Performance as Analysis

In comments to a Crooked Timber post on analytical philosophers, Seth Edenbaum criticizes Derrida for attempting to create literary art at the same time he was analyzing literature.

Derrida was trying to “perform” language while analyzing it.
If my lawyer ever tried to be a legal philsopher while defending my case, I’d fire him. You can’t be an actor and a theater critic at the same time.

Reading this reminded me of an attempt to perform music analysis and a claim that all performances are analyses, and all analyses are performances.

First is Hans Keller's Functional Analysis. Keller believed that each musical piece is controlled by a 'basic idea,' a cellular motive, somewhat related to Schoenberg's concept of Grundgestalt. His goal was to identify that basic idea and show how it was transformed to produce the entire work. The unusual aspect was his idea to perform the analyses. From the Grove Music Online article on Analysis (II, 1945-1975):

In 1957 Keller took an even bolder step than Schenker had when abandoning the word for the graph: Keller abandoned word and graph for sound, by preparing an analytical score which demonstrated what he saw as the background unities of Mozart’s String Quartet in D minor k421/417b entirely in musical sound. Keller’s method involved composing a score, for the same forces as the work under analysis, in which passages of the original are interspersed with aural demonstrations of the links between themes. He claimed for this the advantages that it avoided the transition between musical and verbal thought, that the through-composition of the analytical score led along purely musical lines, and that the subjectiveness of verbal description was eliminated. Several such analyses were prepared and broadcast in Britain and on the Continent, but none of Keller’s scores was published until much later (1985, 1995).

To the second association I had, we turn to David Lewin's seminal article in Music Perception, "Music Theory, Phenomenology, and Modes of Perception" (1986, Vol. 3, No. 4). Lewin spends much of the article describing an application of Husserl's theories of phenomenology to music theory, which is fascinating in its own right. But I'm moving to Part V of the article, "Perception and the Productive Modes of Behavior." During this section Lewin argues that the creation of music, whether performance or composition, is "a mode of musical perception," a reaction to previous experiences with music. So Beethoven's c Minor Piano Concerto is a response to [an analysis of?] Mozart's c minor Concerto (p. 382). Lewin then suggests that critical statements can also be performances, inspired by Jonathan Culler's views on poetry and criticisms of Bloom (The Pursuit of Signs, 1981).

Edenbaum could justly point out that Lewin is using Derrida's views of literature, which Edenbaum is criticizing in the first place. But there is something appealing about the idea that the best descriptions of music should be musical. There are plenty of quotes about music starting where words fail, and of the difficulty in writing about music. I certainly find it difficult to lecture about music without a piano nearby to use in illustrating points. I use music to explain music. Does that make my whole lecture musical?