The aim of this study is to secure the phenomenon of metaphor, and to defend a distinctive theoretical account in view of two challenges that threaten metaphor’s status as a worthwhile object of study for philosophy : the non-cognitivist challenge, and the deflationary challenge.
Part I of the investigation deals with the first of these two challenges, and comprises three chapters. In Chapter 1, we shall take a look at Max Black’s Interaction Theory of Metaphor, a classic formulation of a philosophical approach that aims at vindicating metaphor’s uniqueness. Black’s proposal has it that metaphor features a cognitive yet unparaphrasable content over and above the literal meaning of the sentence uttered. In Chapter 2, we shall look at two criticisms that Donald Davidson famously brought to bear against Black’s cognitivism about metaphor : First, the effects of metaphor have the wrong format to count as meaning. And second, they are brought about in the wrong way to qualify thus. We will find fault with the first claim while conceding that, in the special case of conceptual innovation, the second claim may hold water. Conceding that much, however, does not force us down the path of non-cognitivism about metaphorical meaning. Or so we shall argue in Chapter 3 by drawing on Paul Grice’s notion of speaker’s meaning as well as by reading against the grain some of Davidson’s later remarks on the pragmatics of interpretation. The upshot of our discussion will be that crafting and apprehending metaphor goes hand in hand with a certain degree of conceptual and pragmatic competence that confers more than a modicum of rationality to most uses of figurative language.
In Part II, we shall pursue this Gricean line of thought, over the course of three further chapters, up to the point where such pragmatic accounts of metaphor give way to deflationism. Chapter 4 is dedicated to fleshing out what a comparison theory of metaphor would look like that takes its lead from Grice’s theory of conversation while incorporating some of Davidson’s key insights. While all may seem well and good at this point, we will show in Chapter 5 that this Gricean-Davidsonian Comparison Theory of Metaphor runs into serious trouble, both empirically and theoretically. For contextualists in general, and Relevance Theorists in particular, produced arguments showing that metaphorical content is not conveyed indirectly, via conversational implicature, but is directly expressed by what is said. What does the trick is to allow for pragmatic processes such as modulation or ad hoc concept construction, which operate at the local, subsentential level of lexical items. Backing up this sort of view is the observation that, more often than not, the linguistic meaning of words and phrases underde-termine the concepts language users intend to convey. If it were true that metaphor is an instance of such ad hoc concept construction, and if we accept that occasion-specific modulations of meaning are nigh on ubiquitous, we might indeed end up embracing Relevance Theory’s deflationism. In Chapter 6, we will break this chain of reasoning, though, by raising a fundamental question : Is ad hoc concept construction the right model for metaphor ? We shall answer in the negative, notably citing three reasons to that effect. First, this account risks mixing up metaphor with polysemy. Second, it fails to track the peculiar phenomenal quality of metaphor, namely a kind of discrepancy between literal and metaphorical meaning that reasonably sensitive language users typically perceive. Third, Relevance Theorists themselves proposed another, more appropriate model for metaphor, in the guise of what Robyn Carston called the lingering of the literal. The remainder of Chapter 6 is devoted to cashing out a dis-tinctive, multipropositionalist model for metaphor, which takes inspiration from Carston’s proposal, among others.