JS Bach's Partita N°1 in B flat major

[guermantes]

I'm currently struggling with the Allemande of Bach's Partita N°1 B flat major. The problem is identifying the succession of voices in the soprano compound line.  By a compound line, I mean a melody that is notated in one line but that contains more than one voice such as in Bach's Cello suites. 
The only clue to the fact that the soprano is a compound line can be found in the break in the barring of the otherwise normally grouped sixteenth notes in measure 1 (Henle edition). It is essential to the interpretation of this movement to decipher the beginnings and endings of each voice's melody because the phrasing depends on this knowledge. Has anyone dealt with this problem when working on this Parita?
I am aware that the Kalmus Study Scores of the Bach-Gesellschaft edition for Bach's organ works are useful for determining such points of phrase indication when some doubt exists in modern editions, but I do not know where I might find this information for Bach's Partitas.
Any help gratefully received.
Béryl

[allchopin]

From my experience, barring in the soprano line is not the only indicator of compound melody.  Barring is obviously notation for different voices, but range can tell you about how a line is voiced also.  Take, for instance, the first 2 bars:

One voice of the compound line could be realized as the stepwise line C, D, Eb,  D, Eb, F, etc. while the Bb's on the upbeats are an alto line.  Compound voices are usually best identified through their stepwise motion that also follow counterpoint rules.  So any jumps are clues to a change of voices.  Bach's cello suites are good for this because they only contain one line that is constantly changing voices.

[guermantes]

Thank-you Allchopin for your informative reply. I seemed to manage to chomp through all the first movement of Partita N°1 (Praeludium) without too much problem with the compound soprano line by more or less applying your suggestions.
However the Allemande is a very different beast. If you have a chance to take a look at the music, you will notice that the stepwise motion that begins mesure 1 with F as its first note, continues throughout mesures 2, 3 and 4. That's all well and good. On even closer examination, this stepwise motion would appear to be the LAST note of a phrase in the soprano. (In mesure one, the last three sixteenth notes of the fourth beat constitute the following snippet of the next appearance of the soprano voice which finishes on the first note of measure two and so forth for the other two appearances of this soprano voice.)
Now the soprano voice has been identified, what are the implications of this discovery ? It would seem that since the first note of these four measures is in effect the last note of a phrase, then from the point of view of interpretation, the first beats of these measures will not receive the stress of a normal downbeat since the last note of a phrase is not normally stressed. The first note of the next beat in each of these measures is within another voice, shall we say the alto. That note has been decorated by a lower neighboring tone just previously. I would hasard the thought that it is the second beat in these measures that will be played more strongly than the first beat. This observation indicates a conflict with the metrical rythme indicated by the barlines.
I was hoping to carry on identifying the voices of the following bars in this Allemande in order to verify if this rythmic conflict continues throughout (I think it does except for the perfect cadence in F Major mesure 18 and its preparation mesure 17) and I am having problems doing this.
This discovery implies hand and arm mouvements compatible with usual first-beat stress displaced to the second beat and consequent impact on phrasing.
I would appreciate your comments.
Thank-you in advance.
Béryl

[allchopin]

Two of the last three sixteenths of the first few bars I would not consider part of the soprano line.  I have identified what I would most likely consider to be the soprano in green, and another alto voice in red (though in no way definitive). 


Only the last sixteenth of the last beat would make the soprano, as a rising, stepwise line, which is usually the best indicator.  Some of the parts where I only circled the highest notes could also include some of the stepwise notes coming into it but I just stuck with the highest notes to be safe.  But most of the compound melody in this section is actually in the left hand rather than the right.

Not every voice has to be stressed, necessarily, to be played as a separate voice.  If it happens to land on an upbeat, just play it as an upbeat - the usual rules of stresses should apply to the bar as a whole, but don't worry too much in sub-divided beats.  As you count each group of four sixteenths you should feel a 1 2 3 4 pulse.  I'm not an expert on compound melody and this is just my interpretation of the voicing - there are many good websistes about this topic and others here who are more knowledgeable who can help more than I probably can.

[guermantes]

Dear Allchopin,
Thank you for taking so much trouble to correspond with me, in particular for the illustrations of the music. I am trying to create a file with my software to illustrate what I'm talking about and am finding it difficult. I use Encore to create the score, export to Powerpoint where I can annotate the score, and then somehow get back to you.
Still on the same subject of phrasing in Bach scores, I came across a very interesting discussion of this topic in "Organ Technique, A Basic course of Study" by Oswald G. Ragatz, Indiana University Press. He goes into a lot of detail with illustrations from organ scores, where barring is used to determine phrasing. From p.168-9 of his text :"While the Baroque composers did not use the slur mark to indicate a phrase, as have composers and editors during the past hundred years or so, there was a device to indicate phrasing. This device too often has been overlooked by modern editors (with a notable lack of responsibility!) who did not understand the separation of the barring of eighth and sixteenth notes, and consequently merely barred all notes of a beat together." He prints an example from Komm, heiliger Geist. "Only by checking the Bach-Gesellschaft edition, which fortunately is available in miniature score from Lea Pocket Scores and the Kalmus Study Scores at a reasonable cost, is one able to determine such points of phrase indication." He prints further examples from the Prelude in d minor S-554 and the Prelude in e minor S-533 with barriing brakes that indicate a change in voice.
Nikolaus Harnoncourt in his two books on musical discourse in baroque music also has some interesting insights on the phrasing and articulation appropriate to this period of works.
Anyway, the Henle edition has several such bar breaks and I am trying to prepare the score to include in a message to this forum but guess it will have to wait until tomorrow...
Thanks again for your interest in my question.
Béryl

[guermantes]

I found the Allemande particularly difficult to start because I did not initially take into account the phrasing suggested by the break in the barring of the music (Henle edition). Once I was able to handle the first four measures, the rest seemed to fall into place without problem even at speed. Suggested phrasing for Bach’s Allemande Partita N°1 follows with a few comments and list of practice sessions.

•   m.1-4 Because the initial phrase finishes on the first beat of m1, it might be better not to give the first note of this and the following 3 measures the accent associated with the first note behind a bar. A phrase does not usually end on an accent. The first strong beat in these measures would be the second beat that occurs within the following phrase. The F – G – A and Bflat which are the first notes of these measures are an obvious step-progression but the phrasing may not make it desirable to highlight them.
•   m.19-20 Similar to 1-4 in phrasing.
•   m. 9-11 Different articulation could distinguish the two voices in the left hand. Voice 1 : F – A – C – Bflat Voice 2 : Bflat – D – F – E. Could also choose to juxtapose one of the  voices in the right hand with one of the two in the left for a bit of variety that is : LH : F – A – C – Bflat and RH : D – C – Bflat – A – G to imitate a conversation.
•   m.24-27 Similar to m. 9-11
•   m. 14-15 In the phrase played by the left hand, it might be possible to highlight the notes that are circled by accenting them or by prolonging slightly their rhythmic value (that is, holding the note while playing some or all of the three following notes). If a crescendo or diminuendo is desired for this passage, it is much easier to put into place by playing only those circled notes in the left hand while playing all the right hand. In the right hand, the trills seemed easier to negotiate with a slight break in the phrasing just before playing them. In Baroque music, trills may start new phrases so there is some justification...
•   m. 29-30 Similar to m.14-15

Practice sessions Allemande

01   36-38
02   32-36
03   32-38
04   29-32
05   27-29
06   27-32
07   27-38
08   24-27
09   21-24
10   21-27
11   19-21
12   19-38   End of 2nd part
13   17-19
14   14-17
15   12-14
16   12-19
17   12-38
18   09-12
19   07-09
20   07-12
21   07-19
22   07-38
23   04-07
24   01-04
25   01-07
26   01-19   End of 1st part
27   01-38   Everything

Maybe this will be useful.

Béryl