A panel discussion with Lyn Hejinian, Ron Padgett, Carl Phillips, and Kay Ryan.
This is only a small part of the discussion, but it seems to get right to the heart of the subject at hand: is experimentation with language (as a multi-faceted object of sound, emotion and meaning) justified as a poetic end in and of itself; does it need to be contextualized within a more conventional sense of poetics; or are conventional semantics more important (and, hence, vital to poetry) because words, first and foremost, are units of meaning?
On a fairly regular basis, I run into people who criticize poems they don't understand or comprehend, and frequently this criticism seems to stem more from an unwillingness to appreciate a use of language that moves outside a restrictive, traditionally meaningful norm. Such an opinion might assert that poems of this sort are incapable of being understood, and that they violate the norms of language in a way that diminishes, rather than expands, the potential of poetry, i.e. they don't mean anything so they are of considerably less value as a poem.
But, on the other hand, meaning can also be said to derive from experience, and the meaning of sound is not restricted to lexical potential alone. For instance, instrumental music can be very meaningful in the complete absence of vocal language. The cry of an eagle, or the sound of trickling water, or wind through the trees, can be equally powerful. In fact, onomatopoetic language has a long history of suggesting sense to the human mind rather than reason.
So, if a poem can be created that avoids lexical meaning yet manages to overflow with a sense of non-rational experiential meaning (at least for those who are open to this potential) wouldn't that be quite an accomplishment? And wouldn't it be worth noting that not only has it been done before, but it is also a practice worth continuing? Or that 'quality' in poetry is not necessarily limited to the accomplished use of rhetoric, narrative, or lyrical imagery?
It's something to think about.