Imagine you have been asked to review the reference letters provided by the candidates for a lectureship in philosophy. One reads: ‘My former student, Dr Jack Smith, is polite, punctual and friendly. Yours faithfully, Professor Jill Jones.’ You would, I assume, interpret that Jones is implying that Smith is a bad philosopher and unsuitable for the job. But how did she convey this? By what she left out. Jones does not say (literally) that Smith is a poor philosopher. Nor does it follow logically from what she says. It could be true that Smith is polite, punctual, friendly and an excellent philosopher. Yet somehow Jones gets the opposite message across.
When we convey a message indirectly like this, linguists say that we implicate the meaning, and they refer to the meaning implicated as an implicature. These terms were coined by the British philosopher Paul Grice (1913-88), who proposed an influential account of implicature in his classic paper ‘Logic and Conversation’ (1975), reprinted in his book Studies in the Way of Words (1989). Grice distinguished several forms of implicature, the most important being conversational implicature. A conversational implicature, Grice held, depends, not on the meaning of the words employed (their semantics), but on the way that the words are used and interpreted (their pragmatics).
Grice argued that conversational implicatures arise because speakers are expected to be cooperative – to make contributions appropriate to the purpose of the conversation in which they are engaged. More specifically, they are expected to follow four conversational maxims, which can be summarised as: (1) give an appropriate amount of information (the maxim of quantity); (2) give correct information (the maxim of quality); (3) give relevant information (the maxim of relation); and (4) give information clearly (the maxim of manner). According to Grice, a conversational implicature is generated when an utterance flouts one or more of these maxims, or would do so if the implicature weren’t present. In such cases, we can preserve the assumption that the speaker is being cooperative only by interpreting their utterance as conveying something other than, or additional to, its literal meaning, and this is its implicated meaning.
Jones’s letter is an example. Since she bothered to write it, we assume that Jones is trying to make a cooperative contribution. But the information she gives is obviously insufficient, flouting the maxim of quantity. Hence, we infer that she is trying to convey something else, which she doesn’t wish to say directly, and the obvious conclusion is that what she’s trying to convey is that Smith is unsuitable for the job. (The example is adapted from one of Grice’s own.) For other examples, think of saying: ‘That’s a nice way to behave’ (flouting quality) to convey that someone behaved badly; pointedly changing the subject (flouting relation) to convey that a remark was tasteless; or describing something in an unusual way (violating manner) to convey that it is unusual in some way (eg, calling a broken-down horse a ‘steed’).
In the examples just given, the speaker actually flouts a maxim, but sometimes an implicature arises in order to prevent a flouting. Suppose you need petrol and someone tells you: ‘There’s a garage around the corner’ (another of Grice’s examples). If the speaker did not believe that the garage was open, then their reply would violate the maxim of relation, so to preserve the assumption that they are being cooperative we must assume that they do believe it is open. Their utterance therefore carries this implicated meaning in addition to its literal one, even though it was not stated and doesn’t follow as a matter of logic from what was stated.
In Grice’s view, the link between utterances and any conversational implicatures they carry is a rational one, and conversationally implicated meanings can be inferred, or calculated, from the assumption that the speaker is following the maxims, together with the literal meaning of their words, details of the context, and background knowledge. Grice does not claim that listeners necessarily go through this process of inference whenever they spot an implicature – they might grasp the implicated message intuitively – but he insists that conversational implicatures can in principle always be calculated from the maxims.
Grice highlights several distinctive features of conversational implicatures, including being cancellable, nondetachable and indeterminate. Conversational implicatures can be cancelled by adding a further statement that makes it clear that the speaker had temporarily ‘opted out’ of following the maxims, and that the implicature is not to be inferred. For example, Jones could have added: ‘But that doesn’t mean he’s a bad philosopher; far from it.’
Conversational implicatures are nondetachable in that they are (with some exceptions) derived from the content of what the speaker says, not from the manner in which they said it, and would still have arisen if the speaker had chosen different words to express it. Finally, conversational implicatures can be indeterminate because there might be an open-ended number of different interpretations that would each preserve the assumption that the speaker is being cooperative. For example, in saying: ‘Juliet is the sun,’ Romeo could be interpreted as implicating that Juliet is beautiful, radiant, life-giving and so on, and all of these interpretations contribute to the implicated meaning. Metaphors and other creative uses of language are typically designed precisely to generate such rich indeterminate implicatures.
The distinction between what is said and what is conversationally implicated isn’t just a technical philosophical one. It highlights the extent to which human communication is pragmatic and non-literal. We routinely rely on conversational implicature to supplement and enrich our utterances, thus saving time and providing a discreet way of conveying sensitive information. But this convenience also creates ethical and legal problems. Are we responsible for what we implicate as well as for what we actually say?
Consider a real-life example. During a bankruptcy hearing in 1966, the American movie producer Samuel Bronston was asked if he had ever had a bank account in Switzerland. He replied: ‘The company had an account there for about six months, in Zurich,’ which implicated that he himself hadn’t had one. (If he had, he should have said so, so we infer that he hadn’t.) And this implicated message, it turned out, wasn’t true. Did Bronston commit perjury? He didn’t actually say anything false. (In the event, he was convicted of perjury, but the conviction was later overturned by the US Supreme Court.)
Similar issues arise with consent. ‘Shall we go upstairs?’ often implicates an invitation to sex. Has the speaker thereby consented to sex? What if they didn’t notice that their words carried the implicature? (And what if Bronston denied noticing the implicature that his reply carried?) To avoid disputes and confusion, perhaps we should use implicature less and communicate more explicitly? But is that recommendation feasible, given the extent to which human communication relies on pragmatics?
Not all philosophers and linguists accept Grice’s explanation of how conversational implicature works, but everyone agrees that he highlighted a real phenomenon that is pervasive in human communication. Grice provided us with an important thinking tool and, with it in your mental toolbox, you will find conversational implicatures, and the issues they raise, everywhere.
This article was originally published at Aeon and has been republished under Creative Commons.