Metaphors in Philosophy

Until recently, I thought metaphors had little place in philosophy. It's not that I subscribe to a literalist view of metaphor, wherein a metaphor means nothing more than its semantic, propositional content, but that I think metaphors are not satisfyingly truth–verifiable (typically, they are not at all truth-verifiable). To be clear about what I'm talking about: take the phrase (this is an example from some academic's paper): "You are the cream in my coffee." By a literalist view, this sentence is nonsense. The indexical 'you' requires a human referent and humans are not "cream". Simple as that. Now for some more interesting examples, and the reason I'm writing. In a paper entitled "Sex, Breakfast, and Descriptus Interruptus," Kenneth Taylor argues for a crossover view of semantics and pragmatics in which semantic contents (what the words 'mean' irrespective of context) may sometimes be indeterminate without some amount of pragmatic interpretation. Specifically, he suggests that "context-independent ingredients of sentence meaning more or less tightly constrain the to–be–contextually determined values of either suppressed or explicit parameters.” (55) He calls this view 'parametric minimalism.' His point in arguing it is that, though there may be prepropositional pragmatic externalities (instances where semantic meaning must be determined by reference to context), the potential meanings of the utterance are constrained by semantic (ie. context independent) factors. This is important because otherwise we might be able to claim (as Récanati does) that all context is important and we might then be lost in a willy–nilly orgy of free association. Intuitively, I agree with Taylor. However, in order to give his argument a semblance of factual underpining, he suggests that sentences typically set "up a semantic scaffolding which constrains, without determining, [their] own contextual completion.” (53) Metaphor, in the form of 'semantic scaffolding,' is already creeping in, but so far, it's alright. I'm happy to take his 'scaffolding' to be an explanatory description and not an assertion of a determinate proposition. However, he soon goes on to construct some perhaps dodgy metaphors. These metaphors creep in as he tries to account for the semantic parameters that constrain utterance meaning (an utterance being an instance, within a specific context, of a particular sentence.) As he says, “sometimes the to-be-contextually determined parameter is explicitly expressed in the syntax." Examples are explicit indexicals, demonstratives and verb tenses. However, Taylor also claims that the parameters are sometimes suppressed. As he admits, saying where they are hidden is difficult, but, nonetheless, he goes on to give us two juicy metaphors to help locate them: "some unexpressed parameters hide in what we might call the subsyntactic basement [first metaphor] of suppressed verbal argument structure." As an example, Taylor says that the verb 'to rain' “has a lexically specified argument place which is θ–marked THEME and that this argument place take places as values.” (53) This is to say that “the subatomic structure [second metaphor] of the verb ‘to rain’ explicitly marks rainings as a kind of change that places undergo.” (53) Taylor says we have a ‘tacit recognition’ of this, and intuitively, I agree, but I don't really think there is any such 'sub-atomic structure'. I would instead advance the idea that our background knowledge of the verb 'to rain' constrains the "to-be-contextually determined" proposition to such a degree that it may seem that there exist context independent semantic parameters only because, with the verb 'to rain', the background knowledge is widely–shared and very well–established. Admittedly, this does not hold up so well to post-modern indeterminacy, but I'm happy to simply assert it as fact. (I mean, hey, Thomas Jefferson and the US Congress did just that sort of sleight of hand when they asserted the "self–evident" truth of the "unalienable Rights" of man. Similarly, the French National Assembly when they passed the "Declaration of the Rights of Man" in 1789.) Unfortunately, when I argue against Taylor in this way, I undermine such assertions. Clearly, I need to become a philosoper–king.

Back to Taylor: A week ago, I would have taken a much harsher view of Taylor's metaphors. I would have said that they assert no proposition (ie. they are not truth-verifiable), and that they obfuscate more than they explain. However, I am now much more sympathetic, and, I think, for good reason. Let us first remember that the philosophy of language is written in a meta–language (thank you Colin Cherry), and that the phenomena of our object language (in this case English) are apparent to us only to the extent that our meta-language contains identifying words. In other words, we can only talk about things for which we have words. As Cherry says, "metaphors arise because we continually need to stretch the range of words as we accumulate new concepts and abstract relationships." (Signs, Language, and Communication, 74) Thus they are often the first step in identifying and discussing a heretofore unrecognized phenomenon. Taylor's "subsyntactic basement" or "subatomic structure" (I prefer the former) may just be such a case. To his credit (unlike Récanati and his 'Availability Principle') Taylor is careful throughout to make clear that his ideas are merely hypothetical. So yeah, let us embrace the philosophical metaphor (sometimes, and with due reservations).