ABSTRACT
Suppose all you know about John is that, in uttering a certain sentence yesterday at 5 p.m., he either gave permission to Mary to marry a linguist, or wished Mary to marry a linguist, or asked whether Mary is going to marry a linguist, or predicted that Mary will marry a linguist, or objected to Mary’s marrying a linguist. Could you validly infer, from this piece of disjunctive knowledge, that, no matter which one of these five things John might have done in uttering the sentence, what he would have meant in uttering it would be the same? You certainly could not. And this suggests that, in order to identify what a speaker means in uttering a sentence of his language, it is not enough that you should know which individual he thereby purports to identify (for example, Mary) and which property he thereby purports to, truly or falsely, ascribe to that individual (for example, the property of getting married to a linguist at some point in the future) – to put it more generally, it is not enough that you should know which proposition he purports to be expressing in uttering the sentence he utters. What is required, in addition, is that you should know what is the meaning-determining act in the context of which he expresses that proposition – whether, for example, he expresses it in the context of an act of giving permission, or in the context of an act of giving a wish, or in the context of an act of asking a question, or in the context of an act of making a prediction, or in the context of an act of raising an objection, and so on.
