I started this post several weeks ago and saved a draft of it with the intention of fleshing it out further with some other material I had been thinking about. I've been away from it too long, however, and I don't remember what I wanted to add to it. So if I don't post it now, I'll probably never get around to it.
The concept of poetic "torque" depends on the same metaphor as "flow": the verse is a current, or objects in a current, and that current pursues an inexorable direction. The sentence is one unit in which one can observe this figurative motion: the mechanism formed by the subject, main verb, and auxiliary parts of speech describes a transfer of energy that is imagined as physical matter in motion. For example, the subject throws a ball by means of the verb, and if there is an object, the object catches the ball. Or the subject is a faucet, out of which issues a stream of action in form of the verb; the stream is directed by means of adverbs, prepositions, participles, and other channeling devices, perhaps to an eventual object-reservoir (which may in turn have another nozzle). This notion of syntax as a transfer of fluid or electrical energy is nothing new. It predates even Fenollosa, I'm sure.
In verse--especially verse that does not conform to normative syntactical structures--the metaphor is complicated but remains frequently applicable. In my previous posts, for example, I have argued that Creeley's "I Know a Man" relies for much of its effect, as does a great deal of older and more traditionally "formal" poetry besides, on techniques of enjambment to redivert the reader's grammatical focus continually, resulting in a sense of "flow" that is artfully "torqued." One could make similar arguments about poetry that is even more disruptively asyntactical, such as Clark Coolidge's short untitled poem from Space (1969):
ounce code orange
Even in the absence of a clear grammatical structure, it is still difficult not to read even abstract linguistic assemblages of this sort on the model of normative syntactical connections between words. One might read the first line as three listed nouns in a series, or alternately one might treat ounce as subject, code as verb, and orange as object; either way, the three words seem to stand in a prefatory relationship to what follows, so that they take on a subject-function of sorts. In this way they set the terms for our processing of the rest of the poem, even though those terms will invariably differ in their details from reader to reader. (One could object that some readers might not respond on even this rudimentary of a syntactical level: that to these readers, the poem is no more than a random assortment of signifiers. This may be true. I would call these "bad readers." They do exist, and their incompetence is not limited to poems of this kind.)
Indeterminacy in poetry is not absolute. Or, if it is, that poetry stops being intelligible as articulated groups of words altogether, and becomes something else: purely abstract verbal sculpture, an affectless list, a meaningless accidental mess (to what degree any of these conditions are actually possible is another question). Poems like Coolidge's are radically overdetermined, but they are not totally indeterminate. By following this line of thought, I am also implying that determinacy, indeterminacy, and overdetermination are functions of syntax (this should not be a particularly controversial claim) every bit as much as of individual words. Hence, to speak of "flow" or "torque" in poetry is to speak of a greater or lesser degree of determination, but always some degree.
When determinacy is so thoroughgoing as to render all considerations of means of expression unnecessary, there can be no question of poetry. Conversely, at that point where determinacy drops out altogether, so the possibility of meaningful evaluative criteria drops out as well, and the designation "poetic" again loses its force. Poetry thus relies upon the notion of (syntactic) closure even as it relies upon the disruption of that closure. At either extreme end of the spectrum--absolute singular determination or absolute indeterminacy--the context for poetry is lost. The term "spectrum" is misleading, however, as we do not witness a smooth gradation between the two extremes. If we did, the formula for separating good from bad poetry would be very simple: poetry would always be most effective when it attained to some elusive middle zone of artfully blended determinacy and indeterminacy. Indeed, many great poems do operate within parameters that could fairly be described in such a way--I think, for example, of any number of pieces by Wallace Stevens or others, poems which "resist the intelligence almost successfully." But the fact is that this accounts for only a small percentage of good poetry, and there are plenty of poems one could characterize in this way that aren't very good at all. Most poems written throughout recorded history, good or bad, contain very little indeterminacy (I say "very little," not "no"), and only a very small portion of all the radically indeterminate poetry written in the last half a century or so actually repays repeated reading. If it seems that the bulk of recent interesting poetry tends toward the indeterminate side of the scale (and of course it doesn't seem that way to everybody--this is a biased account), I would argue that this has less to do with the extremity of its indeterminacy, and more to do with the mediocrity of the poetry it often positions itself (or is positioned by others) against. That this mediocre poetry, in turn, often tends toward extreme determinacy is similarly a coincidence of historical conditions largely external to questions of intrinsic artfulness.
It is true that during times when artistic activity is polarized between opposing "camps," and when that polarization coincides with formal markers such as those centered around determinacy and indeterminacy, the formal markers themselves become a convenient shortcut for addressing the real deficits and advantages of either side's actual production. I call this "bad criticism." I have been as guilty of it as anyone. Still, I want to maintain that those formal concerns are not irrelevant to evaluative criticism. As I have been trying to suggest, it is just that we cannot reduce them to crude systems of aesthetic measurement, to be applied like tape measures to formica countertops.