Re: [F_MINOR] glenn gould's goldberg recordings

[Brad Lehman <bpl@UMICH.EDU>]

At 06:21 PM 3/31/2004 +0200, Jörgen Lundmark wrote:
> > > In the case of the Goldberg -81, one must not forget that the Yamaha has
> > > a harsher, more metallic sound than a Steinway. That's something we
> > > can't change in hindsight. Gould didn't seem to mind.
> >
> > Indeed.  But, in the "State of Wonder" analog issue, that Yamaha at least
> > sounds like a real Yamaha piano.  In the standard digital issue it's
> > harsher than they really are.
>
> Are you sure?

Yes.  All the way through high school and college I played on Yamaha grands 
and uprights; I'm quite sure that in real life they sound more like the 
tone of the 1981 analog recording than the digital version of it.

Sure, it's also possible that Gould's particular Yamaha is harsher in tone 
than some others; granted as a possibility.

> One of the main charateristics of Gould's recordings is
> the extreme closeup. That will no doubt enhance any instrument specific
> sound. For example, the less than perfect intonation of this instrument
> is clearly audible due to the close miking. The fact that the analog
> version has less bass clarity doesn't make a compelling case for it
> being a more "realistic" reproduction of the Yamaha at hand. Whether you
> prefer the harsher sound/more bass line clarity or softer/less clear
> bass is another question altogether.

Agreed.

> > The digital version
> > strikes me as off-putting, as if Gould didn't care much about beautiful
> > tone anymore.  The analog one reveals that he still did, or at least that
> > he was *able to* produce beautiful tone.
>
> I think that is quite a harsh view of the -81 version. I can perhaps
> understand the view that this is more a Gould verison than a Bach
> version (which on the other hand is a rather strange view, giving the
> fact that piano playing should be regarded as more or less re-creative;
> another and more far-reaching subject, I know). When I listen to -- say
> variation 6 or variation 25 -- I cannot for my life understand anyone
> claiming Gould to achieve an off-putting piano sound.

Perhaps my view of it *is* harsh.  But I say it as a person who has played 
the Goldberg Variations on harpsichord for 19 years, and who has more than 
30 other recordings of it.  The way Gould handled rhythm (as in, being too 
regular within phrases) in this performance really bothers me.  Maybe I'm 
hyper-sensitized to that, because rhythmic and articulative nuance are the 
main ways to be expressive on harpsichord.  But, all of Gould's earlier 
performances 1954, 55, 59 seem to me much more natural in his graceful 
treatment of rhythm, letting things be gently irregular.  In 1981 it's all 
squared off and machine-like, by comparison.

In Gould's own terms, it's the "too much piano-playing" of his early career 
(i.e. playing the piano as a piano, and with intuitive musical nuances) 
that he scrubbed out of his approach in his middle and late career.  He 
became more analytical, and treating the piano less as a piano than as a 
generic instrument.  He wrote about that.


> To me the -55 is
> much more forced in that department, mostly because Gould chooses such
> extreme tempi. (And yes, I know var. 6 is supposed to be a gigue, which
> is hardly what Gould plays)

That marking was only discovered in 1975, with the discovery and 
publication of Bach's handwritten corrections in his own copy of the 
print.  I don't think we can hold it against Gould for not knowing that in 
the 1950s.  :)  He worked from the 1938 Kirkpatrick edition, mostly.


> >   The digital one strips away some
> > of his nuances, sounding (by comparison) stark and mechanical.  In my
> > opinion, of course.  That's after 20 years of listening to the digital
> > one
> > and disliking it, on the grounds that I feel it's musically shallow.
>
> This is also an extreme exaggeration. You have all the right to dislike
> the -81 recording. But to say it lacks musical depths does strike me as
> very odd. Then you need to define that term very closely.

At the risk of sounding mumbo-jumboish: the music is in the spaces between 
the notes.  Yes, the notes and rhythms are the same in the digital and 
analog versions; those are merely facts about pitch and duration.  The 
balance among voices is slightly different, too, but again not terribly 
significant.  Here's the mumbo-jumboish part: the "aura" of the performance 
seems totally different, to me.  In the digital one it seems to me coldly 
mechanical, like a mental outpouring of a preconceived interpretation, not 
really having much to do one way or another with piano sound.  In the 
analog one, though, I sense that Gould was listening to his own sound and 
reacting to it...much more than the impression I get from the digital 
one.  It seems more like a human being playing the piano, than like a 
disembodied mind.  That's why I'm warming up to this performance in the 
analog but not the digital.

Feel free to disagree, but that's what I meant by "musically shallow" in 
the digital.  Lacking some of the human side (a guy really sitting there 
playing the piano) that I hear in the analog.


> >   Now,
> > in analog, I hear more depth in the interpretation.
>
> I still claim that the difference between the two versions isn't that
> sensational (do a blindfold test yourself with adjusted soundlevels).
> Also, a change in this department cannot alter the main musical content
> of a performance. Gould's artistic choices -- no matter what we think of
> them -- is clear no matter how we filter his recordings.

Granted, my own A/B comparisons in autumn 2002 were not as carefully 
controlled an experiment as yours.  I did the switching back and forth 
myself, manually, and always knew which one I was listening to.  Still, the 
analog one *always* seemed to me to have more of Gould's human presence in 
it; something that can't really be quantized by measuring with an 
oscilloscope or whatever.

And, I'm pretty sure that an artist's musical choices include more than 
pitches and rhythms; filtering does matter.  I spent a couple of months 
last fall working on the production of one of my own recordings (pipe 
organ, in this case).  The engineer sent me several different equalizations 
of the same recording, and the musical effect was (to me) quite different, 
sounding less real in some cases than others.  Most notably, two of the 
tracks were [accidentally] recorded in mono while the rest are in 
stereo.  We tried half a dozen different ways to make the mono seem like 
stereo, but *every time* I felt that the music was diminished, that it 
compromised my interpretation.  We finally ended up going with the 
straightforward mono for those two: where it gives the most accurate 
musical effect in my artistic opinion as the guy who played (and wrote) the 
music.  Anything more artificial in it *changed* the artistic choices, far 
too much.

Everybody's experience is of course different.

Something else to keep in mind on the 1981 Goldbergs: compact discs were 
never commercially available in Glenn Gould's lifetime.  We don't know if 
*on CD* he would have preferred the analog or digital version, if even 
given a choice between the two as we have now.  Digital recording in the 
late 1970s and early 1980s made LPs sound cleaner and quieter, and with 
greater dynamic range...still going back to the analog tracking of a needle 
in a groove.  Digital was the new and up-to-date fad at the time, and it 
got rid of tape hiss.  Who's to say that Gould's choice of digital for the 
LP release was *entirely* motivated by the musical/artistic effect it would 
make, and not just convenience (easier editing) and staying _au courant_?

Hope this helps to clarify my earlier remarks,

Brad Lehman 

**************************************************
Signoff instructions, and user preference interface:
http://email.rutgers.edu/cgi-bin/wa?SUBED1=f_minor&A=1

F_minor Website (with early archive):
http://www.rci.rutgers.edu/~mwatts/glenn
**************************************************