Denis Fodor replies to me:

>>If music were a language (and I mean human music, not what birds do),
>>then it must have a grammar, which is simply a description of how
>>language means.  This leads to all kinds of trouble in the case of
>>music, particularly with avant-garde and even not-so-avant-garde music.
>
>This leaves me feling a little uneasy.  Why? Just pondering the meaning of
>the word "grammar" suggests an affinity to the ways of music.

That's the problem - an "affinity" but not necessarily an identity.  I
believe that many use the word "grammar" when they mean "system" - an
interrelation of constituent parts.  Grammars are systems, but not all
systems are grammars.

>Grammar, a branch of linguistics, deals with the inflexion of language,
>its phonetic system, and with the arrangement of words in a sentence.

I learned it as a description of *how* the language means, as opposed to
*what* it means.  My suspicion is that the definition of musical meaning
is so vague or so elusive that we can't talk about its grammar.  We can
certainly talk about constituents of music and their interrelation, but,
especially since 1912 and since the growth of ethnomusicology, one finds
so many different kinds of music, many of them proceeding from different
assumptions, that the systems become too different to compare fruitfully.
I essentially agree with Jim Tobin, that the language metaphor is exactly
that.  It allows us to talk in certain convenient ways about a piece of
music without having to engage in technical esoterica.  I certainly use the
metaphor a lot myself, but I try to remember that it is indeed a metaphor
and, as such, has certain limitations.  The chief one is that people start
talking about how music appeals to us because our brains are hard-wired
a certain way and that music which appeals only to very small numbers of
very small numbers (read music by "Schoenberg, Webern, Varese, Boulez,
Stockhausen, Nono, Carter, Berio, and so on") cannot possibly *be* music
because it ignores this deep grammar and physiological restrictions.  These
assertions are made on the basis of almost no real evidence whatsoever,
beyond the fact that the speaker doesn't happen to like this kind of music.
Which is fine, but somewhat sinfully proud to raise a personal dislike to
a universal principal.

The thing is, if music is a language, then we can talk about meaning.
If we can talk about meaning, then we can talk about meaninglessness.
It would make sense, in this case, to separate music into sheep and goats.
We would have a fairly firm aesthetic principle, one that lay to a large
extent outside the realm of our personal taste.  I, for example, might
continue to enjoy Stockhausen's Gesang der Juenglinge, but I would have
to admit I'm enjoying a kind of yammering.  If I didn't like it, I could
satisfy myself through whatever grammatic tests that would justify never
listening to it again.

Right now, I listen to music I like and to music I don't like.  I do so
for several reasons.  First, I've had the effort pay off, notably in the
cases of Beethoven and Brahms, who for decades put me to sleep (literally).
This is what keeps me listening to a lot of Mozart.  Second, I like variety
in my listening, and I like the thrill of discovering for myself something
wonderful.  Third, my dislikes limit me.  The more I like, the more I feel
I understand -- ego, if you will.  I like to think catholicity of taste a
sign of intelligence.  I like to think of myself as intelligent -- I admit
it.  But I'm also lazy.  I'd love to have a non-subjective test that would
give me permission not to waste my time.  I just don't see one out there.

Steve Schwartz