I'll call the operative content of the utterance (OC, for short) what, in the Gricean framework, constitutes the input for the inference of implicatures of an utterance. After Grice, this is usually known as 'what is said' by the utterance (namely, what is said by the speaker in making the utterance), but this is somewhat misleading, since it is then considered to correspond (in the case of simple utterances involving singular terms) to a single kind of proposition: either a singular proposition (i.e., a proposition involving an individual) or a general proposition (i.e., a proposition involving not an individual but a mode of presentation of it). I'll argue that the OC of an utterance can be any of a variety of contents, including, but not limited to, those two kinds of contents. Grice (1967) famously distinguished between what a speaker says and what she implicates by uttering a sentence. Think about Anne and Bob talking about their common friend Carol, who both know that she recently started working in a bank. Anne asks: “How is Carol getting on in her job?”, and Bob replies: “Oh quite well… She hasn't been to prison yet.” Bob is clearly suggesting something here; something related to Carol's tendency to yield to the temptation provided by her occupation, as Grice would put it. But that's not something Bob said, but something he implicated in saying
* Previous versions of this paper have been presented at the ILCLI Seminar on Language and Communication and at the First Conference of the Latin-American Association of Analytic Philosophy. I am grateful to the audiences in both places and especially to Eros Corazza, Maite Ezcurdia, Joana Garmendia, Eleonora Orlando, María Ponte, Glenda Satne, Rob Stainton and Larraitz Zubeldia. John Perry deserves special thanks for being a constant source of inspiration. This work has been partially supported by a grant of the University of the Basque Country (GIU 08/23) and the Spanish Ministry of Science and Innovation (FFI2009-08574). what he said.* But what did he say? Grice’s remarks suggest that his concept of ‘what is said’ can be taken as equivalent to ‘the proposition expressed’ or ‘the content’ of the utterance.† He claims that to know what someone said by uttering a sentence one has to know (i) the conventional meaning of the sentence uttered; (ii) the disambiguated meaning of the sentence in that particular occasion of use; and (ii) the referents of referential expressions.‡ (Grice 1967: 25)
There is a debate about the amount of pragmatic ‘intrusion’ that Grice allows into what is said,§ but there seems to be a wide consensus that it roughly corresponds to the proposition expressed or the content of the utterance. The question is, what kind of proposition is that?
Limiting our attention to utterances of sentences containing singular terms --that is, proper names, demonstratives, indexicals and (some uses of) definite descriptions---, traditional philosophy of language offers two general, seemingly incompatible, answers: the proposition expressed is either a singular proposition involving an individual referred to by the singular term (the referentialist view) or a general one, involving a mode of presentation of the individual, provided by its linguistic meaning (the descriptivist view).
Well-known arguments about the subject matter of utterances, same-saying and * Grice included in his overall picture of meaning and communication non-linguistic ‘utterances’ like gestures and movements, but we will limit the discussion to linguistic utterances.
† This more technical terms used by philosophers are not trouble-free, since they can suggest that implicatures are not contents of the utterance, or that they are not propositional. Gricean implicatures (at least conversational particularized ones) are also full-blown truth-conditional (though more or less indeterminate) contents of the utterance, but I’ll ignore this issue here, and follow common practice using ‘content’ only to talk about the contents that are on the ‘what-is-said’ part of the Gricean divide.
‡ He leaves it open whether a speaker using a proper name, e.g. ‘Harold Wilson’, and another one using a definite
description denoting the same individual, e.g. ‘The British Prime Minister’ say the same thing or not
§
Later on, when he discusses the nondetachability of implicatures, he observes that “Insofar as the calculation that a particular conversational implicature is present requires, besides contextual and background information, only a knowledge of what has been said (…) it will not be possible to find another way of saying the same thing, which simply lacks the implicature in question.” (idem, p. 39) Except in the case of conversational implicatures that are to be inferred invoking the maxims of manner,** the nondetachability of implicatures implies that it’s what is said that matters, and not the way in which it has been said.††
Take, for instance, our initial example of Bob telling Anne “She hasn’t been to prison yet” about their mutual friend Carol, thereby implicating that she might have started to steal money from the bank she is working at. Any way of expressing the proposition that Carol has not been imprisoned by time t, including various ways to ** Grice’s supermaxim of manner says ‘Be perspicuous’, and the maxims concern avoiding obscurity, ambiguity, brevity and order. See section 6 below.
†† To be sure, Grice makes also the following statement: “(…) the implicature is not carried by what is said, but only by the saying of what is said, or ‘by putting it that way’” (ibid., p. 39). This is important. The (propositional) content does not carry any implicature, it is the utterance with that content which carries it. Still, it seems clear that one is supposed to identify what has been said to start with the ‘calculation’ of the implicature. That’s what we understand by what sometimes is dubbed as ‘input’ for the inference of implicatures. refer to Carol, would carry the same implicature. This seems to agree with a referentialist view on singular terms.
Nonetheless, we are not going to directly address the debate between
referentialists and descriptivists, or examine Grice’s view about the debate. Our aim is
to call attention upon a specific task Grice imposes to the concept of what is said within
his theory of implicatures: a demand that seems to be somehow present in all the
previous remarks and others like the following one:
“He has said that p; there is no reason to suppose that he is not observing the
maxims or at least the CP [cooperative principle]; he could not be doing this
unless he thought that q; (…)”
The calculation, inference or working out of conversational implicatures requires the identification of what is said. Among the information required to ‘calculate’ or infer what the speaker implicated, the hearer must identify, to begin with, what the speaker said. Or, in other words, the proposition expressed by the speaker is the input for the inference of implicatures. One may wonder what this demand can tell us about the kind of proposition that best plays that role in the case of utterances containing singular terms: is that a singular proposition, as a referentialist would independently defend, or is it a general proposition, as a descriptivist would expect? Our aim in this paper is to answer that question, leaving aside other arguments about what the best characterization of what is said, or the proposition expressed, or the content of an utterance is. We’ll call the proposition that constitutes the input for the inference of implicatures the ‘operative content’ of the utterance (OC for short) and we will start by considering the following issue. In the case of utterances involving singular terms, is the OC of an utterance a singular or a general proposition?
We’ll start by discussing some examples suggesting that the OC of an utterance
of a sentence containing a singular term seems to vary between a singular proposition
and a general proposition; and, moreover, the general proposition can involve not only
linguistic but also psychological mode of presentations or ‘cognitive fixes’ on the
referent. Then, we’ll introduce the theory that provides the adequate framework to deal
with our approach to the OC of an utterance: