As we know today, the six sonatas for organ which Bach compiled around 1730 into a fair copy (still extant today) do not represent a work cycle that was newly composed in its entirety. Some of the altogether 18 movements were based on earlier material that was revised for its new context and new scoring. Sonata No. 4 in E minor (BWV 528) seems the clearest in this respect: a cantata sinfonia of 1723 (BWV 76/8) documents an early version of the first movement; the middle section has come down to us as a single-standing organ piece (BWV 528/2) in at least two early versions; and the closing movement is associated with one of the various versions of the Prelude and Fugue in G major BWV 541. The theory that BWV 528 is a compilation of various independent movements – a seemingly obvious and oft-defended theory – appears to be further supported by the fact that the middle movement is in the dominant minor, which Bach rarely uses in instrumental cycles. Upon closer look, however, this conjecture proves to be unfounded. On the contrary, the three movements are so intricately interconnected with respect to form, themes and proportions that there can be no doubt that they belong together.
Contrary to the view held until now, the Sinfonia that opens the second part of the cantata “Die Himmel erzählen die Ehre Gottes” (BWV 76) is not a new work, but was borrowed from an earlier piece that, judging from stylistic criteria, can be dated at ca. 1714 (the most closely related work is the Sinfonia of the cantata “Tritt auf die Glaubensbahn” BWV 152, written in December 1714). The style of the middle movement of the Trio Sonata, which is closely related in form and style to the first movement and whose “Weimar” background can be clearly discerned through the earliest of the transmitted D minor versions (see “Rekonstruktionsbericht” [reconstruction report]), allows the same chronological classification. The final movement of our sonata, in contrast, is transmitted exclusively in its definitive form as BWV 528/3. Although it is unmistakably the most mature movement, there can be little doubt that it originated in the same stylistic environment as the other two. The reconstruction of its original form presented here is – contrary to that of the first two movements – of a more hypothetical nature (further information can be found in the “Rekonstruktionsbericht”).
This edition thus brings to light an ensemble sonata from the middle Weimar period, which was conceived for oboe, viola da gamba (or viola?) and basso continuo. It is all the more valuable as it stems from a period of Bach’s life from which only few instrumental chamber works are transmitted. Up to now, only the Fugue for Violin and Basso continuo in G minor BWV 1026 and the Sonata for Violin and Basso continuo in E minor BWV 1023 could be assigned to this period with any degree of certainty. The Trio Sonata BWV 528a thus represents a major contribution both to honing our image of Bach’s Weimar oeuvre and to increasing the repertory of works with mixed instrumental scoring.
The question of the work’s original key is closely connected with the reconstruction of the original scoring and its time of origin. The sonata shares with Bach’s (early) Weimar cantatas the problem of the various tuning standards between strings and organ on the one hand, and woodwinds on the other. While Bach’s Weimar organ and the stringed instruments adapted to it were tuned in the traditional organ pitch (about a semitone higher than the present pitch: a¹ = ca. 465 Hz), the new woodwind instruments from France played in the lower, “French” chamber pitch (about a whole tone lower than the present-day pitch: a¹ = ca. 392 Hz). The resulting difference of a minor third can be observed in the sources of many Weimar cantatas (in our context, especially in Cantata BWV 152) and no doubt also formed the basis for the original version of the Sonata BWV 528; here the viola da gamba and basso continuo played in E minor, while the oboe was notated in G minor. This mixture of keys is documented in the autograph transcript of the Sinfonia BWV 76/8, where Bach exploited the fact that the new Leipzig oboe d’amore was tuned a minor third lower, which allowed the piece to be played without difficulty in Leipzig as well, where all the instruments were uniformly tuned to chamber pitch (about a half tone lower than the present-day pitch: a¹= ca. 415 Hz). The oboe d’amore thus continued to play in G minor (finger notation) and the other instruments in E minor.
Thanks to this truly special background, we can derive the following possibilities of performing this sonata, which are all offered in this edition:
A: the “mixed” version in G minor / E minor, with the oboe (a¹ = ca. 392 Hz) in G minor and the viola da gamba and basso continuo (a¹ = ca. 465 Hz) in E minor;
B: an E minor version with oboe d’amore, viola da gamba and basso continuo (here the continuo player performs from the [figured] continuo part);
C: a G minor version with oboe, viola and basso continuo (here the continuo player performs either from the score or from the [figured] continuo part).
Possibility A, which presumably corresponds to the original Weimar version, can no doubt be realized today only in exceptional cases. However, the other two possibilities constitute very plausible alternatives, which can be legitimated through the history of the sonata’s different versions. In the G minor version (possibility C), the gamba part is sometimes too high and can be better played on the viola – an instrument, it should be said, that might very well have been intended as the original middle voice. Solution B can, however, claim greater authenticity, as it follows A more closely; this version with oboe d’amore is confirmed for the first movement through its transmission in Cantata 76. This then novel instrument permitted Bach to play the entire sonata in Leipzig from the Weimar parts without any problem – and he most probably did so! Finally, one should also point out the possibility of replacing the oboe with a violin, thus reflecting Baroque practice (or with a flute, when using modern instruments), or of combining the oboe and basso continuo parts on the harpsichord, which gives rise to an effective version for viola and harpsichord obbligato; both variants conform with the G-minor version of the sonata printed in the score.
In the musical text, each movement is based on a different primary source in keeping with the heterogeneous state of the documentation. The “Rekonstruktionsbericht” provides detailed explanations of the editorial decisions. This edition has been made possible through the kind authorization of the Staatsbibliothek zu Berlin – Preußischer Kulturbesitz, Musikabteilung mit Mendelssohn-Archiv, as well as the Musikbibliothek der Stadt Leipzig.

Culemborg, Summer 2011

A first recording of the present reconstruction was released in 2009: Johann Sebastian Bach: Sonatas with Viola da Gamba and Harpsichord – first complete recording; Alfredo Bernardini (oboe d’amore), Cassandra Luckhardt (viola da gamba), Pieter Dirksen (harpsichord); EtCetera KTC 1365.

We offer a PDF of the complete text with all notes for download.