Continue from Part 6

Playing basso continuo without any rehearsal or prior practice

Thomas Braatz wrote (February 15, 2007):
Here is further evidence/documentation that Bach's continuo players did not have a chance to rehearse or study their continuo parts in advance before a performance. Only an 'Incipient' [Heinichen's term for a beginner] is at a stage where he still has insufficient 'Praxis' to sight-read properly any continuo part put before him. A player who knew and understood how to read basso continuo parts was required to read these parts at the performance without preparation and supply the harmonies ad lib or off the cuff.

From Johann David Heinichen's "Neu erfundene und Gründliche Anweisung zu vollkommener Erlernung des ,General-Basses', Hamburg, 1711

p. 1 Vorrede .und was heisset endlich 'General-Bass' spielen anders / als zu der vorgelegten eintzigen 'Bass'-Stimme / die übrigen Stimmen zu einer völligen ,Harmonie ex tempore' erdencken / oder dazu, componiren'?

("Foreword.and what does it really mean to play the basso continuo part other than to take the single bass-line part and to invent all the remaining parts to make them into a complete harmony created ex tempore (off the cuff, ad lib) or to compose these parts in addition to the existing bass part.")


p. 1

"Ja / wenn auch endlich ein 'habiler 'General-Bassiste' den vorgelegten ,Bass' ohne ,Species' oder ,Signaturen' muß ,tractir'en / und folgbar diese / und des ,Componi'sten Meinung selbst ,judicir'en und errathen können.."

("Indeed, when finally that stage is reached when a well-practiced continuo player is required to play the continuo part which is placed before him [no rehearsals, no previous practice sessions with the music], a part which has no figures or indications of harmonies, and when as a result he must judge and guess at what these figures might be and what the composer may have had in mind..")


p. 20

[it will take a lot of practice] ".ehe das 'Clavier' so begriffen wird / daß man wie nöthig / 'extemporir'en / und alle Sätze und ,Accorde' gleichsam blindlings finden/ ja offt errathen kann und muß"

("It will take a lot of practice before the keyboard will be understood so well that you, as necessary, can play ad lib and, as it were, blindly find all the movements [harmonic progressions?] and chords, and often even can guess or have to guess them.")


p. 284

>>§. 23.
Endlich aber und zuletzt / muß ein Lehr-Begieriger / so lange er sich nicht 'capabel' befindet / eine 'Cantata' oder 'General-Bass extempore' zu 'accompagnir'en / ein und die andere 'Cantata' vor sich nehmen / und die 'Signatur'en nach denen gegebenen dreyen 'Principiis', nehmlich aus der Singe=Stimme / denen gegebenen 'General-' und drittens aus denen 'Special'- Regeln zusammen raffen; selbige allenfalls drüber schreiben / und solchergestalt durch eigenen Fleiß sich täglich in der ,Perfection' höher schwingen."

(" And finally, as the last item to mention: as long as one who is eager to learn [to play continuo parts] is not capable of accompanying a cantata or playing any continuo part part ad lib, then he should select this or that cantata and try to pull together the required figures according to the principle sources of information: 1. from the vocal parts, then 2. from the general rules and 3.from the special rules, then he should write the results over the notes and in this manner, by means of his own diligence on a daily basis move toward the level of perfection.")

Chris Rowson wrote (February 15, 2007):
Thomas Braatz wrote:
< From Johann David Heinichen's "Neu erfundene und Gründliche Anweisung zu vollkommener Erlernung des ,General-Basses', Hamburg, 1711 >
This may be OK for the ordinary music of 1711, but not for the extraordinary chromaticism of BWV 3/1.

This piece is more or less bound to sound horrible if played by someone who has never worked with it before because it is completely impossible to guess many of Bach´s marvellous harmonic movements.

Cara Emily Thornton wrote (February 15, 2007):
[To Thomas Braatz] I have two keyboard players in my ensemble, one of whom has a degree in harpsichord performance and is a very fine continuo player with now several years of experience, able to realize a figured part at sight. To the best of my knowledge, however, she still writes in the figures if they are completely absent. So I think it would require probably at least 10 years of experience to have any hope of doing away with that and realizing even unfigured parts at sight.

Alain Bruguières wrote (February 15, 2007):
[To Cara Emily Thornton] Although I have neither practice nor academic training whatsoever - being, as Brad nicely put it, a 'naive (though presumably well-meaning)' chap, I will once again make so bold as to respond to a professional musician with a praxis.

Cara Emily Thornton wrote:
< I have two keyboard players in my ensemble, one of whom has a degree in harpsichord performance and is a very fine continuo player with now several years of experience, able to realize a figured part at sight. To the best of my knowledge, however, she still writes in the figures if they are completely absent. >
This seems to confirm that musical training was very different then from what it is now, with a priority on the ability to play with very little or no preparation and to fill in the gaps, rather than on achieving near-perfection in reproducing faithfully a pre-existing model.

< So I think it would require probably at least 10 years of experience to have any hope of doing away with that and realizing even unfigured parts at sight. >
However one cannot simply ignore the testimony of contemporary specialists...
Your conclusion takes for granted that musical education was the same as ours, going through the same steps. My guess would be that an apprentice musician was required to become autonomous and able to take part in a public performance much earlier.

Here is for free one more provocative suggestion:
There were no phonographs in Bach's time. A performing musician was not so much an 'artist' as a craftsman whose rôle encompassed the rôle of a modern phonograph. Do you expect a phonograph to rehearse 3 times on its own before it plays the music for you?

Chris Rowson wrote (February 15 2007):
Alain Bruguières wrote:
,,, This seems to confirm that musical training was very different then from what it is now ,,,
Certainly Baroque training was different, but it did not include telepathy. Howeverr, telepathy is the only way anyone could play the continuo for BWV 3/1 without previously having heard or worked with the piece.

It is just not possible to guess the extraordinary things that Bach writes here.

The more I think about this, the more this piece appears to be clear evidence that there were rehearsals.

Julian Mincham wrote (February 15, 2007):
Chris Rowson wrotte:
< The more I think about this, the more this piece appears to be clear evidence that there were rehearsals. >
But less clearly evidential if Bach himself played the continuo part?

Alain Bruguières wrote (February 15, 2007):
[To Chris Rowson] Nobody claimed that there were never any rehearsals.

As far as the context is concerned, I am now convinced that in Bach's time in Germany performances without rehearsals were customary and they were the norm rather than the exception.

Bach's music is particularly taxing and we cannot simply extend this generality to the performance of cantatas, even if Xavier's testimony indicates that a sight-reading performance is possible (at least for the cantatas his group performed).

Now what do we mean by rehearsal? Apparently, a general rehearsal with everybody present and the whole piece played throughout. If so, I think that in view of the evidence we have gathered so far, it is unlikely that Bach could afford that on a regular basis at least during the 2 hectic years presently under consideration. This doesn't mean that every performer discovered their scores at the last minute. Everybody agrees that Bach was a well-organizprofessional; surely if certain pieces required special work from the performers, he would have written these early in the week so that they could acquaint themselves with the difficulties, and perhaps work in small groups if they thought it necessary. Let's not forget that Bach's team was a close-knit group, Bach probably met the performers on a daily basis and could have warned them against such difficulties. If one piece really required a full rehearsal, he would certainly have one. Though I wonder whether this really required everybody being present and playing their instrument. How about if he played it on a clavicord in the presence of the people concerned, explaining the point as he played?

I believe that a more open (and relaxed) approach to this question would be more productive than a crystallization upon adamant positions.

I insist on the word 'question'.

Cara Emily Thornton wrote (February 15, 2007):
Alain Bruguières wrote:
< Although I have neither practice nor academic training whatsoever - being, as Brad nicely put it, a 'naive (though presumably well-meaning)' chap, I will once again make so bold as to respond to a professional musician with a praxis. >

:) Comments follow.

Cara Emily Thornton wrote::
<< I have two keyboard players in my ensemble, one of whom has a degree in harpsichord performance and is a very fine continuo player with now several years of experience, able to realize a figured part at sight. To the best of my knowledge, however, she still writes in the figures if they are completely absent. >>
Alain Bruguières wrote:
< This seems to confirm that musical training was very different then from what it is now, with a priority on the ability to play with very little or no preparation and to fill in the gaps, rather than on achieving near-perfection in reproducing faithfully a pre-existing model. >
Pre-existing model in what sense? Continuo players to this day improvise their harmonies on the basis of what they see in the score (and indeed, often everyone in the ensemble plays from the full score, especially if it is a small ensemble playing one person on a part). However, it is true that they do attempt to duplicate the principles used back in Baroque times for 'constructing' that improvisation.

On the other hand, the principles used back in Baroque times for playing continuo, from what my keyboard player tells me, are somewhat freer than what would apply to composition. In particular, I refer to something called 'voice leading'. If adhered to strictly, the rules of voice leading dictate such matters as the direction in which a given voice may move in given harmonic situations, or which note of a chord may be doubled, or the intervals at which two voices may or may not move together in parallel motion. All of these rules are often 'bent' while realizing continuo.

I think what happens in continuo playing is that the person realizing the continuo focuses more on the horizontal aspect - whether the individual voices flow nicely - than on the vertical aspect, which is governed by the figures, showing intervals that must appear vertically above the bass line, but not necessarily by specific harmonic progressions in the style of so-called 'functional harmony' (which is supposedly not a real concept, but rather a sort of fiction serving, as far as I can tell, the education of beginners in the area of harmony, the idea being that the phrase contains a progression of chords which is like a sentence, and each different chord has its 'function' in the sentence). No doubt there are people on this list who are more knowledgeable in this matter than I am, I shall leave any further commentary on 'functional harmony' to them.

But I digress. The focus on the horizontal aspect, the freedom in use of 'the rules', it seems, applies equally to Baroque era realizations of continuo, and to modern realizations of continuo undertaken according to the directions given in the Baroque era literature on the subject of how to realize continuo. So I am not sure that in either case we are speaking of 'realization of a pre-existing model'.

<< So I think it would require probably at least 10 years of experience to have any hope of doing away with that and realizing even unfigured parts at sight. >>
< However one cannot simply ignore the testimony of contemporary specialists... Your conclusion takes for granted that musical education was the same as ours, going through the same steps. My guess would be that an apprentice musician was required to become autonomous and able to take part in a public performance much earlier. >
All of the above having been said, it is very true that I did not take into account possible differences in the manner of education prevalent in the Baroque era vs. in modern times. So, thank you for pointing out this matter. I do not have material at my immediate disposal giving detailed descriptions of the educational process used in Baroque times to teach continuo playing. Perhaps someone else on the list does?

< Here is for free one more provocative suggestion:
There were no phonographs in Bach's time. A performing musician was not so much an 'artist' as a craftsman whose rôle encompassed the rôle of a modern phonograph. Do you expect a phonograph to rehearse 3 times on its own before it plays the music for you? >
No, but I do expect the people making the recording to rehearse beforehand ;;) Maybe I am 'just being modern'? ;)

Alain Bruguières wrote (February 15, 2007):
[To Cara Emily Thornton] Thank you very much for your comments.

You are right, the sentence
<< This seems to confirm that musical training was very different then from what it is now, with a priority on the ability to play with very little or no preparation and to fill in the gaps, rather than on achieving near-perfection in reproducing faithfully a pre-existing model. >>
< Pre-existing model in what sense? >
is rather confused (and perhaps I was confused when I wrote it).

Let me clarify.

It is my impression that (most? all? some?) modern baroque musician do not learn baroque music from the start. They being with a classical musical education, then they turn to baroque music. Is this correct?

Now a classical musical education is very technical, one learns to play exactly what is written on the score, according to a well-established tradition of interpretation. This leaves comparatively little room for inventiveness, for personal initiative.(This is what I called reproducing faithfully a pre-existing model).

Whereas in a baroque musical education, at least for a continuist (does the word exist?), one must learn to play notes which are not written, which is completely at variance with the classical training.

Of course, a classical training includes composition, but composition and performance are completely kept apart. Is this correct?If this is correct, I think that a classical musical training doesn't develop the abilities required of a baroque continuist, andwhen one turns to baroque music, one has to un-learn part of what one has learned.

Besides, other cultural approaches to learning music are different, and I think they involve much more learning by actually playing in real performances, from scratch (which doesn't mean that there's no personal work between real performances, of course). Which is completely untinkable in our classical tradition, of course, since only one goal is acceptable : artistic perfection! But those who learn that way tell you that there's no better way of learning!

<< Here is for free one more provocative suggestion:
There were no phonographs in Bach's time. A performing musician was not so much an 'artist' as a craftsman whose rôle encompassed the rôle of a modern phonograph. Do you expect a phonograph to rehearse 3 times on its own before it plays the music for you? >>
< No, but I do expect the people making the recording to rehearse beforehand ;;) Maybe I am 'just being modern'? ;) >
I wouldn't dare to suggest such a thing ! ;)

From a disc being recorded to be played hundred thousands of times, naturally one expects near-perfection. This is precisely half of my point; the other half being: in the abof phonographs and in a society where music plays a crucial rôle, music must be readily available for all sort of occasions, and then you have to make compromizes. In the time of Buster Keaton, you had a pianist who played music in the theater to accompany the heroe's pranks; I doubt that he rehearsed at all, yet he would do a very good job of it, I'm sure. In Bach's time, I suppose that any musician had to be good at that kind of things on a level which hardly anybody could achieve nowdays, simply because the need was there and is no longer here.

Julian Mincham wrote (February 15, 2007):
Alain Bruguières wrote:
< In the time of Buster Keaton, you had a pianist who played music in the theater to accompany the heroe's pranks; I doubt that he rehearsed at all, yet he would do a very good job of it, I'm sure. >
This is an interesting analogy and one that I can talk to from a position of some experience--NOT I hasten to say because I am old enough to remember them but because for some years it has been a sideline of mine to play the piano for showings of early silents--Harold Lloyd, Laurel and Hardy and Chaplin as well as Buster Keaton.

The comparison is interesting for two reasons 1 because it does relate to the art of improvising and not just regurgitating from the written note and 2 because of my 'first performance' scenario touched on in an earlier email.

There are a lot of cliches one is expected to fall back on e.g. romantic, villain and chase themes and a good pianist would also need to know a lot of popular songs of the day and a smattering of appropriate classics e.g. William Tell Overture and things like 'Home Sweet Home' But it's only on the first viewing (which was usually the first public performance too) that s/he could pick up the pitfalls and surprises e.g. a prat fall, falling downstairs, gunshots etc which the pianist needs to emphasise or underline for best comic effect.

So you can fall back on a lot of known and memorised repertoire but there is still a large element of improvisation and reaction to events (I know that no two performances for the same film are ever quite identical.)

So first performances are 'touch and go' and not necessarly best effort. After a couple of weeks of half a dozen performances a day the pianist has become so familiar with the film that (I have read) some of them read a magazine while playing the piano. Not entirely improbable--I have read that, on occasions Charles Rosen also read a novel and practised at the same time.

I'm not quite sure what illumination of Bach performances this analogy brings except that I would bet that first performances of both these films and the cantatas were 'belt and braces' affairs in which a number of details could be radically improved at later plaings.

Gosh I think this is my 4^th email to the list today--I'd better stop!

Cara Emily Thornton wrote (February 15, 2007):
Alain Bruguières wrote:
< Thank you very much for your comments. >
You're welcome :)

< You are right, the sentence >
<<< This seems to confirm that musical training was very different then from what it is now, with a priority on the ability to play with very little or no preparation and to fill in the gaps, rather than on achieving near-perfection in reproducing faithfully a pre-existing model. >>>
<< Pre-existing model in what sense? >>
< is rather confused (and perhaps I was confused when I wrote it). >
;;)

< Let me clarify.
It is my impression that (most? all? some?) modern baroque musician do not learn baroque music from the start. They being with a classical musical education, then they turn to baroque music. Is this correct? >
For the most part, yes. I don't personally know any exceptions to this 'rule', although I suppose they are theoretically possible.

< Now a classical musical education is very technical, one learns to play exactly what is written on the score, according to a well-established tradition of interpretation. This leaves comparatively little room for inventiveness, for personal initiative. (This is what I called reproducing faithfully a pre-existing model). >
Well, there is some truth in that, although there are varying approaches to what to do with the historical, theoretical, analytical information one might bring to bear while preparing a given piece, some of them more aimed at recreating, and others more aimed at producing a result which is in some way new, within a certain framework.

That having been said, there is always the exception of the concerto, which until the 19^th century basically without exception included cadenzas - i.e. one place in each of the first and second movements, and three or so places in the third movement - which were historically intended for extended, unaccompanied improvisation, weaving in and decorating motifs from the movement in question, displaying maximal virtuosity in the process. Even throughout the 19th century, it seems to me that relatively few concerti were supplied with 'standard' cadenzas by the composer (Sibelius comes to mind). So that in those cases where one was not supplied, to this day in principle one could improvise cadenzas in the appropriate spots.

Once upon a time, improvisation was elevated to the form of a high art so that, for example, the audience would be encouraged to give the performer notes to form a motif, upon which the performer would then improvise. This sort of thing, from what I understand, persisted well into the 19^th century, but then died out, so that nowadeays people play composed cadenzas, either of their own authorship or, most commonly, someone else's.

So I would suggest we speak rather of modern musical education, which is a relatively recent phenomenon, not much more than 100 years old at most.

< Whereas in a baroque musical education, at least for a continuist (does the word exist?), >
I don't recall ever having heard it myself, but never mind. At least we know what is being spoken of....

< one must learn to play notes which are not written, which is completely at variance with the classical training. >
Not entirely. I don't know how it is at conservatories, but if one studies music at a university, there are courses in ear training which include a keyboard component and in at least some cases require the student to learn how to realize continuo parts. Now, that having been said, I would probably never have gotten away with the sort of freedom in bending the rules of voice leading of which early music specialists might avail themselves. We definitely had to follow the rules strictly, and to this day, it has its effect if I am required to sit down and write out the realization of a continuo part (because not everyone can do it at sight, and no one has the time to work out the realization - and that means that I, being the chief handmaiden, have the privilege of either writing out the realization myself or ordering the organ part from Carus Verlag ;;) ). Fortunately, the person I am normally writing for has similar tastes, an ear which naturally prefers a realization that follows the rules more closely...

< Of course, a classical training includes composition, but composition and performance are completely kept apart. Is this correct? >
Also not entirely. It depends on the student, really. I have had teachers who encouraged me to compose cadenzas for myself. Composers normally also receive training in performance on an instrument, and often write music for themselves. Back in the first half of the 20th century, Polish composer and violinist Grazyna Bacewicz went to the Wieniawski Competition armed with, among other things, material of her own composition, and received an honorable mention in the competition, not inconsiderably because of the quality of her compositions...

< If this is correct, I think that a classical musical training doesn't develop the abilities required of a baroque continuist, and when one turns to baroque music, one has to un-learn part of what one has learned. >
But it is quite true that classical musicians who are not early music specialists do not normally learn improvisation, with only one exception: to my knowledge, organists are the only classicalinstrumentalists who are normally required to learn improvisation; indeed to even be accepted for study, they are expected to show evidence of at least some predisposition in this direction. And to this day, organ recitals often include at least one improvised item (for example, on some more or less well-known theme). Other than organists, yes, it is very true that the methods of education in use in modern times definitely do not encourage improvisation.

< Besides, other cultural approaches to learning music are different, and I think they involve much more learning by actually playing in real performances, from scratch (which doesn't mean that there's no personal work between real performances, of course). Which is completely untinkable in our classical tradition, of course, since only one goal is acceptable : artistic perfection! But those who learn that way tell you that there's no better way of learning! >
There are varying theories on that. Some teachers believe that fewer recitals is better, because then the student can develop his/her abilities in peace and quiet, without the pressure of public performance (or something to that effect). Others believe more is better.

<< No, but I do expect the people making the recording to rehearse beforehand ;;) Maybe I am 'just being modern'? ;) >>
< I wouldn't dare to suggest such a thing ! ;) >
;;)

< From a disc being recorded to be played hundred thousands of times, naturally one expects near-perfection. >
Yes, there is that. See, part of the problem with recording is that while it is being carried out, certain things become audible which would not be in live performance. That not completely inaudible slide... That finger brought down a bit too hard on the fingerboard... So with the advent of recording, now everyone has to really clean up their technique, and of course remember to stand still enough for the mike to pick them up. And then, of course, if the result isn't 100% perfect, we re-record that little bit until it's just so... And if all else fails (because the performer is too nervous to get that note way down at the end of the fingerboard properly), we can even synthesize that one note (supposedly they had to resort to that when Isaac Stern recorded the Dvorak violin concerto). These effects have progressed to such a degree that apparently now people sometimes prefer recordings to 'the real thing'. They have been spoiled, they want to be dazzled with perfection, and the live performance is something of an anticlimax to their expectations...

< This is precisely half of my point; the other half being: in the absence of phonographs and in a society where music plays a crucial rôle, music must be readily available for all sort of occasions, and then you have to make compromizes. In the time of Buster Keaton, you had a pianist who played music in the theater to accompany the heroe's pranks; I doubt that he rehearsed at all, yet he would do a very good job of it, I'm sure. >
Probably something like that would be improvised. But who am I to talk, I am too young to remember the silent movies...

< In Bach's time, I suppose that any musician had to be good at that kind of things on a level which hardly anybody could achieve nowdays, simply because the need was there and is no longer here. >
Well, on the one hand, the little imperfections would not have been so audible, but such things as singing out of tune, which are bad enough in a conventional concert hall, are magnified many times over in the church setting, because of the acoustics of the space in question. That false note which lasts, say, one second in a conventional concert hall, will last at least twice that long and, on top of that, dissonate (yes, I know, it's a neologism, but do I look like I care? ;;) ) with the notes that are being played subsequently. So Bach would have had to perhaps not pay so much attention to little errors, but much more attention to clean intonation than we would nowadays.

There is also something to be said for simply being in the habit of preparing music quickly (whether out of necessity or out of preference). Once it becomes normal, one doesn't think of it as anything special. I personally have not gotten into the habit of sight reading performances, however :D

Enough for now.

Cara Emily Thornton wrote (February 15, 2007):
----- Original Message -----
Cara Emily Thornton wrote:
< That having been said, there is always the exception of the concerto, which until the 19^th century basically without exception included cadenzas - i.e. one place in each of the first and second movements, and three or so places in the third movement - which were historically intended for extended, unaccompanied improvisation, weaving in and decorating motifs from the movement in question, displaying maximal virtuosity in the process. Even throughout the 19^th century, it seems to me that relatively few concerti were supplied with 'standard' cadenzas by the composer (Sibelius comes to mind). So that in those cases where one was not supplied, to this day in principle one could improvise cadenzas in the appropriate spots. >
PS When I talk about concerti with cadenzas, I have in mind from the Classical period onward. Baroque concerti are a different matter.

Julian Mincham wrote (February 15, 2007):
Cara Emily Thornton wrote:
< That having been said, there is always the exception of the concerto, which until the 19^th century basically without exception included cadenzas - i.e. one place in each of the first and second movements, and three or so places in the third movement - which were historically intended for extended, unaccompanied improvisation, weaving in and decorating motifs from the movement in question, displaying maximal virtuosity in the process. >
Additionally it is generally accepted that in a number of Mozart's piano concerti (particularly 2nd movements) the piano part written is at times only a skectch and needs to be filled out.

Chris Rowson wrote (February 15 2007):
[To Julian Mincham] Then there are the Handel organ concertos, where large tracts of the solo parts were not written down until they were prepared for publication. Even then, there still some "organo ad libitum" passages.

Douglas Cowling wrote (February 15, 2007):
Alain Bruguières wrote:
< In the time of Buster Keaton, you had a pianist who played music in the theater to accompany the heroe's pranks; I doubt that he rehearsed at all, yet he would do a very good job of it, I'm sure. >
My grandmother played the piano for the silent films in the 20's. She said her pattern was to watch the film during the matinee and play whatever music she felt appropriate -- she had an extraordinry repertoire of popular tunes under her fingers. For the evening show, which was a much dressier affair, she laid out a sequence of printed music and then improvised bridges between.

One great difference between the 19^th and 21^st century is the fact that training in improvisation is no longer a part of keyboard pedagogy. The only exception is organists who are still required to improvise in church music. Gven the choice between a professional organist and a harspichordist to play continuo in a modern performance, I would always opt for the organist because they know how to respond to the moment and not just play the notes of someone's realization.

Although I think it defies common sense to assert that Bach's musicians never rehearsed, the fact that Bach's musicians worked with him every day as teacher and mentor gave them insight into his music that is denied to us. We should also remember that, despite their youth, Bach's keyboard players were probably whiz-kids. You can still see this in English cathedrals where the sub-organists are often barely out of their teens and show astonishing musicianship in both technique and interpretation.

As always, I think that we should look to contemporary cathedral and collegiate choirs rather than concert hall ensembles and conservatories if we want to imagine what daily music-making was like under Bach. Bach's methods are still very much living pedagogies.

Cara Emily Thornton wrote (February 15, 2007):
Douglas Cowling wrote:
< One great difference between the 19^th and 21^st century is the fact that training in improvisation is no longer a part of keyboard pedagogy. The only exception is organists who are still required to improvise in church
music. Gven the choice between a professional organist and a harspichordist to play continuo in a modern performance, I would always opt for the organist because they know how to respond to the moment and not just play the notes of someone's realization. >
But then again, you live in Canada, if I am not mistaken. Here in Poland, the harpsichordists are better at continuo. The organists don't normally ever learn continuo per se, just improvisation, while the harpsichordists are required to have extensive instruction in improvising continuo parts.

< Although I think it defies common sense to assert that Bach's musicians never rehearsed, the fact that Bach's musicians worked with him every day as teacher and mentor gave them insight into his music that is denied to us. We should also remember that, despite their youth, Bach's keyboard players were probably whiz-kids. You can still see this in English cathedrals where the sub-organists are often barely out of their teens and show astonishing musicianship in both technique and interpretation. >
It's the same here in Poland. Many of the sub-organists, and even some of the chief organists, are still in conservatory.

< As always, I think that we should look to contemporary cathedral and collegiate choirs rather than concert hall ensembles and conservatories if we want to imagine what daily music-making was like under Bach. Bach's methods are still very much living pedagogies. >
Yes, it is true. Especially small collegiate choirs, I would say. I've been in at least one organization of that type which really did expect people to be able to sight read Bach cantatas perfectly at tempo in order to pass the entrance audition. So it sounds like the expectations were pretty similar. Now whether the actual pedagogical methods were the same in this organization as in Bach's time, I will not speculate...

Ed Myskowski wrote (February 15, 2007):
Cara Emily Thornton wrote:
< No, but I do expect the people making the recording to rehearse beforehand ;;) Maybe I am 'just being modern'? ;) >
Only about as modern as the 1940's, when 'rehearsed' performances were recorded direct to wax ('Max Making Wax' is a great Max Roach title!). Then came editing.

Truly modern might include this procedure, as an example: five run throughs, I suppose only the first would be actually 'unrehearsed', and a recording splicing together the best bits from all of them.

Or Frank Sinatra sings duets with his friends. Reportedly, the friends mostly mailed in their contributions.

Ed Myskowski wrote (February 15, 2007):
Alain Bruguières wrote:
< In the time of Buster Keaton, you had a pianist who played music in the theater to accompany the heroe's pranks; I doubt that he rehearsed at all, yet he would do a very good job of it, I'm sure. >
Sure, but there are two quantum discontinuities here:

(1) An individual improvising is much different from an individual sight-reading.

(2) An individual sight-reading is much different from an orchestra and chorus sight-reading.
<>

Rick Canyon wrote (February 15, 2007):
<>

Bradley Lehman wrote (February 15, 2007):
< One great difference between the 19th and 21st century is the fact that training in improvisation is no longer a part of keyboard pedagogy. The only exception is organists who are still required to improvise in church music. Gven the choice between a professional organist and a harspichordist to play continuo in a modern performance, I would always opt for the organist because they know how to respond to the moment and not just play the notes of someone's realization. >
Ummmm...hey, man! In general I sort of agree with this, that some organists are better improvisers than some harpsichordists (just from having to do some of it every Sunday, and for some organ degree programs and certifications requiring it).

But on the other hand: one of the courses I took for a doctorate in harpsichord did require us lowly harpsichordists to improvise without anybody's written-out realization. For half of the course we improvised various types of solo music, some chorale-based (variation techniques, extemporized harmonizations and elaborations etc) and some contrapuntal. For the other half of the course we played straight from unrealized examples, mostly Bach cantata arias and recitatives, of all levels of difficulty. Some fully figured, some only sketchily figured (i.e. as they stand in the sources), some unfigured. We were expected to work our parts between class sessions, and then to play them differently in class every time. We also had to do some of it as wholly impromptu sight-reading, because that too is sometimes a real-world situation.

For other coursework I was required to take several years of ensemble classes where we actually did all this stuff, with various instrumentations and voices, in class sessions and in group rehearsals between classes. We played from unrealized parts there, too, as a rule.

I agree that some harpsichordists know little beyond relying on somebody else's written-out realization. But, on a curmudgeonly day, I wouldn't call them real harpsichordists, but rather "keyboardists who happen to play at harpsichord sometimes". :) [The kind of people I do hired tunings for: very good pianists and organists....]

One of my colleagues a couple of years ago, after we had just rehearsed a bunch of Bach vocal music for a recital, turned to me and remarked about all the unwritten stuff I had played: "How do you do all that poodling?!"

Another colleague, just this past Saturday, was working with me on some Telemann and wanted to rehearse the ending of one of the movements a few more times. She asked me, "Do you have some rolled chord on that last downbeat or what?" I replied, "I frankly have no idea what I'm going to play, or what roll it might or might not be there, because it's different every time. I'm responding to what you are doing. My dynamics and added notes are all instinctive to how it's going each time."

Which is exactly how I play that stuff. Every bit of "poodling" is highly trained instinct, to automatically do something that sounds appropriate in the moment. If the piece needs something that's crescendoing, I play an accelerating and thicker batch of notes. If there's a ritard, or a phrase break, the poodling also has to reflect that character and obey the phrasing/dynamics of the bass line. I consider this to be basic (rudimentary) basso-continuo skill, bringing out the character and accentuation of the continuo melody; and there are certainly other Baroque-keyboard specialists who do that better than I do, because they get more opportunities to play.

< Although I think it defies common sense to assert that Bach's musicians never rehearsed, the fact that Bach's musicians worked with him every day as teacher and mentor gave them insight into his music that is denied to us. We should also remember that, despite their youth, Bach's keyboard players were probably whiz-kids. >
Some of them, anyway; and certainly CPE Bach. ...Whose own book about all this is indispensable, and more applicable to Bach's music than Heinichen's is!

As for Heinichen: I have right here on my desk the book by George Buelow, Thorough-Bass Accompaniment according to Johann David Heinichen, 462 pages, published 1966/1986/1992. It goes through Heinichen providing translation, commentary, musical examples, and hundreds of footnotes out to other contemporary sources (some in agreement, some not). And all of chapter 9, "A Practical Demonstration", goes through Heinichen's own example: working up 90 points (90!) where Heinichen turns an unfigured Alessandro Scarlatti cantata part into the improvised realization strategy that he himself would use. Every one of these 90 points/rules from the example is something that we continuo players (improvisers!) need to think of as we do our work: and with enough practice that it all becomes second nature and happens spontaneously. Buelow'sAppendix A, then, takes that same Scarlatti/Heinichen example and Buelow has composed a complete sample realization (pp 294-306) with footnoting back to Heinichen's 90 points.

This terrific book: Amazon.com

Another indispensable book for doing it Bach's way, going through his own basso-continuo exercises: Amazon.com

A shorter version of those is also in the back of the older Bach Reader (1966), but not in the New Bach Reader because this Poulin book came out in the meantime and did it better.

Both Heinichen and CPE Bach, by the way, recommend that continuo experts should sometimes add octave-doubled basslines for even greater sonority.

There is also the essential technique, described variously by Heinichen, Geminiani, and others (plus the written examples in JSB's sarabandes...), of crushing non-harmonic passing notes into improvised continuo chords. A great introductory article about that is Peter Williams's "The harpsichord acciaccatura: theory and practice in harmony, 1650-1750" (Musical Quarterly 1968, 21 pages).

All of which, of course, requires understanding of the music and its style, and working it out in rehearsals!

Bradley Lehman wrote (February 15, 2007):
< But on the other hand: one of the courses I took for a doctorate in harpsichord did require us lowly harpsichordists to improvise without anybody's written-out realization. For half of the course we improvised various types of solo music, some chorale-based (variation techniques, extemporized harmonizations and elaborations etc) and some contrapuntal. For the other half of the course we played straight from unrealized examples, mostly Bach cantata arias and recitatives, of all levels of difficulty. Some fully figured, some only sketchily figured (i.e. as they stand in the sources), some unfigured. We were expected to work our parts between class sessions, and then to play them differently in class every time. We also had to do some of it as wholly impromptu sight-reading, because that too is sometimes a real-world situation. >
And since I already know what's likely coming in haughty response to this, let's pre-empt it right now. That is, the usual twittery such as: <But harpsichords weren't PLAYED much in Bach cantatas, so all of this is irrelevant, and Lehman is just a stupid misguided moron for saying any of it!...because even non-musicians (or at least one) know better than he does!!!!!>

The same improvisatory type of playing is also done on organ, as to accompanying Bach cantata parts (arias, recitatives, "choruses"). The chief differences in organ vs harpsichord continuo are:

- Sticking more closely to strict four-part style, rather than adding/subtracting so many notes per the moment.

- Generally simpler realization, with less of the "poodling" stuff between beats, and fewer (maybe none) rolled chords.

- Generally shorter and less-legato chords for acoustical reasons (let the space resonate them out), to avoid making the texture too muddy; crisp chords can punctuate the texture (...see especially Mendel's 1950 article...) to aid the dynamic delivery of the bass line, not simply pouring an undifferentiated legato gravy over everything. Finger-substitution techniques are pretty much worthless in continuo playing, since they encourage too much legato.

- Sometimes holding the bass line longer than the right-hand chords; not only because some of the 18th century sources say so, but for acoustical reasons.

- No need to restrike the notes, as sometimes has to be done on harpsichord, because they'll simply keep sounding as long as the keys are held down; which (again) doesn't have to be for full notated value, but just long enough to establish the harmony and provide a clear direction to the motion of the piece.

- Generally stay lower on the keyboard, in the right hand, since some organ stops are simply too loud or prominent in the higher octaves; the whole point is to listen to the other parts and support them with a firm and rich sonority, but not compete with them.

- Do the same types of things that one does, improvisatorially, in accompanying hymn-singing in church: listen at every moment to provide the right balance between accentuation (keep the people together - shortish/crisp chords) vs enough sonority and clear-enough melody...without being overwhelming. Encourage them to sing, not to bask in a wash of sound! Vary the texture from stanza to stanza, and maybe even a little bit from phrase to phrase, if that seems appropriate both to the meaning of the words and to the way the performance is going at the moment.

Bradley Lehman wrote (February 15, 2007):
A bit ago, I wrote:
< And since I already know what's likely coming in haughty response to this, let's pre-empt it right now. That is, the usual twittery such as: < But harpsichords weren't PLAYED much in Bach cantatas, so all of this is irrelevant, and Lehman is just a stupid misguided moron for saying any of it!...because even non-musicians (or at least one) know better than he does!!!!! > >
I apologize. I should have made it more clear, right up front, that I was referring specifically to the following exchange from 2/13.

I had posted a rather long opinion about how useful it is to rehearse Bach's music on clavichord, because the music is difficult and because that instrument forces the player to work hard, thinking out and practicing every detail of the interpretation. Time spent working at the clavichord is worth (at least to me) double the time working on the other keyboards, because it demands such concentration and care.

And somebody then quoted back a small part of it, like this:

BL: >>CPE's whole book is about going way way way beyond the given notes, to discern and bring out a composer's intentions as to musical effect. And this is especially hard to do on CPE's favorite instrument: clavichord. If any keyboard instrument militates against sight-reading facility, it's this one.<<

...And then dismissed the whole thing glibly, like this:
< Clavichords were not used in the performance of Bach’s cantatas. >
So much for the valuable experience of actually working hard on Bach's music, to get it ready for performance. All kicked in the head. And never mind the possibility that Bach himself, or any of his students, may have had cantata-music rehearsals (arias, recits, etc) that did use clavichords because they were such a handy (and lower-cost!) instrument. Especially so, if we accept Christoph Wolff's conjecture that the Thomasschule practice areas did have some manner of keyboards.

Thomas Braatz wrote (February 15, 2007):
Bradley Lehman wrote:
>>I apologize. I should have made it more clear, right up front, that I was referring specifically to the following exchange from 2/13....
And somebody then quoted back a small part of it, like this:
BL: >>CPE's whole book is about going way way way beyond the given notes, to discern and bring out a composer's intentions as to musical effect. And this is especially hard to do on CPE's favorite instrument: clavichord. If any keyboard instrument militates against sight-reading facility, it's this one.<<
...And then dismissed the whole thing glibly, like this:
< Clavichords were not used in the performance of Bach's cantatas.< <<
This is an example of the utter confusion caused by Brad Lehman's insistence on not following the guidelines for this group. It is simply common courtesy to other readers to attribute statements to those who originally wrote them.

Who is this 'somebody'? Is this some ill-begotten attempt to be sensitive to my feelings or to remain completely 'objective' about a matter under discussion?

This is anything but objectivity! What we neis simply identification of the speaker. This might even be helpful for your own records when you do a search and have to refresh your memory on what had been stated.

Bradley Lehman wrote (February 15, 2007):
< Who is this 'somebody'? Is this some ill-begotten attempt to be sensitive to my feelings or to remain completely 'objective' about a matter under discussion? >
Why name any names? It's the BEHAVIOR that's the problem! My appreciative posting 2/13 about the positive value of clavichord practice got kicked in the head: with the glib assertion "Clavichords were not used in the performance of Bach’s cantatas," as if that makes my contribution worthless.

Thomas Braatz wrote (February 16, 2007):
Bradley Lehman wrote:
>>Why name any names? It's the BEHAVIOR that's the problem!<<
Whose BEHAVIOR? I simply did not see any connection between cantata accompaniments and clavichords. Now I see that you mean practice clavichords which might have been available for practice in the school. Thanks for the clarification.

Thomas Braatz wrote (February 16, 2007):
Bradley Lehman wrote:
>>All of which, of course, requires understanding of the music and its style, and working it out in rehearsals!<<
This is precisely where there is a great difference between what happens today (graduate courses teaching what Bach's teenage boys at the Thomasschule had already mastered and used when they were asked to sight-read Bach's continuo parts for each new Sunday cantata). What has become high-level study to attain expertise today was back then considered a reasonable goal for musically talented teenagers to achieve before they even considered attending a university.

Bradley Lehman wrote (February 16, 2007):
>>All of which, of course, requires understanding of the music and its style, and working it out in rehearsals!<<
< This is precisely where there is a great difference between what happens today (graduate courses teaching what Bach's teenage boys at the Thomasschule >
...were responsible to know and do: learning and performing the music given to them. But instead, it continues thus:

< had already mastered and used when they were asked to sight-read Bach's continuo parts for each new Sunday cantata) >
...that's a gratuitous insertion of a principle (sight-reading in church service) that they were not necessarily responsible to know or do! It's your premise, and you're never letting it go, and it's anything but "tentative" (your oh-so-objective-looking word from this week).

If you're not prepared to do the task of sight-reading continuo parts, especially from incompletely figured or unfigured parts, to see how impossible it is to do accurately in Bach's cantata arias/choruses: how about setting your personal insistences and hobby-horses aside, and simply learning from the commentary of those who do have an informed opinion on this subject? (For example, Chris Rowson and I, both experienced keyboard players, who have both commented separately about the difficulties/impossibilities in movement 1 and movement 5 of this week's cantata, BWV 3? Difficulties in playing from incomplete parts, and looking only at the incomplete part? Things that could be worked out only if rehearsals exist?)

< What has become high-level study to attain expertise today was back then considered a reasonable goal for musically talented teenagers to achieve before they even considered attending a university. >
You're still missing part of the point. Some of us today (myself included) were already playing keyboard accompaniments and solos for church services at age 13, 14, 15 -- way ahead of attending university. I've also been composing since about 14, and improvising; and an excellent sight-reader since I was 7. Musically talented teenagers are musically talented teenagers, whether 280 years ago or now. What's off-putting in your presentations (among other things) is this: the insinuation, nay, YOUR INSISTENCE, that those kids at the Thomasschule were necessarily smarter and better at all this stuff than ANY modern people.
<>

Douglas Cowling wrote (February 16, 2007):
Thomas Braatz wrote:
< This is precisely where there is a great difference between what happens today (graduate courses teaching what Bach's teenage boys at the Thomasschule had already mastered and used when they were asked to sight-read Bach's continuo parts for each new Sunday cantata). >
I don't know whether to laugh or cry.

Chris Kern wrote (February 16, 2007):
I think the main problem with the "no-rehearsal" idea is its inherent illogic. That something is illogical does not necessarily make it wrong, but it generally does mean that it can't just be supported by circumstancial evidence and lack of negative proof.

For instance, I could claim that we actually do not lack any of Bach's music -- in cases where the music is "lost", what actually happened is that the orchestra and singers were given the text, and they improvised a cantata on the spot.

Does that sound absurd? Well, much the same arguments being used for the lack of rehearsals can be used to support this. We know that even modern bands can improvise music if they work well together and are familiar with the general musical idiom of the style they are working in -- as Bach's performers would have been. Lacking the scores, we have no proof that any music was ever written. We know that composition and improvisation were both standard parts of a musical education and that Bach's concertists would have had some facility in it. Finally, we know that the general attitude towards music was different enough that an ad-libbed "jam session" (so to speak) like this might have been acceptable.

But all this conjecture and lack of negative proof doesn't really add up to much. The more illogical a statement seems, the more positive proof necessary to make the claim.

Xavier Rist wrote (February 16, 2007):
[To Chris Kern] Perfect post, Chris, I couldn't agree more!

Douglas Cowling wrote (February 16, 2007):
Chris Kern wrote:
< For instance, I could claim that we actually do not lack any of Bach's music -- in cases where the music is "lost", what actually happened is that the orchestra and singers were given the text, and they improvised a cantata on the spot. >
This is not as absurd as it sounds. When Handel was in Italy, it was a common aristocratic salon game for musicians to improvise secular cantatas. I drew on this practice in an imaginary scene in my children's CD, 'Vivaldi's Ring of Mystery" Four schoolgirls are arguing and Vivaldi has them turn it into an operatic scena complete with hair-pulling. http://www.childrensgroup.com/sections/classical/classical_index.html

Keep those retail sales high.

Alain Bruguières wrote (February 16, 2007):
[To Chris Kern] Thank you for stating your argument so clearly.

Of course I agree with the sound principle :

< The more illogical a statement seems, the more positive proof necessary to make the claim. >
I'm just a bit worried about the word illogical, I'm not sure what it means here, perhaps 'unlikely'? or 'contrary to common sense'?

Why do you think the 'no-rehearsal idea' is illogical?

You seem to think that Bach's ensemble would have been able to improvise a cantata on the spot (and Doug's post indicates that such things happened in Italy at the same time). (Of course I understand that such an improvized cantata would be very different from a cantata written in advance).

Still, if they were able to do that, why should they not have been able to play a cantata already written, whose most difficult parts they could have worked on beforehand?

And it they were able to do it, how is the hypothesis that they did, illogical?

Of course I lack experience, but I fail to see the logic in this.

Chris Kern wrote (February 16, 2007):
Chris Kern wrote:
<> For instance, I could claim that we actually do not lack any of Bach's music -- in cases where the music is "lost", what actually happened is that the orchestra and singers were given the text, and they improvised a cantata on the spot. >>
Douglas Cowling wrote:
< This is not as absurd it sounds. >
But one of my points was that saying something is possible is not the same thing as saying that it occured. Even if we could find concrete proof that Bach's players and singers were able to improvise a whole cantata, that would not automatically mean that any "music lost" cantata was actually improvised instead.

By the same token, "they were good at sight reading and could play parts on first glance" is not the same thing as "there were no rehearsals."

Douglas Cowling wrote (February 16, 2007):
Chris Kern wrote:
< But one of my points was that saying something is possible is not the same thing as saying that it occured. Even if we could find concrete proof that Bach's players and singers were able to improvise a whole cantata, that would not automatically mean that any "music lost" cantata was actually improvised instead >
Actually, I added the posting about improvised cantatas to reenforce what you were saying about the possible and the probable. The inherent problem we face in discussing Bach's performance practice is that we are constantly trying to assert probablities from very difficult sources. Whether we are arguing from incomplete documentary evidence, indirect contemporary collateral evidence or an uncertain continuum of modern praxis, I'm afraid that dogmatic certainties are beyond our grasp. Our discussions need to be on a possibility-probablity spectrum.

Continue on Part 8

Articles: The “Shortening of the Supporting Notes” in the Bc of Bach’s ‘Secco’ Recitatives [Thomas Braatz] | Playing Plain Recitative in Bach's Vocal Works [Bradley Lehman]
Discussions of Recitatives: Part 1 | Part 2 | Part 3 | Part 4 | Part 5 | Part 6 | Part 7 | Part 8 | Part 9 | Part 10 | Part 11 | Part 12 | Part 13 | Part 14 | Continuo in Bach’s Vocal Works: Part 1 | Part 2 | Part 3 | Part 4 | Part 5 | Part 6 | Part 7 | Part 8