9
USING LANGUAGE —
This is a self-clearing cafeteria' ran the notice in the student cafeteria. One might expect the plates and cups to put themselves away, judging from other similar phrases in the language, such us selfcleaiiing oven, sclf-ra king flour, self-righting lifeboat Yet the majority of students interpreted the phrase as meaning that they, the customers, were expected to clear away their plates. Why? The obvious answer is that they used [heir common-sense and knowledge of the world to come to the most plausible interpretation in the circumstances, which was not necessarily the one which was most corisistentwith' the linguistic structures,
Pragmatics is die branch of linguistics which studies those aspects of meaning which cannot be captured by semantic theory. In brief, it deals with how speakers use language in ways which cannot be predicted from linguistic knowledge alone. Li a narrow sense, it deals with how listeners arrive at the intended meaning of speakers. In its broad est sense, it deals with (he general principles followed by human beings when they communicate with one another. It is therefore sometimes light-heartedly referred to as '(he waste-paper-basket of semantics'.
Pragmatics overlaps with discourse analysis, which deals with the various devices used by speakers and writers when they knit single sentences together into a coherent and cohesive whole.
These fields ate still fairly new, and there is no general agreement yet as to how to deal with them. This chapter outlines a number of recent approaches which have proved helpful.
--The cooperative principle-
An American philosopher, Paul Grice, is sometimes regarded as the 'father of pragmatics'. Grice emphasized that human beings — 93 —
communicate efficiently because they are by nature helpful to one I another. He attempted'to specify the principles which underlie this * cooperative behaviour, and proposed foil r'maxims' or rules of conver- ■ sation which can jointly be summarized as a general principle: 'Be co operative'. These are given below. I
1 Maxim cf Quantity \
Give the right amoun 1 of in fomvatio n when you talk. If someone al a {
party asked 'Who's that person with Bob?', a cooperative reply would
be That's his new girlfriend, Alison'. An uncooperative reply would be
an over-brief one, such as 'A girl', or an over-long one, such as That's '
Alison Margaret Jones, bom in Kingston, Surrey on 4th July 1370,
daughter of Peter and Mary Jones... etc'
2 Maxim of Quality
Be truthful. For example, if someone asked you the name of an unfamiliar animal, such as a platypus, reply truthfully, and don't say 'It's a kookabu rra', or 'It's a duck', if you know it's a platvpus.
i
3 Maxim of Relevance j
Be relevant. II someone says, 'What's for supper?', give a reply which fits (lie question, such as 'Fish and chips', and not Tables and chairs' or'Buttercups are yellows
4 Maxim of Manner J
i
He clear and orderly. For example, describe things in the order in which they occuired: The plane taxied down the runway, and took off to the west' rather than The plane took off to the west and taxied down the runway', which might confuse people as to what actually happened. i
At this outline level, the cooperative principle seems like common- J sense. It becomes more interesting when we consider how often people apparently break it In answer to the question: "What's for supper?' one is likely to receive a reply such as: 'Billy fell downstairs', which i
U S" I IV O LANUUAUt
doesn't answer the query. In answer to a question 'Why don't you like Pamela?' one might get the response; 'Pamela's an elephant', which is patently untrue.
Such replies are not evidence against die cooperative principle. On the contrary, they simply show how strongly it works: people are so convinced that the other person in a conversation is being cooperative, that a superficial breakdown in a conversational maxim is treated as important and informative. For example, if someone said: "What's for supper?' and the reply was the superficially irrelevant one: 'Billy fell downstairs', the hearer is likely to assume that the information about Billy was somehow important, and will fill in the gaps with assumptions such as 'Since Billy was supposed to cook the supper, and he's fallen downstairs, I assume that there isn't any supper ready'. Similarly, if someone told an overt lie, such as 'Pamela's an elephant', the listener would n o 1 j u st think, That's impossible', he or she would cast around as to why the speaker had made this comment In brief, listeners interpret what people say as conforming to the cooperative principle, even when tl li s piii iciple is overtly broken. They draw implications from the utterance which arc not strictly there in the linguistic meaning.
'Ilie main problem with these Gricean maxims is that they are fairly vague, and the conversational implicatures or conclusions which can be drawn are wide and numerous.^Some recent work therefore has attempted to specify how humans manage to disentangle what is relevant from die mass of possible inferences they could make.
------Speech acts--
When a person utters a sequence of words, the speaker is often trying Lo achieve some effect with those words, an effect which might in some cases have been accomplished by an alternative action. The words 'Get back!' might convey the same notion as a push. A judge's statement: *I sentence you to five years imprisonment' is not a mere string of words, but has the same effect as if the judge had marched a man along to a prison, and locked him up. In brief, a number of utterances behave somewhat like actions. If this line of reasoning is taken further, one could argue that all utterances are acts of some type. Even an ordinary utterance such as "Violets are blue' might be regarded as a special type of act, the act of making a statement
(I state that) Violets are blue.
L J N G U f S TICS
This overall approach is known as speech act theory, and it is another method by which philosophers and linguists have tried to classify the ways in which humans use language, in this case by treating it as parallel tn o ther actions which humans perform.
Proponents of speech act theory try, in the first place, to list the various possible speech acts which a speaker might, attempt to perform -statements, requests, queries, commands, promises, placing of bets, and so on. The lists vary from writer to writer, though the overall core tends to be similar. At the heart of the list come statements, questions and commands:
CI state that:) It's cold.
(I ask you:) WJiat's the time?
(I command you:) Co away.'
These are examples of direct speech acts: the act is expressed overtly by the most obvious Jingoistic means. Hut many speech acts are indirect, in that they possess the syntactic structure more usually associated with another act. For example, the following might all he intended as commands, yet only the first has the typical command structure:
Co to bed!
Isn't it past your bedtime?
You should have been in bed long ago.
The first is therefore a direct speech act, but the second two are indirect speech acts.
But how do people know which speech act is intended, if each act can use the syntactic structure typically associated with one of the others? A possible answer is to specify happiness conditions or felicity conditions — circumstances under which it would be appropriate to interpret something as a particular type of speech act. For txampln, if a genuine command has been given, the hearer must be physically capable of carrying it out, and must be able to identify the object(s) involved. Even this partial statement of the felicity conditions for commands would probably enable someone to identify Tick up that book!' and That book oughtn't to be on the floor' as genuine commands, and 'Go jump hi the lake!' and 'Gird up thy loin si' as pseu do-con mi amis.
If we could fully identify the felicity conditions for each type of speech act, then we would have moved some way towards understanding how humans use language.
USING LANGUAGE
Remembered frameworks
The field of artificial intelligence (AD has provided a further approach to how people understand one another. AI makes proposals about how tn simulate intelligent systems on computers. The original problem was one of finding out how computers could be made to cope with inexplicit and superficially irrelevant conversations;
Sidesman: Fink sinks are the latest fashion, madam.
Customer: My dish-washer's red.
A solution proposed for the computer might also be one utilized by humans. Knowledge, it was suggested, might be stored in the form of stereotypical situations, or frames. These memorized frameworks arc adapted to fit in with present reality, so they are altered as required. So, for example, a person might have a frame representing a typical kitchen, and would have 'slots' in the frame for a sink, a cooker, a dish-washer and so on. A superficially disjointed conversation, such as the one above, would become quite coherent when considered in relation to the 'kitchen frame' in a person's mind. Furthermore the speakers ii i this conversation clearly have a certain amount of mutual knowledge, in that they both have a similar outline kitchen frame. Another way of dealing with human interaction, therefore, is to specify both the relevant frames and the mutual knowledge held in common by the participants.
Discourse analysis
So far, we have concentrated on cases in which people made sense of quite strange disjointed utterances. However, when we use language, we do not necessarily do so in a random and unstructured way. Both conversation and written texts have various devices for welding together miscellaneous utterances into a cohesive whole. Compare the two accounts of George's meal below:
A George ate the curry with delight. Curry had always been George's favourite food. The curry was subtly flavoured. George detected hints of cumin and coriander in the curry. Cumin and
coriander are G eorge's favourite spices.
IS   George ate the curry with delight. This type of food had always
9Q
LINGUISTICS
been his favourite. The dish was subtly flavou red, and in it he detected hints ofhis favourite spices, cumin and coriander,
The two versions are more or less She same as far as semantic content is concerned, and the syntax is fairly similar. Nevertheless, there is a lot of difference between the two. The second is both stylistically better, and more normal-sounding. The first appears to have been written sentence by sentence, without any attention to the overall eCfcct. In the second, various devices have been used in order to link the sentences together into a cohesive whole: after its first occurrence, the word curry has been replaced by alternative words this type of food, the dish, and by the pronoun it. Similarly, George has been replaced by het and in some places, the order of words has been altered so as to maintain the smooth connections, as when in it was brought to the front of its clause. In addition, some of the original sentences have been joined together.
Discourse analysis is the study which deals with this topic. It overlaps with stylistics, the study of linguistics and literature. Devices which maintain die smooth flow of communication are particularly important in written language, where there is no one available to clarify unclear points. However, many of these devices are also used in ordinary conversation. Consider two versions of the same dialogue:
A Edna: Someone ought to lock up Fred. Minnie: Fred is a disgrace.
Edna: Someone caught Fred peeping at the new lodger through the bathroom window.
Minnie; What is the name of the new lodger? Is the name of the new lodger Arabella or Annabel?
B Edna: Fred ought to be locked up. Minnie: That man's a disgrace.
Edna: He was caught peeping through the bathroom window at the new lodger.
Minnie: What's her name? Is it Arabella or Annabel?
'Ihe first version sounds stilted and odd, even though by itself, each sentence Is well-formed. The second version sounds far more like an ordinary conversation. It contains devices similar to those used in the piece of prose about George and his curry: after the first occurrence of Fred, the alternative phrase that man and the pronoun he was used. The third sentence has been changed into the passive, in order to keep Fred at the centre of attention. And so on. The overall result is thaL the whole
USING LANGUAGE
dialogue becomes linked together into a cohesive whole, something that people who know a language do automatically - though people learning a second language usually have to be taught this skill, as the devices used vary in their details from language to language.
Taking it in turns
Conversation is not just a case of linking sentences together. On a more basic level, it fits into a conventional framework. Consider the 'dialogue' below:
Mother: And how's my pretty little darling then? li2.by.Ugh...Ugh.
Motl ier. O what a nice bit of wind that was} You must be feeling better/ Baby: Goo, goo.
This brief snatch of 'conversation' illustrates one important fact about human speech: people take it in turns to talk. Even if one of the participants cannot speak, the other one pretends that the non-talker has taken their turn. But we can go further than simply noting the phenomenon of turn-taking. We can, in addition, describe how a typical conversation might proceed. The speakers are taking part in a social ritual partially prescribed by convention. In a dialogue, utterances very often occur in pairs, which are sometimes known as exchanges or adjacency pairs:
Question: WJiat's the time? Answer. Ten past three.
Greeting; Hi,Jo. Greeting: Why hallo Bill.
Offer: Would you like a cup of coffee? Acceptance: Yes, please.
Apology: I'm terribly sorry. Minimization: Please don't mention it.
Paired utterances are not, of course, inevitable, and triple utterances are also frequent:
Question: What's the time? Answer: Ten past three. Acknowledgement: Thanks,
- 90   - -
- 99 -
LINGUISTICS
USING LANGUAGE
Conversations, then, typically follow a predictable format, exchanges are selected from a number of commonly used types. The options chosen by a particular speaker on a particular occasion depend on the social situation.
--— Repairs-
Conversations do not necessarily run smoothly. People cannot always explain things properly. Or they make a mistake. Or the person they are talki ng to makes a mistake. Tliese minor breakdowns, if noticed, have to be 'repaired'. So-called repairs can give additional insights into tile way in which humans comprehend one another.
Repairs sometimes involve self-repair, when a speaker spontaneously notices a problem and solves it:
Couldyau hand me a spoon? A teaspoon, thatis. Morion arrived on Saturday — sorry, I mean Sunday,
Sometimes they involve other-repair, when someone is not quite sure about what has been said, or suspects that the other person has made a mistake:
/ assume you mean a teaspoon.
Did Marion realty arrive on Saturday?. Wasn't it Sunday?
However, humans do not usually confront one another directly, so other-initiated self-repnir is very common. In such cases, a listener mildly queries the speaker, who then repairs the original utterance:
Speaker A: Alan's taken a course in deep-sea diving.
Speaker B: Alan ? Has h e really?
Speaker A: Sorry, I don't mean Alan, I mean Alec,
As this example suggests, humans tend to be polite to one another, so politeness can radically affect the structure of conversations. Lotus consider this topic further.
- Politeness
Shut the door/
I wonder if you'd mind shutting the door? There's quite a draught in here.
-100—-
';, ■ If you waim-l someone to shut the door, you could En theory use any of the sentences above. But in practice, the first, a direct command, would uttered perhaps only to a young child. To anyone else, it would snem somewhat rude. This avoidance of directness is partly culturally based: *Why did that man look offended when I said, "Pass the salt"?' asked one puzzled visitor. She was even more bewildered when tnld that it would be belter to say: 'I wonder if you could possibly pass (he salt.1 Why such a fuss* she queried, about a small quantity of salt? But in spite of cultural variation, the idea that it is politer Lo say things indirectly may be universal.
Humans everywhere tend to be polite in similar ways, based on two basic social requirements: 'No criticism' and 'No interference'. Humans want to be approved of, and they do not want to be imposed upon. Consequently, anyone with social know-how will minimize criticism of others and will avoid interfering with their liberty, at least overtly.
These requirements of'No criticism' and 'No interference' have an effect on language. Any criticism or interference will be a social risk. Therefore speakers have to balance up the advantages and disadvantages of 'straight talking'. They must tot up the social distance between themselves and those they are talking1 to, the power relationship, the cultural norms, and make a decision.
Suppose a colleague was chinking too much whisky. The speaker could say:
Stop drinking} but would be more likely to say tactfully:
/ wonder i/wc should keep our heads dear for tomorrow's meeting? Oi they might even make a joke of it:
Even if everybody else goes bankrupt, the whisky manufacturers will
survive!
And of course, if offending a colleague was really too much of a risk, tliL' speaker could just have kept quieL
But suppose someone had an urgent request, and felt obliged to 1 impose on another person, what happens? There are various strate-
gies which are used to soothe the situation. For example, anyone imposing is often pessimistic:
I don't suppose you could lend me a pound, could you?
Or they might try to minimize the imposition:
-101-
LINGUISTICS
I won't keep you a minute, but... Or they might just apologise:
I'm terribly sorry to bother you, but...
The various strategies occur worldwide, but they are not necessarily found in every language. Each culture has its own preferred strategics. This type of study therefore overlaps with sociolinguistics, the topic of the next chapter.
QUESTIONS
1 What is pragmatics?
2 What   four   conversational   maxims   form   the cooperative principle?
3 Whatis speech act theory?
4 What are frames?
5 Explain two ways in which a person might, make a piece of prose more cohesive.
6 Explain what Is meant by adjacency pairs. Give two examples.
-102-