Remarks on Impure Quotation*

Quotation marks are ambiguous, although the conventional rules that govern their different uses are similar in that they contain quantifications over quotable expressions. Pure uses are governed by a simple rule: by enclosing any expression within quotation marks one gets a singular term, the quotation, that stands for the enclosed expression. Impure uses are far less simple. In a series of uses the quotation marks conventionally indicate that (part of) the enclosed expression is a contextually appropriate version of expressions uttered by some relevant agent. When the quotation marks have this meaning, it is tempting to think of them as contributing that indication to the truth-conditional content of the utterance. I adopt a cautious attitude towards this hypothesis, for the evidence in its favor is inconclusive. In other uses the quotation marks conventionally indicate that the enclosed expression should be used not “plainly” but in some broadly speaking “distanced” way, or that it is being so used by the utterer, and typically context makes clear the exact nature of the “distance” at stake. In these cases the quotation marks do not even appear to contribute that indication to the truth-conditional content of the utterance. I Quotation marks are used for a variety of purposes. One that is especially significant is the purpose of referring to expressions. In cases where the role of quotation marks is just to help in this purpose we can speak of purely referential quotation, or simply pure quotation. In typical utterances of the following sentence, the quotation it contains will be used with a purely referential purpose: (1) “Socrates” has eight letters. In earlier work (Gómez-Torrente (2001)) I have made two main claims about pure quotation. The first is that it can be understood as governed by a simple semantic principle, apparently first identified by Tarski (1933): by enclosing any expression within quotation