Recitatives in Bach’s Vocal Works
Discussions - Part 14

Continue from Part 13

Belting out recitative accompaniment

Bradley Lehman wrote (January 28, 2005):
< (I'm swimming against it; I want to get the cantatas back on the concert programs of modern symphony orchestras, with at least 24 voice choirs, and grand pianos to belt out some crunchy diminished 7^th chords etc, in the recitatives.) >
Diminished 7^th chords sound a lot MORE crunchy and exciting when played on a good harpsichord, than on piano. A good resonant harpsichord with classic design (wooden frame), not a tinkly iron-framed heavy but paltry model from an assembly-line factory. Piano tone doesn't have the strong upper overtones that good harpsichords and organs do. Italian harpsichords are especially good at providing that dramatic crunch, due to their tonal design.

I like that general idea of "belt out", though. Strong drama there in the recitatives, not apologetic little stuff that is merely polite or twiddly. Needs enough organ tone in there, with some of the higher-pitched stops, to give enough presence to it. Harpsichord too, optionally where available. And bass-line players of strings and winds with the chutzpah to dig into it, making some bold strokes and not merely following along like a faithful spaniel. A strong bass line as a line of accented strokes, different weights and lengths, does some amazing things for the drama of the music. Helps the singer get across the meaning of the words, like exciting declamation punctuated by a team that is right there with every moment of it. The whole range of emotions and dynamics should be available there, for deployment at any given moment according to the thrust of the words.

The recitative genre, fundamentally from the realm of opera and "Seconda prattica" expressionism 100 years before Bach, is not a polite little devotional bit of quiet musing. It is fiery drama. It is in-your-face flips of Affekt, all the way from tender cajoling to sword-strokes. Notation on the page is mere shorthand for "Do something powerfully dramatic here using this bass line motion and these harmonies, in this order." All the rest of it is up to good musicians doing their jobs. The question of instrumentation is secondary here. Engage musicians who understand the genre and play it to the hilt.

Neil Halliday wrote (January 29, 2005):
Bradley Lehman wrote:
<"I like that general idea of "belt out", though. Strong drama there in the recitatives, not apologetic little stuff that is merely polite or twiddly".>
Like this one from Koopman, with organ in the continuo?
http://www.bach-cantatas.com/Mus/BWV8-Mus.htm

The 'C-9' example. I hope you can hear this without broadband. If not, rest assured your description: "apologetic little stuff that is merely polite or twiddly", fits this to a tee; and let's face it, this is currently the norm for recitative performance in Bach's church cantatas.

On the question of drama, do not forget that a piano can play loud and soft chords; a piano can, for example, increase the volume/force of a succession of chords from soft to loud, or vice-versa, in a very dramatic fashion. This a harpsichord cannot do, and it's why pianos were invented. (Perhaps we can agree to differ on the question of the effectiveness of the two instruments' timbres, for conveying the pitch-information of complex/dramatic chords; and in any case I also enjoy the distinctive bright overtones characteristic of the harpsichord).

With organ in recitatives, a quiet registration and/or a shortened-note method is required, because long-held forte chords on the organ are overpowering except in a solo piece. [In contrast, piano (and harpsichord) chords begin to 'die-away' immediately, no matter how loudly they are struck]. This is why organ continuo recitatives often (not always) fail in the drama stakes, as in the above Koopman example.

Seems to me, the piano was designed for dramatic recitative accompaniment.

Doug Cowling wrote (January 29, 2005):
Neil Halliday wrote:
< Seems to me, the piano was designed for dramatic recitative accompaniment. >
I'm not convinced that this is the case. The piano was the instrument of choice for power and expressiveness in the Romantic period and yet it never made an appearance in opera. Mozart acccompanied his operas from the piano and yet there is nothing in the scores which indicates that he used the keyboard for anything more than harmonic support. Mozart certainly knew the instrument's expressive power but seems indifferent to the possibilities of obligato accompaniment. When we look ahead to the Lieder of Schubert and Schumann, secco recitative really disappears as a dramatic device -- the conclusion of "Erlkönig" is the only example I can think of. Accompanied recitative stays around a lot longer -- Wagner's 'Das Rheingold' has many passages which is really recitative.

I really prefer discretion in recitative accompaniment. I don't want to hear the Evanagelist in the Passions constantly competing with an over-fussy and elaborate realizations. I remember one performance with harpsichord which was full of melodramatic arpeggios, ten-note block chords and all manner of ornaments. The drama should be in the vocal line not the keyboard.

Neil Halliday wrote (January 29, 2005):
Doug Cowling wrote:
<"I really prefer discretion in recitative accompaniment">
This is the opposing view; and while we don't really know what Bach had in mind, it means all those wonderful instrumental harmonies/modulations indicated in the figured-bass, become virtually inaccessable to the listener, in any dramatic sense.

My suggestions are an attempt to make the music live, for those like myself who find that extended unaccompanied recitative quickly becomes tedious, by allowing the fullest experience of the instrumental drama (indicated in the figures), as well as the vocal drama.

Of course, I recognise this is not a problem for others, including yourself.

Bradley Lehman wrote (January 29, 2005):
< I really prefer discretion in recitative accompaniment. I don't want to hear the Evanagelist in the Passions constantly competing with an over-fussy and elaborate realizations. I remember one performance with harpsichord which was full of melodramatic arpeggios, ten-note block chords and all manner of ornaments. The drama should be in the vocal line not the keyboard. >
Who said anything about "competing"? IMO, the drama should be in both the vocal line and the accompaniment, where the accompaniment is responding to things the vocalist is doing, and urging him/her on in brilliant collaboration.

A top-notch accompaniment still can't save a performance, if it's saddled with a singer who refuses to spike the ball that the accompaniment has placed in perfect position six inches above the net. (Pardon the volleyball metaphor.) Or an alley-oop in basketball. Or a brilliant combination play in soccer. Whatever. It's still the singer's job to pound the ball home.

And, back to the volleyball one: how can the singer spike the ball at all if the too-discreet accompaniment never gives it a set anywhere near the net?

Thomas Braatz wrote (January 29, 2005):
Doug Cowling wrote:
>>When we look ahead to the Lieder of Schubert and Schumann, secco recitative really disappears as a dramatic device -- the conclusion of "Erlkönig" is the only example I can think of.<<
Try looking at some longer ballads such as Schubert's setting of Schiller's "Der Taucher" [D77] and his settings of Ossian's Gesänge [D375 and D534.]

help needed on interpretation of recits

Indra Hughes wrote (November 11, 2005):
Forgive me if this has been covered here already - I am new to these boards.

Here in New Zealand we are having the Bach Cantatas played on the radio on the day for which they were written (or as close as possible thereto). Aren't we lucky!

By default, however, our station (Concert FM) chooses the Rilling/Stuttgart recordings so week after week we are getting these versions rather than any of the alternatives. I suppose the station must own a set of the Rilling comprecordings. There are many things that disappoint me about Rilling's recordings but that is not why I am writing.

I notice in these performances that in the recits Rilling has the continuo sustain their notes continuously, all the way through. I know that more often than not the recits are notated like that, with lots of tied semibreves in the bass part and so on.

But pretty much every other performer seems to do the recits 'secco' even though the bass lines are notated with long notes.

I have always preferred this 'secco' reading of recits, but that may be only because it is what I have always been used to.

I would like to know whether there is evidence of Bach's intentions and/or practice, one way or the other. Why has Rilling chosen to sustain the chords and bass lines when other performers have not? (Incidentally, Rilling's singers tend to deliver their recits in something very close to strict time as notated, with little or no freedom, and that may be as a result of Rilling taking the notated rhythms of the recits pretty strictly. Is he assuming that, because the cantatas were 'churned out' in short order week after week, the singers would have been virtually sight reading and would have stuck pretty closely to the notated rhythms, not having time to work on the sophisticated 'speech rhythm' delivery we hear from conductors like Herreweghe etc? I know from personal experience just how long it takes to achieve that instinctive synchronicity between the soloist, cellist and organist/harpsichordist.)

I always assumed that Bach wrote a load of semibreves in his recits because they are the fastest notes to write down and he probably had not the time or inclination when writing the cantatas to fill up bars with heaps of rests. I have always assumed when playing the organ in these recits and working with singers that the recits were a case where Bach did not necessarily mean exactly what he wrote....

If anyone can enlighten me, or point me in the right direction, I would really appreciate it.

Thomas Braatz wrote (November 12, 2005):
Indra Hughes wrote:
>>Forgive me if this has been covered here already - I am new to these boards.<<
We forgive you if you read carefully everything found at: http://www.bach-cantatas.com/Topics/index.htm

Then look under

Movements

and subcategory

Recitatives (14 parts!)

If you still have questions after reading this, I am certain there will be individuals who would like to reopen this unfinished (never-ending it seems) discussion.

John Pike wrote (November 14, 2005):
[To Indra Hughes] You are opening something of a Pandora's box by raising this question on this list since opinions are highly polarised. i suggest you read "Bach's Continuo Group", by Laurence Dreyfus, Harvard UP, which has an excellent chapter on this matter. It can be difficult to obtain outside the US. I suggest you try a Google search or www.Abebooks.com or an interlibrary loan.

Indra Hughes wrote (November 15, 2005):
THANKS

Thank you to those who have replied to me both on and off the list. I am grateful to have been pointed in the right direction. Of course I should have remembered about Dreyfus' book - I see it all the time at the University library!

I have no wish to re-open Pandora's Box on here so will not express an opinion!

to sustain, or not to sustain, that is the question

Nils Lid Hjort wrote (November 28, 2005):
A little while ago Indra Hughes asked about the practice of sustaining or not sustaining long notes in the basso continuo parts of the recitatives. We were then informed that this topic has been discussed at great length earlier on this list, without any clear consensus: http://www.bach-cantatas.com/Topics/index.htm
then "Movements" then "Recitatives" (in 14 long parts)

I have been skimming through these, with interest, and have just two simple questions, without asking anyone to open Pandora's box again (unless they wish to do so):

1. I've noticed that Suzuki (with the Bach Collegium Japan) at least for some of the cantatas uses the "stop chords early, let the singer be more alone" approach, perhaps particularly when using the organ. Does he do this consistently, through all the Cantata volumes (so far Vols. 1-28)? Has he commented on this specifically, in some of his "notes on performances"?

As someone pointed out in these earlier discussions, it appears to be Harnoncourt in the 1970ies who is the "modern initiator" for the practice of stopping chords early?

2. These earlier BCML discussions about recitatives discuss aspects of Mattheson's "Der vollkommene Capellmeister" (Hamburg, 1739) in detail. What is known regarding how well known Mattheson's book was at that time? Would must German musicians (or, say, conductors) have been exposed to that book, at around say 1750? Or was it seen as a more obscure or "for specialists only" publication?

Nils Lid Hjort
... who would also like Indra Hughes to tell us a little about his ongoing PhD project.

Neil Halliday wrote (November 28, 2005):
Nils Lid Hjort wrote:
"I've noticed that Suzuki (with the Bach Collegium Japan) at least for some of the cantatas uses the "stop chords early, let the singer be more alone" approach, perhaps particularly when using the organ. Does he do this consistently," >
Nils, have you entered this site into your 'favourites' folder? (kindly given by Roar Myrheim recently): http://www.bis.se/index.php?op=people&pID=2396

You can listen to samples of all Suzuki's cantata movements (You may have to register with Naxos, but it's free).

In short (!), Suzuki has adopted the `HIP' shortened chord method for secco recitatives (except sometimes with the first chord), which results in a marked contrast in form with all the other movements of a given cantata, and quickly becomes tedious, as a form of `non-music', to my ears. (Notice the apparent lack of treble clef material
in the short chords in this week's cantata BWV 194.)

I much prefer good examples of Rilling's method, as in this week's cantata BWV 194, where harpsichordist Martha Schuster develops tasteful, artistic accompaniment on the harpsichord, in conjunction with long-held (as notated) notes on the cello, resulting in a form of accompanied recitative, ie, a continuo recitative, that is much more in keeping with the overall character of the other movements that one finds in a Bach cantata (ie, voice(s) plus instruments). [Undoubtedly a little more phrasing in Rilling's seccos would help, in some other examples of his].

I have also heard pleasing examples from Leusink, using organ, with mostly full length accompanying chords, in BWV 185/4; BWV 136/4; and BWV 69/2 and 4.

[And maybe it's possible that one day an ensemble that is not exclusively devoted to chasing the HIP rabbit might discover what an effective instrument a piano is, in continuo recitatives].

Personally, I much prefer, being a lover of instrumental harmonies as well as vocal music, to hear these movements as continuo recitatives, and like continuo arias, they ought to be accompanied by the continuo, not presented as a strange mixture of recitation and shortened, disjointed instrumental accompaniment.

Unfortunately, the historical record is unclear, but in any case today's possibly more satisfactory organ tunings may allow for better results with long held chords (on organ) than those that might have been possible in the 1720's when mean-tone tunings probably were still rearing their ugly heads on many of the available instruments.

In passing, I regard it as one of those strange things in life, that other people (claim to) prefer the shortened chord method, especially many people do not even appear to have noticed or cared about the matter, before those of us familiar with pre HIP examples raised it.

John Pike wrote (November 28, 2005):
[To Nils Lid Hjort] While I agree with Neil that Rilling and Leusink can provide pleasing accompaniment for secco recitative, I personally do not object to shortened accompaniment (indeed, I enjoy it) and I was persuaded by the evidence for it in "Bach's Continuo Group" by Laurence Dreyfus, Harvard UP.

This is an excellent book, with all sorts of fascinating information in it. It is very well written and easy to read regardless of one's musicological background knowledge. I think this is an area where there is no substitute for reading the book and forming one's own mind up on the issue, since opinions on this list are so polarised anyway.

Thomas Braatz wrote (November 28, 2005):
Neil Halliday wrote:
>>Personally, I much prefer, being a lover of instrumental harmonies as well as vocal music, to hear these movements as continuo recitatives, and like continuo arias, they ought to be accompanied by the continuo, not presented as a strange mixture of recitation and shortened, disjointed instrumental accompaniment.
Unfortunately, the historical record is unclear, but in any case today's possibly more satisfactory organ tunings may allow for better results with long held chords (on organ) than those that might have been possible in the 1720's when mean-tone tunings probably were still rearing their ugly heads on many of the available instruments.<<
This is an astute observation with which I can agree along with the mechanical problems [see below] that could arise back then more than now. [a personal aside: when I was about 8 or 9, my first lessons on a church organ which I also used for practice were given on a trakker-action organ with a flat pedal board more like the type Bach would have encountered. It took a lot of stretching to reach the low and high notes without losing balance on the organ bench. It was very noisy compared to the very slick and practically noiseless modern pedal boards that I played on later. I can easily imagine how using the old-style pedal board could be disturbing to a performance with a single voice!]

It appears that the 'can of worms' needs to be opened once again. Without going back to what I have previously written, I have once again examined the crucial texts/sources involved which can shed light on the decision which is the subject of this thread:

Let's examine the Niedt quotation [which has been referred to numerous times before on this list - do a search on "Niedt" and "Recitative"]:

or look at page 79 in Dreyfus' book if you have it before you:

Niedt states:

>>§15.
Von ersteren [reference to organists and members of the bc group] bitte ich mir aus / daß NB. wenn ein 'Recitativ' vorkommt / und zwey bis drey gantzer 'Tacte' halten gesetzt ist / sie nicht mehr thun, als bey jeder neuen 'Not'e / die da vor kommt / einen Anschlag oder Anstoß zugeben / und dann so lange einhalten / bis wiedrum eine neue 'Not'e erfolge. Ferner / daß sie bey denen 'Caden'tzen die 'Not'en nicht so lange aushalten / als sie geschrieben stehen / sondern gleich zur folgenden schreiten.

§16.
Wo übern 'General-Bass' in denen 'Recitativ'en Ziffern gesetzt sind / so 'observire' der 'accompagnir'ende 'Bassist' wohl / ob der Sänger bey denselben und dem 'Accord' feste bleibe / da er dann solche zu 'exprimir'en eben nicht nöthig hat; Wo aber der Sänger aus dem Thon oder 'Accord' fällt / kan er derselben Ziffern ihre Bedeutung aufm 'Clavier' oder Orgel berühren / damit der Sänger sich wieder auf den rechten Weg helffen könne; als Z. E. es stünde folgendes 'Recitativ':

Niedt's Illustration

so müsten die 'Bass-Not'en angestossen werden / so / wie sie hier im 'Bass' schwartz gezeichnet stehen; und solches muß bey allen 'Recitativ'en wohl /wohl 'observi'ret werden / soll es anderst recht klingen / und nicht als ein altes Mühlen=Rad klappen.<<

My translation with commentary included:

§15.

Regarding the former group which I had mentioned earlier, the organists and members of the basso continuo group, I would request that they, (note this well) when they encounter a recitative where there may be passages with notes held out for 2 or 3 measures, should do nothing more than 'touch' (simply sound out/attack) the new note and then hold on to this note until another new/different note follows/appears in sequence. In addition (I would request) that they [this unclear reference can be deduced from examining the example below - Niedt is referring to only certain members of the continuo group who play in the 16' range: violone, organ - pedal board] should not hold out the notes for the full value as indicated in the score, but should rather immediately proceed to the next (note.)

§16.

Wherever the composer has placed (figured bass) numbers in the recitatives, the accompanying 'bass instrument player in the continuo group' should observe carefully whether the singer remains on pitch and sings the (broken) chords/harmonies securely (without losing his sense of tonality and pitch), because then it will simply not be necessary for the bass instrument player in the continuo group to add ('express') any additional chords or play additional notes (in order to help the singer along;) but when a singer does lose his place/pitch/tonality/sense of harmonic structure, then he (mainly the harpsichord or organ player) may add [to their full extent] these additional chords indicated by the figured bass so as to help the singer 'get back on track' again. For example, here in the following recitative you can see:

[Illustration - see Dreyfus, p. 79]

in this manner the bass notes should be sounded/played as indicated by the black notes in the bass (bottom line.) This method must be carefully observed in all recitatives if the music is to sound different [as Niedt would have it: properly] and not resemble the noise created by an old mill wheel. [Elsewhere, on p. 43 in the same book, Niedt similarly complains about the annoying interference created by the pedal board each time a note is played. He explains that if you play using only a single 16' stop [probably a 'Gedackt'], and you play a short note, you will hear more pedal noise than the normally expected note [it takes time for the sound to develop in such large
pipes of this type.] He refers to organists who attempt to play the pedal notes too fast as "'Pedal'-Quäler" ["Pedal torturers."] Why would Niedt want, then, to create a method for playing a shortened organo accompaniment where the shortened notes simply exacerbate the problem he so desires to solve: 'torturing the pedals' and the 'grating mill-wheel effect'?

One thing is quite certain: Niedt and J. S. Bach would have been seriously at odds with each other on this point (fast passages in the pedal part) as well as Niedt's injunction never to allow any fugues to appear in a church cantata. And yet Niedt is called upon in desperation by those who wish to use his comments as definite proof for the theory regarding 'shortened accompaniment in church-style, secco recitatives.'

* 'einhalten' according to the DWB [equivalent to the complete OED in English] can have various (almost contradictory) meanings in German:

There is intransitive usage where the meaning is:cessare currendo, loquendo, canendo (to stop running, speaking, singing)

and a number of other meanings equally important with various transitive objects, accusative and dative:

1. inhibere, cohibere, reprimere (to hold in/back one's breath

2. servare, observare (to hold one's word) to have resolved to do something and not give up/in

3. occupare, possidere, (to hold onto and defend [a fortress, etc.])

4. continere, enthalten, (to hold on [uphold] a law and continue doing so

5. retinere, zurückhalten (I lent him 4,000 golden guilders, but when I asked for them back, he kept holding on to them [and refused to give them up.]

Now the question might be raised: where is the object (accusative or dative) in Niedt's sentence? The answer is that this may well be a not uncommcase of ellipsis where the object is understood from just a little earlier in the same sentence: "und dann so lange [die Note] einhalten / bis wiedrum eine neue 'Not'e erfolge" ("and then to hold the note so long [to continue holding the note] until another [different] note appears in sequence.")

Note: Examine carefully Niedt's illustration and you might come to the conclusion that Niedt is not asking the entire continuo group to play in the manner described. He is not saying: "take the composer's indication with the long-held notes and feel free to put them down an octave and shorten them, or even make up additional notes and play them even though they were not in the original. What Niedt has done here [his explanation of this and his unclear references do leave very much to be desired] is to indicate deviations from the original score for the following reasons:

1. a loud sounding violone at 16' (octave below what is indicated in the score) would be too strong and too loud. It is necessary to make these 'bass' players become sensitive to the volume of sound that they create. They must not overwhelm the singer, particularly since other instruments (the violoncello is holding out the notes as written) provide a sufficient foundation for the singer.

2. an organist who makes use of the pedal board needs to be mindful that the noise generated by old trakker-action organs is so disturbing that curtailing the playing of notes, even short ones, is a priority (Heinichen [see below]is quite concerned that in holding any long note on an organ there might occur a 'cipher' [the note sticks for various mechanical reasons and destroys any performance until the problem
can be solved.]

3. this leaves the harpsichordist to fill in necessary chords to help keep the singer on pitch and in the correct tonality, but the harpsichord can not sustain the long notes without using devices such as continually sweeping arpeggios which are frowned upon in a church-style recitative.

4. as mentioned above, an instrument, most likely the violoncello/viola da gamba as a key member of the continuo group does sustain the notes as written with the organ (on the manual{s}) holding the chords, but sensitively releasing along with the violone the 16' sound which could easily overwhelm the singer.

Conclusion regarding the illustration:

The top line in the bass line (the original notation with long, held notes by the composer) is still being played as written by the violoncello/viola da gamba (as the sustaining sound) while the pedal notes (if they are played at all by the organist) must not be too short, but also must not be sustained for their full value because they would otherwise be too loud for a single voice.

The same applies to the violone. It reads (or knows that it should read) the long, extended notes and octave lower than written with much shorter values as indicated in the lower half of the notes given in the bass clef. The organist will also sustain for full
value the chords associated with the figured bass.

It is Johann David Heinichen in his "Neu erfundene und Gründliche Anweisung zu vollkommener Erlernung des General-Basses," Hamburg, 1711, p. 226, who makes the clearest statement about the manner in which church-style secco recitatives are to be played:

".man schläget die 'Noten' meist nur platt nieder / und die Hände bleiben hierbey ohne weiteres 'Ceremoniel' so lange liegen / biß ein anderer 'Accord' folget / mit welchen es wiederum / wie zuvor /gehalten wird."

["you simply press the 'notes' [the keys which make the notes sound] down flat [all the way through so that all the notes/keys can produce the desired sounds] and, in/while doing so, your hands remain down [with the keys depressed creating the sounds you want] without any further ceremony [no showing off with special embellishments or wide ranging arpeggios] until another/the next chord follows, after which you follow the same procedure as before."]

Nothing could be more absolutely clear than this. Heinichen continues with some extraordinary circumstances when it might be necessary to use a different technique/method of playing, but as a general, primary rule which covers the situations that a keyboard player [in this case, obviously an organist
is meant] might encounter in playing church-style secco recitatives, this is the best historical source to follow. It is, however, the one source which HIP musicologists such as Dreyfus, attempt to explain away by declaring Heinichen's statement as a 'puzzle.' On pp. 76-77 of Dreyfus' book you will find evidence of how much effort is expended by Dreyfus in attempting to come to terms with the clear language used by Heinichen. Dreyfus accuses Heinichen of being 'contradictory' 'getting it backwards' 'catering only to the novice' of 'not knowing that, on an organ, arpeggios do not work as well as on a harpsichord. What an audacious statement on the part of a musicologist who is 'not in the same league' as Heinichen in understanding what performance practices were acceptable or not during Bach's day!

Decide for yourself what kind of scholarship this is that has to go to such extreme lengths to uphold a very shaky theory that certain HIP protagonists had already taken a liking to. Dreyfus' accusation that Heinichen is indecisive can only spring from the bias with which Dreyfus approaches Heinichen's clear statement.

Nils Lid Hjort wrote (November 29, 2005):
Thanks to Thomas Braatz for opening the can of worms again and for lucid discussion of relevant historical sources. I do not have Dreyfus' book and can't judge his (Dreyfus') opinions regarding Nietz and Heinichen accurately, but Braatz' analysis of what Johann Heinichen actually meant in his "Neu erfundene und Gründliche Anweisung zu vollkommener Erlernung des General-Basses" (1711, Hamburg, page 226) appears convincing. The problem for me is our Bach community's prospective leap from

(1) "ok, granted, then, so this is indisputably the way this particular authority wanted continuo players to perform, when he gave his opinion in Hamburg, 1711, and yes, Dreyfus is wrong regarding this detail, " to

(2) "this is the [only] way all secco-style recitatives must be played, in all Bach cantatas 1707--1748 ".

We still do not know (I presume) whether Bach in Leipzig 1724 agreed with Heinichen in Hamburg 1711, so to speak (neither do we know whether Bach knew about Heinichen or his book in the first place?). I'm making the simple point that (1) does not (in itself) imply (2).

The list's continued continuo discussion reminds me of the Bigger Continuo Sound that is sometimes used for other baroque works; I don't know whether Bach occasionally called for More Forces for the continuo parts, in his cantatas or elsewhere? E.g. in some of the secular cantatas?

I'm e.g. quite fond of Freiburger Barockorchester's version of Le Quattro Stagioni (with Gottfried Graf von der Goltz in the lead, DHM, 1997), which employs a jolly band of some 17-18 musicians (they're not easy to count) in one Big Continuo Group:
2 violincelli, violone, violo da gamba, lirone, contrabasso, leuto, 2 citteri, arciliuto, 3 guitars (!), tiorba, cetarone, clavicembalo, organo legno/regale, arpa doppio, arpanetta.
This group easily outnumbers (but does not out-sound) the "orchestra", with 10 strings plus vd Goltz. They spell out all their historically informed reasons for allowing themselves such a mixed continuo band in the CD booklet. The result is splendid.

I'm mentioning this example also to make the elementary point that rather different interpretations of "how to do the recitatives" can all be effective and successful. This should of course not stop us from further enquires into "how Bach really meant it", however.

Douglas Cowling wrote (November 29, 2005):
Nils Lid Hjort wrote:
< I'm e.g. quite fond of Freiburger Barockorchester's version of Le Quattro Stagioni (with Gottfried Graf von der Goltz in the lead, DHM, 1997), which employs a jolly band of 17-18 musicians (they're not easy to count) in one Big Continuo Group:
2 violincelli, violone, violo da gamba, lirone, contrabasso, leuto, 2 citteri, arciliuto, 3 guitars (!), tiorba, cetarone, clavicembalo, organo legno/regale, arpa doppio, arpanetta.
This group easily outnumbers (but does not out-sound) the "orchestra", with 10 strings plus vd Goltz. They spell out all their historically informed reasons for allowing themselves such a mixed continuo band in the CD booklet. The result is splendid. >
The Big Band Continuo sound can be quite exciting. The recent telecast of Opera Atelier's production of Lully's "Persee" used a huge contunuo group with everything imaginable that could be plucked, played or strummed. I know that is very different from Bach's preferences, but I've always wondered if the St. John Passion (BWV 245) used organ, harpisichord and lute in varying combinations -- they were certainly all present in the choir loft. The only aspect of B.B.C. I dislike is the wearying sound of 16' Violone in secco recits.

Eric Bergerud wrote (November 29, 2005):
[To Nils Lid Hjort] This subject came up a few months back. Below is a post made by yours truly at the time:

I'm a great fan of the Harnoncourt/Leonhardt cantata cycle but can hardly voice any opinion on musical issues raised recently. However, I do believe that Harnoncourt touches on some of the matters raised in his essay written in 1970 that accompanies his (to my ears) wonderful SMP (BWV 244).

"The difference between the notation of the Evangelist's recitatives in the score and in the holograph organ part is rather striking.

This discrepancy has led to a great deal of confused speculation. The score was written after 1741, and the parts, it seems, shortly afterwards. As is well known, it is standard musicological practice to regard the chronologically latest source as an expression of the composer's final will. In this case the version contained in the parts is taken to represent an emendation of the score. However, apart from the fact that it would have been most unusual if Bach had wanted to make such significant alterations after having devoted more than 15 years to this work, it is quite implausible that he would suddenly have wanted to introduce a new style of accompanying recitatives in the St. Matthew Passion (BWV 244). In all the sacred and secular cantatas and in the St John Passion (BWV 245) he had notated the recitatives as in the score of the St. Matthew Passion (BWV 244).

In the booklet for our recording of the St John Passion (BWV 245) we pointed out the difference between what is notated and the actual performance of the recitatives. In secco recitatives, going by rules that were repeatedly written down, each bass note was only allowed to be played briefly (by the cello and accompanying keyboard instrument). This convention was well understood by ever continuo player at the time. However, the notation had to show the correct harmonies between the vocal line and the bass, whereby in practice the bass note continued to sound only in the listener's imagination. In this way it was always possible to understand the text quite clearly. Similarly, there are differences between what is written and what is played in the case of final appoggiaturas (here, in the above example, the two c1 had to be notated on "aber" because a dissonance would be incorrect at this juncture. However, going by the rules, the singer sings d1 c1). In the continuo part Bach exceptionally notated what was actually played and not the normal and orthographically correct long bass notes, as in the score. He probably wanted to ensure that the differences between the short notes in the Evangelist's recitative and the full note values in the recitativo accompangnato of Christ's recitatives, which were hardly noticeable in the part, would not lead to confusion. There is in fact no difference between the original scores of the St Matthew Passion (BWV 244) and the St John Passion (BWV 245). Differences only exist in modern reprints because the parts are incorrectly interpreted as being Bach's revisions.

Bach was clearly at pains to write down everything as precisely as possible for the musician. This ran counter to the freedom usually accorded to performers in the 18th century, when it was the practice to allow singers and instrumentalists to improvise embellishments in solos and sometimes even in accompanying parts. Bach did not want to leave such things to chance in his works, and thus wrote out all the embellishments in full. Many of Bach's melismas and coloratura passages must be understood as written-out ornaments, and these of course have to be played far more lightly than essential melody notes.

One arrives at the natural tempo by extrapolating the actual motif in its unembellished form..."

Thomas Braatz wrote (November 29, 2005):
Eric Bergerud wrote:
>>I'm a great fan of the Harnoncourt/Leonhardt cantata cycle but can hardly voice any opinion on musical issues raised recently. However, I do believe that Harnoncourt touches on some of the matters raised in his essay written in 1970 that accompanies his (to my ears) wonderful SMP (BWV 244).<<
There are some remarkable statements here (not all of them factually true based upon research since that date), but this is a relatively early date in his career of performing Bach. The Bach Cantata cycle had not yet begun and he changed his mind on many issues reversing himself from earlier positions that he had taken. It is great to see such flexibility in thinking, but imho he changed in the wrong direction, forgetting what he had quoted about Bach putting down in notes everything he wanted to hear. By 1995, Harnoncourt was writing as follows: (Was is Wahrheit? 1995, p. 26) "Buchstabentreue ist nicht Werktreue" essentially 'following and performing exactly what Bach put down in the score is not being faithful to the work itself -- the creative freedoms and license granted to the performers by the performers themselves when they believe they are being faithful to Bach's intentions is the most important element in performing his music.'

Do a search on 'Harnoncourt' and 'recitative' on the BCW and you might be able to find my previous discussion of where Harnoncourt later used anachronistic sources for trying to prove that the theory of shortened accompaniment for secco recitatives was documented clearly by a Jean Baumgartner c. 1774. Baumgartner does not even distinguish between church and chamber or opera style of secco accompaniment. Based on shaky information such as this, Harnoncourt then followed what he thought was the manner in which secco recitatives were performed under Bach's direction.

William Rowland (Ludwig) wrote (November 29, 2005):
[To Nils Lid Hjort I do not understand the logic if there is any of such a large continuo group.

Personally, I prefer to limit my continuo players to no more than 5 at max and prefer 1 either an organist or a harpsichordist. If the literature concerning Bach's instrumental players is true then we can safely assume that Bach's continuo players were reduced to less than 4. Harmonicourt also seems to keep his continuo players to less than this. These days unless one is well versed in Jazz; a continuo group is likely to fall on it's face one of the better reasons to limit continuo to one player anc certainly no more than 5.

Neil Mason wrote (November 29, 2005):
[To Thomas Braatz] I'm intrigued that you think Harnoncourt has changed his mind, but for the wrong reasons, when it seems that your views on HIP performance turn on the fact that you don't like Harnoncourt's early performances.

I'm inclined to believe that someone of Harnoncourt's world-class musicality might change his mind for musical reasons, in other words to make the sound more musical.

Neither am I convincedthat Bach wrote down all he wanted to hear. One only has to consider the recitatives to realise that there are implied appogiaturas etc that are not notated in the music.

Thomas Braatz wrote (December 1, 2005):
Neil Mason wrote:
>>I'm intrigued that you think Harnoncourt has changed his mind, but for the wrong reasons, when it seems that your views on HIP performance turn on the fact that you don't like Harnoncourt's early performances.<<
If you were to check out (on the BCW) my comparisons of various recordings of a given Bach cantata where Harnoncourt's recording was among his first dozen or so of Erato series, you will find that I generally found these early Bach cantata recordings by Harnoncourt to be among the best in the series. After this, imho, I perceive a steady decline as Harnoncourt attempted to 'standardize' his approach by shifting away from some of his earlier practices to include more of those which tended towards extremes in shortening notes unnecessarily and paying much less attention to obtaining a 'beautiful, cantabile sound' than chopping Bach's wonderful phrases into many tiny bits. The change, as he documents in his writings, was toward emphasizing the 'ugly' because the 'beautiful' was too boring in his estimation. The 'beautiful' was identified with what the ordinary listener had come to expect from good classical music in performance, but now Harnoncourt decided it was time for something else. There are still a number of performers today who follow Harnoncourt's lead in presenting 'ugly' sounds because they think that this type of performance has authentic validity, but nothing could be further from the truth as far as Bach's own performances of his sacred works is concerned.

Continue of this part of the discussion, see: Harnoncourt & Leonhardt - General Discussions - Part 7 [Performers]

Neil Mason wrote (December 1, 2005):
[To Thomas Braatz] Thanks for clarifying your views.

I still find this intriguing, for it does seem to me that Harnoncourt is now branching out into a wider repertoire than before.

Eric Bergerud wrote (December 2, 2005):
[To Neil Mason] I would be more impressed if Harnoncourt (or Leonhardt) recanted or issued some indication of regret for their earlier cycle: a kind of "My Bad" declaration. Harnoncourt, especially, emerged from the cantata cycle with his reputation quite unscathed. I have Harnoncourt CDs from the 80s and later that roam very broadly and are very good indeed. I don't take a broadening or changing of horizon as a repudiation of other work.

I for one remain very grateful to Harnoncourt and Leonhardt for their great accomplishment. It is quite clear that nobody else is going to use all male choir - one of the few practices we know without doubt that was followed by Bach - and thus their cycle will remain distinctive, unique and to my ears beautiful.

Continue on Part 15

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 | Part 15 | Continuo in Bach’s Vocal Works: Part 1 | Part 2 | Part 3 | Part 4 | Part 5 | Part 6 | Part 7 | Part 8