Dave Lampson replies to me:

>>So here goes: I too believe that a score should carry great weight.
>>However, it's arrogance of another kind to justify repeats or not repeats
>>on the basis of knowing the composer's intention.  One should be able to
>>justify one's decision on the musical implications of the score.
>
>Of course, and this is by far your best (and perhaps only) rational
>argument for not taking a repeat.  If something works musically, then no
>further justification is required.  But who decides what "works"? This is,
>as a matter of practicality, a completely personal reaction.  I say it
>works, and you say it doesn't.  Or you say it works and I claim it fails.
>Now what? What does this tell us about the role of the repeat in the
>composition?

Well, each of us decides what works, as usually happens.  What it tells us
about the role of repeats in the composition depends on the reasons why it
does or doesn't work for us.  I admit that this isn't very solid, if only
for the very good possibility that our minds can change.  Again, we're not
talking about something carved in titanium or something wholly "out there,"
but to a great extent within us.  We can support our conclusions by
referring to the score - probably the best way to support it, in fact.  In
that sense, we learn something about how the music works on ourselves, at
least.  There's no guarantee that it works for others the same way.  Should
it?

>What torpedoes any further discussion is this idea of intentions.  I must
>admit I've become quite confused over where you stand on this.  Forgive me
>if I mis-characterize here, but at times you seem to admit that the score
>captures at least part of the composer's intent, and at other times you
>seem to claim we can know nothing about the composer's intent.

I'll say it as plainly as I can.  Neither one of us can refer to a
composer's intent, since all we have is a score, and the two aren't
necessarily identical at every point.  I don't see why particularly we have
to justify according to intent at all, and given the problems of presuming
intent (because that's what it is), we're better off justifying on some
other grounds, most likely our view of the score's musical structure.
(This is where it gets hard) I distinguish the musical structure from the
formal structure, as, incidentally, Sessions does in Conversations with
Roger Sessions.  For example, the last movement of the Franck violin sonata
is formally a 2-voice canon with free accompaniment.  Musically, however,
it's simply call-and-response, with the head of the next phrase merely
overlapping (usually) the ultimate note of the previous one.  Yes, it can
be and has been analyzed as a canon, but the overall musical effect is
something much less contrapuntally grand.  What this may have to do with
repeats is this:  the repeat may not only be musically unnecessary (though
of course failing to follow it changes the formal structure of the score),
it may even harm the musical structure of the piece.  In other words, I
find it very hard to believe that several generations of musicians of
greater skill and culture than most of us here were insensitive boors when
it came to playing Mozart.  I'm not saying that observing all the repeats
wouldn't produce something even more wonderful - after all, I've never
heard it that way, and I like to believe I keep an open mind - but I'm
not consigning Walter and Szell to the dung heap just yet.

>Let's take a hypothetical example (but one for which we could probably
>find a close, real-life parallel).  We have the composer's notebooks, his
>sketches, and the fair copy of a work in the composer's own hand.  Further
>we have a letter from the composer to a long-time confidant that the work
>is complete, and the composer is comfortable that the piece could not be
>improved by changing a single note.  When published, the composer once
>again writes to his friend that though several mistakes crept into his
>score when typeset by the publisher, he has now corrected all of these
>errors and is now content that the published scores represent his best
>efforts as a composer.  Further, we can compare the manuscript we know
>to be the composer's fair copy with the published version, and they match.
>Finally, we have yet more correspondence from the composer describing
>rehearsals and his reactions to the interpretations of the performers.  In
>this example, I think we know a great deal about the composer's intentions
>for the work.  We can't have perfect knowledge of anyone intentions, so
>that's not an issue.  What is at issue is whether a score captures some of
>the composer's intentions (within the limitations of the notation, etc.),
>and what are the implications of following or not following instructions
>in the score.

Here's a wiseacre reply:  the composer was lying or merely delusional and
didn't know what he really had.  Here's a more serious answer.  I expect
performers to give me not the composer's intention, but their best effort
toward realizing the score, to put the score in the best light.  Often they
do this by observing every tick in the score and doing more besides.  Other
times, they take certain liberties, usually with tempi and dynamics and
occasionally with dropping repeats.  My main point has always been that
you can't justly condemn what you haven't heard.  If you have a bunch of
criteria beforehand that give you permission not to listen at all, so much
the worse for your criteria.

This doesn't mean that the performer is always right or has a better
idea than those in the score.  It's up to the performer to convince you,
particularly if you know the score well.  On Kunzel's Gershwin set, the
pianist (William Tritt?) improvises parts of the Rhapsody in Blue, since we
know Gershwin improvised certain passages at the work's premiere and only
later set certain passages.  We also know that Gershwin rarely recorded his
music the same way twice, from which we might conclude that he wasn't all
that interested in a "definitive" performance.  Here we can argue intention
both ways until we're blue in the face.  Why argue intention at all? The
real problem is that Tritt doesn't improvise all that well, compared to
what's in the score, although I admire his guts.  It may also be that
somebody may do better than what's in the score.  Is this still Gershwin?
Where are we then? Probably in the same place we would be about cadenzas in
Mozart's concerti where the composer usually didn't supply them.  We don't
consider the concerti any less Mozart's because Beethoven supplied a
cadenza.

>In general, I think the idea that a score captures musical intent -
>at least some of it - is unassailable.

I'd agree.  It's awfully hard to conceive of somebody absolutely unaware.
But we're not talking about a whole score, really.  Are we?  We're talking
about bits and pieces - one decision here, another there.

>What might be in question is providence of the score.

{Dave gives the example of the uncertainties of Vivaldi scores.  I agree
with his point.  In fact, I tried to make it once, myself.}

Let me give a literary example.  Not too long ago, I finished reading
Fagle's translations of Homer.  No known manuscript of Homer exists.
It may well be that Homer didn't write at all (although Fagle discusses
the uncertainty even of this).  We might reasonably imagine that several
versions of the epics co-existed after Homer and that what we now recognize
as the standard text (with little disputes here and there) might well be
the result of people consciously or unconsciously replacing the original
with something they liked better.  Now, I can't judge Fagle's translation
on the basis of Homer's words, because I don't know them for sure.  I *can*
judge the translation compared to the Greek text Fagle used.

Closer to us, none of Shakespeare's major plays survive in his own hand.
The First Folio (on which almost all standard editions are based) is the
result of actors recalling their parts and, possibly, from prompt books and
from cheap editions, whose authority is questionable.  It may well be that
all these hands in the pot have picked up not what Shakespeare wrote or
intended, but what they wanted to pick up.  In short, we can't judge
Shakespeare's intent, only the texts we have.

I'd expand this a little to include both sides of the argument.  The
score is important, not because the dead composer morally owns it, as the
talk about the composer's "rights" implies, but because it's the standard
against which we judge the performance.  I don't mean standard in the sense
of excellence, but in the sense of a standard reference.  It makes sense to
try to come up with a score as close as possible to what the composer has
left us, since the composer is the music's source.  But once we have that
reference, the composer becomes much less important (to listeners) than
performers of the piece.  The performers realize the score.  We listeners
judge that realization by referring to the score, *not* to the composer.
In short, intention is pretty much beside the real point of how we listen
and evaluate.

Steve Schwartz