Whole pie

One special notion entertained by the Chicago critics is that the only alternative to a genre criticism is an escape into the psychology of the author or reader. Thus they execute what I shall call a double or slant antithesis, one which is not a true or exclusive opposition. They say that, if criticism leaves the species or kinds of poetry to search for the general definition of poetry itself, then criticism leaves behind also the cognitive object and passes into the realm of general psychology. They seem to assume that there cannot be a general objective discussion of any topic. You have to divide it into its parts or species in order to remain objective. If you try to pick up a pie whole, it will melt into a plan for a pie or the taste of a pie. If you cut it into slices, it will remain pie and whole pie. The argument might gain something of plausibility if we were able to entertain that literal equation of poem with physical artifact which has been criticized above. In the realm of artifacts, if we were somehow constrained to talk only about "artifacts" and never or little about "houses" or "hammers," the objects of our discourse might conceivably tend (at least if we were not on guard) to lose tangibility and become just qualities or, by a further slip, just human aims and methods. But poems, as we have said, are not physical artifacts; they are, to start with, verbal acts. Their definition and defense have to be drawn along a different line.
-William Wimsatt, "The Chicago Critics: The Fallacy of the Neoclassic Species" in The Verbal Icon

This is what Wimsatt calls the "fallacy of the neoclassic species"--perhaps the least memorable fallacy featured in The Verbal Icon because it hovers between the two others: the famous "intentional fallacy" and the other (and equally important) "affective fallacy." Perhaps the fallacy here is actually closest to the latter. This fallacy is where something like generic convention is taken to be the ultimate determiner of the form, and thus the form gets hardened (in a way that it wouldn't if it were seen to produce concretion or complexity says Wimsatt--obviously anticipating his important essay "The Concrete Universal") into something identical with the shape of the poem on the page. Thus, it should be clear that when Wimsatt opposes taking a poem as a physical artifact, and affirms the verbal act, he actually means that we shouldn't think of the poem as a materialization of some ("doughy" he says--Wimsatt likes confection-metaphors) abstraction called language (this position would be resolutely against the linguistic determinism--and indeed sham materiality--we find in Paul de Man). Such a view (combining what he rightly says is an Aristotelian view and, when pressed hard as to what makes the artifact, a Crocean view of poetry) encourages considerations of the "whole object"--that is, considerations that relate any minor point back to the rest of the poem, producing a sort of "tension" that defines the work or requires that any local change (in a novel, the reinterpretation of one character's action) be registered throughout the work (by a reinterpretation of the whole plot, say)--where this whole is defined only as the poem on the page. In short, this makes form too static, or "tension" too defining, at the same time as it makes local "ambiguity" too important. Two indicted critics here would be Cleanth Brooks on the one hand (dear to most New Critics) and William Empson (dear not only to deconstructionists like de Man but indeed to most current critics and theorists) on the other--though it is their tendencies as combined and put to work by the Chicago critics (R.S. Crane included) that are under consideration here. As you can see from the quote above, this all is involved in the general movement of over-avoiding the affective fallacy at the same time as one commits the intentional fallacy (and covers up the deed).