[Smt-talk] Palestrina and Jeppesen, was: MISSING THEORY COMPONENT?

[Olli Väisälä]

First, my thanks to Richard Hermann for summarizing some of those  
features of Jeppesen's works that make Peter Schubert's view of J. to  
appear surprising. (I could not have done this as easily since I  
don't have the English translations at hand.)

Then, some comments to Richard's more general points:

>
> This discussion raises a broader epistemological issue: is the only  
> acceptable evidence from theorists of the day? Can we reasonably  
> ignore extensive evidence from the scores themselves in this case?

Where I teach, it is taken for granted that a primary tool in any  
"model composition" is the intensive analysis (and more or less  
analytically oriented playing and listening) of "the scores  
themselves." Of course, analysis can and must be informed by relevant  
concepts, and those of contemporaneous theorists can be most  
revealing. Nevertheless, it is not at all realistic to think that  
those theorists covered all significant aspects of the music. Music  
is hugely multi-dimensional, and there can be several dimensions that  
the contemporaneous theorists did not have motivation or insight to  
describe. (We might compare some dimensions of musical organization  
to linguistic syntax. Just like we speak fluently our languages,  
observing syntactic principles without ever being aware of this,  
composers may have fluently observed musical principles that they or  
their contemporary theorists never made explicit.)

>
> I would welcome statistical studies periodically redone as  
> theoretical concepts and categories become more refined, changed,  
> or even rejected. These, of course, could not simply and directly  
> be used in either model composition or analysis from the  
> repertoire, but it would provide a better set of starting points.  
> They would not be theories but rather “teachings” in the sense of  
> Schoenberg’s use of “Harmonielehre” with the limited kind of  
> support that empirical studies provide. Certainly knowledge of the  
> theorists of the day should be continually consulted in devising  
> lists of what should be counted but they also should be open to  
> critique in light of findings. The discussion, I hope, will remain  
> open. There is so much yet to learn about this repertoire.

I also welcome statistical studies, but it seems to me that at their  
present stage such studies tend to be a bit elementary when compared  
to the multi-dimensionality of music. For example, David Huron  
mentions in his book (Sweet Expectations) several statistical studies  
concerning melody formation in Palestrina and other sources. One very  
general melodic principle is "step inertia," according to which a  
step in a certain direction tends to be followed by another similar  
step. However, the statistical study Huron quoted (by whom, I don't  
remember) did not take into account meter, i.e., the difference  
between strong-weak and weak-strong steps. I suspected that there is  
a big difference between these two situations and tested my  
assumption with a small informal statistical study of a sample of  
chorale melodies. It appeared that I was right: strong-weak steps had  
a significantly higher statistical tendency of being followed by the  
step in the same direction than weak-strong steps.

This is just one very simple example of multi-dimensionality: for  
describing melodic principles, it may be necessary to allow for the  
interaction of pitch and meter. Needless to say, the questions get  
much more complicated when proceeding towards larger Gestalten.

Olli Väisälä
Sibelius Academy
University of the Arts, Helsinki
ovaisala at siba.fi


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.societymusictheory.org/pipermail/smt-talk-societymusictheory.org/attachments/20140526/3f7c68ff/attachment-0002.htm>