Re: GG: Triple Fugue, etc.

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

[long opinionated comparisons again...]

On Mon, 22 Sep 1997, Andrew J Hrycyna wrote:
 
> As someone who has recently discovered with pleasure the Monsaigeon/Gould
> films on Bach, I want to take exception with Bradley Lehman's dismissal of
> Gould's performance of the final unfinished Contrapuntus from The Art of
> Fugue.  (I'm assuming he means the performance recorded for the film An
> Art of the Fugue, music recently also issued on CD.  Also: if this is
> digging up an old F-minor thread, I hope people will forgive the
> repetition; I'm a new subscriber). 

Yes, that version which is now on CD.  I've seen that film only once, and
it was several years ago, so my assessment was based on the evidence on
the CD.  It doesn't contain GG's pre-game color commentary, or of course
the visual experience of seeing GG labor away in his inimitable manner.... 
:)

> Ok: I love this performance.  Introducing it in the film, Gould discusses
> the way that the Art of Fugue (unlike the WTC) works with a fluid sense of
> key association that harkens back to old Flemish counterpoint and that
> perpetually postpones strong cadences, giving a feeling "of an infinitely
> expanding universe."  For me the performance of the first and third
> sections of this long triple fugue--as well as that of the first
> Contrapunctus, from the same sessions--completely captures this feeling of
> purposeful floating through harmonic space without ever coming to rest. 

I listened to the recording again yesterday (see below) and heard the
"floating through harmonic space" as you mention it, but I didn't pick up
the "purposeful."  To me it sounded a bit aimless by GG's standards; he
usually sounds more decisive about how he wants to play a piece, but in
this one he doesn't sound settled.  Instead (in my opinion) he falls back
on a series of local effects, highlighting some of the compositional
structure (which he generally does well), without giving the whole piece
an overall sense of inevitable flow. 

> By taking it slow and building carefully controlled climaxes in each
> section, Gould emphasizes this sense of gradual unfolding.  He plays 
for 5
> minutes in the first section and it almost feels like one elaborate,
> continuous breath.  Gould quotes Schweitzer talking about this music
> as evoking "a still and serious wold without light, in tints of grey." 
> That's praise for someone with GG's color preferences, and it points to
> another aspect of the mood he's after and, I think, creates.  I find it
> dignified and dramatic.  For me, Gould sustains a pace and texture that
> convey the "devotional" quality that he attributes to the music.

That "devotional" quality does come across, but I think it stems from GG's
imposed concept of the piece, rather than being inherent in the music.  He
downplays (maybe even deliberately undercuts) the cumulative drama and the
momentum; I think this strategy is a miscalculation.  The piece is a
valedictory culmination of contrapuntal techniques...it gets *more*
intense as more and more subjects are combined, and as there is less free
material.  The composition becomes increasingly compact, as Bach has to
fit his chosen subjects together; there are fewer combinations available. 
The energy builds *up* steadily from section to section, and logically the
piece will be over when enough of these are explored, when there are no
more permutations.  (Like "change ringing" of bells.) But what does GG do
instead?  He dissipates the energy, especially in the third section;
doesn't let the piece grow in visceral excitement.  He makes it sound (I
think) like Bach is wasting away, "mostly dead" (cf. Monty Python), or
certainly at least not very physically fit.  This *might* be a defensible
(though over-romanticized) interpretation if Bach had written the piece on
his deathbed or something, but this movement was actually written when he
still had about ten years left in him. 

> And it's not just monochromatic pianissimo throughout.  He uses a variety
> of articulation and color that highlights the structure of this immensely
> long fugue and brings out the drama of it.  First, he builds
> (pianistically) in each of the three sections to strong dynamic 
climaxes. 
> That the last fugue breaks off, unfinished, in the middle of one of those
> inevitable-seeming crescendos makes the end of the music that much more
> poignant and intense, to my ears.  Second, he plays--between the first and
> third fugues, with their sustained, choral textures--a middle section with
> stacatto, almost hicuppy articulation.  It's decidedly an instrumental
> rather than vocal articulation, and I must say that at first I found it
> weird--a little unmotivated, even jokey.  But I've come to hear as making
> sense.  I think it's an excellent answer to the good question Gould is
> quoted as asking about this piece: "How do you keep it going?"  How do you
> make an engaging, listenable experience out of this long and potentially
> signpost-less piece?  By giving the middle part a very different texture,
> he highlights the structure and the distinctive properties of that second
> subject.  He introduces variety.  He also makes the transition back to the
> understated quiet of the B-A-C-H fugue's opening that much more dramatic.

Another way to keep the piece going is simply to play it faster, playing
for the bigger picture.  His tempo seems self-absorbed, uncomfortably
slow...that's why he has to indulge in all those strange articulations to
make up for the suspension of normal time.  (Sort of like his approach to
Beethoven's "Appassionata," or the slow movement of the Sonata #7, or
Wagner's "Siegfried Idyll," or the Brahms "Edward" Ballade, or the
deconstruction which he and Leonard Rose did of Bach's first gamba sonata. 
The suspension of normal time can be a magical effect, sure...but it's
also a willful distortion of the music.  Why not just play some other
piece where it's already written in, for instance something by Morton
Feldman?) The variety is right there in the fugue already: in that second
section, Bach starts using running eighth notes as a default rhythmic
profile, where the first section had been quarter notes.  There is no need
to make *more* of it yet by means of artificial articulations (I have the
same complaint about other pianists, too, see below).  Bach knew what he
was doing, dramatically.  Why overrule him? 

I agree that GG's second section sounds weird and jokey.  And then when he
gets to the third section, slows down the tempo, and plays as quietly as
possible, it just sounds timid and tentative to me: like a Bach who's
afraid to assert himself when he quotes his name.  Why be so understated
and "devotional"?  This is Bach at the pinnacle of his creativity, writing
complex music *because he can* better than anyone else....  To make it too
introverted and introspective takes a lot of the sheer rhythmic fun out of
it. 

>  
> GG says that the music of this Contrapunctus moves him more deeply than
> any other, and I must say that comes intensely through.  I certainly find
> that the performance is consistent with his articulated vision of the
> piece, whatever you might think of that vision.  In short, I don't find
> the performace "creepy" at all, but musically well-conceived and
> emotionally intense. 
>  

I agree that it's consistent with a GG vision, and a valid musical
approach.  But that's also my main complaint, as with the 81
Goldbergs...the performance becomes a portrait of GG (and draws attention
to itself) more than a performance of what's naturally in the piece.  GG
takes over the controlling role of structure and direction, overruling
Bach's compositional mastery and substituting something else.  Whenever I
hear these performances, I'm listening to GG playing the piano (and being
creative to the point of distraction), rather than to Bach's thought. 

> Let me make a confession: Gould's performance of the piece is the only one
> I know.  Manifestly no expert on the Art of Fugue, I'm coming to these
> pieces after years of hearing occasional excerpts and admiring the idea of
> the whole work, but always from afar:  I never knew quite how to listen to
> it, or maybe just didn't have enough patience.  (I certainly can't play it
> through.  My fledgling adult-beginner keyboard abilities are still at the
> Invention level.)  I don't have a complete recording of the Art of Fugue,
> so I have no other performance-- or performance traditions--of the last
> Contrapunctus in my ears. 
>  
> It's Gould who--as in so many other things--has gotten me excited by this
> music.  And (so) I feel a real affection for this performance.

That's certainly a valid reason to enjoy it, and I've been there myself
with some of GG's other recordings.  I think that's one of the things that
GG was especially good at: pedagogically getting people interested in and
excited about things which he thought were worthwhile.  Sort of an
evangelistic function for the general public, through his recordings and
writings and television/radio (as Bernstein did in a different way for the
middle class in the US).  GG was a good catalyst.  Trouble is (at least in
my experience), when one then goes further to explore other approaches to
the same material, these evangelistic approaches by GG and Lenny and
others eventually seem excessive and artificial, even distorted.  The
music doesn't necessarily need as much of the performer's intervention as
they gave it (even though in doing so they brought much larger audiences
to it, which is a good thing).  

All this is to say, I guess I liked GG's general approach to Bach a lot
better 15 years ago than I do now...it was helpful as an introduction,
certainly, but ultimately for me his distortions are more distracting than
fulfilling or convincing. 

>  
> I wonder how Bradley and others would react to this long-winded 
defense. 
> I'd be happy to be led to other recordings that give different views of
> this difficult, wonderful music.
>  

Read on!  Your posting inspired me to do a direct comparison again of many
different recordings.  I pulled out all the piano recordings that I have,
plus half a dozen in other instrumentations for contrast (out of the 30+
choices). 

On Tue, 23 Sep 1997, John P. Hill wrote in response to Andrew:

> FWIW, I completely agree with your assessment of the final triple fugue
> from Die Kunst der Fuge.  I find it truly awesome;  Gould playing what
> was arguably *his* favorite Bach piece with such complete committment
> and understanding.  But I'm a late-period Gould fan and some others on the
> list clearly are not.(...)

OK.  But then again, what about GG's admission to his producer right
before the session that he still wasn't sure which overall approach to the
piece he wanted to use?  The resultant performance suggests some of that
indecisiveness, too...or maybe I'm just reading something into it, when
the second and third sections seem to drift rather aimlessly, and when in
the first section I hear him building up local crescendos and decrescendos
or thundering out the first two notes of the subject, to make something
happen (more than he would need to do).  Sure, he had an intellectual
contrapuntal understanding of how the piece is constructed, that's very
clear from the way he plays it.  But he's not so clear projecting a
consistent big picture.  I suspect that if this had been a commercial
recording (sound only) instead of a film, he would have used some
"take-twoness" months later to fix that. 

-----

On to the comparisons:

This brief review is primarily of piano recordings, even though I believe
the modern piano is inherently the *least* effective keyboard instrument
for the Art of Fugue.  Its attack (necessarily mushier than the clear
pluck of a harpsichord) diffuses the rhythmic profile of each movement. 
Its tone is relatively bland and colorless, mostly fundamental when
compared with harpsichord and organ (which usually have more prominent
high harmonics).  And its sustain envelope (somewhere between that of
harpsichord and organ) seems to offer less textural clarity overall than
either.  The piano's equal temperament is not a plus here, either. 

Pianists are tempted to make up for these handicaps by imposing artificial
articulations or by playing some voices louder than others, especially
whenever the subjects enter.  Part of the fun of listening to a fugue
repeatedly is finding hidden entrances and subtleties in the
countermelodies, not just hearing a subject regularly thundered out with a
spotlight on it.  Such an approach of over-italicizing with dynamics or
articulation doesn't allow Bach to be very subtle; what if the listener
wanted to hear something in one of the other voices, but the pianist has
pushed it into the background? 

An organ similarly obscures some of the rhythmic motion, but makes up for
that deficiency with a different strength:  letting long notes sound
through to the end, bringing out lines and clashes.  A strange thing is, a
harpsichord brings out the lines at least as well as the organ does, even
though the tones decay; the note events draw the ear to listening
attentively in that pitch range, until the next event in that same voice. 
The player can direct the listener's attention to a specific place in the
texture, using subtleties of the space in that voice immediately before
the entry (as an organist does), and subtleties of staggering among the
voices to accent certain notes.  Internal details of the writing make it
self-evident that Bach intended the Art of Fugue more directly for
harpsichord than for organ.  (Playability by two hands throughout, without
pedal; restriking of long held notes; parallel thirds in low tenor and
bass; the 2-harpsichord arrangements of the mirror fugues; etc., see
Leonhardt's published essay.)

Anyone ever heard a recording of the Art of Fugue on clavichord?  I
haven't, but I've played through some of it on clavichord for my own
enjoyment...might be an even more ideal instrument for some of the piece
than the harpsichord is.  Attacks are clearly defined, yet singing; decay
is quick, giving clarity; it is possible to play long notes slightly
louder than others to make them last longer, and even to color them with
vibrato.  Trouble is, all this is extremely difficult to control when the
piece is already so tricky to play!  The harpsichord offers more margin of
error, is more forgiving when one plays difficult notes or leaps with an
uneven amount of force (just to get them at all).  The piano is also more
forgiving of this, despite allowing dynamic variation note to note,
because one is playing with much more overall force (digging into it,
playing into the bottom of the keys, etc.) than one does on a clavichord.
Clavichord requires extreme control within a narrow range. 

How about synthesizer options?  Anything goes, and I've worked up an
experimental version of the three-voiced mirror fugues...sequenced, not
played in real time.  OK, onward to the reviews.  This is the order in
which I listened through (just to the unfinished fugue, not to the entire
Art of Fugue this time). 

Evgeni Koriolov (Tacet LC 7033, c1990), 11'15": This interpretation seems
very much patterned on GG's.  As I listened, it seemed to be described
very well by what Andrew wrote (above), as to tempo, articulation, and
mood.  It really sounds like some vague carbon copy of GG's performance,
but without the humming, and is a little more forthright: less emphasis of
localized events, clearer sense of big picture.  The beginning tempo is
steady and slow, and a hushed mood leading into a very long crescendo. 
The second section is played almost all staccato, which I thought sounded
unnatural, though it does provide very clear contrast.  The third section
is *very* quiet and slow, but the underlying general pulse still seems to
be there...more rhythmic excitement than GG's.  Koriolov suddenly picks up
the tempo at the point where all the subjects are combined (the section
which GG doesn't play, breaking off before it).  A bit jarring there, but
I think it makes an important point about the piece going on in the
imagination after the last note is released.  GG's ending, by contrast, is
more sudden and startling. 

GG (Sony, recorded 1981), 12'19": first section seems very slow, like the
Aria of the '81 Goldbergs.  GG brings a lot of dynamic rise and fall all
the way through, and often accents the first two notes of the subject. 
Humming is quite loud.  His occasional staccato sounds musically arbitrary
and mannered, but it also serves a structural purpose, pointing out the
beginnings of episodic sections (no subject present).  So it makes sense
in retrospect.  Second section: tempo picks up slightly, but still too
slow.  Interesting mix of paired notes and default staccato single
notes--gives good clarity when the two subjects are combined, the first
more solemn and the second skipping along.  But in this section I also had
trouble hearing the overall direction of the piece, for all the local
interest within the texture.  Very loud ending to the section,
oppressively so: overdone.  Third section: suddenly quiet and slower,
jarring transition, like suddenly changing TV channels.  This section
seems especially directionless, like Bach drifting without a sure sense of
purpose.  Very odd: Bach out of focus.  "Creepy" as I wrote earlier.  With
the performance overall, I get the impression that GG is not settled in
his mind where he wants the piece to go.  And the physical rhythm of the
piece is so downplayed throughout, as he doesn't allow momentum or energy
to accumulate, the performance comes across as a series of dissipations,
wispy.  Strange.  Fear of the body, or something.  Not at all what I think
of as a quality of Bach's music. 

Charles Rosen (Odyssey LP), 9'08": the entire piece flows along at a
common and well-chosen tempo (not an easy thing in the performance of this
piece, catching exactly the single tempo that makes the different subjects
work well individually and together).  In the first section, he often
brings out the principal subject by making it louder within the texture. 
That's an OK idea, but might also be overstating the obvious.  There is a
gentle overall crescendo through the section, seeming more naturally
derived from the contrapuntal excitement than imposed from outside.  In
the second section he lets Bach's new texture of running eighth notes
provide enough natural contrast, instead of imposing a different default
articulation or tone color.  The third section continues in this same
tempo, with the momentum from the earlier sections, so that when all three
subjects are combined, the piece is still building up naturally.  The
ending then trails off into the imagination.  Very successful performance,
because he doesn't try to do too much to the piece; he lets Bach speak,
and it works fine. 

Grigory Sokolov (Opus 111 9116/7, 1982/91), 12'31": *Very* slow,
reverential tempo: even slower than GG.  He italicizes events along the
way to a similar extent as GG, but in a different manner: uses more
variety of weighting and articulation phrase by phrase, rather than GG's
more sectional approach.  General, slow crescendo throughout the entire
piece.  Second section: even more variety.  Preserves the same tempo,
which sounds unnecessarily slow and affected here.  I especially dislike
the plunky staccato.  Third section: louder and firmer, still in same
tempo.  He plows through to the end, with a reasonably clear texture and
some buildup of excitement.  I still don't like the heavy plunking which
he uses as one of his colors. Overall, the consistent tempo (though far
too slow) makes the piece hold together better than the Koriolov/Gould
approach. 

Millette Alexander & Frank Daykin (Connoisseur Society CD 4203, 1994),
8'38": they play the Erich Schwebsch arrangement (1937) for two pianos. 
The performance has an easy flow, unfolds naturally (like the Rosen).  The
octave doublings of the arrangement provide color interest, so they don't
fall back on any tricks of articulation or pounding out the subject.  The
second section continues in the same tempo, with a natural legato.  This
section builds to an exciting big finish through a long crescendo.  After
a short pause, the third section begins in a slightly slower tempo (from a
different take?).  The energy soon builds back up.  They speed up a bit
where the three subjects combine.  As with the Rosen, this performance
makes Bach's thought come across clearly. 

Louis Bagger (Titanic TI-151), 7'56", harpsichord: After all these piano
recordings, for relief I had to hear the piece again on the proper
instrument in one of my favorite recordings.  Ahh, rhythmic profile, so
clear on harpsichord!  This performance has easy flow in one consistent
tempo, and natural clarity; the ear can pick out any part to follow for as
long as one wishes.  In the second section, Bach's writing adds natural
brightness as there are more fast notes happening.  In the third section,
Bagger adds the 4' stop for further brightness; the piece naturally
continues to crescendo, with no sudden lessening of energy as some of the
other performances above did.  Bagger does a strange thing when he gets to
the point where the print ends but the manuscript continues for a short
section:  he erases the free soprano voice of the continuation, and
substitutes the principal subject from the other Art of Fugue movements. 
It combines well there, and this possibility is implicit in the other
three subjects, as Tovey had pointed out.  This strikes me as a strange
thing to hear at this particular point in the piece, however.  We get this
new subject, offering promising new development, and...the piece is
suddenly over. 

Bernard Lagace (Analekta AN 2 8218-9, 1995), 9'45", organ: full
registration throughout, massive effect.  His interpretation is
straightforward, with an easy flow, except for using some ritards and
pauses to mark beginnings/ends of large sections. The third section is
slightly slower as it gets rolling, comes back up, then slows down quite a
bit when the sixteenth notes begin (right before the place where all three
subjects are combined in stretto).  That takes down the momentum a
bit...more fun just to barrel through them!  Perhaps he did it for clarity
in the very resonant acoustic, with that registration. 

Wolfgang Rubsam (Naxos 8.550704, 1992), 12'59", organ: slowest performance
of all, yet it doesn't seem as slow as the piano recordings do.  On the
organ there is more room for this spaciousness, and it seems more natural. 
He uses a quiet and gentle registration throughout: sounds like a
Principal 8 and a Flute 4 only, until he uses pedal more prominently in
the last section, adding 16 and a gentle reed there.  Melancholy mood
throughout, and a probing and plastic approach to tempo: he highlights
many events by dwelling on them slightly.  The phrases all have their own
breaths, and the flow of the music sounds improvisatory.  In the second
and third sections he stretches things more and more as they go along. 
Interesting concept, as the piece sounds like some form of evolution.  It
is clear that the piece is becoming more complex, as he requires more time
to bring out everything. In a way, that builds up its own kind of energy,
quite different from motoric momentum.  Sort of a mental multitasking
energy rather than physical. 

This time around I also listened to two saxophone performances (Berlin and
Los Angeles quartets), and the Canadian Brass (dedicated to GG), and
Musica Antiqua Koeln's (played by Baroque string quartet), and Savall's
(four viols), and Herrmann Scherchen's chamber-orchestra version...won't
go into detail on them here.  They all bring very different characters, of
course.  And all are in generally consistent tempos, rather than the
segmented approach of some of the piano recordings. 

Anyway...those are *some* approaches to this fugue.  And again, anyone
else's mileage may vary.  I hope those other descriptions help explain why
I'm not as excited about the GG recording as some others of you seem to
be.  Generally, I think he imposes too much artifice on the music.  He
expends effort "making things happen" instead of letting things happen
which Bach put there. As with the '81 Goldbergs he grabs the listener to
follow it in *his* didactic way, instead of allowing the listener to
discover the piece's subtleties privately.  He draws attention to himself,
or to something else outside the notes of the piece. It's like the
difference between aggressive hard-sell hype marketing (Gould, Koriolov,
Sokolov) and allowing the customer to seek out and appreciate unobtrusive
quality (Rosen, Alexander/Daykin, Bagger). 

But I'll also say, I think *overall* the Art of Fugue sets by Koriolov and
Sokolov are quite successful, convincing, and delightful to listen to. 
They catch emotions and deploy contrasts very well, and are consistently
interesting. Remember, I was comparing here only their isolated choices in
one particular fugue. 

For this piece, at least, I think the unhyped approach is ultimately more
rewarding.  Let the listener bring something different and find something
different every time, exploring the possibilities which Bach has offered.
Give the listener a clear texture in which there are several levels of
choices ("OK, I'll follow the life and times of the alto voice this
time!"), not just one brightly illuminated path paved on top of the notes. 
When you go on a walking or bus tour of some tourist attraction, do you
look at *only* the things which the tour guide is talking about, or do you
enjoy letting your consciousness wander in and out of the script, trying
some tangents?  When you hear a lecture or read a book, is it more
exciting if the presenter tells you everything, or sparks your imagination
to go all sorts of directions on your own?  When you hear a familiar piece
of music, do you want to have exactly the same experience as last time, or
let yourself find some new aspect of it? 

This is my favorite piece of music, any music.  Sometimes I want to hear
it for the piece itself: Bach's pure ideas and compositional techniques. 
Other times I want to hear what someone does with it or to it, listening
for the interpretation more than for the music: i.e., novelty.  In
general, I most respect the approaches which allow me to find something
new every time (Bagger, Leonhardt, Rubsam, Rosen, Savall, the sax
quartets, etc.), as long as they're not *so* generic and uninflected (e.g.
Moroney, Canadian Brass, and some others) that they don't hold the
attention for long.  It's that tough balancing point between objectivity
and subjectivity, and there are so many variables in a piece this complex! 

And listening to it, no matter how great the performance, is always less
engaging than playing it oneself.  The feeling of the piece in the hands
brings more senses into play. 

Bradley Lehman ~ Harrisonburg VA, USA ~ 38.44N+78.87W
bpl@umich.edu ~ http://www-personal.umich.edu/~bpl/