I am learning the D minor prelude from book 1 of the well-tempered clavier. I had been following the Henle Verlag Urtext edition, but lost access to it and fell back to this Ricordi edition from IMSLP. I noticed some differences between the two scores, as follows:

  • m. 14: in the left hand, the Henle has the first note of the measure (Bb) an octave lower than the Ricordi
  • m. 14: in the right hand, the triplets over beats 2+, 3, and 3+ are (E C F), (Eb B F), and (D A F), whereas in the Ricordi they are (E C A), (E D F) and (Bb A F)
  • m. 25: the Henle writes the chord as (B D F G#)/(D F G# B), while the Ricordi has (B D F# G)/(D F# G)

Looking around at some of the other scores available on IMSLP, it seems like most (including this manuscript) are in agreement with the Henle version on all counts. The Henle version also sounds correct to my ears (both in itself and as compared to recordings.) However, this version edited by Czerny agrees with the Ricordi.

My questions:

  1. Which version, if any, is deemed canonical?
  2. Which version is played on most recordings (Gould, Richter, Tureck, Schiff, Pollini, ...)
  3. Why is there such a divergence between versions?
  4. Why is this divergence not noted in any of the versions I've seen, or discussed more broadly (I couldn't find any information about it online)?

I would be very surprised if the answer to 2 was not "the Henle version," but it can be hard to discern precisely what some of the faster players are doing in measure 14.

  • 1
    If you'd like to return to an Urtext edition, and don't want to pay money, IMSLP has the Barentreiter, and it seems to be public domain without restriction by region. Commented May 14 at 14:35

1 Answer 1


The preface to the Henle edition points toward the likely source of the discrepancies.

By an ill stroke of fortune, three of the most important manuscript copies that deliberately sought to imitate the layout and handwriting of Bach’s original were long thought to be autographs as well. This mistake had dire consequences: editors saw themselves confronted with a total of four allegedly autograph manuscripts that differed from each other in myriad details, and every deviation was thought to represent a legitimate alternative reading.

It stands to reason that the Ricordi edition was based on one of the "bogus" autographs.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.