Skip to main content

9) Pragmatics and Grice's Cooperative Principal

We have so far considered how words are formed and classified, and how words are grouped to form phrases. We have considered as well how words are formed vocally through the manner and place of articulation. We have even explored the murky realms of semantics to consider the difficult issue of how all these factors combined to make a language capable of communicating meaning.

In our last entry in this series, we considered several facets of semantics, and touched briefly on the figure of Ferdinand de Saussure. Saussure recognised that language works in a variety of different ways, and is principally known for his analysis of the linguistic sign:

The other thing Saussure has become known for, is for his identification of two central branches of language – those of langue and parole. Langue refers to the systems of language, like the linguistic sign, morphemes, phonemes, grammar, etc.. Langue means the system of rules by which language works. Parole, on the other hand, refers to language in use – it is the actual speech acts, and manifestations of language in practical, concrete circumstances.

Initially, Saussure was more interested in the systems of language than in speech act theories – which were necessarily more flexible because they were so dependent on the behaviours of social human beings. However, many theorists since Saussure (like M. M. Bakhtin and J. R. Searle) have been far more interested in this point of language in use.

It is this idea of parole or language in use, which we are going to be considering here, and the theorist we are going to use for this exploration is Paul Grice.

The problem is that when we talk about systems of language (langue), we tend to assume that all that is needed in order to be able to communicate effectively is to ensure that people are jointly aware of those systems. For example, in terms of the linguistic sign Saussure used the example of a ‘tree’. Here, the signifier is the word itself – ‘tree’. What is being signified by the word though, is the idea of a tree itself:

The sign itself is used to point to a referent which is real and specific. For example, consider the sentence:

The tree in my back garden

Here, the sign ‘tree’ works exactly as we seen above. However, the sign itself is being used to point to a specific object (the referent) which can be clearly identified because it shares the same qualities as the signified in the sign:

The signified has to be generic. The signified is not a particular tree, but the idea of tree-ness. There are, though, enough similarities between this idea of tree-ness, and the object in my back garden, for me to be able to know that when you say ‘the tree in my back garden’ you mean the thing with a brown trunk, branches and leaves, and not the dustbin in the corner.

This idea of tree-ness is doubtless going to be different for each individual. When I say the word ‘tree’, an idea of a picture of a tree will pop up in your mind – but that picture is always going to be unique to you. The person sat next to you will have a slightly different picture of a tree in their mind:

However, as we have already seen with our exercise in drawing a car, although the image of the signified in our minds differs from person to person, there are always going to be generic similarities between them. This means that when somebody points to an object and says ‘look at that tree’, you do not need to have exactly the same picture of a tree in your mind to be able to match your image against the reality being described:

This is all well and good for proper nouns (nouns which refer to specific, concrete realities). However, what happens when we attempt to use the same mechanism for understanding abstract nouns (nouns where refer to abstract concepts). For example, consider the word ‘culture’. We know that the word ‘culture’ is the signifier, but what is the signified?

If we consider some of the ways in which the word ‘culture’ is used, we might begin to see some of the difficulties in identifying any particular system of langue which can effectively define it.

Consider each of the following sentences.  How would you define the meaning of the word ‘culture’ in each one?

“There are enormous cultural differences between Asia and America”

“She is such a cultured person”

“Pop music is often used by sub-cultures to assert their identity”

“There is a danger that mass culture may destroy the values of society”

“This module will examine eighteenth-century literary culture”

“This is an interesting cultural artefact”

“Disneyland fosters a distinctive culture based on certain values”

The truth is, the systems of langue only go so far to explaining how communication takes place. We explored some of these briefly last week when we discussed pragmatics, but we can see again here the importance of different contexts.

Consider the sentence:

Dogs must be carried

This sentence contains signs which are relatively straightforward. However, it can mean different things depending on the context in which it is made. For example, if you were to see this written on a sign next to an escalator, how would you interpret it? Would you assume that you are not allowed to travel on the escalator unless you are carrying a dog?


However, consider this alternative sentence:

Passports must be carried

If you were to see this sentence on a sign in an airport, would you interpret it any differently to the sentence about carrying dogs? If so, why? The structure of the sentence remains the same. The only thing that changes is the contexts.

If so far we have suggested that systems of language create problems, then the trouble is that parole does not necessarily make life any simpler because language in use is effected by more than simply the geography of the utterance. After all, as Geoffrey Finch says:

We also convey meaning through our bodies, by gesture, posture, and looks, that is, by non-verbal communication, and through our voices, by intonation and rhythm. All of these can have paralinguistic functions, in other words, they can run alongside the words contributing to the total meaning of the communication, either by reinforcing the word meaning of the communication, or, sometimes, contradicting (Finch, 2003. p. 38)

These paralinguistic functions are an accepted part of normal communication, and in fact communication would be very difficult without them.

To communicate the sentence ‘I want to talk to you’ in such a way as to imply that you are cross or angry with the person you are talking to, you would adopt an appropriate tone and appropriate body language:

However, to communicate ‘I want to talk to you’ in such a way as to imply that you are attracted to them and want to ask them on a date, you would apply a very different tone and body posture:

Get these logical structures the wrong way around, and communication breaks down (not to mention your relationships!). Again, Geoffrey Finch puts this nicely:

In order to comprehend what is said to us, then, we need to understand more than the literal sense of an utterance. We need to understand its force … The force of an utterance is the meaning it has in a particular situational setting or social context … One of the ways in which we interpret the force of utterance is by paying attention to // the intonation pattern of speakers. It’s often the case that we object to the manner in which something is said rather than just what is said (Finch, 2003. pp. 55-6).

Finch illustrates a simple sentence, the meaning of which can be easily changed through the force of its utterance:

I can’t drive there

If we change the stress of this sentence, we can change the implied meaning. For example:

I can’t drive there: Here, the emphasis on the word ‘I’ implies that although the speaker may not be able to driver there, somebody else might.

I can’t drive there: Here, the stress on ‘can’t’ makes the statement itself adamant. There is simply no way I will be able to drive there.

I can’t drive there: Here, the stress on ‘drive’ suggests that although driving might be out of the question, it might be possible to walk or go by public transport.

I can’t drive there: Here, the stress on ‘there’ implies that although it is out of the question to go to this particular place, they might be willing to go to another place instead.

Language, then, is composed of many different systems all working, or relating together – whether they be systems of sound (phonetics) or words (grammar) or meaning (semiotics), and whether they be linguistic or paralinguistic. The fact that there are so many openings for miscommunication throughout each of them is what is known as linguistic indeterminacy. It is something that, as linguists, you are going to have to get used to – just as Saussure did.

In fact, there are so many ways in which language could be crucially (even tragically) misinterpreted that Paul Grice began to realise that cooperation between linguistic participants is one of the most essential factors in any effective system of language. Whenever we engage in a conversation with somebody, we make an unspoken agreement to accord by the rules of a set of systems and to accommodate the patterns of language of the other participant.

Think about these questions:

Do you sometimes find that you modify your language to fit in better with those around you?  For example, do you speak more ‘properly’ in class than you do with your friends?

Have you ever modified the way you speak in order to emphasise difference from those around you?  For example, when you were younger did you ever speak in a way more like your friends and less like your parents deliberately to show you parents that you were independent of them?
If you answered ‘yes’ to either of these questions, then what you are recognising is the principal of accommodation, by which two speakers in language cooperate by modifying their language to increase the common basis for understanding. This, essentially, is the central principal of Grice’s theory – which Grice himself defines as this:

Make your 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, 1991. p. 26)

In other words, fit your language to the requirements of its contexts: Do not say more than you need, or less than is necessary; make sure what you say is relevant to conversation, and delivered in a way most likely to be understood by the other conversational participant; and only say what you believe to be true.

Grice broke this principal down into four maxims: The maxim of quantity, the maxim of relation, the maxim of manner and the maxim of quality.

The Maxim of Quantity:

As you can probably deduce from the name of this maxim, it has to do with the amount of information which is appropriate to any conversational exchange. For example, imagine you ask somebody:

You: Do you have any family?

They could answer by saying ‘yes’, and finish there. If they did this, however, you would surely feel as though the answer was inadequate to the question. In terms of formal logic (the purely logical deconstruction of the meaning of the words, as presented in that particular order), the answer ‘yes’ is an entirely fair one – but there is something more at work. There is a natural logic or a convention – a sub-textual or cultural understanding – which if understood informs us that the question is asking for more details.

This natural logic is, despite its name, not natural at all. Children are notoriously bad at recognising the conventions which require a certain amount of information. A grandparent who rarely sees their grandchild might ask them:

Grandparent: How is school?

To which the grandchild might answer ‘s'alright’, and then go off to play a computer game. The grandchild does not recognise or appreciate that the question is not requesting a positive or negative response, but is the consequence of a desire to engage in the day-to-day experiences of the grandchild as an expression of the love of the grandparent.

Equally though, a conversation can provide too much information. If you imagine somebody working on the tills in a supermarket, they may well ask a customer:

Cashier: How are you today?

The likelihood is that your question is merely a formal one which is made in order to be polite and welcoming to the customer. However, the customer may respond:

Customer:  Actually my bunions have been playing up, and my cousin had a blazing row with his girlfriend last night, so she comes round to my house crying, and he calls demanding to know if she’s there, and it was all bit chaotic and stressful.  On the plus side though, I won twenty quid on the lottery and we got a letter from the management the other day saying we were all getting a 2% pay rise.  I says to my friend Jane, ‘2% isn’t much, but at least its not a deduction – should be enough for a drink or two, eh?’

In this circumstance, too much information has been given. The contexts of the conversation determine that only a limited about of information is required for the sake of politeness.

Both too much information and too little information can be symptoms of a limited linguistic competence – perhaps because of age, or perhaps because of differences in cultures where such competencies differ. However, the maxim of quantity can as well be flouted for deliberate effect:

  • Too much information can be a deliberate ploy to avoid addressing an issue which is more problematic. Politicians commonly do this in television interviews – where an awkward question is raised, they spend an inordinate amount of time answering some minor detail of it in the hope that time will run out before they are forced to address the main issue.

  • Too little information can be a deliberate statement about the relationship between conversational participants. ‘I know you want more information, but frankly I do not want to talk to you’.

In either case, the maxim of quantity forms the basic principal on which communication takes place – and as such if Grice’s first maxim in his cooperative principal.

The Maxim of Relation

This maxim recognises that there is an implicit agreement between conversational participants to make their utterances relevant to the contexts and to the conversation itself. As one conversational participant is talking, we may be forming in our minds how we might respond to their utterance in a way which is relevant to it:

Participant 1: I had a terrible flu the other week.

Participant 2: Yes, I had a friend who had the flu last week too. Must be something going around.

If you were participant 1, and participant 2 responded to your statement with:

Participant 2: Arsenal won at the weekend.

Then you would doubtless think that something strange was happening. In fact, so sure are we that participants in a conversational exchange will ensure their contributions are relevant, even if participant 2 did respond with ‘Arsenal won at the weekend’, we would be more likely to assume there must be some hidden connection between your flu and the football result.

Participant 1: I had a terrible flu the other week.

Participant 2: Arsenal won at the weekend.

Participant 1: Yeah, they did. And you’re right – things do get better in the end.

Again, this maxim can be flouted (deviated from deliberately). It can be used as a strategy of avoidance:

Participant 1: Did you drink my tea?

Participant 2: I’m just popping out to the shed – back in a moment.

It can also be flouted for comic effect – as in this example from Finch (2003):

Lecturer: You should have been here this morning.

Student: Why? What happened?

The lecturer’s statement can be taken as either an exclamation (‘Wow! You should have been here this morning! You’d never believe what happened!’) or as a reproof (‘You were supposed to be here this morning. What is your excuse?’). The contexts tell us which is the correct interpretation – but here the student has deliberately chosen to interpret it wrongly in order to avoid having to answer the more pertinent question.

The Maxim of Manner

Imagine you are asking somebody for directions. You would expect the person giving those directions to organise them in a sequential order:

Example 1:

• First, you go to the roundabout and take the third exit.

• Then, you will come to a set of traffic lights. Take a left there.

• Then, take the third turning on the right, and head round the corner.

• Your destination is opposite the pub you will see just ahead of you.

What you would not expect, is something more like:

Example 2:

• There is a pub and your destination is opposite it.

• Before that you need to turn right at the traffic lights.

• There is a corner – follow it round.

• It’s third exit on the roundabout

• Third turning on the right.

In the case of example 2, you would be very unlikely to reach your destination, because the information has not been given to you in the correct order. The mixed-up order means that you will almost certainly misunderstand the communication – probably by taking the third turning on the right after the roundabout instead of after the traffic lights.

Consider the following two sentences, and notice how in terms of clause structure the meaning is changed radically if the order is not maintained:

• Laura ran up to the top floor and jumped.

• Laura jumped and ran up to the top floor.

Most conversations do not rely so rigidly to a specific order, but there is always somehow a sense that an order underpins any conversational exchange, and that we follow that order in order to make our contribution as clearly comprehensible as possible. When we are recounting how our day has gone, we tend to do so in a chronological order. When responding to a question about what we have been up to, we tend to list things in a categorical order. For example:

Participant 1: Did you like the film?

Participant 2: I liked some things. The acting was very good, and the scenery was amazing. Thought the plot was a bit thin though.

In this instance, participant 2’s response of ‘I liked some things’ implies (by natural logic rather than formal logic) that there are certain things that they liked and certain things that they did not like. They then proceed to describe those things in a categorical and ordered fashion.

As with all the other maxims, the maxim of manner can be flouted. The violation of this maxim communicates something. A certain disorderliness of conversation is often a symptom of emotional distress or anger, and as the cartoon above indicates some artists have used it as a means of challenging the preconceptions of order which underpin our view of the world. In a strange way, these violations only serve to emphasise more strongly just how central the idea of order is to the ways in which we use language.

The Maxim of Quality

It is always assumed that when we engage in a language exchange then what we are communicating is what we believe to be true. There is no obvious purpose in asking the question ‘what did you think of the film?’ if you have no expectation that the other participant is going to tell the truth.

All conversation is based on the initial premise that participants cooperative in ensuring their contributions are true. A conversation which is composed of a succession on non-truths is entirely un-productive and fails to communicate anything.

For example, consider the following conversation:

Participant 1: Did you see the Harry Potter film?

Participant 2: Yes.

Participant 1: Who was your favourite character?

Participant 2: Luke Skywalker.

Participant 1: What was your favourite bit?

Participant 2: When he rubbed the lamp and the genie appeared.

Participant 1: Oh. My favourite bit was when the dinosaur jumped in at the end and ate the bad guy.

The conversation itself serves no purpose. If the purpose of language is to communicate meaning, then the fact that the maxim of quality has here been flouted means that it has failed in that purpose.

Actually, there is a problem here. We have already seen that all of the maxims can be flouted for deliberate or specific effect, but of all maxims this one is perhaps the most consistently flouted in everyday use. To flout the maxim of quality is to lie, and this is something we do all the time, whether for good reasons or for bad.

Imagine you have arrived in College and have met with a number of students. You have been asked the following questions:

• Did you have a good weekend?

• How are you today?

• I bought this new dress at the weekend – do you think I look ok in it?

• Oo! You have a Wispa bar! Do you mind if I have a bite?

If you are brutally honest, what are the chances that your responses to these questions are going to be entirely honest? Even if you were the one asking if you looked ok in your new dress, you may well be asking the question because you want re-enforcement or encouragement – not because you want to be confronted with a cold hard reality.

To flout the maxim of quantity in relation to these questions is not to render language abortive. It is to fulfil some other maxim in which determines that ‘white lies’ are a necessary expectation in any cooperative communication. This other maxim has been labelled by Geoffrey Leech the maxim of politeness (1983) – the function of which Finch describes well:

The politeness principal enjoins people to be tactful and polite unless there is a specific reason not to be (2003, p. 63).


• Finch, G. (2001). How to Study Linguistics, 2nd edn.. Basingstoke: Palgrave.

• Finch, G. (2003). Word of Mouth: A New Introduction to Language and Communication. Basingstoke: Palgrave.

• Grice, H. P. (1975) Logic and Conversation. In; Cole, P. and Morgan, J. L., eds., Syntax and Semantics, Vol III. New York: Academic Press, pp. 41-58.

• Grice, H. P. (1978) Further Notes on Logic and Conversation. In; Cole, P., ed., Syntax and Semantics, Vol IX. New York: Academic Press, pp. 113-127.

• Grice, P. (1991). Studies in the Way of Words. Harvard: Harvard

• Leech, G. (1983). Principles of Pragmatics. London: Longman.

Popular posts from this blog

2) Introduction to morphemes

So does language begin with words?

No. Language begins with sounds. It is important to understand this first and foremost. We have already raised this point, but it is worth raising again – language begins with sounds!

If I appear to be emphasizing this with a rather bizarre desperation, it is because it would be easy to think that since we are beginning our exploration of language and linguistics with words that this is where language begins. When you think about it logically though, all words are composed of various sounds grouped together. The word ‘cat’ is composed of three distinct sounds - /c/, /a/ and /t/.

So why aren’t we starting with looking at how sounds create language?

Well, in the not-too-distant past, when European football used to be free on the telly, Manchester United or Arsenal would jet off to Spain for a titanic contest with Barcelona. When the commentators referred to Barcelona, they would pronounce it ‘Bar-se-low-nah’ (bɑ:sɜ:ləʊnæ). After a few years th…

6) Places and Manners of Articulation

Place of Articulation
The place of articulation refers to “the point in the vocal tract where the speech organs restrict the passage of air in some way so pro¬ducing distinctive speech sounds” (Finch, 1999). As with manner of articulation, places of articulation are more frequently used to describe consonants than vowels. The following are the principal terms used in linguistics to describe these:

Bilabial.Sounds formed by both lips coming together” (Finch, 1999).Examples include /b/, /p/ and /m/.

'It's owned by Elsevier': Why this is relevant when choosing referencing software

At my University we are currently discussing how to provide support for software that can help students and staff manage their references and sources.  There are of course many different options available on the market - some free, and some not.  During discussions I have made no secret of my preference for Zotero - which I believe offers the most intuitive and comprehensive functionality.  To this end, I have done some showcases of Zotero for various academics - which appear to illicit one of three responses from them:

Oh, brave new world that has such software in it!  I had no idea - and I want it now!We already use it.  Have been for years.  So why are you telling us about it now?But don't we already have Mendeley in our official software catalogue?

I fully expected the first response - but was surprised at the number of people who came back with the second and third.  It is really rather nice to be able to tell academics who fight tirelessly each year to teach academic referen…