The week before Derrida died, I was working on a position paper surveying several different theories of metaphor. The inquiry seems to arc towards translation problems; it abruptly stops because I ran out of time to work on it. I think it is the matter of intersemiotic translation that makes such inquiries seem inconclusive. My inquiry also halted because I was reading Wilson and Sperber’s Relevance Theory, and hit their claim that metaphor is a poorly defined mode of inquiry. Metaphor, for them, is simply one aspect of the normal inferential process rather than any sort of special case. The major challenge of speaking about metaphor at all is that it cannot be spoken of without recourse to metaphor—using the term as a tautology, in definition of itself.
Speaking of metaphor at all assumes that it exists. Metaphor, in its broadest sense, is a framing device—a juxtaposition which results in the shift in meaning of a reference, either a word-reference (metalanguage) or an object reference (catachresis). While I was writing about it, I couldn’t help but think about Derrida’s The Truth in Painting. The obvious reference is to the distinction between ergon (content) and parergon (frame). Max Black’s terminology defining the pairing of concepts in metaphor—focus and frame—at first suggested an optical metaphor to me. Actually, though, this is more of a spatial metaphor suggesting an inside surrounded by a border (context).
The context (border) shifts across time and space in usage—therefore meaning shifts and is indeterminate. We can only speak of contingent meanings in metaphor, or, as Derrida would have it, in general. To attach a “protocol” to two terms—such as subject and predicate, signifier and signified, focus and frame— we dictate in advance the range of possibilities for the explication. Understanding the relationship between things means hypothesizing that a relationship exists; broadening the relationship lessens its explanatory properties when applied to contingent situations. Metaphor, in Wilson and Sperber’s opinion, does not define a specific relationship: “Metaphor plays on the relationship between the propositional form of an utterance and the speaker’s thoughts” (243). All language does that—if we define language as the translation of thought into signs.
If all signs are polysemous and context-dependent, then how is metaphor different? The answer, from Aristotle forward, is to brand metaphor as a form of transgression. Classically, it is a word-level nominal transgression—the application of an improper name. In the twentieth-century, it seems to have expanded into a propositional idea—metaphor is the application of an absurd comparison. These transgressions require some concept of a normative meaning that the metaphor is measured against, otherwise there can be no transgression. These definitions require the application of a rigid and specific frame.
The opening section of The Truth in Painting, labeled “Passe-Partout” uses the polysemous nature of language to render the concept almost funny. The phrase literally means “passkey” but it also signifies the type of overlay one uses to frame a picture—a mat. As Derrida mentions in passing, mats are often beveled. Instead of a rigid frame of taxonomy (inside vs. outside), the transition between frame and focus (or concept) is filled with an odd sort of slippage which threatens to make either the object of study, or the frame, disappear.
This brought me, in my twisted mind, back to William Blake’s indictment of the “muses of memory”—Blake called for artists to work through the “daughters of inspiration” (which might be literally translated as the inhaling of the Holy Spirit, in his context) rather than being slaves to memory. It also made me remember August by Tom Verlaine, a song sadly unreleased to the modern marketplace:
But the final vision has
Grown at least a hundred-fold,
I see your disillusion gone
A quiet room, not long ago,
Beneath the silent room deliver me,
No metaphor, no memory,
Above the hours always shining through.