W.V.O. Quine’s project in Word and Object is the derivation of a formal language suitable for scientific hypothesis in a way that does not rely, like logical positivism, on protocol sentences that are taken to communicate observations in impartial and objective terms, or the existence of “meaning” independently from language. Instead of building his system from the ground up out of logical primitives, Quine derives something quite similar to the language of propositional logic from empirically-testable regularities in our natural ways of speaking.
What’s crucial to Quine’s project is the idea of paraphrase. In Quine’s terms, we accept a sentence as a paraphrase of what we’ve said not because the two sentences express the same underlying meaning, but because we judge that the second would have served whatever our purpose was just as well. Quine sets out in detail a procedure for transforming arbitrary sentences into clear logical form using only operations that normally, at least when we are trying to speak objectively, produce acceptable paraphrases. The operations he chooses to use are not meant to apply with certainty in every single case, but their applicability is supposed to be empirically testable in a reasonable range of circumstances. As I’ve noted before, Quine emphasizes that it’s the original speaker who must finally judge whether a paraphrase is acceptable. Any supposition we might make about what the other person is getting at can ultimately be no more than an educated guess unless we ask them and have reason to trust in the truthfulness of their answer.
Since Quine’s primary interest is science, he proceeds in his regimentation of language by using operations that he thinks would normally produce acceptable paraphrases in scientific discourse. What happens if we apply a similar approach to literature? A canonical form that arises out of, say, poetry, might look quite different from what Quine comes up with for science. I don’t have the space to try this out in this post, but I will sketch out how the process might go.
First off, arguments about authorial intent would have to be off-limits in a Quinean approach to literature. The theory that Quine sets out puts the focus on the speaker, but he expects that, if we are in doubt of what someone’s getting at, we can actually ask them to confirm our paraphrase. To apply his approach to literature, we would have to set the standard for paraphrase in reader-centric terms based on the effect that the work has on us, so that we have a reasonable epistemic standard. An acceptable paraphrase of a poem, then, would be one that produces approximately the same response in us as the original. With this definition in mind, we would proceed by examining what changes we could make that produce acceptable paraphrases, looking for regularities in the sorts of operations that tend to leave a poem’s effect intact and perhaps learning something, thereby, about how a poem does what it does.
Those who’ve made a fair effort at understanding both analytic and continental philosophy won’t be surprised that, by applying Quine’s approach to literature, we’ve gotten quite close to deconstruction. Both approaches avoid having to talk about meaning by focusing on the relationships between different pieces of language. There is a difference in valence, though, and I think the Quinean approach has an advantage – it would, I think, be able to avoid making everything look like a “surface.” The question that a deconstructionist asks is what difference it would make if another word were used in a particular spot – she aims to get everything she can out of the little textual details, with the assumption that every word has an equal chance at significance. Instead, we would be looking for changes that wouldn’t make much of a difference, with the aim of stripping out all but a skeleton. In the process we would be building a logic of poetry in terms of connectives and quantifiers like the structuralists tried to do, but we wouldn’t be building them from the ground up. We would be enmeshing our theory with a living body of poetry.
This approach has an obvious affinity for the Imagists, and in particular the eliminative editing style of Ezra Pound, but I don’t think it would take us back to the Modernist distinction between surface and essence that has been rightly criticized from so many quarters. Instead, it provides an explanation of what Pound was doing if he wasn’t finding poetry’s essence. It would also provide, if it were to succeed, a class of literary universals that consists not of supposed “objective correlatives” between image and emotion, but, instead, of invariancies under transformation.