Concept of the week: Grice’s “implicature”

[I am sometimes labouring here under the fixed belief that I’ve reviewed books/articles or that I’ve put down notes which in fact I haven’t. Grice’s implicature, the concept on which I will turn in a moment, is such an unmentioned, though notwithstanding referred to specimen. For those of you who aren’t familiar with the CoTW section, please note that what follows is merely an introduction (and a non-critical one). Now, to hamper the yawns of those who are closely acquainted with the term implicature, let us begin.]

 

 

In 1975, Herbert Paul Grice (more often, H. P. Grice), a British philosopher of language, published an article called Logic and Conversation, in which answers are given to two long-lasting philosophical questions: (1) how do we mean more than we say? (2) how does the hearer understand what we say, when we mean more thanwe say? Grice’s explicit aim was to give an account as to how/why some “formal devices” (~, ∨, ∧, ⊃  etc.) are not completely identical in formal language and in natural language[1] (not, and, or, if etc.) He also seemed specifically averse to the lofty positivistic thesis that “the possession by the natural counterparts of those elements in their meaning, which they do not share with the corresponding formal devices, is to be regarded as an imperfection of the natural language” (Grice, [1975]1989, p. 23).

So he proposed the term IMPLICATURE. A speaker implicates something (hence, making that something, an implicature), when what he means is not part of what he said. For example, if I say

(1) Even this blog has about 60-70 visitors a day. [which he does! not. ignore this],

I am implicating, by the use of the word even in this context, that it would be the least likely for this blog to have more than 60 visitors a day, or that the normal expectation is below 60. Let’s take two other examples:

(2) A: Do you know what time it is?
B: It’s half past twelve.

(3) A [from an obviously immobilized car]: I’m out of petrol  
B: There is a garage round the corner                                        (Grice, 1989, p. 32)

In the (2) example, speaker A is literally asking a yes/no question (the answer “Yes, I do know what time it is” is linguistically appropriate) but speaker B understands that A actually wants information (to know the time). But the difference is that, while the implicature in example (1) is embodied in the semantic description of the word ‘even’ – it is, thus, CONVENTIONAL –, in the (2) and (3) examples it is generated by the context, i.e. calculable not from the meaning of words, but from the mutual goal of the dialogue in which the utterance is being put forward. It is, thus, CONVERSATIONAL. While (2) is a more generalized version of the implicature, (3) is a more particularized one. So, if we were to picture this sort of analysis, it would look something like:

 

---

TOTAL
SIGNIFICATION = What is SAID
OF UTTERANCE                +
                               What is IMPLICATED = [either] CONVENTIONALLY (”conventional implicature”)
                                                                                          [or]
                                                                            CONVERSATIONALLY = [more or less] GENERALIZED/
                                                                         (”conversational implicature”)                      PARTICULARIZED

---

 

According to Grice (1989, p. 39), a pretty good method to determine whether the implicature is conventional or conversational, is this: conversational implicatures are (a) “not part of the meaning of the expressions to the employment of which they attach”, (b) cancellable and (c) non-detachable. For an implicatum to be cancellable, it means that there should be some means of annulling or invalidating what is implcated (for example, I should be able to say, in the example (3), “I’m out of petrol,” … “but it’s ok, I’ve just called someone”. However, all things being the same, I cannot ‘detach’ the implicature in that specific context: “it will not be possible to find another way of saying the same thing, which simply lacks the implicature in question”.

In some way, we can say we’ve seen how the speaker means more than he sais, but we have actually only pointed out names for different types of information. In the next CoTW post, we will tackle the second question: how is it that the speaker and the hearer understand each other, how does the former encode and the latter decode the implicated meaning.

---

[1] To select one of the examples of imperfect analogy between formal constants and their natural counterparts “It has, for example, been suggested that because it would be incorrect or inappropriate to say ‘He got into bed and took off his trousers’ of a man who first took his trousers and then got into bed, it is part of the meaning, or part of one meaning of ‘and’ to convey temporal succession” (Grice, 1989, p. 8)

 

Technorati Tags: Herbert Paul Grice,H. P. Grice,Paul Grice,conversational implicature,implicature,conventional implicature,Principle of Cooperation,Logic and Conversation,Studies in the way of words,meaning,utterance