Sonata, que me veux-tu?

In a review of a new anthology of sonnets, Ron Silliman (boy, do I learn a lot from his blog) writes of poets who

...have seen in the sonnet precisely the dynamics of constraint that elsewhere drives Oulipo toward its amazing proliferation of forms. The point of the sonnet therefore is not to put oneself up against the likes of Shakespeare or Ben Jonson, but rather to see the sonnet for our time as a series of powerful literary devices that can open the present up completely.

Isn't this amazing proliferation of forms quite like that, for the sonata, found in Cage's Sonatas and Interludes, Lou Harrison's early Six Sonatas for Cembalo, or Gordon Mumma's Sixpac Sonatas, or in the profligate sonatas of John White or Boudiwijn Buckinx (also here)? One of the reasons that these pieces are so rich in spite of their brevity — and especially in comparison with the longer modern sonatas based on late classical and romantic models — is that they take the form to its roots, prior to the establishment of the tonal model of the classical sonata movement, as a canzona per sonare (an instrumental work in which sonic quality is emphasized) or the elementary binary form in which the fundamental compositional problem is that of the repeat which is able to lead both back, Ouroboros-style, to its own beginning, as well as forward to something else.