Re: On the Grammar of Music

[Martin Braun <nombraun@xxxxxxxxxxxxxxx>]

Dear Robert,

thanks for your reply. You raised interesting questions.

You wrote:
"Surely you don't need an ERP study to tell you that there is some sort of
grammar of chords in tonal music, or more generally a rule-based system in
music. All you need is to demonstrate that people are able to notice a
"wrong note" in a passage of music they have never heard before. (This has
been shown many times by lots of different people) When that happens,
doesn't it demonstrate that there's an underlying structure or syntax which
the nervous system is able to abstract, and therefore notice when it has
been violated?"

Answer:
No, it doesn't.

1) A "grammar of chords in tonal music" does not exist. Concerning the
functions of chords in European tonal music there are as many views as there
are music theorists. If we had the same situation concerning the grammar of
languages, we would not be able to talk to each other.

2) You are right, we can detect a "wrong note" in an unknown piece of music.
But why can we do so? I once detected a wrong note in a piece of Mozart that
had been unknown to me before, when I started playing it myself. In this
case it was even possible to see that it was not a misprint but an error of
Mozart himself. The wrong note was logical in a close-range context, but
utterly wrong in the given long-range context of the piece. Mozart often
wrote at such an incredible speed that things like that were bound to
happen.
The answer to your question is that we detect "wrong notes", because they
are wrong within one particular piece, not because they are wrong within
European tonal music in general.

3) The authors of the study did not detect a violation of chord-grammar
rules, because such a grammar does not exist. In fact they detected a
violation of tone material. Two of the three tones in a "violating" chord
were new in a given sequence of chords, and also, they were not part of the
tone scale used for the sequence.


You wrote:
"Actually, their results seem to suggest that there may be two parallel
systems with similar localization of critical regions, but with different
hemispheric weightings, which is very interesting."

Answer:
In fact, the authors collected data from only six subjects. In four of the
six subjects there was a stronger effect above the right hemisphere. Taking
the average of all six subjects there was no hemisphere effect.


You wrote:
"To what extent these results are specific to rule-based systems in speech
and music, as opposed to more general sensitivity to patterned events and
violations of expectancies remains to be seen, in my opinion."

Answer:
I can only support this opinion.


Martin


Martin Braun
Neuroscience of Music
Gansbyn 14
S-671 95 Klässbol
Sweden
nombraun@post.netlink.se



----- Original Message -----
From: Robert Zatorre
To: AUDITORY@LISTS.MCGILL.CA
Sent: Wednesday, April 25, 2001 11:33 PM
Subject: Re: On the Grammar of Music