Scarica Text Analysis and Discourse: Understanding Text-Context Relations e più Sintesi del corso in PDF di Linguistica Inglese solo su Docsity!
Discourse Analysis
H.G. Widdowson
1. Language in use
What is a TEXT? A text is an actual use of language; it is produced for a communicative purpose.
SENTENCE →abstract unit of linguistic analysis (words combine to form this grammatical unit). We would generally need to know the language a text is in to be able to interpret it, but we also may know what the language means but still not understand what is meant by its use in a particular text. ex. “KEEP OFF THE GRASS” →Knowing the meaning is not the same as knowing what this public notice refers to. The definite article the signals that what is being referred to is a matter of shared knowledge. We establish reference by relating the text to the context in which it is located. But does it refer to the particular patch where the notice is placed or to other patches nearby as well? Or to the whole park? →We make assumptions on the basis of what we know. In other words, we relate the text not only to the actual situational context in which we find it, but to the abstract cultural context of what we know to be conventional. And by relating text to context we infer not only what the notice refers to, but also what its purpose is. We cannot make sense out of the text simply by focusing on the language itself.
Text and discourse
Not all texts extend beyond the sentence, but many of them do: travel guides, newspaper articles, poems, and so on. Some have an obvious utility function but others are meant to serve a range of different social purposes : to give information, express a point of view, provide entertainment, and so on (ex. a travel guide may provide information + it also promotes the attractions it describes). People produce texts to get a message across, to express ideas and beliefs, to explain something, to get other people to do certain things or to think in a certain way, and so on. We can refer to this complex of communicative purposes as the discourse that underlies the text and motivates its production →At the receiving end readers/listeners have to make meaning out of the text to make it a communicative reality: they have to interpret the text as a discourse that makes sense to them. In this view, texts do not contain meaning, but are used to mediate it across discourses. No matter how explicitly we think we have textualized what we want to say, there is always the possibility that it will be interpreted otherwise. So the term discourse refers both to what a text produces meant by a text and what a text means to the receiver.
Spoken and written text
Texts are the perceptible traces of the process of mediating a message. In conversation these traces disappear as soon as they are produced (they can be recorded though). Written text is typically designed and recorded unilaterally in the act of production by one of the participants, the writer, as a completed expression of the intended message. The text is then taken up and interpreted as a separate process. The mediation is delayed and this makes a convergence between intention and interpretation more difficult to achieve. Moreover, when people communicate they do not only produce linguistic texts: in speech, they make use of not only of language but of paralanguage (tones of voice, pauses, stress, facial
expression or gesture); in written communication a text is given a particular shape by choice of typefaces, arrangement on a page + it may be multimodal if related to other modes of communication (pictures, diagrams). When we use language we do not just present the meanings that are encoded in it ( semantic meanings : those described in dictionaries and grammar books), we exploit them as a potential resource for making meaning of our own (communicative intentions, the semantic meaning serve a pragmatic purpose ).
Illustration from a crowded train
Let us suppose that you overhear the following utterance in a conversation between 2 people in a crowded train: “ He has put it in a safe place and it will not be found ”.
- Grammatically it is a complete and well formed sentence. Present perfect ( has put ) + passive ( will … be found ). Agreement between the singular pronoun he and the verb. Correct word order.
- However, who is 'he' and what is 'it'? These pronouns have an established denotation: he encodes the semantic features of singular and inanimate. But although a knowledge of these denotations, it does not tell you who or what is being referred to → pragmatically elusive. This utterance is only a fragment of conversation, one piece of the text. The 2 people have established the context of shared knowledge. The linguistic trace of their discourse is not enough to interpret what they mean by what they say.
Conclusion
To summarize. When people communicate with each other, they draw on the semantic resources encoded in their language to key into a context they assume to be shared so as to enact a discourse, that is, to get their intended message across to some second person party. The linguistic trace of this process is the text. In the case of conversation, the typically disappears once it has served its purpose. In the case of writing the text is unilaterally produced and remains as a permanent record. But it is still only a discourse trace, and what is meant by it has to be inferred by interpretation. In the normal circumstances of use we only pay attention to text in order to realize its discourse function. But although we normally experience text as part of the discourse process, it is perfectly possible to focus on the text alone →e.g. proofreaders scrutinize a piece of writing to identify typographical errors, or wordings that do not conform to established code conventions.
2. Communication
Grammar and communication
Meaning is formally encoded in lexis and grammar. The description of such properties can be said to be an account of what people know of their language, of their linguistic competence. It was this competence that enabled us to describe the grammatical features of the utterance in the train in the preceding chapter. Knowing the grammar of a language is not the same as describe it-that is the business of the grammarian. But the point is that anybody competent in English would recognize that the utterance conforms to the encoding conventions of the standard language, that it exemplifies a well formed sentence in English, which it would not do if it had taken the form, for example: “ They has it in a safe place put and it will not to find”. So on hearing this remark in the train, one judgement we can make on the basis of our linguistic competence is whether it is grammatically and lexically possible or not, that is to say, in accordance with the encoding conventions of the language.
Four aspects of communicative competence
What is possible in the language, in the sense of what can be encoded in it, and what is appropriate in its use are two factors that are included in the well-known account of communicative competence proposed by the American scholar Dell Hymes. He includes two other factors in his account:
- Feasibility → Our competence in a language includes our recognition of the extent to which a particular expression can be readily processed. A piece of language might be a grammatically and lexically possible encoding in the language, but difficult to decipher. For instance: “ This is the corn the rat the cat the dog chased killed ate ”. Though this can be taken as a possible structure in English, since it conforms to encoding conventions, it will also be recognized as relatively unfeasible. The point about feasibility can also be illustrated by kinds of structure that are by no means uncommon, namely those we recognize as ambiguous.
- Ambiguity occurs when two distinct structures converge into one single sequence of sentence constituents, as in the example: - Two sisters were reunited after eight years at a checkout counter. - The stolen painting was found by a tree. Perfectly possible expressions in a language may be relatively lacking in feasibility. Conversely, an expression may be entirely feasible in that it is easy to process, but not be grammatically well formed and so not possible with regard to the encoding conventions of the language. The general point to be made is that there is no direct correspondence between feasibility and grammatical well-formedness, any more than there is between grammatical well-formedness and appropriateness to context. The extent to which expression is possible, processible, and appropriate are three separate things. We have a fourth factor to consider. Hymes suggests that to be communicatively competent in a language we must be capable not only of recognizing how far an instance of it is possible, feasible, and appropriate but also how far it is actually performed or produced, the extent of its frequency of occurrence. The language code provides the basic resources for communication, but we do not, of course, make use of these resources in equal measure. For example: the words walk and perambulate both will be listed in a dictionary of English, but we know that walk is of relatively common usage, and perambulate relatively rare. In this case, we are consciously aware of relative frequency. But in innumerable other cases, we are not. Thus we may know perfectly well that the words big and large are semantically synonymous, as are the words little and small, but, if asked, we would not be able to say with any certainty which word in each pair is more frequently used. Nor what other words they tend to keep company with. In the usage of any language there is a tendency for words and structures to co-occur in patterns of relative probability. Some co-occurrences (or collocations) are fixed and familiar, as in idiomatic phrases like to and fro, by and large, by hook or by crook , and so on. Some we know as proverbs: Too many cooks spoil the broth , and so on. By means of the computer, linguists can now provide details of the frequency and co-occurrence of words and structures which are simply not accessible to intuition. But since our usage is informed by these frequencies and patterns of co-occurrence, we must know them in some other subliminal sense and this must be a covert part of our competence in the language. We have procedural knowledge of them, and what linguistic description can now do is to make this knowledge overt and declarative. Hymes identifies the possible, the feasible, the appropriate, and the performed as four distinct aspects of communicative competence, and in the analysis of language in use we can focus attention on any one of them.
Conclusion
When we use language in the normal circumstances of everyday life, all of these four factors come into play in complex and interdependent ways. But the determining factor is that of contextual appropriateness. It is the discourse we want to achieve that regulates how we draw on the encoded resources of the language to make a text. The language that we process is not designed to demonstrate what is possible or feasible commonly occurring but to realize a discourse purpose.
3. Context
Conditions of language use
We experience language not as something separate but as an intrinsic part of our everyday reality. We put our linguistic knowledge to use to give shape to our internal thoughts and to give external expression to our communicative purposes. We only produce language when we have the occasion to use it, and the occasions for use occur in the changing contexts of our daily life.
Context and shared knowledge
These contexts can be thought of as situations in which we find ourselves, the actual circumstances of time and place (ex. The chalk is over there/I like the look of that/Is that the time? ). People make sense of what is said by making a connection between the language and the physical context of utterance. The language serves to point out something which is present in the perceived environment, and the listener can only understand what the speaker means by the utterance by making the necessary connection. When such utterances are isolated from this shared situation, they have nothing to point to, and so lose their point. But being present in the same physical situation is not a guarantee that listeners will make the required connection: they may still fail to identify just what is being indicated ( Over there.. where do you mean?, I like the look of that.. the look of what? ). Context, then, is an abstract representation of a state of affairs. This may be constructed directly from the immediate concrete situation, but it can also be entirely independent of such situational factors. The context is the common knowledge of the two people concerned, which will have been established in their previous conversation. Of course, the situation can be contextually relevant. But the point is that the situation is made contextually relevant. It has no necessary relevance of its own.
Text-activated context
Context, then, is not what is perceived in a particular situation, but what is conceived as relevant, and situational factors may have no relevance at all. This is particularly clear with written communication. There are exceptions, but typically where and when a written text is read is quite different from where and when it was actually produced. Although there can be no appeal to a common situation, however, there must be an appeal to a common context of shared knowledge or otherwise no communication can take place at all. Some of this context will be created by means of the text itself. The following, for example, is the opening paragraph of a magazine article: “ With 300 million native speakers scattered across 20 countries, Arabic is the world's sixth largest language. Yet British ignorance of and indifference to the Arab world remains startling: of 737 postgraduate students in Islamic or Middle Eastern studies funded by the Economic and Social Research Council last year, 12 were British nationals. ” →In the opening sentence, the writer provides information to establish the context of shared knowledge. Notice, however, that she assumes that this will activate knowledge that is not made explicit in the text: that the Arab world is to be identified with Islam and the Middle East, for example. Unless such knowledge is activated, the text makes no sense.
Conclusion
The contexts that texts, whether spoken or written, are designed to key into are constructs of reality as conceived by particular groups of people, representations of what they know of the world and how they think about it. Although, as we have seen, some of the knowledge that the text producer assumes to be shared is of particular things, events, persons, either within the immediate situation of utterance or not, these particulars are typically related to more general schematic structures of knowledge. If readers cannot ratify these assumptions, they will be at a loss to know what discourse the writers intend to mean by their texts in terms of their reference, force, and effect. Communication is not simply a matter of bringing kinds of knowledge into correspondence, but of bringing them into a degree of convergence, and this may call for quite complex negotiation.
4. Schematic conventions
Context and situation
Context is an abstract representation, a mental construct. It may be abstracted from the immediate situation of utterance, as when reference is made to something that is directly perceptible by both parties in an interaction. So if somebody asks me to close the door, for example, I can readily infer that what is being referred to is a particular door in the room we are in. This is a case of what is called deixis → the pointing out of something immediately and perceptibly present in the situation of utterance: that door there, this door here. But context is obviously not confined to what is situationally present in the here and now. The language we produce is part of the continuity of our individual and social lives, and so always related to the context in our heads of what we know and believe. This context in the head is a schematic structures of knowledge.
The concept of the schema
A schema is a construct of familiar knowledge. The psychologist Bartlett introduced the concept over 70 years ago. In one of his experiments a group of British students were asked to read a North American Indian story called The War of the Ghosts and then rewrite it from memory as accurately as possible. What they did in their versions of the story was to change the events so that they corresponded more closely with their own conventional and customary reality, very different from that represented in the original. In other words, the discourse they derived from the text was one that suited their preconceived schematic expectations. The text that these students were asked to read and retell came, of course, from a culture very different from their own →cross-cultural misunderstanding. Schematic assumptions are socially shared by a particular community; they are indeed cultural constructs. Schematically shared among smaller groups of people. The process of making sense by taking schematic bearings applies to the interpretation of all texts. Everything new has to be related to what is given.
Frames of reference
What happens in text interpretation is that the language triggers off the recall of some familiar state of affairs, some schema or other, and this sets up an expectation of what is to follow. → An example is the first sentence of an article from a recent edition of a British news magazine: “In the past it took a disaster to bring the Olympics to London”. The language here activates schematic knowledge about the Olympic games and how different cities bid to host them every four years. So it indicates a frame of reference and at the same time
projects the reader's attention forward to what is to come next. This process of keying the language into an appropriate frame of reference to make sense of text comes so naturally to us that we take it for granted, and it is easy to suppose that the meaning is actually in the text itself and not derived from it by this kind of schematic inference. Notice that once a frame of reference is established, the use of definite articles becomes appropriate ( 'the hymns', 'the prayers' , and so on) because the phases refer to what is common schematic knowledge. Of course readers can be mistaken and invoke the wrong frame of reference. They can be deliberately misled and their expectations thwarted. → e.g. “Rocky slowly got up from the mat, planning his escape. He hesitated a moment and thought. Things were not going well. What bothered him most was being held, especially since the charge against him had been weak. He considered his present situation. The lock that held him was strong, but he thought he could break it.” What is being described here? The text is ambiguous in that it admits of two interpretations. If we take 'Rocky' as referring to a prisoner, then what we have here are his thoughts about escaping from his cell. We could, however, take 'Rocky' to refer not to a prisoner but to a wrestler and then what we have are his thoughts about how to get the better of his opponent. The ambiguity of the text here is sustained by the use of words which call up two possible and competing frames of reference. Word forms frequently encode more than one semantic meaning and these are recorded in a dictionary. When put to pragmatic use, words function as schema activators, but of course they are not usually on their own: they connect with other words in a text, and the connection will usually have the effect of directing the reader's attention to just one schema by elimination. In the case of ambiguity, readers do not know what to eliminate and they have to keep more than one possible schema in mind at one and the same time. Of course, in reference to the text about Rocky, if you know nothing about wrestling, there will be no wrestling schema to invoke, and words like escape and lock are then likely to bring only prisons to mind. Schemata are representations in the mind of what is familiar or customary. But all this is relative: what is familiar to one group of people may be unknown to another, and customs vary across communities. So what such schemata represent are culturally different ways of ordering the world, different versions of social reality.
Frames of reference
These schemata are cultural, taken-for-granted constructs, and they become so firmly entrenched in our consciousness that we often find it difficult to envisage any alternative ways of thinking. When we encounter a text that does not fit our culturally schematized world, we naturally find it hard to make sense of it. → e.g. “A man was taking a walk with his son one day and as they were crossing the road, a car came round a corner unexpectedly and hit the boy, injuring him badly. An ambulance was called for and the boy was taken to the nearest hospital and into the operating theatre. On seeing the boy, the surgeon suddenly let out a cry of horror: 'My God this is my son!'” Anybody who reads this text for the first time may well be mystified by it. Suppose the surgeon is a woman. As soon as that possibility is suggested, then the mystery disappears. The idea that the surgeon is a woman does not (for many people at least) come immediately to mind is because the mind is naturally inclined to interpret things by relating them to what is schematically established as normal and customary. A reader with different schematic expectations, someone living in a society where it is common for surgeons to be women, would not find this text mysterious in the least.
Genres
The interpersonal schemata we have been looking at so far are those which find expression in the regular routines of everyday spoken interactions. There are also interpersonal schemata which are much more extensive. So it is that we recognize certain stretches of interaction as speech events of specific kinds, or genres, and we have names to attach to them: meeting, interview, cross-examination, debate , and so on. → e.g. A committee meeting of a formal kind, for example, follows a certain agenda which specifies the order of business, with a chairperson who sees to it that the order is respected, and who controls how discussion proceeds by nominating those who are to take turns at talk, even requiring them to address what they say to the chair rather than to other people present. Not all meetings, and certainly not all other kinds of genre, are as ritualistic or schematically rigid as this, of course, but there will always be a set of assumptions and expectations about what behaviour is appropriate to the genre, about what is the 'done thing' and what is not. The kinds of genre (meeting, interview debate) are all speech events which involve turn taking. There are others which do not: speeches, for example, sermons (in one tradition at least), which consist of only one, long (sometimes all too long) turn. Speeches and sermons are examples of speech events which are single-turn genres. The most obvious examples of single-turn genres, however, are those of written language use. Although there are times when the interactants overtly participate in written communication (in an exchange of notes, or email messages). If I open up my newspaper to read a football report, or look up a menu in a cookery book, I will have some idea of what to expect because I know how they are typically written. Or, if I venture into writing myself and decide to offer an article to a journal, reporting on a piece of research, I know that I will be expected to structure it in conformity to the genre which has been established by the discourse community of scholars in this area as appropriate for this kind of written communication.
Conclusion
Genre conventions are rigidly fixed and always adhered to: they are naturally subject to variation and change because there will always be some room for individual manoeuvre. People will have schematic knowledge of what is typical of particular genres and this will prime their expectations. But these expectations may need to be subsequently adjusted.
5. Co-textual relations
Information structure
One of the things we do when we use language is to formulate a proposition, to make reference to some state of affairs. Let us suppose that we want to express a proposition about a certain event, for example the actions of the police in dispersing the crowd. English allows for the possibility of expressing our proposition in different ways: “The police dispersed the demonstrators” or “The demonstrators were dispersed by the police”. If we think of these as linguistic forms, we recognize the first as an active and the second as a passive sentence with the police being the subject in the first case and the demonstrators in the second. But if we think of these expressions as utterances, they are textual variants, different ways of distributing the propositional information. In terms of textual structure, the first piece of information, which here takes the form of a subject noun is said to be the theme, and the rest of the utterance the rheme.
But how would we order the information in the next utterance?
- We could follow the phrase, same pattern and start with the same theme (T): “The police (T) dispersed the demonstrators. Some of the law officers (T) …”
- Or we might reverse the order and thematize the previous rheme: “The police (T) dispersed the demonstrators (R). The banners they were carrying (R → T) … “
- Or we could, of course, introduce a different theme altogether: “The police (T1) dispersed the crowd. The motorcade (T2) …” It needs to be noticed, however, that whether a theme is different or not is by no means easy to determine. It may be a matter of simply identifying a semantic connection (recognizing police and law officers as synonymous →1), but schematic know ledge may also be involved. There is, for example, no semantic link between the words demonstrator and banner (2): to recognize the thematic connection here, you need to invoke what you know about demonstrations. The alternative thematizations we have been considering are brought about by changes in grammatical structure, and the use of active and passive forms is only one way of signalling theme and rheme. There are other possibilities, for example the use of sentences like: “What the police did(T) was to disperse the demonstrators.” But thematization can also be brought about by changing the sequence of constituents in a sentence without affecting its structure, as in: “The police(T) dispersed the demonstrators early in the day.” →“Early in the day(T), the police dispersed the demonstrators.” Theme and rheme can be easily identified as a feature of text, but what is it that motivates the use of one textual variant rather than another? What communicative intention lies behind them' What might their discourse function be? An item of information might be given the status of theme because the first-person text producer, P1, assumes it to be given, that is to say, already known by the second-person receiver, P2. In this case, the theme simply confirms common knowledge and sets the scene for the new information to be provided in the rheme. OR. Conversely, the theme could signal the main topic that P1 wants to talk about, with the rheme representing comment on that topic. It is clear that thematization plays a crucial role in organizing information in a text.
Text linkage
Theme/rheme assignment is a general way of organizing information and carrying reference over from one proposition to the next. But the linking of theme and rheme across parts of text depends on the identification of other more specific and small-scale connections to establish text continuity. As communication takes place, in speech or writing, what is said at a particular point naturally makes reference to what has been said before and a context is created in the mind and signalled in the text in the process of its production. The pronoun acts as a pro-form , that is to say, it stands in for the fuller expression that precedes it. This kind of text-internal, or co-textual connection is known as anaphora.
Anaphora and pro-forms
Take the pronoun he. This encodes the semantic features of singular and masculine and it works anaphorically only if these features can be traced back to realize a reference in common with a previous mention; while its encodes the semantic features of singular and non-human.
Communication generally operates on a least effort principle and we only use as much language as we need to make the required contextual connection. The problem is always to know how to regulate the degree of co-textual explicitness by the judicial choice of pro-form. How much do writers need to spell out cohesive links, how far can they count on readers making sense of the text without them?
Cohesion and coherence
Cohesive devices are only aids to understanding and can only be effective to the extent that they enable readers (or listeners) to construct meaning that makes contextual sense to them, in other words to the extent that the cohesion in the text enables them to derive a coherent discourse from it. It follows from this that it is possible for a text to be cohesive but incoherent (es gatto Nuccorini?) How far you can make coherent sense of a text depends, then, on how far you can relate it to a frame of reference. A frame of reference is what was referred to in the preceding chapter as an ideational schema. A passage can also be cohesive as text but lack coherence as discourse because it does not key the reader into any familiar schema of an interpersonal kind. Consider the following: “We spent our holidays in Romania. This is a country where grapes are grown. They are a kind of fruit. So are bananas. Fruit contains vitamins, and these are essential for a healthy life. So is regular exercise. Jogging is good for you. We do it every day ...” There is no shortage here of cohesive links across the different parts of the text. They is a pro-form that links to grapes, so is a pro-form that serves as an anaphoric token of both a kind of fruit and essential for a healthy life, jogging makes a semantic connection with the preceding exercise, and so on. But cohesive though it is, it is a very odd kind of text, and it is difficult to make any coherent sense of it. Unlike the previous passage, it is not that we cannot key the text into a familiar frame of reference, but rather that the frame of reference keeps shifting. The problem here is not so much that we cannot understand what is being said but that we cannot see the point of saying it. In other words, we do not recognize its purpose because it does not correspond with the communicative conventions of any genre we are familiar with. And so we find it incoherent as discourse.
Conclusion
No matter how cohesive a text may be in terms of internal co-textual links that can be identified, the extent to which it is interpreted as coherent discourse will always depend on how far it can be related externally to contextual realities, to the ideational and interpersonal schemata that readers are familiar with in the particular socio-cultural world they live in.
6. The negotiation of meaning
Systemic and schematic knowledge
The process of communication involves the engagement of two kinds of knowledge:
- schematic knowledge →people make sense of text, spoken or written, by relating it to what they know of the world they live in, the ideational and interpersonal schemata that represent the customary and conventional ways in which their socio-cultural reality is structured.
- systemic knowledge →knowledge of what is semantically encoded in the language itself in order for this schematic knowledge to be pragmatically activated as appropriate. Thus, a reader of a text has to know that particular pro-forms have a semantic relationship with other expressions in a text in order to recognize them as cohesively linked. It is important to note, however, that neither kind of knowledge is fixed. They provide us with a necessary starting point in making meaning, but as communication proceeds they are subject to
modification. So schematic knowledge projects provisional expectations which are subject to on- line revision as we proceed in deriving discourse from text. Although the schemata that we are familiar with will dispose us to interpret a text along certain lines, they do not therefore determine our interpretation.
Communicative convergence
For communication is always a matter of negotiating some kind of common agreement between the parties in an interaction. The first-person party, the sender (P1), formulates a message by drawing on systemic and schematic knowledge and the second-person party, the receiver (P2), brings similar knowledge to bear in interpretation. Communication is effective to the extent that there is some convergence between the two. Problems might arise if P1 uses items of language outside P2's competence, or refers to an ideational framework that P2 does not know about, or follows interpersonal conventions that P2 is unfamiliar with. Such problems can be resolved by negotiating meaning 'on-line': P2 can ask for clarification, or elicit additional information, or let the problem pass in the hope that it will get sorted out as the conversation develops. P1, if sensitive to the problem, may try to resolve it by subsequently elaborating on the message, or reformulating it in different terms. When, as in written language use, there is no possibility of such reciprocal on-line adjustment, P has to somehow anticipate what problems P2 might have. Communication, then, is a matter of the parties concerned negotiating a measure of convergence. But it is important to note that the degree to which the parties actually converge does not just depend on how far they are able to do so; it also depends, crucially, on how far they want or need to do so. So the degree of convergence we seek to achieve is regulated by the purpose of our communication. Convergence is always only partial. Mutual understanding is never, and never can be, perfect but only approximate to purpose.
Negotiating convergence
For communication to take place, then, PI and P2 have also to be parties to an agreement to co- operate in negotiating a convergence, a meeting of minds, a mutual understanding, whereby meaning is achieved as required by their purpose in communicating. There are different kinds of pragmatic meaning that are to be negotiated in communication, and the two parties may negotiate one successfully but not another. So P2 may understand what PI intends to refer to by saying something, but not what illocutionary force is intended, or may recognize the intended force but not grasp what effect it is meant to have.
The co-operative principle
The philosopher Paul Grice has proposed that when people converse they tacitly subscribe to what he calls the co-operative principle. Grice suggests that the co-operative principle can be expressed in terms of 4 maxims that parties in an interaction will subscribe to.
1. The quantity maxim
“Do not provide more, or less, information than is necessary.” There is no need to provide information by means of language if it is already common knowledge. If the P1 underestimates how much context is shared and so over-textualize by producing too much language, then what they say will be heard or read as pointlessly wordy, or verbose. If, on the other hand, they overestimate the extent of shared contextual knowledge, and so under- textualize, then what they say will be heard or read as obscure. Notice again that contextual knowledge includes a knowledge of schematic conventions. So
utterance has a significance over and above its apparent meaning. Compliance with this maxim is again, as with the others, regulated by convention.
4. The maxim of manner → “be clear, avoid ambiguity and obscurity.” This maxim has to do with what Hymes refers to as feasibility. Unintentional violations of this maxim can have comical consequences, as in the following examples of ambiguous newspaper headlines: DRUNK GETS NINE MONTHS IN VIOLIN CASE / RED TAPE HOLDS UP NEW BRIDGE By convention, the quantity maxim is appropriately applied in newspaper headlines to concentrate as much information into as few words as possible. This can result in expressions which are, in the Hymes sense, not normally grammatically possible (noun phrases without determiners, for example) or not normally appropriate (simple present tense used to refer to past events as in gets, holds in these headlines). As we can see from the ambiguity in these examples, keeping appropriately to the quantity maxim can result in the unintentional violation of the manner maxim. But not all violations are unintentional, of course. They can also be exploited to produce deliberate ambiguity attention. We said earlier that for communication to take place, the parties involved have to co-operate in negotiating some degree of agreed convergence, and that these maxims can be taken as a set of ground rules for doing so. And yet, as we have seen, people do not always keep to the rules. If people on purpose, what is the purpose? In communication two parties co-operate to converge on common ground. For this to happen there has to be some give and take on both sides: each party has to concede some ground of their own. This ground represents his/her own individual reality, a sense of self, a personal territory of identity, which it is their natural instinct to assert and protect. Co-operation necessarily involves some encroachment on this individual life space, and the area of convergence is always a potential site of contention between self and other.
Co-operative and territorial imperatives
So, for communication to take place at all, you have to co-operate, but, on the other, there is always the risk that this will compromise your own individuality. So this co-operative imperative is countered by another that acts against it → a territorial imperative, a need to preserve and protect one's own space. So it is not just meaning that is negotiated in communication but human relations. P1 and P2 are not just parties seeking an impartial agreement about shared knowledge but individual personalities competing to establish their own position in the area of convergence. They project a personal attitude or point of view. So one reason for disregarding the maxims is to assert territorial rights. But there is another reason. Co-operation involves encroachment, and this will often need to be tactfully managed. An incursion into the other's space may not be welcome. It may involve an adjustment the other is not prepared to make, or a threat to face , and this may cause offence, or embarrassment, a “ loss of face ”. The other may react in ways that threaten your own face. It is generally in our interests to maintain good relations by a mutual respect for face and the territorial rights of the other. This kind of co-operation which, paradoxically enough, calls for the non-cooperative departure from the maxims, goes under the general name of politeness. So there will be occasions when we avoid telling the truth because we want to reinforce the other's self-esteem, when we are deliberately uninformative because giving a direct answer might cause offence.
Conclusion
Although the violation of the maxims of the kind we have been considering have to do with small- scale adjustments that are made in conversational negotiation, they are symptomatic of how the
co-operative and territorial imperatives operate in discourse as a whole, written as well as spoken, and generally of how language is used for persuasion and the control of opinion.
7. Critical analysis
Positioning
Communication is not only a matter of the parties concerned finding a form of words which will serve most efficiently to indicate what reference and force they intend. It also involves finding words that will have the desired effect–that is to say, words which are tactically effective in regulating the position of self in relation to the other. For all communication is an exercise in control, an attempt to assert one's own position and to persuade the other to accept it. When somebody says, or writes, something it is with the intention of getting the addressee to think or feel or act in a certain way, and the maxim violations we considered in the last chapter can be seen as tactics that are used for the purpose.
Terms of reference
If you have the required schematic knowledge, you will understand when different wordings have the same referent. But then, why not just use one of them? Why the variation, and what motivates the choice of one wording rather than another? Avoiding redundancy and not saying the same thing twice, thereby avoiding a possible implicature: the reader might infer some significance in the repetition. But then, why not use the pronoun he to make a normal anaphoric link? We can locate these different terms of reference on a scale of increasing deference or respect. → e.g. Tony-Tony Blair-Mr Blair-The Prime Minister (But this does not exhaust the number of terms that might be used in reference to Tony Blair. Even less respectful terms might be our Tony.) What motivates the use of one expression rather than another? Every use of language involves selection and so every text can be rewritten in other terms, terms that could have been chosen, but, for one reason or another, were not.
Alternative wordings and persuasive purpose
A language will always provide the resource for alternative wordings: there will always be different grammatical structures and different lexical items available for referring to the 'same' thing in a variety of ways, thereby allowing for the expression of attitude, personal evaluation, point of view. In some cases, such expression is allowed for by different connotations that are assigned to lexical items by convention. Thus we can find in English terms that correspond in denotation, but are marked for negative or positive evaluation. For example, the verbs withdraw, decline, and negotiate are usually positive; retreat, refuse, and haggle negative. It might be supposed, then, that if you want to indicate attitude, the lexical means are available to you to do so. But things are not so straightforward. To begin with, the vocabulary of English is not all conveniently marked for attitude in this way. And anyway, words are not put to use in isolation but incorporated into lexical and grammatical patterns in texts where they are acted upon by other words in complex and unpredictable ways. And the texts themselves contract complex and unpredictable relations with context. What people mean is not always apparent from what they say.
maxim in that information has been withheld. With the passive, the grammar does not constrain you to provide an agent. The active alternative, of course does: “The police drove back the refugees and injured many of them” The agent as grammatical subject becomes thematized, and so now, we might suggest, agency (and so responsibility) is foregrounded as the topic, so if you wanted to play down the active role of the police, you would presumably want to avoid using this construction. Only when we look at textual continuity can we decide on whether or not the writer is conforming to the quantity maxim. Let us, for example, suppose that there is some text that comes before the structure we started with: “Hundreds of protesting refugees then began to move in the direction of the parliament buildings, where the police confronted them. The refugees/They were driven back and many of them were injured.” One cannot argue that to use the agentless passive here would be to withhold information about agency, for this information is already given, explicitly provided in the preceding sentence. How a particular part of a text is understood depends on its connection with what has gone before. So it is that we understand the pronoun them in the second sentence of this text as referring anaphorically to the refugees (and not to the parliament buildings). There is no need to spell this out. And similarly there is no need to spell out the agency role of the police in the last sentence. The cohesive devices we considered there follow a least effort principle. As such they serve an essentially co-operative purpose, for their function is to regulate information in relation to what is already given or known to make it easier to process.
Conclusion
Clearly language is widely used, and abused, as a means of control and persuasion, and it is one of the main purposes of discourse study to develop a well-grounded understanding of how this is done. But just as clearly, it makes little sense to assign ideological significance to the occurrence of lexical and grammatical features as such without regard to the co-textual relations they contract with other features.
8. Text analysis
Actually attested language
In this case, we treat the text simply as a collection of linguistic elements. But texts are normally only produced if there is an intention to communicate a message of some kind and the other three kinds of judgement have to do with this normal communicative functioning of text. Thus we can say how far a particular text is (2nd) feasible (how easy it is to process), and this will depend on how it keys in with shared knowledge. How far it is (3rd) appropriate will also, crucially, depend on how the text can be related to context to bring about reference, force, and effect. The 4th kind of judgement has to do with whether and to what extent a particular text is attested as usage. Here we are concerned with conformity not to linguistic rule but conventions of usage. As has already been pointed out, communication involves the selective use of the encoded resources of lexis and grammar. But we need to consider which selections are generally preferred by convention. Though all lexical and grammatical elements are equally possible in the linguistic code, they are obviously not all equally common in actual usage.
Norms of usage
With the advent of the computer, vast quantities of text can be collected into a corpus and electronically analysed. As a result, it is now possible to establish the relative frequency of occurrence of words and structures either in particular domains of use or more generally across
domains to provide profiles of frequency in the language as a whole. → e.g. if somebody in a conversation were to talk about circumventing a hindrance rather than getting round a problem , you would suppose that they were trying to impress or striving for a comic effect. Corpus analysis can tell us not only about the overall frequency of words, but also about their range , that is to say their distribution in different domains of use. The words circumvent and hindrance , for example, uncommon though they are, might well be found quite often in certain kinds of legal genre. They would then constitute a local norm of appropriate usage. So relative frequency can be taken as having a schematic significance in that certain words mark particular genres or discourse domains. → e.g. Thus the words customer, consumer, product, marketing, retailer, sales will be shown to have a much higher concentration in business domains than in any other genre.
Patterns of collocation
It is not, however, only the simple frequency and range of single items that is revealed in the corpus analysis of text but also, more interestingly and significantly, the frequency and range of their patterns of co-occurrence with other items. Certain items tend to keep company, or collocate, with others: a kind of mutual attraction that draws them together. → e.g. So the word unforeseen will attract the word circumstances , crying will attract shame , etc. But collocation goes beyond the relationship between two lexical items in a noun phrase to include many other recurring combinations in phrases like: as a matter of fact, all things considered, when all's said and done, all things being equal, for better or for worse, and so on. Such formulaic phrases are easy enough for proficient speakers of English to cite: they are aware of them as familiar idiomatic features of their usage, and no corpus analysis is needed to reveal them. But there are innumerable other textual patterns and phrases that emerge from corpus analysis that the language user is not aware of. It is easy enough to demonstrate this by means of a concordance , which displays all the occurrences of a particular word in lines of text so that one can see at a glance where co-textual combinations recur. Concordances for words which are semantically related can be compared and their collocational differences identified. → e.g. the words big and large , little and small , are shown not only to occur with different overall frequencies, but to have rather different collocational preferences + the words amazing , astonishing , and surprising not only vary considerably in frequency but also in the way they can combine with other words in lexical phrases or bundles: it is not surprising that …; and not: It is not amazing that… What corpus analysis reveals is that the constituents of texts are not so much separate words and structures as patterns of language, collocations and lexical bundles, of variable flexibility. This suggests that producing a text is, to some extent at least, a matter of assembling it from pre- fabricated parts, making whatever adjustments are necessary. The use of recurring patterns of language, stored ready for use in the mind, is in this case motivated by the co-operative principle. To deliberately disregard them would be to violate the manner maxim.
Semantic prosodies
But what motivates the use of particular collocational combinations? One answer might be that it is simply a matter of custom, a kind of ingrained habit and that's that. Alternatively we could look at whether there is anything in the semantic meaning of these words which might explain this
