Reasons to doubt the present evidence for metaphoric representation.

In his reply to my article, Raymond Gibbs provides further information about the thinking underlying claims for metaphoric representation. As he points out, I was not able to review some of the theoretical rationale underlying claims for metaphoric thought, such as the issue of embodiment. He also describes further empirical evidence for the notion of metaphoric concepts. To summarize my rebuttal, I will argue that the empirical evidence Gibbs cites is not very conclusive in supporting metaphoric representation. Furthermore, the problems with linguistic evidence remain in spite of his defence. Much of the problem is that the notion of metaphoric representation is too vague at this point to be empirically confirmed or disconfirmed. However, I agree with Gibbs that the idea that parts of a concept must be completely consistent is likely not to be fully correct, and I conclude that further work is necessary on this question. One topic I won’t discuss in much detail is that of embodied cognition. Gibbs points out that theories of metaphoric cognition have often referred to underlying bodily processes and experiences in order to explain why we have some metaphors rather than others. (This cannot be the basis for all metaphors, since some of them, like LOVE IS A FINANCIAL TRANSACTION, are not very bodily – in fact, love is much more embodied than financial transactions are.) Although the notion of embodiment is an interesting one, it is not an empirical finding, but is itself a theory that does not yet have wide acceptance. I for one am not sure how bodily sensations of containment, say, can be extended to understand abstract containment relations such as text being in a book, or an object being in a category (Lakoff, 1993, p. 212). In any case, I think that the argument from embodied cognition must await further development of this theory.