Студопедия — INTRODUCTION 3 страница
Студопедия Главная Случайная страница Обратная связь

Разделы: Автомобили Астрономия Биология География Дом и сад Другие языки Другое Информатика История Культура Литература Логика Математика Медицина Металлургия Механика Образование Охрана труда Педагогика Политика Право Психология Религия Риторика Социология Спорт Строительство Технология Туризм Физика Философия Финансы Химия Черчение Экология Экономика Электроника

INTRODUCTION 3 страница






I have said that the utterance of a sentence is not necessarily a sentence. This is readily illustrated with reference to the utter­ance of 'I have not seen Mary'. Let us suppose that we are faced with the following text, either written or spoken:

(3) Have you seen Mary? I haven't. Peter hasn't either. She is never here when she should be.

It comprises four segments, or text-units, only the first and (possibly) the fourth of which would normally be described as complete sentences. The second and third are what would be traditionally regarded as elliptical sentence-fragments. And yet, in this context, I haven't is just as much the product of the utterance of the system-sentence (1) as is the text-sentence (2) in other contexts. (This is empirically demonstrable by asking informants, as teachers of traditional grammar in school would ask their pupils, to make I haven't into a full sentence.) And its prepositional content cannot be identified unless we are able to identify the sentence that has been uttered in the performance of the locutionary act of which I haven't, in this context, is the product. The same goes, of course, for Peter hasn't either with respect to 'Peter has not seen Mary' (and perhaps also for She


262 Text and discourse; context and co-text

is never here when she should be with respect to 'Mary is never here when she should be here').

It is important to realize that, although I have introduced a certain amount of technical terminology to handle the requisite theoretical distinctions, the distinctions themselves are real enough in our everyday experience of the use of language. We have no difficulty in deciding that I haven't has the prepositional content of 'I have not seen Mary' in one context, of 'I have not been to Switzerland' in another, of 'I haVe not got any money' in yet a third, and so on. In fact, out of context I haven't is infi­nitely ambiguous. In context, I haven't loses it ambiguity only in so far as it is possible to say which of the infinitely many sentences of English (with the appropriate grammatical structure) has been uttered.

I will continue to use the term 'sentence' in both senses, rely­ing upon the notational distinction between single quotation-marks and italics to make clear what kind of units I am referring to. Most linguists, as I have said, do not draw a clear conceptual distinction between system-sentences and text-sentences; the fact they they do not introduces a good deal of confusion into the discussion of the relation between sentence-generating gram­mars and the production (and interpretation) of texts. Argu­ably, it has also vitiated much of the work done in text-linguistics from the point of view of generative grammar. This will be evident from what follows. The sense in which texts are generated (i.e., produced) in particular situations is different from the sense in which sentences (i.e., system-sentences) are generated as abstract, mathematical, objects by the rules of a generative grammar.

9.2 WHAT IS AT EXT? AND WHAT IS TEXT?

Considered from the viewpoint of semantics (and pragmatics),
text and context are complementary: each presupposes the
other. I will come to context presently. But what is a text and
what is text? As we shall see, these are two distinct (but related)
questions. Let us start with the former: what is a text?


9.2 What is a text? And what is text? 263

One answer that is often given is that a text is a sequence of sentences. As it stands, this is clearly unsatisfactory - if 'sen­tence' means, as it must in this context, "text-sentence". True, there are some texts that would satisfy the definition, notably texts of a more formal character. But the vast majority of every­day colloquial texts are made up of a mixture of sentences, sen­tence-fragments and ready-made locutions. However, this defect in the definition of 'text' that has just been given is only one aspect of a more serious deficiency: its failure to make expli­cit the fact that the units of which a text is composed, whether they are sentences or not, are not simply strung together in sequence, but must be connected in some contextually appropri­ate way. The text as a whole must exhibit the related, but distin­guishable, properties of cohesion and coherence.

Roughly speaking, the distinction between cohesion and coherence has to do with the difference between form and con­tent; and some such distinction, however it is drawn by different schools of linguists, is both intuitively attractive and theoreti­cally justifiable. To return to our sample text, (3): that the pro­ducts of the utterance of 'I have not seen Mary' should have the form / haven't, rather than / have not seen Mary, is a matter of cohesion. So too is the use of either in Peter hasn't either and the use of the pronoun 'she', rather than 'Mary' in the first clause of She is never here when she should be. Cohesion is destroyed if the first three text-units are put in a different order, such as:

(4) Peter hasn't either. I haven't. Haveyou seen Mary?

It is also destroyed if we replace each of the text-units with the corresponding full text-sentence.

(5) Have you seen Mary? I haven't seen Mary. Peter hasn't seen Mary
(either). Mary is never here when she should be here
.

It is evident that (5) does not have the same kind of connected­ness that (3) had. For this reason it is less easy, though not impossible, to take the sequence as a text, rather than as a string of unconnected (or disconnected) utterances. Ellipsis and the use of pronouns, as well as the use of particular connecting par- tides and conjunctions (therefore,

so, etc.) commonly serve to

 


264 Text and discourse; context and co -text

create and sustain that kind of connectedness to which the term 'cohesion' is applied. Languages differ considerably with respect to the degree to which they permit or oblige their users to connect text-units in sequence by means of explicit indications of cohesion.

The other kind of connectedness - coherence - is a matter of content, rather than form. In default of any contextual indica­tion to the contrary, what is being said in any one text-unit is assumed to be relevant to what has just been said in the imme­diately preceding text-units. For example, in (3) the preposi­tional content of the fourth text-unit

(6) She is never here when she should be

will normally be taken to be relevant to that of the preceding three. In particular, 'she' will be understood to refer to Mary (by virtue of the kind of cohesion that is called anaphora) and the general statement that the speaker is making about Mary will be understood as a comment upon her absence at that time, rather than as the expression of some totally unconnected pas­sing thought. Similarly, if one heard or read the following sequence of two text-sentences,

(7) The whole family went to town last Saturday. Veronica bought a
dress, while John kept the children occupied in the toy-shop
,

one would normally assume that Veronica was one of the family, and presumably the mother; that she bought the dress in town; and that the toy-shop was also in town. None of these proposi­tions has been explicitly formulated, still less asserted; and any one of them might be contradicted, in specific contexts of utter­ance, by other propositions that are part of the speaker's and hearer's background beliefs and assumptions. We shall return to the question of coherence and relevance later in the chapter. Meanwhile, there are three points to be mentioned here and given due emphasis.

First, as we have already noted, the question "What is a text?" " differs from the more general question "What is text?". What are commonly referred to as texts, whether written or spoken, are deliberately composed by their authors as discrete wholes


9.3 Utterance-meaning and context 265

with determinate beginnings and ends. And, like (3) and (7), they are more or less readily divisible into text-units, some of which (though not all) can be classified as (text-)sentences. Moreover, longer texts, such as novels or plays, can usually be divided hierarchically into larger and smaller whole units (chap­ters and paragraphs, or acts, scenes and speeches), each of which is internally cohesive and coherent and can be analysed into smaller, sequentially ordered, units: chapters into sequences of paragraphs, paragraphs into sequences of (text-)sentences, and so on. Most of the text that we produce in our day-to-day use of language is not organized in this way.

The second point to be noted is that, as I am using the term 'text', individual text-sentences, sentence-fragments and fixed locutions all count as units of text in relation to their context of utterance, regardless of whether they are embedded in larger stretches of text or not.

Finally, it must be emphasized that the account that I gave of speech acts in the previous chapter is intended to cover in prin­ciple all aspects of the production of text. Speech-act theorists have been concerned primarily with the production of text-sen­tences (without drawing the distinction that I have drawn between text-sentences and system-sentences). But the utterance of a sentence, in practice, always involves its contextualiza-tion — the process of making the product of utterance both cohesive and coherent in relation to its context. As I have said, text and context are complementary. What then is context? And how does it relate to utterance-meaning? We shall begin by discussing the relation between context and utterance-meaning.

9.3 UTTERANCE-MEANING AND CONTEXT

Context determines utterance-meaning at three distinguishable levels in the analysis of text or discourse. First, it will usually, if not always, make clear what sentence has been uttered - if a sentence has indeed been uttered. Second, it will usually make clear what proposition has been expressed - if a proposition has been expressed. Third, it will usually make clear that the


266 Text and discourse; context and co -text

proposition in question has been expressed with one kind of illocutionary force rather than another. In all three respects, context is relevant to the determination of what is said, in the several senses of 'say' that were identified in the preceding chapter.

But utterance-meaning goes beyond what is actually said: it also includes what is implied (or presupposed). And context is highly relevant to this part of the meaning of utterances. In this section, we shall restrict our attention to what is said: to the locu-tionary and illocutionary aspects of utterance-meaning. We shall rely initially upon an intuitive everyday notion of what context is. That context may tell us what sentence has been uttered is obvious from our discussion of locutionary acts. As we saw, tokens of the same utterance-type can result from the utter­ance, on different occasions, of different sentences. In such cases, the utterance-inscription itself will usually be either gram­matically or lexically ambiguous (or both). For example,

(8) They passed the port at midnight

is lexically ambiguous. However, it would normally be clear in a given context which of the two homonyms, 'porti' ("harbour") or 'ports' ("kind of fortified wine"), is being used - and also which sense of the polysemous verb 'pass' is intended. Polysemy, unlike homonymy, does not give us grounds for distinguishing one sentence from another (on a traditional view of sentences). But it may none the less give rise to lexical ambiguity. In colloca­tion with 'port2', the most salient sense of 'pass', in most contexts, is undoubtedly the one in which it means "hand from one to another". But it is easy to see that in an appropriate context 'pass' meaning "go past" can be collocated with 'port2' just as readily as it can be collocated, in other contexts, with 'port1'.

We do not know what propositional content is being expressed unless we know what sentence is being uttered. Moreover, if the sentence contains one or more polysemous expressions, we do not know in what sense they are being used. Context, therefore, is a factor in the determination of the propositional content of particular tokens of utterance-inscriptions on different occasions of utterance. Usually, we operate with contextual information


9.3 Uttemnce-meaning and context 267

below the level of consciousness in our interpretation of every­day utterances. Most of the ambiguities, whether lexical or grammatical, therefore pass unnoticed. For example, the phrase 'the vintage port' would normally be interpreted as referring to wine, and 'the busy port' as referring to a harbour. From time to time, however, we are made aware of such ambiguities, precisely because our contextual beliefs and assumptions differ from those of our interlocutors. We may then either fail to understand what they are saying, hesitating between alternative interpretations, or misunderstand their utterances by taking them in the wrong sense. The second of these two possibilities is often exploited by humorists and comedians, who deliberately set up the context in such a way that their audience will unconsciously assign one interpretation to an utterance-inscription and then, in the so-called punch-line, suddenly reveal to them, more or less indir­ectly, that they have been led up the garden path.

In some cases there is no need to set up the context specially for the purpose. The out-of-context saliency of what is subsequently revealed to be what may be referred to as the garden-path inter­pretation will suffice. To take a rather hackneyed example: if

(9) Three strong girls went for a tramp is followed, after a brief pause, with

(10) The tramp died,

the comedian will probably secure the desired effect, simply by virtue of the out-of-context saliency of the sense of 'go for a tramp' in which it falls, semantically and syntactically, with 'go for a walk', 'go for a ride', 'go for a swim', etc.

Both contextually determined and out-of-context saliency are, of course, exploited for more serious purposes in literature, where readers may well be expected to hold two or more interpretations in mind simul-taneously and either to hesitate between them or to combine them in some way, in order to construct a richer composite interpretation. Ambiguity is commonly described by philosophers and linguists as if it were of its nature pathological - something which gets in the way of clarity and precision. This is a highly prejudiced and unbalanced view of


268 Text and discourse; context and co-text

the matter. Not only is it frequently, and erroneously, associated with the view that all sentences have precise and determinate meanings; it is based on the equally erroneous assumption that clarity and the avoidance of vagueness and equivocation are always desirable, regardless of genre, style and context. Nothing that is said about ambiguity in this section, or anywhere else in this book, should be taken to imply that ambiguity is, or should be, avoided in all contexts.

Let us now turn to the second of the two levels at which con­text determines utterance-meaning: let us take up the fact that context can make clear, not only which sentence has been uttered (and, in the case of polysemous sentences, with what meaning), but also what proposition has been expressed. In Part 3, I drew a distinction between 'proposition' and 'preposi­tional content', and a corresponding distinction between 'refer­ence' and 'referential range' (or 'referential potential'). I pointed out that, whereas the propositional content of a sentence and the referential range of its component expressions can be established without appeal to the context of utterance, it is not generally possible to establish what proposition is being expressed, without knowing in what context the sentence is uttered. We can now relate this point to the immediately preced­ing discussion of text and context.

As we have seen, I haven't can be put into correspondence, by means of the notion of contextualization, with any one of an infi­nite set of sentences. In our sample text (3), it can be identified as the product of the utterance of the sentence 'I have not seen Mary', which contains two referential (i.e., potentially referring) expressions: ' I ' and 'Mary'. What do they refer to? Obviously, out of context there is no way of knowing. If we make certain assumptions about the production of the text, we can say that the speaker or writer - more generally, the locutionary agent - is referring to himself or herself by means of ' I ' and to some third person (i.e., to some person other than himself or herself and his or her addressee) by means of 'Mary'. It is worth noting, however, that we cannot be sure even of this simply on the basis of our knowledge of English. There are circumstances in which speakers may refer to someone other than

 


9.3 Utterance-meaning and context 269

themselves by means of ' I ' (notably when they are acting as interpreters); and there are circumstances in which one may refer to one's addressee by name, rather than by the pronoun 'you', so that 'Mary' could in principle be used to refer to the addressee. In any case, granted that the locutionary agent is referring to himself or herself with ' I ' and to someone else with 'Mary' (and that a proposition is being expressed), we cannot say what proposition is being expressed and evaluate it for truth or falsity without knowing who the locutionary agent and Mary are.

We also need to know when the utterance was produced. The fact that the locutionary agent used the present-perfect form haven't (seen), rather than didn't (see), hadn't (seen), don't (see) (or can't (see)), is relevant to the truth-value of the proposition that he or she expresses. (So too, incidentally, is the fact that in most contexts there will be a tacitly understood reference to the period of time of which the predicative expression 'have seen' is, or is not, true. For example, the speaker may have seen Mary on the previous day, or even a very short time before, and yet be held to have made a true statement in saying I haven't.) In the case of other utterances, we need to know, not only the time, but also the place of utterance, in order to establish what proposition has been expressed. For example, this is so in respect of the fourth text-unit in (3), which, unlike the second and

third, happens to be a text-sentence, rather than an elliptical

sentence-fragment:

(11) She is never here when she should be:

'here' normally refers to the place of utterance, so that the pro-

position "Mary is here" may be true in respect of one place at certain times and false of that place at other times. Questions of this kind will occupy us in Chapter 10. Let us merely note for the present that the vast majority of utterance-inscriptions in most languages are implicitly, if not explicitly, indexical or deictic, so that they express different propositions according to the context in which they are produced. This point has already been mentioned in connexion with the treatment of sentence-meaning in formal (linguistic) semantics in Chapter 7.

 


270 Text and discourse; context and co-text

We come, finally, to the contextual determination of illocu-tionary force. As we saw in the preceding chapter, the same sentence may be uttered on different occasions with different kinds of illocutionary force. For example,

(12) 'I will give you ₤ 5'

may be uttered as a promise or as a prediction. Or again,

(13) 'Sit down'

may be uttered, in what is normally regarded as its most charac­teristic use, as a request or a command; it may also be used in order to grant the addressee permission to sit down. Frequently, but not always, the prosodic contour (i.e., the stress and intona­tion) will indicate to the addressee that the utterance has one kind of illocutionary force rather than another. But whether this is also indicated prosodically or not (in the case of spoken utterances), it will usually be clear, in context, what kind of illo­cutionary act has been performed. For example, it will usually be clear whether the speaker has the authority to order the addressee to sit down or to grant him or her permission to do so. Indeed, much of our day-to-day language activity is so closely integrated with other kinds of social behaviour and activity that the occurrence of an utterance with a particular illocution­ary force is often predictable from the socially identifiable situa­tion in which it occurs. For example, we would not normally sit down in someone else's house or office without being invited to do so. On the other hand, in most situations - paying a call on a new neighbour, coming to see the bank-manager about an overdraft, etc., it will be evident to us and to our interlocutor that at a certain point in the conversation an invitation of this kind should be made. This being so, addressees do not have to calculate or determine the illocutionary force of Sit down, from first principles, in terms of the meaning of the sentence 'Sit down' and their assessment of the speaker's motivation for say­ing what he or she has said. The situation itself predisposes addressees to expect either this very utterance-inscription or another with the same illocutionary force (Won'tyou sit down?, Why don't you take a seat?, etc.). It is arguable that most so-called


9.4 Implication and conventional implicatures 271

indirect speech-acts, of the kind that were mentioned in the pre­ceding chapter, can be accounted for in this way. At any rate, there can be no doubt that, in many instances, the illocutionary force of an utterance is strongly determined by the context in which it occurs.

To recapitulate, then, context determines utterance-meaning at three distinguishable levels in the analysis of text. I have still not said what context is or how it is handled theoretically: we shall come to that presently. From what has been said in this sec­tion, however, it will be evident that the context of an utterance includes, not only the relevant co-text (i.e., the relevant sur­rounding text), but also the relevant features of the situation of utterance. As we shall see later, what is sometimes referred to as the context of situation can, and should, be defined in such a way that it subsumes everything in the co-text that bears upon the question of cohesion, coherence and relevance.

9.4 IMPLICATION AND CONVENTIONAL IMPLICATURES

There is an everyday meaning of the verb 'imply' in which we can, and usually do, imply by means of our utterances something other than what we actually say. For example, asked to give an opinion about a person's character, one might say

(14) He'd share his last crust of bread with you,

Obviously, it has not actually been said of the person in question that he is both kind and generous. But one might reasonably be held to have implied this. Let us introduce a distinction, then, between what is actually said, or expressed, in an utterance-inscription and what is conveyed either by what is said or by the fact of saying what is said.

Much of the information that is conveyed from speaker to addressee in day-to-day conversation is implied, rather than asserted. In some cases, of course, it is not clear whether the speaker intends the addressee to draw a particular inference or not. And this opens the way for misunderstanding and misrepre­sentation, on the one hand, and for the subtle manipulation of


272 Text and discourse; context and co-text

the addressee's opinion, on the other. However, in what one may think of as the standard kind of situation, not only do addressees draw the inferences that speakers intend them to draw, but these inferences are such that the speakers themselves, if asked, would also subscribe to them. I have assumed that this is so in respect of (14). It is easy enough, however, to devise a situation of utterance in which the hearer would not draw the inference that the person referred to is kind and generous. It is equally easy to think of circumstances in which the speaker might insin­cerely and deceitfully intend the addressee to draw this infer­ence.

In recent years, the notion of implicature has been intro­duced into the philosophy of language, and subsequently into linguistics, to bridge part of the gap between the standard logical notions of implication and entailment with which formal seman­tics operates, on the one hand, and the broader everyday notion of implication, on the other. According to Paul Grice in his 1967/8 William James Lectures (see Grice, 1975, 1989), there are two kinds of implicatures: conventional and conversa­tional. The difference between them is that the former depend on something other than what is truth-conditional in the con­ventional use, or meaning, of particular forms and expressions, whereas the latter derive from a set of more general principles which regulate the proper conduct of conversation. Conversa­tional implicatures will be discussed in the following section.

It has been argued, for example, that the difference between the forms but and and in English can be accounted for in terms of the notion of conventional implicature. Those who take this view, including Grice himself, would say that the following two sentences have the same prepositional content:

'He is poor and he is honest',

'He is poor but he is honest'.

If they also identify sentence-meaning with propositional con- tent, they would say that the two sentences have the same mean- ing. Most native speakers of English operating with an everyday notion of 'meaning' would probably disagree (see 6.3). The proponents of truth-conditional semantics can meet


9.4 Implication and conventional implicatures 273

this challenge - if they accept that there is such a thing as con­ventional implicature — by attributing the apparent difference in meaning to the conventional implicature associated with the form but. They can say that the use of but, in contrast with and, indicates that the speaker feels that there is some kind of contrast between the conjoined propositions.

For example, on the assumption that the two sentences are being used to make a statement and 'he' refers to the same person in each of the conjoined clauses, in uttering (16) the speaker might be implicating (though not asserting) that it is unusual for someone to be both poor and honest. But would the implica­tion, or implicature, be as determinate as this? Out of context there is no way of knowing exactly which of several propositions speakers are implicating. They might be indicating (whether voluntarily or not) that they are surprised, not that anyone should be both poor and honest, but that a man should be; or, alternatively, that anyone in this person's circumstances or this person in any circumstances should be. Indeed, they may not be indicating their own surprise at all, but merely their expecta­tion that their interlocutor will be surprised. In fact, there is a whole range of further possibilities, most of which can be sub­sumed loosely under the notion of contrast. But it is remarkably difficult, in most cases, to say exactly what is being implicated by the use of but and impossible to do so without considering in some detail the actual context of utterance.

It is usually taken for granted by those who have discussed the notion of implicature that the difference between and and but c annot be part of the propositional content of the compound clauses in which they occur (and we tacitly accepted this view in section 6.3). But there are circumstances in which speakers can use but and and contrastively within the scope of the verb 'say', and even of the adjective 'true'. For example, they might claim at some point in the argument that their interlocutor is misrepresenting them:

 

(17) I did not say that he was poor but honest, I said that he was poor and

honest. And that's a very different thing. Personally, I don't find it

surprising that anyone should be both. Let us recapitulate then. It is


274 Text and discourse; context and co -text

true that he is poor and honest; it is not true ~ in my view at least -that he is poor but honest. We both subscribe to the truth of the proposition that he is poor and he is honest. We appear to disagree as to the truth of the proposition that he is poor but he is honest.

I have deliberately constructed this passage in such a way that it starts with an everyday use of 'say' and ends with what is a recognizably technical use of 'proposition'. There is little doubt, I think, that it is more natural to use but and and contrastively within the scope of the verb 'say' than within the scope of 'proposition'. And yet the passage, as a Whole, is surely linguistically, if not philosophically or logically, acceptable; and whether it is judged to be philosophically or logically







Дата добавления: 2015-09-07; просмотров: 387. Нарушение авторских прав; Мы поможем в написании вашей работы!



Обзор компонентов Multisim Компоненты – это основа любой схемы, это все элементы, из которых она состоит. Multisim оперирует с двумя категориями...

Композиция из абстрактных геометрических фигур Данная композиция состоит из линий, штриховки, абстрактных геометрических форм...

Важнейшие способы обработки и анализа рядов динамики Не во всех случаях эмпирические данные рядов динамики позволяют определить тенденцию изменения явления во времени...

ТЕОРЕТИЧЕСКАЯ МЕХАНИКА Статика является частью теоретической механики, изучающей условия, при ко­торых тело находится под действием заданной системы сил...

Деятельность сестер милосердия общин Красного Креста ярко проявилась в период Тритоны – интервалы, в которых содержится три тона. К тритонам относятся увеличенная кварта (ув.4) и уменьшенная квинта (ум.5). Их можно построить на ступенях натурального и гармонического мажора и минора.  ...

Понятие о синдроме нарушения бронхиальной проходимости и его клинические проявления Синдром нарушения бронхиальной проходимости (бронхообструктивный синдром) – это патологическое состояние...

Опухоли яичников в детском и подростковом возрасте Опухоли яичников занимают первое место в структуре опухолей половой системы у девочек и встречаются в возрасте 10 – 16 лет и в период полового созревания...

Понятие массовых мероприятий, их виды Под массовыми мероприятиями следует понимать совокупность действий или явлений социальной жизни с участием большого количества граждан...

Тактика действий нарядов полиции по предупреждению и пресечению правонарушений при проведении массовых мероприятий К особенностям проведения массовых мероприятий и факторам, влияющим на охрану общественного порядка и обеспечение общественной безопасности, можно отнести значительное количество субъектов, принимающих участие в их подготовке и проведении...

Тактические действия нарядов полиции по предупреждению и пресечению групповых нарушений общественного порядка и массовых беспорядков В целях предупреждения разрастания групповых нарушений общественного порядка (далееГНОП) в массовые беспорядки подразделения (наряды) полиции осуществляют следующие мероприятия...

Studopedia.info - Студопедия - 2014-2024 год . (0.011 сек.) русская версия | украинская версия