Sperber & Wilson and Grice share the basic assumption th a t communication is an inferential a c tiv ity involving identification of the intended interpretation. However, the differences between the two are significant - they differ particularly in how the possible interpretations of an utterance are evaluated and accepted or rejected.
Grice claims that conversation is a co-operative activity governed by the Co-operative principle and maxims:
Co-operative principle
Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged (Grice 1975: 45)
Maxims of quantity
1 Make your contribution as informative as is required. 2 Do not make your contribution more informative than
is required.
Maxims of aualitv
Supermaxim: Try to make your contribution one that is true.
1 Do not say what you believe to be false.
2 Do not say that for which you lack adequate evidence.
Maxim of relation Be relevant.
Maxims of manner
Supermaxim: Be perspicuous.
2 Avoid ambiguity.
3 Be brief (avoid unnecessary prolixity). 4 Be orderly.
Grice’s biggest contribution to the study of communication is his claim that utterance interpretation is^matter of hypothesis formation and evaluation. His co-operative principle and maxims are indeed the standards by which hypotheses about the intended in te rp re ta tio n are evaluated. In Grice’s framework, the interpretation which satisfies the co-operative principle and maxims should be confirmed and accepted as the one intended. Implicatures are then the beliefs that have to be attributed to the speaker in order to preserve the assumption that the Co operative Principle and maxims (or at least the Co-operative principle) have been obeyed.
However, although Grice’s framework is a much better starting point than the code model, it also has some significant problems. Let me mention three major drawbacks of his fram ework. First, it is not clear where the Co-operative principle and maxims come from, and whether they should be seen as universal or culture-specific (see e.g. Keenan 1976). Second, many of the terms used in the Co-operative principle and maxims are too vague to be useful. For example, the first maxim of quantity says, ‘make your contribution as informative as re q u ire d ’ , and the second quantity maxim says th at the contribution should not be more informative than is required’ . The problem here is that the notion of the ‘required information’ is le ft unexplained and undefined. Similar criticisms can be applied to most of his maxims, and it is especially significant that the maxim of relation is le ft undefined. Finally, Grice’s framework makes apparently wrong predictions about what the hearer will choose as the intended interpretation of an utterance. An adequate account of utterance interpretation should be able to explain which of the various possible
interpretations will be chosen by the hearer. Grice’s framework, however, provides no method of choosing between a range of interpretations which satisfy the Co-operative Principles and maxims. I will illustrate this point by reference to example (3):
(3) a. Do you fancy that Italian restaurant on the High Street?
b. I have eaten pasta three times this week.
As I mentioned, this example cannot be accounted for by the code model, since it ignores or underestimates the role of inference in interpretation. Grice, on the other hand, offers an explanation for how (3) can be understood.
According to Grice, when the speaker apparently violates at least one of the maxims, the hearer will search for an interpretation which satisfies it, and attributes to the speaker any belief required to dispose of the apparent violation. Such beliefs are what Grice calls ‘implicatures’ . In the example above, (3b) is the result of an apparent violation of the Maxim of Relation, since it seems not to answer the question in (3a). Therefore, according to Grice, the hearer of (3b) will start searching for implicatures. As I mentioned before, the most natural interpretation of (3b) is to attribute to the speaker the beliefs in (4):
(4) a. If one has eaten pasta three times in a week, one will not choose to eat pasta again within the same week.
b. Jane doesn’t fancy Spaghetti Junction.
These would then be implicatures of (3b), which restore the assumption that the maxim of Relation has been obeyed.
However, there is a significant problem: Grice’s maxims will apparently be satisfied by various other possible
interpretations of (3b), such as the one in (5):
(5) a. If one eats pasta three times a week, one must love pasta.
b. If one loves pasta, one wants to eat in restaurants which serve it.
c. Jane fancies Spaghetti Junction.
The problem is that, although it is unlikely that the hearer of (3b) will interpret it as (5) suggests, Grice’s framework does not provide an obvious way of rejecting this interpretation.
Relevance theory differs from Grice in various ways, and as a result, it does not encounter the same problems as Grice. In relevance theory, Grice’s Co-operative principle and maxims are replaced by a single principle, namely, the principle of relevance:
Principle of relevance
Every act of inferential communication communicates the presumption of its own optimal relevance
According to Sperber & Wilson, the principle of relevance is not something that has to be known or followed by communicators, and communicators could not violate it even if they wanted to. The principle o f relevance is simply an exceptionless generalisation about what happens when someone enters into communication. Its source, according to Sperber & Wilson, lies in an essential feature of human cognition. Human cognition is relevance-oriented: humans typically pay attention to the most relevant phenomena available; they typically construct the most relevant possible representations of these phenomena, and process these representations in a context that maximises their relevance. As a result, when a communicator requests an audience’s attention, this automatically raises an expectation
that the utterance (or other act of communication) will be relevant enough to be worth the audience’s attention - or, in Sperber & Wilson’s terms, ‘optimally relevant’.
Furthermore, unlike what happens in Grice’s maxims, the most crucial notion in relevance theory, namely, ‘optimal relevance’, is clearly defined:
Optimal relevance
An utterance, on a given interpretation, is optimally relevant iff:
(a) it has enough contextual effects to be worth the hearer’s attention;
(b) it puts the hearer to no unjustifiable e ffo rt in obtaining those effects.
Notions such as ‘contextual effects’ and ‘processing e ffo rt’ are central to relevance theory, and I will discuss them later in this chapter.
Finally, relevance theory is preferable to Grice’s theory in that it offers an adequate method for evaluating candidate interpretations. The following is the pragmatic criterion proposed by Sperber & Wilson:
Criterion of consistency with the principle of relevance An utterance, on a given interpretation, is consistent with the principle of relevance if and only if the speaker might rationally have expected it to be optimally relevant to the hearer on that interpretation.
As Sperber & Wilson show, because o f the way ‘optimal relevance’ is defined, there is never more than at most a single interpretation which satisfies this criterion. I will discuss how the criterion works in actual communication in section 3.
2. Basic Notions
2.1. Communication and Relevance
In order to understand how the principle of relevance actually works, it is important first to clarify the notion of ‘communication Sperber & Wilson are interested in a notion of communication that involves ‘ostensive’ behavior. By ostensive behavior, they mean a behavior intended to attract an audience’s attention to some phenomenon. As noted above, Sperber & Wilson claim th a t ostensive behavior autom atically creates an expectation, or presumption, of relevance, i.e. an expectation that the ostensive behavior will be relevant enough to be worth an audience’s attention. On the basis of the presumption of relevance, once she starts attending to a certain ostensive phenomenon, the audience will try to find the intended meaning of the ostensive behavior. This mode of communication is called ‘ostensive-inferential’ communication and is defined as follows:
Ostensive-inferential communication
the communicator produces a stimulus which makes it mutually manifest to communicator and audience that the communicator intends, by means of this stimulus, to make manifest or more manifest to the audience a set of
assumptions ///. (Sperber & Wilson ibid.: 63)
Within ostensive-inferential communication, there may be an element of coding, as in verbal communication, but it also works on its own, as in many types of non-verbal communication.
Sperber & Wilson argue th a t by attracting someone’s attention, a communicator is asking for some effort to be spent, since paying attention requires a certain amount of effort. They also claim that humans are not willing to spend their effort for nothing - there is always an expectation of some reward. In the case of communication, as in cognition, the reward for attention
and mental effort is measured in terms of relevance. Hence, if a communicator requests your attention, you assume that as a result of paying attention, you will get adequately relevant information as a reward. This fact, that every act of ostensive communication creates an expectation of relevance, is the basis of the principle of relevance, introduced in the last section.
H ow ever, th e re are usually several possible interpretations of an act of ostensive communication, and this is why m isunderstanding often occurs. To succeed in communication, the communicator’s task is to make sure that his intended meaning is the one selected. The audience’s task is to choose the intended meaning, and nothing else. The question is: how can this be possible? To answer this question, we have to understand the notion of ‘optimal relevance’. In defining the notion of optimal relevance, Sperber & Wilson introduce two factors - contextual effect and processing effort - which lie at the heart of their theories both of communication and of cognition.