Context and Expectation -- was Re: CIMESP-Results Fwd:


Subject: Context and Expectation -- was Re: CIMESP-Results Fwd:
From: Kevin Austin (kevin.austin@videotron.ca)
Date: Mon Nov 07 2005 - 11:03:01 EST


At 11:21 AM +0000 11/7/05, Richard Wentk wrote:
>... Or with Wagner, where chromatic meandering continually thwarts
>the apparent tonality.

I think that this may be contextual. As my tonal hearing has
developed I have ceased hearing "meandering", rather significant
(psychological) extension ... (think Parsifal Act II perhaps).

>I won't argue that it's also possible to create tension by slowing
>down and thinning out the music too, which might seem
>counterintutive.

For examples of this, you may want to check the late Shostakovich
string quartets, and the Britten Third String Quartet.

>But in both cases there's still the sense of changes being made with
>respect to some reference of what's happened before, which implies
>creating tension by undermining the perceptual momentum.

But perhaps the "before" is the listener's knowledge. Bach Chorales
may be part of the reference for how the fugue at the end of the
second act of Meistersinger works.

>From the point of view of information theory it's still about
>changes in the default assumptions of what's coming next.

I like the concept of "default assumption" as a way of describing
embedded Markov chains. The broader the base of knowledge, the deeper
and more profound the "understanding" may be. For example, I could
describe the difference between Scriabin and Delius as the difference
between vodka and Earl Grey tea.

>Or in other words perhaps it's the rate of change of information
>density that matters, not whether or not the change is positive or
>negative.

But it is possible (in my experience) to have no change in the rate
of (received) information. The individual's perception may shift
without their being a significant shift in the information stream. If
the shift is "downward", I consider this (for myself) as 'losing
interest', and vice versa.

Best

Kevin



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