Contemporary philosophy of language offers, overall, two approaches to meaning. One is "representational semantics," and the other is "conceptual role semantics". According to representational semantics, the meaning of a word is what it refers to, and...
Contemporary philosophy of language offers, overall, two approaches to meaning. One is "representational semantics," and the other is "conceptual role semantics". According to representational semantics, the meaning of a word is what it refers to, and the meaning of a sentence is its truth conditions. This semantics offers a good explanation of the so-called "evolutionary problem." That is, the reason why human beings have survived successfully so far is because they have had correct representations about how things really are. Admittedly, it is difficult to doubt that our perceptual discourse, unlike a fictional story, engages with the world. Although representational semantics is, in some sense, intuitively plausible, it faces, unfortunately, many serious problems. For example, "phlogiston" is no longer taken to refer to something, it is still meaningful. In addition, contemporary physical theories, like Newtonian mechanics, could turn out to be wrong in the future. Moreover, although "Superman" and "Clark Kent" refer to the same man, these names play different conceptual roles in the psychology of Lois Lane, the heroine in the movie Superman. Due to these problems, conceptual role semantics seems to be an attractive alternative, according to which meaning is conceptual or inferential role. However, conceptual role semantics seems to face the problem of representation, for conceptual role semantics does not seem to explain how the meaning of a sentence is connected to the world. The purpose of my study is to avoid the problems of each of these two semantics, while preserving the intuitions behind them.
How can we explain that the meaning of an expression is about an object in the world, while we take the meaning of an expression to consist basically in its inferential role I think we can find the key to the solution in Robert Brandom’s masterpiece Making It Explicit (1994). According to Brandom, the meaning of an expression consists basically in its inferential role within our language, but its empirical content comes from the world. It is because we use our language in order to obtain information about the world from what others say. For this purpose, our language contains expressions which denote objects in the world, which we causally share. But here we need to note that an expression cannot obtain a meaning merely by this sort of causal connection. For example, suppose that there is a certain parrot that has been trained to utter "This is red" whenever it sees a red object. In this case, we cannot say that this parrot possesses the concept red. For this parrot would not understand that the sentence "This is red" implies sentences such as "This is colored" and "This is not a natural number," and "This is red" is also incompatible with sentences like "This is green." Therefore, merely reliably responding differentially to red things is not yet being aware of them as red. Thus, grasping or understanding a concept requires practical mastery over the inferences it is involved in. Nonetheless, as mentioned before, the reason why our linguistic expressions have empirical contents is because we use them in causal relations with the objects in the world. For examples, in order that you possess the concept red, you must have the disposition to noninferentially judge "This is red" when a red object is in front of you. Since this sort of causal connection is necessary for the meaning of an expression, we may say, on the one hand, that meaning is realistic; on the other hand, since the meaning of an expression is basically given by its inferential role, we may say that meaning is non-representational. This is the basic idea for my non-representational, realistic semantics.