Re: GG: Art of Fugue

[Bradley P Lehman <bpl@umich.edu>]

On Thu, 18 Apr 1996, John P. Hill wrote:

> Re:  Gould and The Art of Fugue
> I've often wondered why GG never realized a complete set.  He obviously
> adored this work and its strange that the only commercial release was
> an incomplete organ set from the early 60s.

I'm even more surprised that he didn't record it on piano instead of
organ.  It probably would have been even clearer (both texturally and
conceptually), and probably would have sold better too.  Yes, Gould played
the organ as a kid, but he hadn't played it recently when he made that
record.  His recording on organ has always struck me more as a novelty
than as a serious effort that he put all his energy into. 

Here are some of my guesses as to why he never finished it, having 
started it on organ:

Was that same organ still available years later?  Maybe not.

Did Gould have a change of mind about playing it on organ at all, instead
of piano?  Would he have had to start over?  (Remember, in those days he
wasn't recording things twice.) 

Those first nine were done while Gould was still in a touring career.  Was
there any other project which he started before his retirement, and
finished afterward?  (Other than concertos, which present a different
problem.)

Maybe Gould never decided what he wanted to do with the unfinished fugue
at the end? 

Contrapuncti 12 A/B and 13 A/B are "deucedly awkward" (to use a Gouldism). 
13 is especially hard to bring off on the organ, unless one starts using
quite a bit of pedal (there are many wide stretches where notes must be
cut short, playing manuals only: this works fine on the harpsichord, but
sounds odd on the organ).  And if Gould had wanted to do the alternate
versions of 13 (rearranged by Bach for two players) instead, it would have
been overdub time; Gould didn't do overdubbing until later, on the Wagner
piano disc and on a bit of the Beethoven 5th.  Back on 12 and 13: 
performance of both those on the organ almost requires the use of pedal
(for some sustained notes) while they work fine on harpsichord (those
notes will have died away anyway, so can be safely released).  Maybe this
would have required Gould to practice more on organ than he wanted to. 
<grin>

> 
> Re:  Art of Fugue as a Harpsichord Work
> Hmmmm.  I've never played the work on harpsichord, but I had been under
> the impression that AOF was essentially a *compositional* treatise and
> was not intended as an instrument-specific work.  Although clearly
> performable on keyboard instruments, I'm wondering about specific evidence
> that the work was intended to be played on the harpsichord.

Leonhardt (essay publ. 1952) argues it very forcefully (and, to my mind,
convincingly).  Tovey (publ. 1931) also argued strongly for it being a
keyboard work: the only way to fully experience the piece is to play it
with one's own two hands.  (Tovey did his arguing for piano, because the
harpsichord wasn't into a big revival yet.) Both their polemics line up
with my own experience as a player.  I play it myself on harpsichord,
organ, and piano (and as of a few months ago, clavichord)...it works best
on harpsichord.  And as I worked on learning the whole thing, I realized
that the movements are not only compositional examples of fugal technique,
but graded exercises in *playing* technique.  In almost every case, each
movement is more difficult physically than the preceding ones.  In my
experience, one has to develop a new technique of practicing, and a new
technique of learning pieces, especially getting into contrapuncti 11-13. 
(And consequently those are especially satisfying to bring off
successfully, like climbing a mountain, I suppose.)

Back to Leonhardt's most convincing arguments: he deals with physical
playability (the otherwise "pure" musical thought being altered a bit to
fit what two hands can manage); notation (open score being a
well-established practice for solo keyboard works); compass (notes being
too high or low for any other existing orchestration except keyboard...for
example, the tenor line going below what a viola could handle); addition
of new voices in the last few measures sometimes (natural on keyboard,
problematic in ensemble); use of the tenor sometimes as a bass, below the
bass line (arguing against organ with 16' pedal playing the bass line);
use of parallel thirds in the deep bass (which sound like mush on the
organ but are clear on harpsichord); and the issue of long-sustained notes
being more easily droppable on harpsichord than on organ (and he makes
reference to this in other Bach works that are clearly for harpsichord). 
The nature of harpsichord tone causes the ear to project voices forward in
time coherently, even though the actual notes may have faded.  (Listen to
a good harpsichord performance of #7, with the subject going at three
speeds, and hear how clear the counterpoint is despite the fact that long
notes have faded out.)

For an obvious example of that last point, long-sustained notes, see the
end of #6, where the D pedal point in the bass has died away by the point
where the left hand needs to move upward off it to play more notes...but
then Bach has the hand suddenly retake that same D in the tenor to get it
sounding again, and from there the hand can hold it through the end.  This
doesn't make sense on anything except harpsichord or clavichord (or more
recently, piano).  It's especially telling because the manuscript
indicates that he likely wrote that little rhythmic lick in later, as a
compositional trick to get the D back.  The use of harpsichord makes
Bach's cleverness have the proper effect; on other instruments those notes
have less reason to exist at all. 

And then of course there's the huge point that the earlier manuscript
version was written in keyboard score in the first place....  If it had
been explicitly an organ work, there might have been clues as to
registration or to what the pedal is to play (if anything).  

The Art of Fugue was most likely intended for music lovers to play
themselves, in their homes...where they would have had harpsichords and
clavichords, not organs.  (And drawing this back to Gould...it reminds me
of his comment about Schoenberg's use of the piano as a quick and
convenient medium in which to try out his new compositional ideas first. 
If a composer is working on a somewhat "pure" compositional approach that
doesn't rely too much on special timbral effects to make its points, s/he
might well cast the piece for whatever instrument is most generic and
easily available and flexible.  In Schoenberg's case, the piano; in
Bach's, the harpsichord and clavichord.)

-----------------------------------------------------------------------
Bradley Lehman, bpl@umich.edu       http://www-personal.umich.edu/~bpl/