Re: CIMESP-Results Fwd:

Subject: Re: CIMESP-Results Fwd:
From: Richard Wentk (
Date: Sun Nov 06 2005 - 09:25:20 EST

At 23:15 04/11/2005, you wrote:

>Ok. I'm proposing my idea as an alternative to the theory that says "when
>we listen to music we hear its form -- ABA,"
>from which we conclude that "the cognitive map of a piece closely follows
>its form." I say this is false --
>the map closely follows the order in which we like what we hear. There's
>no ABA in the mind
>(even if the music is laid out like that) but rather a BA1A2 if we
>happened to like what happened in B.

This is possibly true, but I'd argue with 'like' and replace it with 'found
memorable.' Which may or may not have anything to do with 'like'.

>So the theortical problem is to figure out the "real" forms (or as I call
>them, "popular shapes") that are actually
>used in any music -- ie, the forms that plausibly match a listener's
>cognitive map.

The problem with this is that I'm not sure many listeners care about form
at all. The hierarchy of hearing ole' fashioned music with notes in it
seems to be rhythm, melody, harmony and then form. And the latter is often
only heard in the sense of 'This is a repeat.' Untrained listeners barely
get any sense of form at all.

>For instance, one popular (rough) shape is "climax near the end." Assuming
>the listener found that to be the
>best bit, then his map starts from that point, rather than at the
>beginning of the piece. Eg, it's the
>first thing he remembers when someone mentions the piece.

This seems very oversimplified to me. I can't think of any pieces where I
specifically remember the climax above the rest of the piece.

It's a lot more true of writing, where drama, novels, film scores and so on
are deliberately written with rising action and tighter pace in the last
act. This is so absolutely standard that you'll see jokes about it.

It doesn't seem so true of music. Does a 3.5 minute pop song have a climax
at the end? If not, why has the verse chorus etc structure survived so well
in that genre?

>There are some devious ways to try to figure this out, involving computer
>modelling, but I won't go into that here.\
>Suffice to say that one has to ease out popular shapes by assuming that if
>they occur frequently, they
>must be popular -- people must like them. These are not like patterns of
>pitches but more like specific

In that case the verse chorus etc structure would probably be a better
place to start, with literally hundreds of thousands of examples.

There are other pop rules that are well know: Such as:

During the chorus
Modulate upwards
Add backing vocals
Make the arrangement busier

And so on.

These are all formulaic cliches, but they're cliches because they're
popular and people seem to enjoy listening to them, rather than vice versa.

>So it;s quite clear that there are very many popular shapes in EA,
>suggesting that the theory of listening being
>applied IS like what I'm calling "structural favoritism."

Or maybe it's just rote repetition learned from teachers?

My experience of EA is that the structures seem completely arbitrary. There
are certainly cliches, but I don't see any aesthetic justification for
(e.g.) making something go 'Bang!' every so often.

What interests me more is that two pieces can use very similar musical
language but one will sound successful and above all musical, while the
other will sound like arbitrary noodling. This is interesting because it
seems to be true of other genres too.

In other words, it's not the form that really matters. Nor the conventions
of the form.

As for competitions - if prizes are being awarded for adherence to
arbitrary conventions rather than creativity, that's rather disappointing.


This archive was generated by hypermail 2b27 : Sat Dec 22 2007 - 01:46:14 EST