do think we underestimate the importance of copying music as a pedagogical tool. With the ubiquity of published music today, the notion of learning through transcribing your own copy has all but vanished. And yet it was the standard method until the end of the 19^th century. Perhaps the only place left is the teaching of harmony where the student has to transcribe the melody or figured bass. Over the years I have transcribed hundreds of pages of music and, although it's dog-work, I have always found that I learned and appreciated the music on a level much deeper than just opening up a score and performing.

Thomas Braatz wrote (September 16, 2006):
Bradley Lehman wrote:
>>Hypothesis: the sets of parts (usually only one of each exist, more rarely two) were exemplars. Whether they were actually used in the performance by one or more players, we don't know. But, the thrust of my hypothesis: the student(s) assigned to play/sing each movement would have--as part of the assignment--the task of making HIS OWN copy of the whole thing, by hand, as part of the learning process. And then keep it afterward, not file it back away with the master set of parts and score, in Bach's folders.<<
Reasonable objections to Brad's hypothesis:

1. Paper was expensive!

Consider how Bach made use of any spare space in his scores and on the original parts (he would copy a different part for a different singer/player on the back of another part, yet consider that this single part would be usable during a performance by both instruments!). The various physical arrangements of his autograph scores are testaments to his ingenuity in eking out every possibility to use whatever space was available to him.

Now imagine Bach telling his Thomaner: "Hey kids! You'll need to go out and buy yourself a large supply of paper because every week you will need to copy all of your practice parts from my original parts which I will not let you take home. I am currently planning to compose some very difficucantatas based on the cantata texts that I have had printed. There will be no way for you to simply sight-read these difficult parts. When you do your own copywork, I will first put my original copy part under glass over here at this desk because I do not want you to create any extraneous ink spots or smudges on my original that might be construed as notes that will be incorrect for the next person who copies from it."

2. None of the doublets that have been preserved is copied by a student performer!

The most frequent way that Anna Magdalena Bach participated in the copy process involving the cantata parts was to copy a few doublets. This was probably done soon after the main copyist had finished his copy work and after Bach had checked the accuracy and made any additions or corrections to it.

Certainly at least one of these parts would have survived just as those copies of the WTC did that had been copied by Bach's private pupils.

BL: >>This was also Bach's OWN practice in educating himself about other people's music: start by making his own copy of it by hand.<<
Again, you have not considered the cost of paper in Bach's time, nor have you considered the difference between Bach's preserving a composition for his own use (the value attached to the composition itself, it could be traded or purchased) and the original set of performing parts which could and were often regenerated from the autograph score to accommodate different performance requirements for a repeat performance.

What kind of education will a viola player derive from copying a single Bach cantata part while not having access to the score? What a great learning experience! "I am learning how to copy notes from one page to another although I have no concept of the whole!"

BL: >>And yes, we must first discard the over-romanticized notion of the cantatas being written only as late as a couple days before the gig.<<
The hypotheses you have presented thus far are entirely unconvincing. Faced with continually changing performance conditions (which performer will be ill or suddenly necessarily absent (death in the family, etc), which great instrumentalist will happen to be rather unexpectedly traveling via Leipzig and might be able to perform in a cantata during a short stay there, etc.) Bach would be wise to wait until these conditions could be more fully ascertained despite the fact that the texts had been chosen and that he may already have been working on some ideas in his mind how the music might be composed.

The problem with your hypotheses is that they are based upon the arrogance of viewing Bach's composing and performance conditions from only a present-day viewpoint with only contemporary empirical knowledge which does not sufficiently take into account the evidence that does exist regarding Bach's work habits as based upon a close examination of his autograph scores and parts.

BL: >>Note that this is independent of the number of singers or players assigned to perform any given part. Within this hypothesis, we can't use the number of extant parts (master copies) to extrapolate the number of singers/players, either to many or few; it's an unknown.<<
Aw! Just when I thought I had a means for dispelling the OVPP/OPPP hypothesis by calling attention to all those busy-beaver students copying out all those extra parts that would allow us to consider that each singer/player would have his own physical copy of a part. Now we could, with your help, easily persuade those who believe that only one person can read from a single part that Thomaner singers in Bach's time did perform Bach's figural music with a choir of 12, 16, or more.

BL: >>Corrections to the master parts, if any, would tend not to get written down: that particular part might not be on anyone's music stand at all, if the players were playing from their own copies (where any corrections would get written in -- the copies they kept for themselves afterward). The master part itself might be back in the library during the performances, having served its purpose as exemplar....<<
All of this, of course, is only possible if you can assume that Thomaner boys had access to a paper supply consisting of reams of paper, while Bach, on the other hand as one setting the example for them, resisted change and insisted upon being very chary with paper. Does this type of example-setting on Bach's part make a lot of sense to you?

BL: >>p.s. Note that this hypothesis also offers an explanation why some of the blisteringly difficult continuo parts remain unfigured in the official part, for the archive files with the score in the library.<<
Had you not considered that such an unfigured continuo part would be for other members of the continuo group, particularly violoncello? Are you not aware of the fact that a separate violoncello part is normally not included in the original set of parts? Why would a violoncellist need a figured part unless it were to follow the anachronistic evidence first given by Schering in 1936 that in a Bach secco recitative, according to Baumgartner (a virtuoso cellist who described how operatic recitatives were to be played at the end of the 18th and the beginning of the 19^th century, the cellist should play broken chords instead of simply the notes given in the part?

BL: >>The surviving performance materials on paper diverge every time a piece is seriously worked on by intelligent musicians. That's a natural dynamic process for learning music. It blows away the concept of "Urtext", but so what? That was always a chimera anyway: the notion of fixing some single "best" version that is to become the standard for all time forward, for all players, for all acoustical situations!<<
And yet we have a wealth of indications (documented evidence) that Bach did fix his best versions very carefully and did not want other musicians, some of whom thought they were the greatest, to inflict their own sense of good taste on the performance of his music, a poor sense of good taste with which Bach could not agree since he knew his own sense of good taste in music was better than theirs. Bach wished to be remembered by the best possible performances of his music which he described in great detail much to the dismay of many musicians who would have preferred to change it. Remember in this regard the oft-quoted, well-documented statements from the Scheibe-Birnbaum controversy!

Scheibe's complaint:

"Alle Manieren, alle kleine Auszierungen, und alles, was | man unter der Methode zu spielen verstehet, druckt er mit eigentlichen Noten aus."

("All mannerisms, all the little embelishments, and everything that is understood as being performed according to 'the method' [the usual free style of treating music leaving such things as embellishments, coloratura, etc. entirely up to the performer] he [Bach] puts down on paper in actual notation.") Bach-Dokumente II, p. 286

Birnbaum's answer:

"Es kommt ohne dem in der Music alles auf die execution an."
("Anyhow, in music everything depends upon the how the composition(s) is/are performed") Bach-Dokumente II, p. 302

"...so wäre es eine überflüßige sache, wenn ihnen der componist das in noten noch einmahl vorschreiben wollte, was sie schon wissen. Allein da die wenigsten hiervon genugsame wissenschafft haben; dennoch aber durch eine ungeräumte anbringung ihrer methode die Haupt melodie verderben; ja auch wohl offt solche passagen hinein machen, welche von denen, die um der sache eigentliche beschaffenheit nicht wissen, dem componisten leicht als ein fehler angerechnet werden könnten; so ist ja wohl ein jeder componist, und also der Herr Hof-Compositeur befugt, durch vorschreibung einer richtigen und seiner absicht gemäßen methode, die irrenden auf den rechten weg zu | weisen, und dabey auf die erhaltung seiner eigenen ehre zu sorgen."

[Summary translation] "It might appear superfluous to those musicians who desire the freedom to do as they wish with the music (improvise coloraturas, include additional notes, change/add embellishments, disregard the intended articulation because they have a better idea how it should be performed) for Bach tdictate through his detailed notation how his piece should be performed. However, it is a fact the only a very few musicians have sufficient knowledge to perform this music [with the same good taste that Bach applies in his composition]. And yet they continue unabashed to spoil Bach's melody lines by attempting to apply their own notion of what the 'method' should be like. They even frequently include such passages [variations, coloraturas] which could easily be considered as mistakes made by Bach. For this reason Bach is justified in prescribing in great detail what he considers his own intended 'method' in order to help point out the correct path for those musicians who have lost their way [applying injudicious variations from the score which are not of the high quality standard of Bach's sense of good taste in music]. Bach does this in order to preserve his own honor as composer [he would like to be remembered by his own musical choices over those made by most musicians, many who even consider themselves great performers]. Bach-Dokumnte II, pp. 304-305

Is Urtext important for a truly good performance that resembles what Bach wanted us to hear? It certainly is!

Bradley Lehman wrote (September 16, 2006):
Bradley Lehman wrote:
<< Hypothesis: the sets of parts (usually only one of each exist, more rarely two) were exemplars. Whether they were actually used in the performance by one or more players, we don't know. But, the thrust of my hypothesis: the student(s) assigned to play/sing each movement would have--as part of the assignment--the task of making HIS OWN copy of the whole thing, by hand, as part of the learning process. And then keep it afterward, not file it back away with the master set of parts and score, in Bach's folders. >>
Douglas Cowling wrote:
< This is an attractive hypothesis to account for mutliple copies but I can't imagine that the extra copies would be tossed -- the cost of paper was pretty stiff. >
What's the proof/details that the paper would have been somehow prohibitively expensive? Remember that we're not talking about any pre-printed lined paper, but ordinary plain paper: scrape the stave-thingy across it first and then write in the notes. Wouldn't the students have had some ready source of paper, for all their schoolwork of which this was merely a portion? How would we know if they didn't?

< It is possible that the extra copies were stored away from the exemplars and disposed of over the centuries. >
Agreed, and that's certainly my own experience with a lot of other music.... Whether handwritten or photocopied, what gets stored back to a master folder is not some random handful of detritus and smashed-up parts, but one clean set.

< Still it's odd that not even one such external copy surivives when throwaway items like the cantata libretti do. >
Agreed, but what reason would there be to preserve the stray copy or copies of some individual part? Without the rest of the ensemble parts, it would have sentimental value only (perhaps to the person who wrote it out) and not much utility.

And, some keyboard parts worked out by students do survive. Take, for instance, the written-out part for an aria of BWV 3 (a 1725 cantata), the one discussed in the Peter Williams article about continuo (the one I cite here in this forum often). This same part is also discussed by Robert Marshall in his review essay of the Harnoncourt/Leonhardt series volume 1. The part is written by Christian Friedrich Penzel, who was one of Bach's students in 1749. Would he have written this realization out only for a lesson, or perhaps for a repeat performance of that particular cantata? (The performances of the 1730s and 1740s happen to be not very well documented! We don't know how many times each piece was reused for church.)

< I do think we underestimate the importance of copying music as a pedagogical tool. With the ubiquity of published music today, the notion of learning through transcribing your own copy has all but vanished. And yet it was the standard method until the end of the 19^th century. >
Agreed. And, one of my grad school classmates came from a country (Vietnam) where it was almost impossible to buy good scores or to photocopy anything; so, he had written out several Beethoven symphonies for himself just to have them! He told me he learned a great deal in doing so....

< Perhaps the only place left is the teaching of harmony where the student has to transcribe the melody or figured bass. Over the years I have transcribed hundreds of pages of music and, although it's dog-work, I have always found that I learned and appreciated the music on a level much deeper than just opening up a score and performing. >
Same here. That hands-on work with writing out the notes is a walk in the compositional shoes, and an irreplaceable experience to have. So is the making of new editions, by hand, copied from old manuscripts and working out the clefs etc. We had some required assignments to do that, and ended up with both some fine music and valuable experience. I haven't yet got into the habit of setting much music in software packages (even though I've had them for 15 years), for the simple reason that I can write it out by hand faster and get exactly the results I want, especially with regard to figures. I'm still finding handwritten things littered throughout my files....

Thanks for the response.

Thomas Braatz wrote (September 16, 2006):
Bradley Lehman wrote:
>>What's the proof/details that the paper would have been somehow prohibitively expensive? Remember that we're not talking about any pre-printed lined paper, but ordinary plain paper: scrape the stave-thingy across it first and then write in the notes. Wouldn't the students have had some ready source of paper, for all their schoolwork of which this was merely a portion? How would we know if they didn't?<<
The problem is that you are still (always?) approaching Bach's time from a modern perspective. Have you ever heard of slates? This was a one-time purchase that could be used over and over again for years in school. Only the most important material from an entire year might be entered into a very slim paper notebook for future reference. I do not think that scraping the stave-thingy [das Rastral in German] across the slate would have worked very well either.

Here's a thought: Because paper, even plain paper, was so inordinately expensive in Bach's time, the pressure was placed upon musically talented pupils of their own accord to learn to sightread and memorize music very quickly.

BL: >>Agreed, but what reason would there be to preserve the stray copy or copies of some individual part? Without the rest of the ensemble parts, it would have sentimental value only (perhaps to the person who wrote it out) and not much utility.<<
This is the very reason why the futile, if not very expensive exercise of copying out week after week one's own part, let's say the viola part, in order only to prepare one's facility in reading it properly for just a rehearsal and one or two performances, and then expect the pupil-performer to learn something about the entire composition (rules of harmony, etc) without any reference to the score makes little sense for educating pupils musically either in Bach's time or in ours.

Richard Mix wrote (September 18, 2006):
Thomas Braatz wrote:
< What kind of education will a viola player derive from copying a single Bach cantata part while not having access to the score? What a great learning experience! "I am learning how to copy notes from one page to another although I have no concept of the whole!" >
This reminds me of Benedict Mason's talk on his Symphonia Concertante for 12 violas in which, besides covering most of the extensive repertoire of viola jokes, he advanced the notion that the viola parts of Schumann and Tchaikovsky symphonies, being worked upon by both outer voices, were palimpsests of the whole comosition to which they belonged. I think he might have been leg-pulling... But Thomas, you might try the exercise first before dismissing it out of hand: I know I learned a little copying such parts myself.

I can t awith Brad's theory, though:
"Corrections to the master parts, if any, would tend not to get written down" Wouldnt it be even more important that the exemplars be correct? And, cost of paper aside, wouldnt just the labor of producing an unused set of parts be, well, a little extravagant? Especially considering that the score could serve the same purpose? I'll of course defend to the death your right to indulge in flights of fancy, considering how fruitful some have been!

This morning I had a choir perform Hayden's The Heavens Are Telling from a cpdl edition full of typos. At rehearsal everyone had a pencil behind their ear, but if pencils hadnt been invented yet we would have easily remembered the corrections even without sending someone out for an inkpot. If they were used this way I suppose Bach's parts had less heavy wear than, say, Lutkin's warhorse The Lord Bless You And Keep You in my old Methodist church's files (the Stravinsky copies I ordered looked crisp enough to serve as exemplars and probably still do).

Richard Mix, trying to work out whether I can see Eric's house from here...

Johann van Veen wrote (September 19, 2006):
Douglas Cowling wrote:
< I do think we underestimate the importance of copying music as a pedagogical tool. With the ubiquity of published music today, the notion of learning through transcribing your own copy has all but vanished. And yet it was the standard method until the end of the 19^th century. Perhaps the only place left is the teaching of harmony where the student has to transcribe the melody or figured bass. Over the years I have transcribed hundreds of pages of music and, although it's dog-work, I have always found that I learned and appreciated the music on a level much deeper than just opening up a score and performing. >
You are not the only one: Reinhard Goebel has done the same when he wanted to play music which was only available in manuscript. I preferred it over making photocopies or microfilms. He believes it helps in understanding the piece and why it was composed in a specific way.

Douglas Cowling wrote (September 19, 2006):
Johan van Veen wrote:
< You are not the only one: Reinhard Goebel has done the same when he wanted to play music which was only available in manuscript. I preferred it over making photocopies or microfilms. He believes it helps in understanding the piece and why it was composed in a specific way. >
And it's not just full or keyboard scores. When you copy a single part you really begin to see the whole shape of a movement from one horizontal stratum. Modern singers are so used to seeing all the other choral parts and a reduction of the orchestral accompaniment that it is almost impossible for them to approach a score in the way Bach's singers did with only their part in front of them..

Ed Myskowski wrote (September 19, 2006):
Douglas Cowling wrote:
<< I do think we underestimate the importance of copying music as a pedagogical tool. [...] >>
Johan van Veen wrote:
< You are not the only one: Reinhard Goebel has done the same when he wanted to play music which was only available in manuscript. I preferred it over making photocopies or microfilms. He believes it helps in understanding the piece and why it was composed in a specific way. >
You are both missing the larger point. The world started to go downhill with the development of the printing press. I believe Gutenberg (I did not check for Umlauts) gets the credit, but historical credit is always subject to review. I would bet on some anonymous graduate student.

Before that, Irish monks (and others) copied by hand. Those are the guys who really understood Scripture. Everything subsequent is just a shadow, of text, translation, and interpretation.

Bradley Lehman wrote (September 22, 2006):
A few more comments on a thread from a few days ago:
< This morning I had a choir perform Hayden's The Heavens Are Telling from a cpdl edition full of typos. At rehearsal everyone had a pencil behind their ear, but if pencils hadnt been invented yet we would have easily remembered the corrections even without sending someone out for an inkpot. If they were used this way I suppose Bach's parts had less heavy wear than, say, Lutkin's warhorse The Lord Bless You And Keep You in my old Methodist church's files (the Stravinsky copies I ordered looked crisp enough to serve as exemplars and probably still do). >
That's a good example. Just a few more remarks from me:

Let's look briefly at what a conflationary Urtext edition (e.g. the Neue Bach-Ausgabe, or Musica Britannica) is: it's the fiction of telling us what the composer allegedly should have been thinking/producing as the object, as final or best or ideal preparation, had he been in complete control of all extant and lost source material.

The making of such an edition is a useful process, both practical and scholarly, through minefields of conflicting material; but it remains guesswork next to the faithful reproduction of extant sources, individually. The dynamic process of musicianship, the evolution of a composition across different practical circumstances, is squashed out in favor of a single idealized reading.

Corollary: any detail that doesn't make it into the editor's idealized product is implicitly assumed to be inferior or unreliable -- a mistake within its source material, or something the composer really didn't intend, or (Danger Will Robinson!) "shouldn't have" intended.

That's a philosophical problem in editing. Should one reproduce a source exactly as it stands, even if it's in conflict with other sources of the same piece? Or, should one try to construct a "better" reading that might never have existed in real practice, in any extant or lost source? How far can one go inside the composer's mind to figure out what he likely was and was not thinking, according to extant products and historical context, and with what level of confidence in the results?

As for "performance" editions vs "Urtext" editions, the same basic problem stands: how much or how little should be shown on the page, as to the editor's opinions and organization of the available material, along with practical suggestions? How do we really know one way or another what makes sense, toward a composer's intentions (or lack of specific intentions for variable circumstances!), without actually being that composer?

Whether some handwritten or printed parts have passages that look like mistakes, or not, it still doesn't necessarily reflect exactly what happened in practice. That's because there are similar philosophical problems in performance, too. When playing from any sort of edition or manuscript, how much should one take the liberty to correct anything that looks wrong, tacitly, toward the goal of creating a more convincing-sounding performance? This is akin to having the mind automatically screen out typos or engraving errors, maybe not even noticing them consciously, in favor of flow and meaning.

Everything always spins round to epistemology.......

Douglas Cowling wrote (September 22, 2006):
Bradley Lehman wrote:
< As for "performance" editions vs "Urtext" editions, the same basic problem stands: how much or how little should be shown on the page, as to the editor's opinions and organization of the available material, along with practical suggestions? How do we really know one way or another what makes sense, toward a composer's intentions (or lack of specific intentions for variable circumstances!), without actually being that composer? >
The notion that there is one perfected "Urtext" of a musical work is a Romantic concept. For Bach every work was more of a "work in progress".

I remember when the Tallis Choir of Toronto decided to perform the 1725 version of the St. John Passion and my initial repsonse was "Why did Bach tamper with such a perfect work?" Gone were "Herr Unser Herrscher" and the final chorale. There were angry protests from the choir. But that version with its book-end chorale-fantasies eventually achieved a grudging admiration.

Poor Cecila Bartoldi had acid critical bombs thrown at her a few years ago because she asked to have an alternate aria for Susannna which Mhad written for a particular singer in "Marriage of Figaro". She was basically accused of being a diva and not respecting Mozart's work. I just wish that the upcoming Toronto performance of "Cosi Fan Tutte" would include the alternate aria which mentions Canada!

"... ed egli e certo che glo uguali
non si trovano da Vienna al Canada".

Bradley Lehman wrote (September 22, 2006):
Douglas Cowling wrote:
< The notion that there is one perfected "Urtext" of a musical work is a Romantic concept. For Bach every work was more of a "work in progress". >
Yup. He even kept tinkering with his published scores, on his personal copies....

< Poor Cecila Bartoldi had acid critical bombs thrown at her a few years ago because she asked to have an alternate aria for Susannna which Mozart had written for a particular singer in "Marriage of Figaro". She was basically accused of being a diva and not respecting Mozart's work. I just wish that the upcoming Toronto performance of "Cosi Fan Tutte" would include the alternate aria which mentions Canada!
"... ed egli e certo che glo uguali
non si trovano da Vienna al Canada". >
"Ohime, Canada!"

Somebody could also slip in that tune on an instrument or two, since Mozart's tune for Sarastro in The Magic Flute is so much like it already, anyway! I can't hear that aria without thinking "Ach, Canada!"

Last week I was listening to Madame Butterfly since I really don't know it yet, and here come both the Star-Spangled Banner and Sakura in the first act.... Sort of like Debussy ripping off "God Save the King" for esquire Pickwick, in book 2 of preludes.

Thomas Braatz wrote (September 23, 2006):
Douglas Cowling wrote:
>>The notion that there is one perfected "Urtext" of a musical work is a Romantic concept. For Bach every work was more of a "work in progress".<<
While the latter statement is true from a generalized standpoint, the NBA editors do find that the various stages of development can be reconstructed to account for various versions which Bach used for different performances of a single work. The former statement about Urtext being a single perfected editorial version collated from materials existing for different performances does not apply to the NBA, a truly critical edition, which makes every effort to present these variant versions separately. For instance, the same cantata will be printed in different voice ranges (and changed keys). The NBA also supplies alternate parts when such alternate parts exist. Of course, the NBA KBs include all documented differences or variant forms which can be reconstructed by anyone seeking to find subtle variations or even the 'mistakes' which a performer may want to reinstitute. Staring at a facsimile of an autograph score or using it alone as is for a performance does present some serious problems for such a performance which excludes any reference to the original parts when they are extant: indications of tempo in descriptive words (andante, adagio, etc.), dynamics (forte, piano), articulation (slurs, staccato, etc.), embellishments, if they exist at all on the score, will only be found in much greater profusion in the original parts. In some instances, Bach made changes in the parts which were not reflected in the score which came first. This is certainly a "work in progress", but the editors necessarily conflate the information from the parts with the score in order to present a complete picture which would otherwise be lacking to anyone working only with a facsimile of the autograph score. It makes little sense to 'reinvent the wheel' by presenting only what the facsimile score offers, discovering for oneself where Bach left some unclear, difficult-to-decipher passages or actual, uncorrected errors, and then to attempt to make adjustments to the Urtext which may or may not agree with what Bach had intended as presented in the original parts.

In German musicology, a distinction is even made regarding the following:

1. a Performance Edition: one which lacks much of any kind of critical apparatus which documents how the editorial decisions were arrived at - the editors force upon the resulting notation their own editorial choices including performance practice choices not available in the original scores, as a result, there is no way to distinguish clearly between the editorial changes and additions and the original score(s) from which the resulting printed score is derived

In German these types of editions are called "Bearbeitungsausgaben" ("revised or arranged editions") or "Interpretationsausgaben" ("editions with the personal interpretation of the editor included")

2. an Urtext Edition: one which is based upon the following principles: a critical attempt to restore the original notation to its intended form; a report, usually at the beginning or end of the composition on all attempts at revision; an objective to make as few changes or additions to the original as possible and that these must be easily detectable to anyone who performs from the score.

3. a Critical Edition: one like the NBA which presents a copious critical apparatus (example: the KBs - Kritische Berichte - 'critical' reports consisting of separate volumes, not simply additional materials preceding or following the printed music) which documents just about every conceivable detail that can be gleaned from the original sources. This allows any musician to view and ponder all the differences found in various source materials the lead to the final Urtext notation. A critical edition includes all the requirements of the Urtext edition but goes far beyond it in offering a comprehensive view of all factors which enter the decision-making process before a score appears as published music. The NBA does not supply performance parts for each work it presents.

There are editions found on a scale at various degrees between these three categories as they combine some of the traits of one with those of the nearest category. Some examples would include the recently discussed versions of the Breitkopf Collection of Bach's 4-pt chorales: the Riemenschneider Edition and the Edition Breitkopf ("The only authorized American reprint of the original" - Associated Music Publishers, Inc.) both of which claim authenticity, one retaining the mistakes of the original but reverting to the open score and chorale text inclusions found in the mid-19^th-century Erk publication of the same, the latter, however, making undocumented improvements to the original. Both of these additions fail to be classified fully as Urtext, but they certainly do not fall into the category of a performance edition. Only the NBA with its associated KB can truly qualify as a critical edition. The NBA performance materials also published by Bärenreiter, although based upon the NBA results, lack the critical apparatus. Any commentary added by the editor of the performance edition (not the same one as the one who prepared the original NBA edition) may contain tendentious comments on performance practice as seen in Alfred Dürr's comment with footnote on p. 129 of his "Johann Sebastian Bach: Die Johannes-Passion" Bärenreiter (1988, 3^rd edition 1999):

>>Ich halte es daher persönlich für sinnvoller, die Rezitative entweder durchweg wie notiert - also ausgehalten - zu begleiten oder aber durchweg kurz, allenfalls mit ausgehaltenen Baßnoten. [Footnote to this statement follows]: Die letztgenannte Kompromißlösung bietet das nach der NBA gefertigte Aufführungsmaterial (Bärenreiter, Kassel etc. VEB Deutscher Verlag für Musik, Leipzig), freilich mit der Anweisung, auch die - außer in der Orgelstimme (?) - lang notierten Bassnoten "heutiger(?) Praxis folgend - zu verkürzen". Auch hier wird sich der Aufführende als selbstverantwortlich entscheiden und das Aufführungsmaterial entsprechend korrigieren müssen.<<
[>>For this reason I personally think that it will make more sense either to always accompany the recitatives as written/notated -this means held out for the full value of each note - or to always make them short, but in any case with fully held-out bass notes. {Footnote follows} The just-mentioned compromise solution given in the NBA performance materials {published jointly by Bärenreiter, Kassel and VEB German Music Publishing House, Leipzig} though with the direction/instruction to follow present-day(?) performance practice - with the exception of the organ part(?) -- to also shorten the bass notes notated in long/held notes. Here as well the performer, in making this decision, will assume personal responsibility and will need to correct the performance materials accordingly.<<]

Commentary arising also from the immediately preceding section not quoted here: Alfred Dürr, from his vast experience in working with Bach's original periods and from his understanding of musical practices in Bach's time, expresses his opinion that under no circumstances should the long and/or held notes in the bass line of the continuo in secco recitatives be shortened as they were indicated in the performance parts of the NBA performance edition of the SJP (BWV 245). [Here is evidence that the NBA performance edition is not entirely the same as the NBA critical edition. Even the notation can be changed to suit a performance edition editor's opinion. Dürr recommends that the musician should change these parts accordingly] It is not clear from the latter edition whether the Organo part would be treated similarly; however, in any case, the bass line must be played as written with the option that the chords may be shortened while the single bass notes are sustained for their full values. Dürr points to Heinichen's statement (1911) as the most reliable regarding such recitative accompaniment, a statement pertaining to Bach's performance practices. Dürr gives a rough summary of Heinchen's prescription to play the bass notes as written without getting into the fine details but which Dürr describes as coming down on the side of 'held notes' being held out for their full values, whereas Schröter (1772) is more in favor of shortening the long notes. According to Dürr, this may seem to leave the slight possibility that Bach might have attempted to use the shortening of long notes on a trial basis around 1735 (Ascension Oratorio) and 1736 (SMP).

In conclusion:

The NBA edition with its KBs represent what might be called a 'critical Urtext' edition not to be confused with its associated Bärenreiter performance edition. This distinction should always be considered when comparing the NBA with any other Urtext edition.