Apr 7, 2017

The bizarre chord ?


5th Suite  in C minor  BWV 1011



Bar 25, first bass note: since Dotzauer (1826), most editions (except New Bach Edition, Icking, Henle, etc.) had been taking b♭,

   Dotzauer (the highest note of the chord is written in scordatura, so it sounds a♭):

   Bach-Gesellschaft Ausgabe (1879):

but all the 4 manuscript copies and Paris First Edition (1824) show g unanimously.


   Anna Magdalena Bach:

   Source C:

   Source D:

   Paris First Edition:

It is difficult to understand this chord because there are only three notes (except Kellner, the highest notes are written in scordatura, so the chord is g-d-a♭). But in the Lute arrangement (in G minor), Bach wrote five notes. So we can understand easily this chord (upper staff is in tenor clef, lower staff is in bass clef, then they are d-f-a-c-e♭).

 Transposed in C minor; then we have g-b-d-f-a♭:

Therefore also in Bach's autograph of Cello Suites, the bass note must have been g, not b♭. But in the New Bach Edition, Revised Edition (NBA rev. 4) , the editor (Andrew Talle) wrote "the bizarre chord" (page XLIV) and had changed it again into b♭, although in the New Bach Edition (not revised / 1988), Hans Eppstein wrote it correctly g, for the first time since Dotzauer.

This chord is not "bizarre". I will explain it.

As we see above, first chord is dominant 7th on the tonic, and second chord (as we see in the Lute arrangement) is dominant 7th on the mediant (third degree of the scale). It is less frequent compared to first one, but sometimes is used. These chords resolve into the next chords in small notes.

   Transposed in C major for easy understanding: 

We can see this chord in the other works of Bach (I have found 10 examples, 14 places), but before that, I will show you two examples from French composer, François Couperin's works.

From Second Ordre, Allemande "La Laborieuse", bar 23:

Leçons de ténèbres, Troisième Leçon à deux voix "Mem":

And then, from Bach's works; for example:

The Well-Tempered Clavier, Book 1-7 E♭ major, Prelude bar 18, 21, 50:

The Well-Tempered Clavier, Book 2-3 C♯ major, Prelude bar 11: 1st chord. I transposed it in C major and added the reduction of harmony written by J. S. Bach himself (BWV 872a):

The Well-Tempered Clavier, Book 2-16 G minor, Prelude bar 12: 1st chord:

One more example from The Well-Tempered Clavier, Book 2-21 B♭ major, Prelude bar 58:

Flute Sonata in B minor (BWV 1030), 2nd movement, bar 13 (Bach's autograph):

Another example from St. Matthew Passion: Finale, bar 11 (and bar 23, 91, 103, the same):

Mass in B minor: Gloria, bar 56:

One more example from Mass in B minor: Quoniam, bar 14 and 91 (musically the same; here is a example of bar 91).

So "the bizarre chord" is no doubt composer's intention. We must not change it into b♭.

Feb 13, 2017

New Bach Edition, Revised (NBA rev). 4


A great disappointment

On November 2016, Bärenreiter Verlag published Bach's Cello Suites of "New Bach Edition, Revised Edition" (NBA rev. 4 / Editor: Andrew Talle).

First of all, there is a misprint in the 6th suite, Prelude, bar 24, 9th note: it is f ♯, not e. Of course, it is a little thing, but...

There are three serious mistakes; they are:

1. 1st Suite, Prélude, from bar 33 to 36: lack of double stems.
Of course, they must be played in double-stopping of open A-string and stopped a on the D-string.
See Neglected double stoppings

2. 5th Suite, Allemande, bar 25, 1st bass note: it is not b♭, but g.
In the New Bach Edition (1988), Hans Eppstein wrote g correctly. Why was it changed into wrong note ??? The editor writes "the bizarre chord" (page XLIV), but we can see this chord in the other works of Bach. For example: The Well-Tempered Clavier, Book 2-3 C♯ major, Prelude bar 11, Book 2-16 G minor, Prelude bar 12, St. Matthew Passion, Finale bar 11, and Mass in B minor, Gloria bar 56, Quoniam, bar 14 and 91.
See With Lute arrangement
If Dr. Talle was the editor of these works of Bach, did he change their notes ?

3. 6th Suite, Prélude, bar 91, last note: it is g, not a. Bärenreiter must correct it immediately into this surprising, amazing, astonishing, beautiful, wonderful, excellent, incredible, remarkable, extraordinary 7th note.
See Two neglected g

Why I said they are "serious mistakes" ? Because on these three places, all the sources are unanimous. I can't completely understand why the editor chosen these wrong notes. For what reason did he choose the notes which don't exist in any sources ! (that is which Bach did not write !)

And I must add that Henle edition (2000) chose correct notes in these places. Why after 16 years, must we go back to the 20th century? This is the third edition as Bach Ausgabe (first edition 1879, second edition 1988 and third edition 2016).  When at all can we see the double-stoppings of the 1st Suite Prélude ?

About the sources, the editor write some interesting things (for example, how Kellner has mistaken at the Menuet 1 of the 2nd Suite), but the most important thing about sources is Sources C and D (= Source G) are the descendents of Source A (Anna Magdalena Bach). We find 17 common errors among Sources A, C and D.

   (See also Preface)

I wonder why no Cello Suites researchers refer to these common errors.
Unfortunately we don't have Bach's autograph of the Suites, therefore it need a some knowledge and sense of music for understand these errors, but fortunately, about the 5th Suite, we have Bach's autograph of his own arrangement for lute, so we can understand easily 7 common errors of the 5th Suite, and at these places Kellner (Source B) is the same as Lute arrangement (except last one; there are only first 9 bars in Gique).  

My hypothesis of the relationship among the sources


Other points in question 

1st Suite 


Allemande, bar 29, 10th note: f instead of d; it is not misprint but is in the Sources C and D, Paris First Edition and Dotzauer. I think it is simple error of Source G. If it was f, I think it needs g before.

2nd Suite

Allemande, bar 9, 3rd beat: lack of a on the upper voice. The editor indicate a as ossia in a footnote, but it is obvious that a in 16th note in the 2nd beat is an anticipation, so it requires another a as chord tone (diminished 7th) on the next beat.

3rd Suite 

Gigue, bar 24, last note: the editor chose e according to Sources C and D, but I think it is Anna Magdalena's error. She wrote e and then rewrote d, but the copyist of Source G understood it was e. It is very unnatural as a passage note, because it consists a new chord (g-c-e) with precedent note (g). This case is very similar as the 5th Suite, Prélude, bar 193 (see bellow).

4th Suite

is very good. I agree to all notes.
See also Attention to the flats!

5th Suite

Prélude, bar 170, 1st note: the editor refer to the possibility of a in a footnote; it is admirable but in truth, it is nothing except for a. And bar 193, 3rd note: it is of course a, there is no possibility of g.
See Scordatura, too complicated

Allemande, bar 25, 2nd 4th note: c-d-b is two 16th notes and a 8th note, not a 8th note and two 16th notes.
See With Lute arrangement

Gigue, bar 16, 2nd note: it is d♭, not d♮.


6th Suite

Sarabande, bar 31, 1st bass note: it is g♯ as Kellner wrote it, not g♮.
See Two forgotten sharps

Gigue, bar 8, 4th note: it is c♯, not e. And bar 18, 2nd beat: e-c-e is probably Anna magdalena's error.
See The finale

Feb 21, 2016

How do we play the Gigue of Partita No.6 ?

This article is not for Cello Suites but for


Partita for keyboard, No.6 in E min. (BWV 830)

Bach wrote two Gigues in binary notation, usually written in ternary (3/8, 6/8, 12/8, etc.) : in "French Suite" No.1 (BWV 812) and in "Partita for keyboard" No.6.

The time signature of "French Suite" is 2/2 (2 half notes in a bar) but that of "Partita" is most unusual 2/1 (2 whole notes in a bar) !

    French Suite No.1 (autograph, 1722)

   Partita No.6 (first edition, 1731)

But in the autograph of 1725, he wrote it in 2/2.

I don't know why Bach changed time signature when he engraved it. Probably he wanted to avoid to be played too fast. But in his original idea, both Gigue has same signature, so we must not play that of "Partita" too slowly. But most interpretations of this Gigue are very very slow. Why ? Because they play it in binary rhythm as it written.

No, no, it is a Gigue ! So it must be played in ternary rhythm. Binary notation of gigue is a tradition from the time of Froberger (1616 - 1667) but practically, it was played in ternary.

Very interestingly, Froberger wrote the same gigue in two notations: in binary (4/4) and in ternary (3/4). We can consider the latter is the realisation of the former.
   in Suite VII
   in Suite XXIII

So, we will try to play Bach's "Gique" (according to his spelling) in ternary. I show you the 1st page of my realisation.

It is better you continue it. It will be a very interesting puzzle !  

Complete score of my realisation is in IMSLP.

Audio file by synthesizer (piano)

Of course, my realisation is an example. There will be plural possibilities especially in the series of four 8th notes. 

I can't believe no one played it in this way during probably 200 years ! This is a Gigue, why no one tried to play as a Gigue.

Yes, I know some players (such as Trevor Pinnock, András Schiff, etc.) play it in "ternary". This approach was proposed firstly by Howard Ferguson and after by Meredith Little and Natalie Jenne in their "Dance and the Music of J. S. Bach".

But to be precise, it isn't ternary but senary because a half note is divided in six. So we can say it is still binary because a half note is firstly divided in two and then is divided in three. It is mere notes inégales or swing, not ternary nor Gigue.

Jan 28, 2016

Error of the 20th century

2nd Suite  in D minor  BWV 1008

Gique, Bar 28, last note: most editions of 20th century (such as Wenzinger, Fournier, Tortelier, Henle, etc.) take b-natural according to Anna Magdalena Bach (AMB). But it is probably an error.

Kellner and Souces C and D take e as below. The notes d-e-f draw natural ascendant line.

   Source C:

   Source D:

There is a parallel period from bar 61 but it isn't the same. In the bar 65 (far right), the diminished 7th (c-sharp - b-flat) is divided into two 16th notes, so it is natural to draw descendant line (g - e - c-sharp).

Most editions of the 19th century such as Paris first edition (1824), Dotzauer (1826), Bach Gesellschaft Ausgabe (BGA/1879), Klengel (1900), etc. had taken e. But after publishing Alexanian edition (1929) that includes the facsimile of AMB's manuscript, most editions followed it.  


History sometimes regresses.

By the way, some cellists play as follows:

They are right because they aren't content with AMB, but of course it is an incorrect solution. Composer often avoids to repeat the same thing.

Jan 18, 2016

Scordatura, too complicated

5th Suite  in C minor  BWV 1011



5th Suite is played with cello whose 1st string is tuned in G; whole tone lower to normal tuning (A). Such tuning is called Scordatura in Italian.

Its advantage is to be able to play the chords of C minor easily. But to write the note is complicated.

Only the notes for 1st string are written whole tone higher. So c (real note) is written as d. But the notes for other strings are written normally. So sometimes it is not clear that the notes of higher a or b are for 1st string (G) or 2nd string (D).

So we had mistaken this note (1st note of bar 170) .

   From the copy of Anna Magdalena Bach:

For a long time, we had considered this note is g (real note) for open 1st string. But it is a mistake. It is a-natural (real note) for 2nd string. Because from next bar, we hear lower pedal point of dominant (G), it is strange to hear the same note before. It must be secondary dominant (dominant of dominant / Doppeldominante) chord like previous bar (bar 169).

Yes, indeed, in the arrangement for lute, Bach uses dominant of dominant chord during 2 bars.

   From bar 167:

   Transposed into C minor (from bar 166):

Bach added the bass notes that don't exist in the Cello Suite, therefore, he had changed the 1st note of bar 170, but the chord is no doubt dominant of dominant.

And if this note is g, it is the same as bar 167 (3 bars before), why did Bach write these notes differently? There is no reason to use different string in the bar 167 and bar 170. It is natural to use 2nd string for the 1st notes from bar 166 to bar 170.

This note is no doubt a-natural (real note) for 2nd string. Of course if you play the 5th Suite with normal tuned cello, you can play this note with open A-string.

Sep 18, 2014

The finale (6th suite: Gigue)

6th Suite  in D major  BWV 1012



Bar 8, 4th note: Most editions show e according to Anna Magdalena Bach (AMB) (and Sources C and D):

But it is probably a mistake because the 7th chord of next bar requires c-sharp as a preparation of suspension. Only Bach Gesellschaft Ausgabe (BGA/1879) take c-sharp according to Kellner:


   BGA (but the chord of next bar is a mistake since Paris first edition):

I show you the harmony reduction for easy understanding:

A similar example from the Allemande of the 6th Suite (bar 6, 2nd and 3rd beats):

Bar 18, 2nd half: Some editions take e, c-charp, e in 8th notes according to AMB (and Sources C and D). But I think it is AMB's error. I can't prove it but I can't find any reason to modify from charming figuration same as bar 2 into three 8th notes. They are too poor, dull, unnatural. So I took Kellner's figuration.

   AMB (2nd bar of figure below):