Bach Art CDA67138

Déjà Review: this review was first published in April 2003 and the recording is still available.

Johann Sebastian Bach (1685-1750)
Die Kunst der Fuge, BWV 1080 (arr. Robert Simpson; final fugue completed Sir Donald Francis Tovey)
Delmé Quartet
rec. 1999, Temple Church, London
Hyperion CDA67138 [74]

What is popularly regarded as Bach’s unfinished final work has been enjoyed in concert only since June 26, 1927 when the first orchestration by Wolfgang Gräser was performed at St. Thomas Church in Leipzig. But there have always been two problems with the work. First, for what instruments was it intended? Or was it intended not to be performed at all but merely contemplated in score by composition students? Sir Donald Francis Tovey was one of the first to assert that it was a keyboard work, as, more recently, have Charles Rosen and Gustav Leonhardt. In the eighteenth century complicated keyboard works were occasionally written in open score to make them easier to sight read. So the work has been frequently recorded as a solo work on the organ, harpsichord, and piano. The two-piano transcription by Erich Schwebsch was inspired directly by Gräser’s version. It is performed by (Millette) Alexander and (Frank) Daykin with tremendous drama and rich lyrical sensuality of sound, and without any of the pounding beat usually encountered in piano-duo playing. This underscores the most successful approach to this music—play it as music, not as an essay in multiple voices. And since some parts of the work are marked “for two keyboards,” it’s not unreasonable to play the whole work thus.

The alternate approach is to assign the four voices to different instruments, the most obvious combination being the string quartet. Robert Simpson points out that the ranges of the voices are wrong, unless the work is transposed to g minor, something Bach would have done himself without a second’s hesitation. Bach, whatever else he was, was a performing musician. Simpson, in this arrangement, simply omits the keyboard canons, suggesting (not terribly plausibly) that Bach intended eventually to group them into a separate work.

Goebel presents the work played by strings, but as written (without transposition) and this requires more than four instruments to accommodate the ranges of the voices. He typically uses aggressive original performance techniques and achieves a sound not unlike that of a chest of viols. But 7 of the 22 tracks are played by harpsichord(s). At times you think you’re listening to a harpsichord performance of the work.

Scherchen’s version is for small orchestra including strings, winds, and brass, and is the result of a lifetime of work and study. In 1966 Scherchen was interviewed by French Television and he said that the sense of The Art of Fugue is: ‘… as if four people discussed a same topic. The first one would say, “Oh, life is very difficult indeed.” The second one would say, “Life is not that difficult.” The third one would add, “It is even more than difficult, it is terrible.” And the fourth would then ask, “What shall we then do with life?” Once the counterpoint begins, we start thinking … listening to what the other one has to say …’ Scherchen assigns the rectus fugues to the winds, the inversus fugues to the strings. He obtains overpowering passion from his players and brings the music to a pitch of critical immediacy. This is probably the single most successful arrangement of anything by Bach for modern orchestra.

And here we come to problem number two: the last fugue, already the longest fugue Bach ever wrote, is unfinished. It just stops about 3/4 of the way through, and so do most performances, including Goebel and Alexander/Daykin. Dr. Simpson rightly ridicules this ‘… sentimental practice of allowing the music to trail off into thin air like the spirit of the frustrated composer being dragged off to heaven. …’ One alternative is simply to omit the unfinished fugue, but considering that if finished it would have been one of Bach’s very greatest works in fugue form, this is also unsatisfactory. In 1930, Sir Donald Francis Tovey completed the fugue according to the most informed scholarship available to him. The Delmés use that completion (also including a playing of the work incomplete, for those inclining to sentimental flourish.)

But that’s not completely fair. Tovey’s completion, even considering its exceptional virtues, has problems of its own, and even unsophisticated listeners can perceive them. Simpson says, “it may be reasonably doubted that Bach himself would have ended the fugue very differently …”. One must take slight exception to that statement. First, the scholarly argument is that Bach had already allotted the space on the engraved plates to his completion, and it was about half of the length of Tovey’s completion. Bernstein in his Norton lectures makes the point that artistic genius lies in breaking symmetry; Tovey in his completion observes all symmetries. There are harmonic transitions in Tovey that are awkward, even clumsy, and Bach would have been incapable of such gaucherie. The mood of the beginning of the fugue is tragic, but Tovey’s completion follows a Wagnerian logic, moving gradually to a mood of triumph, even grandeur, then to repose. The Baroque aesthetic would have Bach either stay in the tragic mood to the end, or move abruptly to one of ecstasy. I have loved the Tovey fragment for years, and have prepared an orchestration for full orchestra, and this is the context in which the harmonic texture became evident to me. For the Bach part of the fugue, one follows Baroque orchestration practice quite naturally, but when one moves across the transition into Tovey, one starts hearing Wagner tubas and trombones, even cymbals and bass drums. No, if we had Bach’s version, it wouldn’t be the same, and that is probably what has kept Tovey’s completion, in spite of its many virtues, from entering the performance canon as the accepted version.

What is right about the Tovey completion is that it brings us down from a high pitch of musical ecstasy to a satisfying musical conclusion, something other “completions” fail at utterly. It’s just what we wanted and needed to hear. If we had the Bach version, perhaps some would prefer the Tovey. At the very least the Tovey completion can be appreciated in the same light as those plaster interpolations added to anciently damaged sculpture which allow us to perceive the original sweep and reach of the work without distraction. But what did Bach really have in mind? Is it possible that he intended to conclude the work with a segue into a harmonised chorale? The so-called KDF motto is a slightly varied version of the Christmas song Wie schön leuchtet der Morgenstern. One of Bach’s harmonisations of this chorale includes all of the Art of the Fugue themes at some point or other in the various voices. He may have mentioned to someone that he was going to finish off the fugue with a chorale, and that may be the reason they mistakenly assumed he meant Vor deinen Tron and appended it to the publication. Maybe that’s the reason nobody noticed among his papers his sketched completion of the fugue, that was seen as just another version of Wie schön, of which he wrote so many.

Scherchen deals with the incomplete fugue in a really interesting way; he takes the last few bars of the unfinished fugue to be a cadence, and plays them that way. With his experience in performing atonal compositions by Berg and Schoenberg, he convinces us that there is no problem with the work coming to rest in a remote key on a very strange chord. It works well enough to be taken seriously. It gives us musical “closure” without adding anything Bach did not write.

When the Delmé Quartet play the Tovey completion they avoid the mistakes I’ve made in my performances. Simpson is right: you can’t find the seam at all, the music just goes on by the break. Here when the KDF motto makes its first appearance it’s hardly noticeable. Whereas I make it very conspicuous because I know the “importance” of it, the Delmé sneak it in without a fanfare. Almost subliminally you realise that the fourth theme is there in the mix, and then you become fully aware of it, and then things come down to a very gentle landing. It’s a beautiful, fulfilling satisfying performance.

A fourth solution is just to add a cadential chord after the end of the unfinished fugue, and I’ve heard this done quite sensitively. But my preferred version is still the Tovey completion, and I’m very grateful to him for giving it to us and to Simpson and the Delmé Quartet for having recorded it. In Tovey’s moment of clear sight he briefly stood on the pedestal with Bach and saw what Bach saw. The Gods of music punished Tovey severely for that moment; he endured mental and physical anguish for the rest of his life. At his death his life work with the Reid orchestra and his musical compositions virtually disappeared from the face of the earth as swiftly and quietly as the ending of the Bach fugue.

Paul Shoemaker

Help us financially by purchasing from

Presto Music

*Connoisseur Society recordings are available from: In Sync Laboratories, Inc., 2211 Broadway, New York, NY 10024. Telephone (212) 873-6769.

**The video is all rehearsal, but the first and last (unfinished) fugues, and the canon at the octave harpsichord fugue are presented complete without interruption or comment.