Andrew Radford - Linguistics An Introduction [Second Edition]

speaker b: Is the Pope a Catholic?

396

senten ces

d.

speaker a (a journalist):

Do you think the President was telling the

truth?

speaker b

(a government official):

I have no evidence which would demonstrate

conclusively that he was not telling the truth.

These examples are all a little odd in some way if taken at face value. The writer in

(535a) expresses a straightforward falsehood. But is it a lie? That depends on the context. If the writer is trying to put a brave face on a bad vacation decision, it could be a deliberate attempt to mislead by telling an untruth. On the other hand, if the writer knows that the recipient of the postcard has been following the weather forecasts, it will be properly interpreted as ironic. In (535b), the writer is expressing something that, at first blush, appears to be completely uninformative. The weather ‘could have been better’ most days. As for (535c), we might ask what relationship there could possibly be between the President’s probity and the religious affiliation of the Holy Father. And in (527d), did the official accuse the President of lying?

In a boring and logical world (Mr Spock’s Vulcan world perhaps), we should

replace the examples in (535) with those in (536):

(536) a.

We’ve had very bad weather

b.

We’ve had very bad weather

c

speaker a: Was the President lying?

speaker b: Yes.

d.

speaker a (a journalist): Do you think the President was telling the truth?

speaker b (official):

No.

However, given the right context, the examples in (535) get across the same message as those in (536), only more vividly. How is this possible?

In the 1960s, the philosopher Paul Grice drew attention to examples like those in (535) and argued that ordinary conversation must be governed by a Co-operative Principle according to which interlocutors are required to be helpful to each other. This rules out lying (even white lies) and other deliberate attempts to mislead, as well as boasting, false modesty and so on. Grice maintained that the overriding Co-operative Principle is reflected by conversationalists’ adherence to four conversational maxims governing the way we interact in conversation. These maxims are as in (537):

(537) a.

Maxim of Quality: try to make your contribution one that is true, specifically (i) do not say what you believe to be false;

(ii) do not say that for which you lack adequate evidence.

b.

Maxim of Quantity:

(i) make your contribution as informative as is required for the current purposes of the exchange;

(ii) do not make your contribution more informative than is required.

c.

Maxim of Relation: make your contributions relevant.

Using sentences

397

d.

Maxim of Manner: be perspicuous, and specifically

(i) avoid obscurity;

(ii) avoid ambiguity;

(iii) be brief;

(iv) be orderly.

The point of the Co-operative Principle and the maxims is not to tell people how to behave, of course. The point is that speakers are permitted to flout the maxims in order to convey something over and above the literal meaning of their utterance. The example in (535a), in circumstances where the recipient of the postcard is assumed to be familiar with the bad weather, flouts the Maxim of Quality, and the consequence is that the intended meaning is the opposite of the literal meaning giving the effect of irony. In example (535b), the writer flouts (i) of the Maxim of Quantity, giving rise to understatement or litotes. In (535c), speaker B’s response to A’s query appears to be a completely irrelevant question, violating the Maxim of Relation. But via this flouting – an obvious violation – B invites A to conclude that the President was lying just as surely as the Pope is a Catholic. Finally, (535d) relies for its effect on the fact that the official’s prolixity flouts the Maxim of Manner, and this again invites the addressee to seek an interpretation beyond the literal meaning of what B says.

It is useful to have some way of referring to the kinds of proposition that a speaker intends to convey in this implicit fashion, and the standard term for this is conversational implicature. The implicature is conversational because it only arises in an appropriate conversational context. In different contexts, the relevant utterances in (535) might be given their literal interpretations. For instance, if the official speaking in (535d) had just presided over an exhaustive and independent inquiry into the President’s testimony and wished to convince the audience that the President had in fact (despite all the rumours) told the truth, the utterance in (535d) might be used to support the President (exercises 4 and 5).

Context and coherence

Earlier in this section, we saw how the context of utterance is impor-

tant for determining the interpretation of deictic words such as personal pronouns.

Moreover, it should be clear that context is crucial in the operation of Gricean maxims. A graphic illustration of the importance of context is illustrated by the following interchange which took place between one of the authors (A) and a

colleague (D):

(538)

D: Hmph! If I’d known it was going to be fish, I’d have put in my contact lenses.

A: You don’t like kippers, then.

In (538), A has interpreted D’s utterance in exactly the manner intended by D, because the context provided ample clues. Actually, A then continued ‘Do you realise how incomprehensible your last statement would sound out of context?’

The contextual knowledge needed to interpret D’s utterance is that D was late

398

senten ces

arriving for breakfast at a conference and had not put in his contact lenses in order to save time, only to discover that the only food left was something he didn’t like.

It will probably not surprise readers to learn that we have yet to find anyone who can construct this context without a lot of hints.

One reaction to an interaction like that in (538) is that it is incoherent. A’s response does not seem to fit D’s utterance. We feel that the utterances that make up a discourse should be coherent, though it’s rather difficult to define exactly what we mean by that. In (539), we see another (more famous) example, which in its original form, was presented by the American sociologist, Harvey Sacks:

(539)

mrs smith: I have a fourteen-year-old son.

mr jones:

Well, that’s all right.

mrs smith: I also have a dog.

mr jones:

Oh I’m sorry.

This discourse seems hopelessly incoherent, until we learn that Mrs Smith is trying to rent an apartment and Mr Jones is a landlord.

Various groups of linguists, psychologists, philosophers, computer scientists and others have tried to provide a definition of textual or discourse coherence, and it seems that the essential feature of this property refers to what speakers and hearers believe and what they can sensibly infer. In the dialogue in (539), both participants understood perfectly well that Mrs Smith was looking to rent an apartment from Mr Jones and so Mr Jones’s final response would be taken to

mean something along the lines of ‘I can’t rent the apartment to you’. But this comes about as a result of our knowledge of the restrictions on property rentals.

Notice that this sense of ‘context’ is more general than that we referred to in our discussion of the interpretation of deictic expressions. All that was relevant in the earlier case was readily identifiable factors such as the identity of the speaker and the time of utterance. Here, however, context seems to be embracing the full set of beliefs that speakers and hearers have and inferences that they might make on the basis of those beliefs. Importantly, however, when computing the full meaning of a discourse, we obviously don’t try to deploy everything we know or believe about the world or all the possible inferences that we could draw. We only make use of beliefs and inferences which are relevant to us, and, as we have seen, the notion of