A little more on Schenker. It'll be brief, because it's cold and I don't have any pants on (trying to fall asleep when this occurred to me).
I was thinking of the possibility of doing my dissertation on analyses of the Beethoven sonatas, looking at sonata form in particular. There are a few interesting anomalies: the
strange second key of the first movement of the Waldstein, the false recap in the first movement of opus 10 no. 2 (at which I'm looking right now), the third movement of the same, with a second theme that scarcely does anything at all, etc.
So I thought of why I would want to do Schenkerian analyses of so much stuff. What's my project?
I'm not a fan of hierarchies, as most of you know, and Schenkerian analysis is often criticized and praised for being hierarchical. Part of my project for the topic I'm working on now is to fix that, but to do so, I have to explain what it means to be hierarchical, or how is Schenker's work so?
We might say Schenkerian analysis is hierarchical in two ways. First, there is are the various layers, from the background to the foreground (or vice versa)--this is what I was talking about in my previous post. This, the most common citation for hierarchy, is hierarchy improperly so-called. A more proper designation would be a continuum. Schenker says "always the same, but never the same way," or something like that. That is, every piece of (tonal) music has the same fundamental structure: the triad. But it is always composed out differently. Implicit in this position is that it is the composing-out of the triad that makes music so swell. On the one hand, then, you have the governing background, which is clearly important (for mystical as well as practical reasons, for Schenker); on the other hand, you have the intricate foreground, which, particularly since Schenker was concerned perhaps first of all with performance, is of unavoidable importance. I would submit that these two extremes, mediated by the myriad middlegrounds, are both extremely important. And while Schenker's rhetoric is not consistent on the matter, I would also submit that neither is subsumed under the other, though perhaps they are both subordinate to one another.
This (non) hierarchy does bring us to the true hierarchy, though. This is the primacy of the top voice (or maybe the soprano/bass counterpoint. I need to work out the details there with the original sources and the quite substantial secondary literature). Since "all" music is a composing out of the tonic triad by means of a descending linear progression, all music is then governed by a single coherent melodic line. Consequentially, all material appearing above this line is simply--or merely--an inner voice that has been displaced (and, significantly, which often then has to work its way back to where it belongs).
I propose that a more useful reappropriation of Schenkerism would be to retain his (non) hierarchical reductionism and his privileging of voice leading at the expense of key relations and functional harmony, but to dispense with his mystical reliance on the tonic triad and it's resultant Urlinie. This seems the most reasonable way to allow the second theme group of a sonata to be equally important in both its manifestations (but also markedly different each time), and is supported by Forte and Gilbert's reading of Beethoven opus 10, no. 1, III. It is also contrary to Charles Smith's position, which necessitates the adoption of numerous new fundamental structures, in the attempt to 1) hierarchize that which I argue above is not in fact hierarchical, and 2) to locate all important formal events on the same level. Reading layers as non-hierarchical allows us to see that Smith's move is not necessary: middleground events are not non-structural, nor a-formal, and re-writing the voice leading to make them so is not a desirable solution.