What, exactly, did Michael Dummett mean by “a theory of meaning”?
Firstly, it concerns the “implicit knowledge we have of our language”. Dummett says:
“It must issue in principles which will make explicit the implicit knowledge we have of the language we use, the knowledge we display in our practice, in such a way as to show that these principles do adequately reflect that practice.” (78)
This is not, then, unlike Chomsky’s enterprise of showing us the “implicit knowledge” we must have in order to generate an infinite possibility of grammatical sentences from a limited stock of linguistic material. This is accounted for, in Chomsky’s theory, by our “language faculty” which is built into our brains and thus also our minds.
Dummett is clearly doing something similar in his theory of meaning. Indeed we must have implicit knowledge of “the language we use” in order to account for that use. After all, not many non-philosophers know about truth-conditions, sense, reference, proper names and so on. Dummett must be arguing that we do have implicit knowledge of such things; just as Chomsky argues that we must have implicit knowledge of what he calls “universal grammar” and the “universals” contained in that grammar. This knowledge of truth-conditions, sense, etc. “we display in our practice” (78). Again, we wouldn’t display what we do display without such implicit or tacit knowledge.
These “principles”, however, must “adequately reflect that practice” (78) because Dummett isn't carrying out a normative or revisionary project. He is uncovering what we must actually know – even if only implicitly. Thus he mustn't “ascribe to us knowledge we could not possibly have” (78).
Many philosophers (for example Gilbert Ryle) have also distinguished “knowing how” from “knowing that”. The former is non-propositional or non-theoretical, unlike the latter.
Is our implicit knowledge of language a case of knowing how or of knowing that? Passmore writes that
“we have an implicit knowledge, for example, of the physical principles which are manifested in our capacity to ride a bike”. (78)
Didn’t Ryle and others argue that we couldn't even express or formulate our implicit knowledge of how we can ride a bike even if we were asked to? In that case, would knowing how actually be a case of implicit knowledge at all? Or perhaps we have implicit knowledge without being able to express of formulate it.
In Dummett’s case, does he think, or expect, the layperson to express or formulate his implicit knowledge about his language or does he still have such knowledge even when he or she can't do so? In that case, is implicit knowledge, as it were, mechanical and non-propositional (or non-theoretical)? If that's the case, again, why is it knowledge at all and not something else? Does it matter to Dummett whether or not we can express or formulate our implicit knowledge?
Passmore reiterates Dummett’s position by talking in terms of our understanding. He writes:
“A good theory of meaning, then, will be a theory of what it is to understand, of what one knows when one knows a language, so that, for example, ‘Meaning is use’ could serve as a thesis in such theory only if to know the meaning is to know the use – implicitly or explicitly.” (78)
To put this simply. Why and how is it that we understand what it is we understand when we hear or read examples of our own language? What enables us to understand such examples of our language? Another way of putting this is to ask, “what one knows when one knows a language” (78). After all, we must know something about our language and its sentences and words otherwise we wouldn’t understand our language. And what we know must be more than what we've learned at school. What we know, in a sense, must have come before our formal training (or our adult training) otherwise perhaps we couldn’t have even learned the basics in the first place (never mind the higher levels of grammar and vocabulary).
Interestingly enough, Passmore writes that the “meaning is use” thesis could be a part of Dummett’s overall theory of meaning. However, that would only be the case if “to know the meaning is to know the use – implicitly or explicitly” (78). That is, it's still about our knowledge or understanding of meaning. In this case, what we know is how we use words and sentences. These uses of words and sentences provide us with their meanings. Thus use-theory can become a part of Dummett’s general theory of meaning – that is, if he were to accept it in the first place.