论语境对话篇的影响(txt+pdf+epub+mobi电子书下载)

作者：王全智

出版社：世界图书出版广东有限公司

格式: AZW3, DOCX, EPUB, MOBI, PDF, TXT

论语境对话篇的影响试读：

前言

众所周知，语境既决定语篇的产生，又制约语篇的进展。然而，语境中到底有哪些因素对语篇产生影响，以及这些因素通过何种方式对语篇施加影响，人们则不甚了了。这正是本研究所要探讨的内容。

本研究以系统功能语言学为理论框架，多角度、全方位地探讨语境对语篇的影响。全文共分七章。

第一章讨论本研究的原理、目标、研究方法以及论文的总体结构。

第二章对现有的有关语境和语篇的研究进行了综述。第一部分评述各派语言学家对语境的定义，并将所有派别按照其对语境定义的不同分为两类：客观语境派和认知语境派。作者强调指出：认知语境依赖于客观语境而存在，而非独立于客观语境之外；从而为本研究采纳韩礼德的语境模式为原型来探讨语境对语篇的影响提供了理论依据。第二部分简要介绍各派语言学家对语篇的界定。通过比较，本文确定采用术语text来指代各类语篇，并提出语篇的五个使成条件。

第三章探讨语场对语篇的影响。Eggins以专门性为标准将语场分为专业语场和非专业语场。在肯定其贡献的同时，本文指出其在语场切分方面的局限性使得我们无法采取其语场切分作为探讨语场对语篇的影响的模型。与Eggins不同，Leckie-Tarry从语场的构成着眼，认为语场包括三个成分，即：场景、参与者和主题。本文认为，由于客观世界的无限多样性，语场亦呈现无限多样的特性。故而，Eggins将语场笼统地划分为专业语场和非专业语场不可取。有鉴于此，本章采用Leckie-Tarry的语场概念，通过语场的三个成分即场景、参与者和主题的变化对语篇的作用的分析探讨了语场对语篇的影响。

第四章旨在探讨基调对语篇的影响。在介绍了Gregory，Poynton，Leckie-Tarry，Martin等人的基调概念之后，本文指出基调包括三个参数：权势、接触和感情投入；并依次探讨了各个参数之细微变化可能对语篇产生的影响。

第五章分析语式对语篇的影响。作者指出，语式作为语境因素之一，应当将其看作单独实体，方可探讨其之于语篇的影响。因此，在探讨语式对语篇的影响时，作者不采纳Leckie-Tarry，Martin等人的语式概念。作者认为，语式可以归结为两种：口头语式和书面语式。每种语式又有若干体现形式。本章详细比较了网络即时通信语篇和面谈语篇，并揭示网络即时通信的若干特点。

第六章分析语境因素之间的相互影响以及这种互动可能对语篇产生的影响。作者指出：语境因素不仅单独直接对语篇产生制约，而且彼此之间也相互作用，并表现于语篇之中。

第七章总结了本研究的贡献，同时指出不足之处，并对未来研究提出若干建议。

本书脱胎于本人的博士论文。从选题到成文，撰写过程中每一个环节都得到导师朱永生教授悉心指导。而今付梓之际，又承蒙先生拨冗作序。先生对后学的提携，学生永志不忘。

借此机会，感谢厦门大学的杨信彰教授、河南大学的刘辰诞教授、苏州大学的严世清教授、北京师范大学的苗兴伟教授、中山大学的丁建新教授等。他们一直以来的督促和鼓励，给本人提供了前进的动力。

限于水平，书中谬误在所难免，恳请广大读者批评指正。Foreword

It has been long noticed that context both contributes to and conditions the production of text. However, what contextual factors are influential to, and in what way they exert their impact upon text, remain unclear. It is against this background that this current research is conducted.

The current study mainly explores the impact of context upon text by taking Hallidayan Systemic Functional Linguistics as its theoretical framework.

This research falls into seven chapters.

Chapter 1 discusses the rationale, objectives, research methodology and the overall structure of the research.

Chapter 2 conducts an overview of the existing researches with regard to context and text. In its first part, different definitions of context given by different schools of linguists are reviewed and then classified into two kinds: objective context and cognitive context. Then the current author points out that cognitive context is dependent upon objective context, rather than independent of the latter. In this research, the current author takes the theory of context established by Halliday and other systemic functional linguists as the paradigm in the exploration of the impact of context upon text. In the second part, different definitions of text are given and compared, and the term “text” is chosen to cover the existing terms such as conversation, discourse, and dialogue. Moreover, in order to avoid the confusion engendered by the wide variety of prevalent definitions, the current author proposes five enabling conditions for text.

Chapter 3 addresses the impact of field on text. While pointing out the contributions made by Suzanne Eggins in her division of field into technical and non-technical in accordance with the technicality of one and the same field, the current author argues that its limitation makes it impossible for us to take it as a model in analyzing the impact of context upon text. Different from Suzanne Eggins, Leckie-Tarry approaches field from its composition and contends that field is composed of three variables: arena, participants and subject matter. The current author argues that owing to the infinitude of the objective world, it is not appropriate to conduct studies in just one and the same field. Field can be limitless. Therefore, Leckie-Tarry's field is taken in exploring the impact of field on text. And the three components of field she proposed are redefined. Setting refers to the particular locations of the interaction, both in terms of their inherent features, and in terms of the social institutions which determine them. Participant is mainly concerned with the inherent features of the participants, i.e., their physical and mental attributes and the knowledge they bring to bear on the setting and events. Subject matter refers to what the spoken or written communication is about. Any change in any of these three variables is likely to lead to the corresponding change in the text.

Chapter 4 deals with the impact of tenor on text. Just like field, tenor has also been defined by a large number of systemic functional linguists. In this chapter, the current author has quoted Gregory, Poynton, Leckie-Tarry, and Martin. Gregory divides tenor into two kinds: personal tenor and functional tenor. Poynton, different from Gregory, mainly deals with the components of tenor and argues that tenor comprises power, contact, and affective involvement. Leckie-Tarry points out that tenor involves three elements: formality, role and focus. The elements of tenor defined by Martin are status, contact and affect. Since tenor, according to Halliday, refers to the interpersonal relationship in context of communication, the current author takes the definitions given by Poynton and Martin as the paradigm in probing into the impact of tenor on text.

Chapter 5 investigates the impact of mode on text. Not convinced of Leckie-Tarry's and Martin's analyses of mode, the current author points out that mode, as one of contextual factors, should be dealt with ontologically as an objective entity; and holds that there are two modes: spoken, written. Every mode has its own realizations. For example, face-to-face conversations and telephone conversations are realizations of spoken mode and letters are realization of written mode. In this chapter, the author has conducted a detailed discussion on the difference between face-to-face conversation and Internet relay chat, with the former being a realization of spoken mode and the latter a realization of written mode.

In Chapter 6, the author contends that contextual factors not only exert their influence on text respectively, but they also influence each other, and this influence also finds full expression in text. On this ground, Chapter 6 explores tentatively the interaction between contextual factors and attempts to reveal their impact on text. The interaction between contextual factors can be divided into five kinds: the influence of field on mode, the influence of field on tenor, the impact of tenor on mode, the impact of tenor on field, and the impact of mode on tenor. By depicting how a certain change in the elements of one contextual factor is likely to cause a change in a certain element of another contextual factor and how this change can bring about changes in the ongoing text, the author draws a clear picture of internal interaction between contextual factors and its impact on text.

Chapter 7 summarizes the contributions of the present research and at the same time points out its drawbacks. Moreover, it has made some suggestions for future research.

To conclude, the current research has conducted an extensive study into the impact of context upon text and has revealed that text is subject to the influence of context.Chapter 1 Introduction1.1 Rationale

The twentieth century witnessed the fundamental shift in the history of linguistics — the transition from structuralism to functionalism. In 1916, the Swiss linguist Ferdinand de Saussure fathered structuralism in linguistics, which was widely acclaimed as the Copernican Revolution (Holdcroft, 1991). Moreover, structuralism was later generally taken as the paradigm for various sciences, such as anthropology, architecture, literature, and so on, hence greatly promoting the development of these sciences. Unfortunately, this line of thought dominated the field of linguistics for more than 50 years, with Chomsky's Transformational-Generative Grammar, Pike's Tagmemics, and many other schools as his faithful disciples despite their self-claimed discrepancy from the former. Not until the nineteen fifties when London School came into the limelight and was gradually accepted by the wide public in linguistic arena was this thick ice broken. What structuralism lays store by are the formal units of language rather than their meaning. And structural linguists like de Saussure, Chomsky, Pike and so forth never studied linguistic units extending across a sentence. Despite the fact that de Saussure dichotomizes the relationship between linguistic units into associative and paradigmatic, which he considers equally important in language studies, he quite obviously subjugates associative relationship to paradigmatic relationship (Harland, 1993: 2; Wang, 1998), thus obstructing the studies of linear structure of language events, which explains why there had been no studies of text, discourse, or even paragraphs which span across a sentence before the emergence of London School. What is more, de Saussure never takes any trouble to embark upon the interplay between language and society though he agreed that language is a social phenomenon. In stark contrast, London School, founded by J. R. Firth and now led by M. A. K. Halliday, contends that linguistic studies should not be confined to the studies of isolated units, like words, phonemes, sentences; instead, linguistic studies should be oriented towards the studies of communicative events which are conducted by means of language. In other words, what London School studies are linguistic events that are embedded in a certain situation rather than isolated linguistic units with no link to surrounding situations and social milieus. And these linguistic events are not necessarily limited in length in that they can be a single word, a sentence, and a paragraph, which are all defined as texts by Halliday (1994). Nowadays, functionalism has been expansively accepted in the field of linguistics, and discourse analysis/text linguistics, as an application of functionalist linguistics, is on the upswing. Nevertheless, most of the studies in this area are rather problematic and a large number of questions remain to be answered, which is one of the reasons behind this research.

In the first chapter, we will mainly deal with the significance of this study in a detailed manner.1.1.1 Motivation of this study1.1.1.1 Sterility of structuralist studies of language

As already has been mentioned, the Swiss linguist Ferdinand de Saussure fathered structuralism in 1916, which earned him an everlasting reputation as the founder of modern linguistics. In the following several decades of the 20th century, structuralism held sway in the field of linguistics, with lots of famous scholars both in Europe and in the United States of America practicing linguistics under the influence of his teachings, either directly or indirectly. The Prague School in Czechoslovakia and the American Descriptive Linguistics are the prominent representatives, both of which have made tremendous contributions to the development of linguistics. Here we will probe into their contributions and then their limitations and drawbacks in their research methodology.

1.1.1.1.1 The Prague School's contribution to linguistics

The Prague School was formed approximately in 1926 with Vilem Mathesius as the leader. The Prague School practiced a special style of synchronic linguistics, and the hallmark of Prague linguistics was that it saw language in terms of function. That is why they are called the school of Functional Linguistics. Members of this school think of language as a whole as serving a purpose, which is a truism that would hardly differentiate them from others. But they analyze a given language with a view of showing the respective functions played by the various structural components in the use of the entire language (Sampson, 1980: 104). This differentiates the Prague School sharply form their contemporaries, the American Descriptivists. Prague linguists look at languages as one might look at a motor, seeking to understand what jobs the various components are doing and how the nature of one component determines the nature of the others. As long as they are describing the structure of a language, the practice of the Prague School is not very different from that of their contemporaries — they use the notions of phoneme and morpheme, for instance; but they try to go from description to explanation, saying not just what language is like but why it is the way it is. One fairly straightforward example of functional explanation in Mathesius's own book concerns his use of terms commonly translated theme and rheme, and the notion which has come to be called Functional Sentence Perspective by recent writers working in the Prague tradition. Most (or, at least, many) sentences are uttered in order to give the hearer some information; but obviously we do not produce unrelated pieces of information chosen at random, rather we carefully tailor our statements with a view not only to what we want the hearers to learn but also to what we have so far built up. According to Mathesius, the need for continuity means that a sentence commonly falls into two parts: the theme, which refers to something about which the hearer already knows, and the rheme, which states some new fact about that given topic. Unless certain special effects are aimed at, theme will precede rheme, so that the peg may be established in the hearer's mind before anything new has to be hung on it. Here we can find the origin of Halliday's segmentation of a sentence into theme and rheme.

Another representative scholar in this school who is worthy of being mentioned here is Nikolai Sergeyevich Trubetzkoy, who formulates his authoritative statements in Principles of Phonology in 1939. And the statements he makes in this masterpiece is called by later linguists the functional approach to phonology. Deeply influenced by de Saussure, the Prague School in general is primarily interested in the paradigmatic relations between phonemes, i.e. the nature of the oppositions between the phonemes that potentially contrast with one another at a given point in a phonological structure, rather than in the syntagmatic relations which determine how phonemes may be organized into sequences in a language. Trubetzkoy finds that for each abstract sound, there is a range rather than a point in which speakers are allowed to find a realization different from every other realization by themselves and also distinct from every other realization by other people. He points out that what serves to distinguish the phonological unit is not the sound itself, but rather the sound's contrastive functions. He develops a vocabulary for classifying various types of phonemic contrast: e.g. he distinguishes between (i) privative oppositions, in which two phonemes are identified except that one contains a phonetic “mark” which the other lacks (e.g. /f/~/v/, the mark in this case being voice); (ii) gradual oppositions, in which the members differ in possessing different degrees of some gradient property (e.g. /i/~/e/~/æ/, with respect to the property of vowel aperture); and (iii) equipollent oppositions, in which each member has a distinguished mark lacking in the others, (e.g. /p/~/t/~/k/). He also puts forward the idea of “archiphoneme”, which is useful in dissolving pseudo-problems (Sampson, 1980: 108). What's more, Trubetzkoy distinguishes various functions that can be served by a phonological opposition, such as the distinctive function, by which is meant that a certain phonological opposition can keep different words or longer sequences apart. Here it can be clearly seen that Trubetzkoy's function of phonological contrast resonates with de Saussure's viewpoint that contrast plays a vital role in the determination of the value of a certain linguistic unit among its neighboring units.

1.1.1.1.2 The contributions of American Descriptive Linguistics

Also known as American Structuralism, American Descriptive Linguistics was developed, almost independently of Saussure's impact, by the anthropologist Franz Boas. For this reason, American structuralism is intricately entangled with anthropology from the first day of its existence. The development of American structuralism can generally be divided into three stages: Boasian stage, Bloomfieldian stage and Pike's stage. It is not fair to say that the research methodology and major viewpoint on language of American structuralists remain the same over the long period of development which spans over 50 years, but there is a common vein which makes American structuralism what it is. Just as Sampson (1980) points out, while Bloomfield did much to promote and codify the Descriptive tradition of linguistic analysis, it is fair to say that his theoretical work does not contain a great deal of innovation. The main points of Bloomfield's theories of language description can be found in Boas, though they are often stated more explicitly and with more elaboration by Bloomfield. As for Kenneth Pike's tagmemics, the novelty of tagmemic formulae lies more in their superficial appearance than in any theoretical innovations they represent, and the abstract theoretical writings of Pike and others of this group seem the less valuable aspect of their contribution. What counts is that they maintain the Descriptive approach of subordinating theory to the practice of analyzing unfamiliar languages. In short, for descriptivists, the description of an individual language is an end in itself, or a necessary step towards understanding the wider culture of a particular community. For example, the methods they invent in describing the phonemes of Indian languages include segmentation, comparison, equation, and classification. They also apply these methods to morphological analysis and syntactic analysis of these languages. Leonard Bloomfield's Language is a typical summary of these methods, thus widely acclaimed as the masterpiece in American Descriptive Linguistics. What should be pointed out here is that Transformational Generative Grammar proposed by Noam Chomsky is also a branch of American Structuralism (Wang, 1985), despite his self-claimed distinction from Bloomfieldian School and the new gadgets and methodology he invents in the analysis of language. The reason for this lies in that they all focus their research interest on the formal aspects of language and seldom, if not never, touch upon the meaning aspect.

1.1.1.1.3 A brief summary

It goes without saying that there must be quite a large discrepancy between the branches of structuralism in their views of language and research methodology. All the same, they are all characterized by a striking similarity. That is, all the structuralists concentrate their efforts on the studies of formal aspects of language and turn a blind eye to its social aspects. Ethnographical studies of Indian languages as conducted by the early American Structuralists, Chomsky's Transformational-Generative Grammar and Pike's Tagmemics are all concerned with the formal aspects of language and never attach any significance to semantic aspects and social aspects of language.

The inadequate attention to the social aspects and semantic aspects should all be ascribed to the teaching of de Saussure. It is he who put forward a lot of dichotomies in linguistic study, such as langue versus parole, paradigmatic versus syntagmatic, synchronic versus diachronic, value versus meaning of signs, internal linguistics versus external linguistics, signifier versus signified (Shi, 1996). According to de Saussure, what are worth studying are the former categories of the dichotomies. That explains why the linguists under his influence, whether directly or indirectly, conducted linguistic studies just the way he thinks fit. However in the dichotomies he ignored some very important aspects, if seen nowadays (Yang, 1994), because language is a social fact, just as de Saussure himself points out (de Saussure, 1959: 15) in his Course in General Linguistics. The communication function it plays in human society ascertains the significance and indispensability of linguistic studies with a social perspective. No study of language shall be claimed to be complete without taking into consideration its social nature and the role it plays in everyday human communication.1.1.1.2 Limitations of Z. S. Harris's Discourse Analysis

Although Z. S. Harris's Discourse Analysis was published as one of the papers in the series Papers on Formal Linguistics, and although this book consists of only 73 pages including the contents and appendices, it did arouse people's attention to the study of linguistic units that span across the boundary of sentence. From then on, the linguists’ research interest was no longer restricted to the formal aspects of a single word as in Bloomfieldian era, nor to the transformational and generative aspects of a single sentence, as was the standard practice in Chomskyan era. On the contrary, their research orientation has shifted to the study of text or discourse, which is much longer than single sentences. In this sense, Harris is the forerunner of discourse analysis, and it is he who embarks upon an entirely new field in linguistics study.

However, Harris's paper falls far short of our expectations in spite of its seductive title. Working within the Bloomfieldian tradition, he sets out to produce a formal method “for the analysis of connected speech or writing which does not depend on the analyst's knowledge of the particular meaning of each morpheme.” He observes that in grammar it is possible to set up word classes according to their distributions and produce a class of adjectives A which occur before a class of nouns N; such a statement captures a powerful generalization, even though it is possible to show that a particular member of the class A, “voluntary” may never occur before a particular member of the class N “subjugation”.

Harris argues that a distributional analysis can be successfully applied to a whole text to discover the structure above the rank of sentence. As an example he creates a text containing the following four sentences,The trees turn here about the middle of autumn.The trees turn here about the end of October.The first frost comes after the middle of autumn.We start heating after the end of October.(Coulthard, 1977)

The aim of the analysis is to isolate units of text that are distributionally equivalent though not necessarily similar in meaning, i.e. equivalences that have validity for the text alone. From the first two sentences above one establishes the equivalence of “the middle autumn” and “the end of October”, not because they are similar in meaning but because they share an identical environment “the trees turn here”. The next step is to carry over the equivalences derived from the first two sentences into the next two and this allows us to equate “the first frost comes” with “we start heating”. Thus, in terms of equivalence classes, all four sentences have identical structure, class X followed by class Y.

In the twenty years after this paper was published no one adapted or developed the method, probably because any purely formal analysis above the rank of sentence is impossible. Just as Harris himself observes, it is impossible to describe the structure of a paragraph in terms of sequences of sentences of particular types — the constraints above sentences are stylistic rather than grammatical, and organization and sequence can only be described in semantic terms. For example, a text that consists simply of a sequence of clauses with no obvious patterning from a grammatical viewpoint may be seen to consist of a sequence of question-answer pairs from a semantic or functional viewpoint.

In addition, he is quite mistaken in not taking into account the meaning of text when revealing the global structure, although meaning does play an important part in endowing a discourse with a global structure. And what is more, he turns a blind eye to the social nature of discourse, let alone the impact of contextual factors upon text. Therefore, his analysis remains in general a formal and static one, which fails to give a true reflection of the full picture. This present paper contends that without taking into consideration the contextual factors and their roles in the production of texts, a full and convincing analysis of texts is unlikely to be attained, however delicate the formal analysis of them may be.1.1.2 Insufficiency of Halliday's exploration of context and text

As mentioned above, it is the London School who first arouses people's attention to the relationship between context and text. Halliday divides context into two kinds: context of situation and context of culture. In order to give a detailed analysis of the impact of context upon text, he further divides context of situation into three components: field, tenor and mode. And he argues that each of these components has a bearing on the text. However, he confines his studies to the one-to-one mapping between these contextual factors and the three meanings of text. According to Halliday, field determines the ideational meaning of a text, tenor determines the interpersonal meaning of a text, and mode determines the textual meanings of a text, as shown in the following figure.Figure 1-1　Halliday's one-to-one correspondence between context and text

The current author argues that besides the one-to-one correspondence between context of situation and text, there is also a crisscross between the contextual factors and text. In other words, field can also exert its influence upon the interpersonal meaning of a text; tenor can also exert certain influence upon textual meaning; and so on.1.1.3 Significance of this current study

Owing to the inadequacy of the formal studies of language in revealing the social nature of language, and also thanks to the indispensability of the exploration into the social nature of language, it is strongly contended here that a full exploration of text, i.e. the language used in everyday communication setting, is unlikely to be attained if contextual factors are ignored. On this ground, this paper is oriented towards an exploration of the impact of context upon text. The significance of this current study can be summarized as follows:

i. It will highlight the fact that context contributes to, and conditions, texts so that it can make people more keenly aware of the mechanism of text production.

ii. Aside from the theoretical significance of this study, it is also abundant in pedagogical implications by providing many implications for the teaching of writing, such as arousing the writers’ awareness of contextual factors in composing an article and enhancing their consciousness of the feedback of text upon context and thus greatly facilitates the teaching process.

iii. Based upon detailed studies of texts, this research will provide some guidance and suggestions for the future development of text analysis.1.2 Research Methodology

The qualitative methodology will be adopted in this research. The findings and conclusions will be based on the analyses of the collected data.1.3 Collection and selection of corpora1.3.1 Data collection and selection

In his Doing Conversation Analysis, Have (1999: 47) points out that there are three strategies of getting spoken corpora: copying radio and TV broadcasts; using existing recordings; or making one's own recordings. By the same token, it is contended that there are three ways of collecting the corpora to be analyzed in this research, namely: copying the materials that have been used by other discourse or conversation analysts; using existing materials; and making my own corpora. The most important criterion by which the materials are selected for analysis is authenticity. In order to maintain the authenticity of language materials, a fairly large amount of spoken data here is taken from dramas, plays, and everyday communication, with a limited number of texts adopted from textbooks of discourse analysis, conversation analysis and text analysis.1.3.2 Justification for the adoption of spoken prose as corpora

Spoken prose refers to the dialogue or conversation writers create in novels, plays or dramas. Admittedly, spoken prose is a concentrated and condensed experience extracted from a continuum which is never going to be presented in its entirety to the audience. If compared to real-life talk, dramatic texts generally exhibit no paralinguistic features; similarly, overlaps, false starts and unintended obscurities are absent and, in general, interaction in fiction seems characterized only by medial exchanges, while opening and closing sequences are generally left out (cf. Caldas-Coulthard, 1988). In Burton's words, “drama scripts are markedly tidied-up versions of talk, adhering closely to the two rules that Sacks declares are the most basic conversational rules available, that is ‘one party speaks at a time’ and ‘speaker change recurs’”(Sacks, 1970: 115). Besides, dramatic dialogue is ‘overdetermined’ in a variety of ways “given the audience presence, and the necessities of presentation and participation that this involves” (Herman, 1991: 99). Differently from real talk, speaker's alternation and turn-taking are not locally managed but totally author-controlled (cf. Caldas-Coulthard, 1992). The workable hypothesis in this research is that, notwithstanding obvious differences, spoken prose such as dramatic discourse — especially in the case of modern theatrical works — expresses a close relation with real-life talk and is forged from life itself (Piazza, 1999). Hence the interaction portrayed in a drama reflects the way its author envisages the mechanisms of interpersonal exchange in real life.1.3.3 Transcription of real-life spoken data

In addition to the spoken data taken from dramas and plays, there are also a lot of real-life spoken data. The transcription convention for these real-life corpora is shown in the following table.Table 1-1　Transcription convention1.4 Theoretical framework of this study

This study mainly takes the theory concerning the relationship between text and context in Systemic Functional Linguistics as the basic theoretical framework. With a history of over forty years, systemic functional linguistics has undergone several historical periods, namely Firthian period, Hallidayan period and Matthiessen period. And synchronically as a worldwide influential school of linguistics today, it is also divisible into quite a large number of sub-schools with different scholars as their representatives. Despite their minute differences in their research foci and other minor aspects, they are all systemic functional linguists. They have conducted and are conducting their studies within the theoretical framework of systemic functional linguistics. In the present paper, the current author will mainly adopt Halliday's and Hasan's study in this area as the theoretical framework. However, the analyses conducted in this research will not be confined to Halliday's and Hasan's delineation of the relationship between text and context. Instead, many linguists related or similar to this school will also be drawn on so as to facilitate our research.1.5 Structure of this research

In the following we will talk about the arrangement for the remainder of the research.

Chapter 2 is an overview of the researches of text and context and their mutual relationship, which has been carried out by different linguists of different schools, with a purpose to give a detailed introduction to the existing studies on the current topic. Also in this chapter a clear redefinition of several technical terms, such as text, context many others is forged. In this chapter the author argues that Halliday's three contextual variables, namely field, mode and tenor will be taken as tools in exploring the influence of context upon text.

Chapter 3 addresses the impact of field upon text. After summing up the existing studies on field, the current author proposes a new classification for it in terms of subject matter, which is different from the prevalent dichotomy of field into technical and non-technical. Then the author explores the way in which different aspects of field can exert their impact on text.

Chapter 4 deals with tenor. By comparing all the existing studies in this area, the current author takes Poynton's classification as the model for the exploration of the impact of tenor on text.

Chapter 5 investigates the impact of mode upon text. In this chapter, the current author points out that mode, as a means by which the message is put across, should be regarded as an objective entity in exploring its impact upon text.

Chapter 6 deals with the interaction between contextual factors upon text and how this interaction can result in changes in text. By interaction is meant that contextual factors not only exercise their influence upon text respectively, but also influence each other as well. For instance, a minute change in field may interfere with tenor; a change in mode affects field, and so on.

Chapter 7 summarizes the current research. While highlighting its contributions, the current author also points out its shortcomings and insufficiencies and offers some suggestions for future research on relevant topics.Chapter 2 Literature Review on Context of and Text2.1 Introductory remarks

It would be impossible to make a profound study of the impact of context upon text without a clear definition of text and context. For this reason, in order to pave the way for a detailed study of the impact of context upon text, the first chapter is dedicated to the discussion of their definitions. In the first part, we will deal with the different definitions of context by different linguists and philosophers, and then try to give a workable definition of my own. In the second part of this chapter, we will deal with the different existing definitions of text, and conclude with a workable definition for it.2.2 Context

Context is not something we find in Nature; there are many different ways of using the term “context” in philosophy, linguistics, psychology, theory of communication, problem solving, cognitive science, artificial intelligence, etc. We speak of “context principle” in Frege and Wittgenstein; we speak of “context of utterance” in pragmatics; we speak of “context sensitive” grammars in linguistics, and we also speak of “linguistic context” and “non-linguistic context”.

At present it does not seem possible to give a single, precise, technical definition of context, because the term “context” can mean different things within different research paradigms. The fact that so many investigators recognize the importance of context and are actively involved in trying to unravel how it works is precisely why this concept provides such a productive focus for the current study. In this part, we will make a brief introduction to the different definitions of context given by different linguists and then touch upon the philosophical division of two kinds of contexts, i.e. objective context and cognitive context, and will discuss the cognitive context in more detail in light of the rapid development of cognitive linguistics with a purpose of revealing the insufficiency of cognitive context and its strong dependence upon objective context.2.2.1 Context defined by different linguists2.2.1.1 Malinowski's definition of context

In linguistics, context originally means the accompanying text, the wording that comes before and after what is under attention. In the nineteenth century it was extended to things other than language, both concrete and abstract: the context of the building, the moral context of the day; but when people are talking about language, it still refers to the surrounding words, and it is in modern linguistics that it comes to refer to the non-verbal environment in which language is used. When context is generally used with the new meaning, Catford, according to Halliday (1999), coined a word “co-text” to refer to the verbal environment.

The acquisition of the new meaning of the term “context” was attributed to Bronislaw Malinowski. In his writing in 1923 about what was referred to as a “primitive” (that is, unwritten) language, he wrote:

“In a primitive language, the meaning of any single word is to a very high degree dependent upon its context…. An expression such as we paddle in place demands the context of the whole utterance, … and this latter again, becomes only intelligible when it is placed within its context of situation, if I may be allowed to coin an expression which indicates on the one hand that the conception of context has to be broadened and on the other hand that the situation in which words are uttered can never be passed over as irrelevant to the linguistic expression” (Malinowski, 1923: 306).

Ten years or so later, Malinowski had given up his view that this was a special feature of “primitive” languages; in 1935, he said that “all languages were alike in that the real understanding of words is always ultimately derived from active experience of those aspects of reality to which the words belong” (Malinowski, 1935: 58). By this time Malinowski had extended the notion of context even further: over and beyond the context of situation lies “what we might call the context of culture”, so that “the definition of a word consists partly of placing it within its cultural context” (Malinowski, 1935: 18). What this means is that language considered as a system — its lexical items and grammatical categories — is to be related to its context of culture; while instances of language in use — specific texts and their component parts — are to be related to their context of situation. Both of these contexts are outside language itself.

However, Malinowski did not elaborate on what components both context of situation and context of culture include. This task was partially completed by J. R. Firth and M. A. K. Halliday.2.2.1.2 J. R. Firth's definition of context

Malinowski's notion of context was taken up by his younger colleague J. R. Firth, a linguist and the founder of London School. Firth saw the possibility of integrating this notion of the situation as a kind of context, into a general theory of language. Firth was also interested in spoken language; but not as something quaint or exotic like rural dialects and aboriginal languages. On the contrary, he was concerned with the typical — what he referred to as “typical texts in their contexts of situation” (Firth, 1957: 224) by which people enacted their day-to-day interpersonal relationships and constructed a social identity for themselves and the people around them. A text was an object of theoretical study in its own right; and what Firth did was to map the notion of “context of situation” into a general theory of language. Linguistic analysis, Firth contends, is a study of meaning, and meaning could be defined operationally as “function in context”; so to study meaning one has to take into consideration context of situation, since it is in the context of situation that the text as a whole can be contextualized.

Firth sets up the following framework for the description of context of situation:

1) participants: their statuses and roles;

2) action: the participants’ verbal and non-verbal action;

3) other relevant features of the situation: the surrounding objects and events;

4) effects of the verbal action.

Firth made very little use of the idea of context of culture, although he considered that a language was “embedded in the life and culture of its speakers”, because he was actually very skeptical about general notions like “the language” and “the culture” (Robins, 1963: 17).2.2.1.3 Halliday and Hasan's notion of context

Halliday and Hasan (1985: 46, 99) distinguish between material setting and context of situation, with the former referring to the physical environment in which a text might be created, where speaking, listening, writing or reading might take place, and the latter referring to the immediate environment relevant to the text. The degree of overlap between the two is often at its lowest in writing, whilst in speech, particularly where the role of language is ancillary, the overlap is at its highest.

The context of situation is an abstract representation of the environment in terms of certain general categories relevant to the text. Given the context of situation, the hearer or reader can make guesses at what the speaker is going to mean. The context of situation, according to Halliday and Hasan, can be represented as a complex of three dimensions: the ongoing social activity, the role relationships involved, and the symbolic or rhetorical channel. They are respectively referred to as field, tenor and mode.

Field refers to the nature of the social action that is taking place, in which the text is embedded. It has recognizable meaning in the social system: typically a complex of acts in some ordered configuration, and in which a text is playing some part, and including the subject matter as one special aspect. It is what the participants in the context of situation are actually engaged in doing.

Tenor refers to who is taking part, to the nature of participants, their social statuses and roles. It is concerned with the cluster of socially meaningful participant relationships, both permanent attributes of the participants and role relationships that are specific to the situation, including speech roles, those that come into being through the exchange of verbal meanings. This set of relations includes levels of formality as one particular instance.

Mode refers to what part the language is playing, what it is that the participants are expecting the language to do for them in that situation: the symbolic organization of the text, the status that it has, and its function in the context, including the channel (spoken, written, or some combination of the two) and also the rhetorical mode, what is being achieved by the text in terms of such categories as persuasive, expository, didactic and the like.

With regard to the notion of context of culture, Halliday and Hasan further develop Malinowski's idea. While Malinowski uses it to refer to the total cultural background, Halliday and Hasan use it to refer to the institutional and ideological background that assigns value to the text and constrains its interpretation. Since situations are also culturally constructed, the constraining force of the context of culture is highlighted.2.2.1.4 Roman Jacobson's definition of context

Roman Jacobson (1960) points out that linguistics, just like poetics, is concerned with “the questions of relations between the word and the world”. Linguistics is likely to “explore all possible problems of relations between discourse and the ‘universe of discourse’”. According to him, for any speech community, for any speaker, there exists a unity of language, but this overall code represents a system of interconnected subcodes; each language encompasses several concurrent patterns which are each characterized by a different function. And he proposes six basic functions of language: emotive function, conative function, referential function, poetic function, phatic function and metalingual function. He contends that language must be investigated in all these functions. But an outline of the functions demands a concise survey of the constitutive factors in any speech event, in any act of verbal communication. The ADDRESSER sends a MESSAGE to the ADDRESSEE. To be operative the message requires a CONTEXT referred to (referent in another nomenclature), seizable by the addressee and either verbal or capable of being verbalized; a CODE fully, or at least partially, common to the addresser and addressee; and finally, a CONTACT, a physical channel and psychological connection between the addresser and the addressee, enabling both of them to enter and stay in communication. All these factors inalienably involved in verbal communication may be schematized as follows (Sebeok ＆ Sebeok, 1960: 363):Figure 2-1　Contextual Factors

He argues that each of these six factors determines a different function of language. To be specific, context determines the referential function; addresser the emotive function; addressee the conative function; contact the phatic function; message the poetic function; and code the metalingual function. We can schematize it as the following (Sebeok ＆ Sebeok, 1960: 357):Figure 2-2　Functions of Language

From here it can be clearly seen that for Roman Jacobson, context is an entity or a set of entities that is tangible both to the addresser and to the addressee, outside of language. But lamentably to our regret, he did not elaborate upon the context and never made any attempt to particularize it by making a detailed classification.2.2.1.5 D. Hymes's definition of context

In his Foundations in Sociolinguistics, D. Hymes (1974: 53-62) provides us with a more detailed specification of context, which summarizes contextual variables mnemonically with the term SPEAKING, an acronym coined by taking the first letters of the following eight words standing for contextual variables: situation, participants, ends, act sequence, key, instrumentalities, norms, genres. That SPEAKING consists of eight words does not mean that there are only eight context variables. Rather, the components of context total up to sixteen, as argued by Hymes. These sixteen components are: message form, message content, setting, scene, speaker/sender, addresser, hearer/receiver/audience, addressee, purposes as outcomes, purposes as goals, key, channels, forms of speech, norms of interaction, norms of interpretation, and genres. In the following we will introduce them one by one.

Message form is important in that it can sometimes condition and control content, as the old saying goes, “The manner in which something is said is part of what is said”.

Message content refers to the topic and the change of the topic; in other words, content is concerned with what is being talked about and when what is being talked about has changed.

Message form and message content are central to the speech act and the focus of the text; they are also tightly interdependent. Thus they are dubbed jointly as components of act sequence［A］.

Setting refers to the time and place of a speech act and, in general, to the physical circumstances.

Scene, different from setting, designates the “psychological setting”. Within a play on the same stage with the same stage set the dramatic time may shift: “ten years later”. In daily life, the same persons in the same setting may redefine their interaction as a changed type of scene to analysis of speech genres.

Setting and scene may be linked as components of situation［S］.Speaker, or sender.Addresser.Hearer, or receiver, or audience.Addressee.

These four components are mainly concerned with the participants of the communication. The difference in the participants will inevitably lead to the difference of their use of language. The various components may be grouped together as participants［P］.Purposes as outcomes and purposes as goals.

At first glance, these two terms refer to the same thing, namely what is expected of a speech event. However, they are different in that they are approached from different perspectives. D. Hymes cited a very convincing example to illustrate this point. In a speech event in Venezuela, the two parties reach a decision to marry a woman to a man after a negotiation. Here this event has the same outcome for both parties, i.e. the accomplishment of a marriage contract. But their goals can be widely discrepant. For the fiancé’s family, the goal may be to gain a handsome amount of dowry from the fiancée's family, while the goal of the latter may be just to get rid of her for her age.

In short, these two aspects of purposes can be grouped together by exploiting a homonymy, ends in view (goals) and ends as outcomes［E］.

Key refers to the tone, manner, or spirit in which an act is done. It corresponds roughly to modality among grammatical categories. Acts otherwise the same as regards setting, participants, message form, and the like may differ in key, e.g. mock: serious or perfunctory: painstaking. Key is often conventionally ascribed to an instance of some other component as its attribute; serious, e.g. may be the expected concomitant of a scene, participant, act, code, or genre. The significance of key is underlined by the fact that, when it is in conflict with the overt content of an act, it often overrides the latter (as in sarcasm).

Channels. By choice of channels is meant choice of oral, written, telegraphic, semaphore, or other medium of transmission of speech.

Forms of speech. Even where there is a single language present in a community, that language will be organized into various forms of speech. Three criteria seem to require recognition at the present time: the historical provenience of the language resources; presence or absence of mutual intelligibility; and specialization in use. The criteria seldom coincide. Language and dialect are suggested for the first; codes for the second; and varieties and registers for the third. One speaks of the English language and dialects of English, wherever forms of speech are found whose content is historically derived from the line of linguistic tradition we call “English”. The different dialects are not always mutually intelligible and their social functions vary considerably around the world, from childhood vernacular to bureaucratic lingua franca. Code suggests decoding and the question of intelligibility. Unintelligibility may result when speech is made in a language historically unrelated to one's own speech. In short, some forms of speech are derived from others by addition, deletion, substitution, and permutation in various combinations. Finally, forms of speech are commonly specialized to uses of various sorts. Register has become familiar in English linguistic usage for reference to specific situations; and varieties, or “functional varieties”, have been used in American linguistics in relation to broad domains (e.g. vernacular vs. standard). The criteria of provenience and intelligibility have to do with sources and characteristics of the criterion of use with the functional organization, of the forms of speech.

Channels and forms of speech can be joined together as means or agencies of speaking and labeled, partly for the sake of the code word, partly with an eye on the use of the term instrumental in grammar, as instrumentalities［I］.

Norm of interaction refers to the rules governing speaking. It refers specifically to the behaviors and proprieties that attach to speaking — that one must not interrupt; that turns in speaking are to be allocated in a certain way; and so on. It implicates analysis of social structure and social relationships in a community.

Norm of interpretation concerns itself with the question how the utterances and gestures in communication are evaluated and assessed by the others. The norm of interpretation implicates the belief system of a community.

These two kinds of norms may be grouped together (N).

Genres. By genres are meant categories such as poem, myth, tale, proverb, riddle, curse, prayer, oration, lecture, commercial, letter, editorial, etc.

After putting the code words together, we get the code word SPEAKING, which represents all of the contextual variables, namely settings, participants, ends, act sequences, keys, instrumentalities, norms, and genres.2.2.1.6 Elinor Ochs's definition of context

In an attempt to specify the basic parameters of context, Ochs (1979: 1) notes that the analysts must use as a point of departure “the social and psychological world in which the language user operates at any given time”. According to him, the notion of context should cover the following parameters:

Setting, i.e. the social and spatial framework within which encounters are situated. Neither of them is fixed, immutable, and simply “out there”. Instead, these phenomena, and the very real constraints they provide, are dynamically and socially constituted by activities (talk included) of the participants that stand in a reflexive relationship to the context thus constituted.

Behavior environment, i.e. the way that participants use their bodies and behavior as a resource for framing and organizing their talk. Through spatial orientation and posture, participants both display their continuing access to the actions of others present and frame the talk they are producing. Of particular importance is the way in which postural framing establishes the preconditions for coordinated social action by enabling participants to both project and negotiate what is about to happen.

Language as context. This refers to the fact that talk itself both invokes context and provides context for other talk; in other words, talk itself constitutes a main resource for the organization of context. Gumperz (1982) gives a detailed description of the seminal notion of contextualization cues and the theoretical motivation for such a concept, and even applies the concept to the analysis of specific data.

Extrasituational context. It refers to background knowledge that extends far beyond the local talk and its immediate setting.2.2.1.7 CHEN Wangdao's definition of context

Chinese scholars have also contributed a lot to the definitions of context, with Chen Wangdao as a typical example. In his An Introduction to Rhetoric, he pointed out that language use is intricately entangled with the outside factors, such as the social experience of the language user, the political stance, his or her worldview and knowledge of natural sciences. When talking about the specific factors of a context that play a decisive role in the process of language using, particularly in composing a text, he laid special emphasis upon the following six he's (wh's). The first is why the text is written. This refers to the purpose of the writing of the text. For example, is it intended to persuade somebody into or to dissuade them from doing something? Or is it intended to argue with others about a certain subject? The second is what the text is about. Is it about everyday trifles or about an academic discussion? The third is at whom the text is aimed. This refers to the relationship between the readers and the writer. Are the readers literary youth or just people on the street? The fourth is where the text is produced. Is the text written in a city or in the countryside? The fifth is when the text is produced. In what year and what month is the text written? The sixth is in what manner the text is organized. For example, how the plots of the text are arranged and developed.

Chen Wangdao is not alone in this area. As early as in the Northern and Southern Dynasties, Liu Xie revealed that writing is closely related to context in his well-known Wenxin Diaolong. In the Tang Dynasty, Bai Juyi also emphasized that a text should respond to the changes in the outside world, and a poem should echo the requirements of the time.2.2.2 Cognitive context

With the rapid development of cognitive linguistics, the concept of cognitive context has drawn more and more attention. However, the difference between objective context and cognitive context is not a recent argument; it has existed for a long time in philosophy. Here in this section a brief introduction will be made to the difference between objective context and cognitive context in philosophy; then the concept of cognitive context in linguistics will be elaborated on. After that we will reveal the insufficiency and the lack of interpretative power of cognitive context to highlight its strong dependence upon objective context.2.2.2.1 Philosophers’ statement of objective context and cognitive context

In philosophy, two paradigmatic positions representing these two different points of view on context are the model theoretical tradition developed in the works of Kaplan, Lewis, etc. and the artificial intelligence tradition developed in the works of McCarthy, Giunchiglia, etc (Penco, 1998). These two different conceptions of context can be summarized as follows:

(a) Context is a set of features of the world and we can express it as:〈time, place, speaker,...〉and

(b) Context is a set of assumptions on the world and we can express it as:〈axioms, rules〉

Some relevant quotations are as follows

On (a): “context is a package of whatever parameters are needed to determine the referent ... of the directly referential expressions”; “each parameter has an interpretation as a natural feature of a certain region of the world” (Kaplan 1989);

On (b) “context is a group of assertions (under entailment) about which something can be said” (McCarthy 1993); context is “a theory of the world which encodes an individual's perspective about it” (Giunchiglia 1993).

In “Afterthoughts” Kaplan speaks explicitly of the metaphysical point of view in describing contexts, while in “Notes on formalizing contexts” McCarthy uses a notion of context that leads towards the idea of a subjective point of view on the world (Giunchiglia). Given these differences, we can distinguish these two different conceptions of contexts, namely:

a) Context as an objective, metaphysical state of affair.and

b) Context as a subjective, cognitive representation of the world.

In other words, we have two different interpretations of what a context is: features of the world or representations of features of the world; we might call the two different interpretations:

- “objective” or “metaphysical” (ontological) theory of context.and

-“subjective” or “cognitive” (epistemic) theory of context.

At first sight, the relation between the two contexts is apparent and may be presented in a naïve conception: cognitive context gives the beliefs on objective context. There is nothing wrong with this conception about the relations between objective and cognitive contexts. However, in this conception, some problems remain unsolved. We might be unable in principle to know what the objective context is. Different standards of measurement or standards of precision may be in conflict with each other and may make it impossible to determine in a unique way what we have to take as objective context. Besides, we might be unable to know where and when something happened or some utterance has been produced. Just think of situations where people may rely only on memory and testimony and you can verify this point.

Superficially, the distinction of objective and cognitive contexts is a natural and well grounded one in that it reproduces a general distinction we normally make between representation and what is represented. However, notwithstanding the intuitive distinction between the two above-mentioned theories of context, the line of demarcation is not so easy to draw. One of the reasons is that a concept born to represent the objective features of the world inescapably has a natural tendency to become subjective (even if somebody might say that the background knowledge or the speaker's beliefs still belong to the objective world). On the other hand, the subjective context has to deal with objective states of affair, and with the objective reference of any element of the cognitive context as well. If context is intended as a set of assertions, these assertions will certainly be about something objective; therefore, it is natural to think that they should give a representation not only of subjective points of view, but also of what is represented by them.2.2.2.2 Context in cognitive linguistics

Cognitive linguistics stems from the discontent with the modularized study of language, as proposed by de Saussure. Langacker (1987) points out that meaning could not be identified as a separate level, autonomous from the study of other levels of grammar. Ostyn (1993) further develops this idea and proposes that it is impossible to separate linguistic knowledge form extra-linguistic knowledge. Cognitive linguists often use the term cognitive context to refer to the conventional or stereotypic representations of “knowledge of the world” for the interpretation of discourse. These representations, found in psychological and computational approaches to discourse understanding, are mainly used to account for the type of predictable information a writer/speaker can assume his hearer/listener has available whenever a particular situation is described. The technical terms often used to refer to these representations are scenario, script, schema, frame, plan, mode and so on. They generally refer to the readers’ knowledge of the outside world related to the understanding of the text concerned, in spite of their large differences as claimed by different linguists.2.2.2.3 Problems with the theory of cognitive context

Despite its seemingly persuasive power, cognitive context falls short of expectations of in many aspects. First of all, only when a text is presented to the readers can cognitive context play its part in facilitating their understanding. Put differently, cognitive context only facilitates the understanding of the existing texts and it does not touch upon the production of text. Second, cognitive context fails on one of its major concepts, i.e. relevance, which means the reader/speaker has the ability to retrieve their related knowledge to help understand the ongoing texts. The problem is why these parts, instead of other parts, of knowledge are mobilized. In other words, what triggers the mobilization of certain part of knowledge in the interpretation of the text? It is argued here the mobilization of knowledge or cognitive context is still determined by context as defined by Halliday and Hasan's definition.2.3 Text2.3.1 Text, discourse, conversation

Since the publication of Harris's Discourse Analysis in 1952, the study of texts has always been on the upswing. With the publication of a considerable amount of related works on text in nineteen sixties, for example, Beaugrande and Dressler's Introduction to Text Linguistics, Brown and Yule's Discourse Analysis, the study of text began to draw more and more attention of linguists and thus became the research focus in linguistics. In China, the research on texts dates back to the Han Dynasty, when Liu Xietouched upon “chapter” (zhang) and “sentence” (ju) in his well-known Wenxin Diaolong. In line with their different research foci, different linguists select different terms to facilitate their study. Correspondingly text linguists usually entitle their works on text analysis discourse analysis, text analysis, text linguistics, conversation analysis and so on to show their different research orientation and divergent emphases. Before proceeding to a workable definition, we will have a survey of what definitions the linguists have given to their research object.2.3.1.1 van Dijk's definition of text and discourse

In his Text and Context: Explorations in the semantics and pragmatics of discourse, van Dijk (1977: 5) takes both text and discourse as his research object. He argues that TEXT is a “larger unit” composed of sequences of sentences that are dependent upon each other in meaning, and the term “will here be used to denote the abstract theoretical construct underlying what is usually called DISCOURSE”. In other words, for him, text is an abstract theoretical unit of discourse, which is used to refer to what actually happens in spoken communication. He emphasizes that semantic macrostructure is a criterion in judging whether an actually happening discourse event is a text or not.2.3.1.2 Halliday and Hasan's definition of text

The term that Halliday and Hasan choose for their research object is text. And they use the notion of text to refer to any passage, spoken or written, of whatever length, that forms a unified whole. To expound this point, we quote the following three paragraphs:A text may be spoken or written, prose or verse, dialogue or monologue. It may be anything from a single proverb to a whole play, from a momentary cry for help to an all-day discussion on a committee.A text is a unit of language in use. It is not a grammatical unit, like a clause or a sentence; and it is not defined by its size. A text is sometimes to be envisaged to be some kind of super-sentence, a grammatical unit that is larger than a sentence but is related to a sentence in the same way that a sentence is related to a clause, a clause to a group and so on; by constituency, the composition of larger units out of smaller ones. But this is misleading. A text is not something that is like a sentence, only bigger; it is something that differs from a sentence in kind.A text is best regarded as a semantic unit: a unit not of form but of meaning. Thus it is related to a clause or sentence not by size but by realization, the coding of one symbolic system in another. A text does not consist of sentences; it is realized by, or encoded in, sentences (Halliday ＆ Hasan, 1976:1).

From their definition of text and the relevant quotations, it can be seen clearly that Halliday and Hasan use the term text to cover both the spoken and the written forms of human communication. With regard to the length of text, they argue that a text is not limited in length and can be as short as a cry for help or as long as a novel.2.3.1.3 Brown and Yule's definition of discourse and text

In their book Discourse Analysis, Brown and Yule define discourse as “language in use” (1983:1) and text as the verbal record of a communicative act (1983: 6). It is obvious that they use the term discourse to refer to the actually happening spoken communication. When talking about the criterion of a text, they pointed out that a text must have a topic (1983: 65).2.3.1.4 Beaugrande and Dressler's definition of text

According to Beaugrande and Dressler (1981: 3), a text is defined as a communicative occurrence which meets seven standards of textuality. If any of the standards is not considered to have been satisfied, the text will not be communicative. Hence, non-communicative texts are treated as non-texts.

The following are the seven criteria, or textualities in their terms, they think a text must meet: cohesion, coherence, intentionality, acceptability, informativity, situationality, and intertextuality.

Cohesion concerns the ways in which components of the surface text. i.e. the actual words we hear or see, are mutually connected within a sequence. The surface components depend on each other according to grammatical forms and conventions, such that cohesion rests upon grammatical dependencies.

Coherence, the second standard, concerns the ways in which the components of the textual world, i.e. the configuration of concepts and relations which underlie the surface text, are mutually accessible and relevant. A concept is definable as a configuration of knowledge (cognitive content) that can be recovered or activated with more or less unity and consistency in the mind. Relations are the links between concepts that appear together in a textual world: each link would bear a designation of the concept it is related to.

The third standard of textuality, intentionality, concerns the text producer's attitude as regards whether the set of occurrences constitutes a cohesive and coherent text instrumental in fulfilling the producer's intentions, e.g. to distribute knowledge or to attain a goal specified in a plan.

The fourth standard of textuality, acceptability, concerns the text receiver's attitude that the set of occurrences should constitute a cohesive and coherent text having some use or relevance for the receiver, e.g. to acquire knowledge or provide co-operation in a plan.

The fifth standard of textuality, informativity, concerns the extent to which the occurrences of the presented text are expected or not. Every text is at least somewhat informative: no matter how predictable form and content may be, there will always be a few variable occurrences that cannot be entirely foreseen.

The sixth standard of textuality, which can be designated situationality, concerns the factors that make a text relevant to a situation of occurrence.

The seventh standard of textuality, intertextuality, concerns the factors that make the utilization of one text dependent upon knowledge of one or more previously encountered texts. Intertextuality is, in general fashion, responsible for the evolution of text types with typical characteristics. Within a particular type, reliance on intertextuality may be more or less prominent.

Beaugrande and Dressler argue (1981: 11) that these seven standards function as a constitutive principle of textual communication: they define and create the form of behavior identifiable as textual communicating, and if they are defied, that form of behavior will break down.2.3.1.5 Coulthard's definition of discourse

Published in 1977, Malcolm Coulthard's An Introduction to Discourse Analysis is the first book to conduct discourse analysis in a serious manner. He points out that discourse does not consist simply of a string of grammatically well-formed utterances or sentences. Although he agrees with the distinction made by Widdowson (1973) between contextualized data and decontextualized data, such as usage/use, sentence/utterance, locution/illocution, text/discourse, etc., he uses the term “discourse” to cover both spoken data like conversation and written data such as plays.2.3.1.6 Fowler's notion of text and discourse

In his Linguistics and the Novel, Roger Fowler (1983) contends that the frame for prose fictions runs parallel to the frame for sentences. And he makes the following comparison between sentences and prose fictions:Table 2-1　A comparison between sentences and prose fictions

As can be seen from this table, Fowler uses the term “text” to stand for textual surface-structure, the most “perceptible”, “visible”, dimension of a work, close to the literary critic's sense of a text as a formal object. Text, according to Roger Fowler (1983: 49), can be thought of as a sequence of phrases and sentences leading a reader's attention along the left-to-right “information structure” of a passage — working progressively or disruptively to allow him to retrieve the meaning from the surface structure in an ordered (or disordered) sequence. With regard to “discourse”, he refers to the aspects like “point of view”, “attitude”, “world view”, “tone” and many other modality factors, which indicate the author's beliefs, the character of his thought-processes, the types of judgment he makes for the narrator and the characters within the fiction and the whole network of interpersonal relationships between the author, the characters and the intended reader. And by content he means plot, character, setting, theme and so on.2.4 Text and Context in this research

If a detailed and convincing exploration into the impact of context of situation upon text is to be attained, it is of vital importance to give a clear definition to the two terms in question. For this reason, the following will be devoted to the discussion of these two concepts, namely text and context.2.4.1 Text in this research

In this research, we will use the term text to refer to human communicative occurrences, both spoken and written. In other words, the term text in this research will cover discourse, conversation and text. With regard to the criteria of a text, it is argued that a text must satisfy the following five criteria: informativity, semantic compatibility, associability, cognitivity, and length felicity (Wang, 2001). In what follows these criteria will be addressed one by one.2.4.1.1 Informativity

Conventional wisdom has it that everything, in human society and in Nature, is a unity of form and content. Text, as a tool to transmit ideas and to reflect on objective reality, is no exception. For this reason, it must be in possession of a certain amount of information. And we often utilize the term “empty verbiage” to criticize the texts that fail to supply some new information. The informativity of text is obviously shown in people's understandings of the concept of coherence.

2.4.1.1.1 Coherence embodied as the topic of a text

van Dijk (1977: 95) holds that coherence can be classified into two layers: linear or sequential local coherence and macrostructure. The former refers to the coherence relations holding between propositions expressed by composite sentences and sequences of sentences. It can be further divided into three types: 1) The order of facts or sentences. Except when special pragmatic needs arise to change the order of sentences and facts, the presentation of the events and facts usually corresponds to the factual order of actions and events. For example, in the description of states, where the facts all exist at the same time, it will be assumed that a normal ordering corresponds with the general-particular and the whole-part relations between facts. Disobeying these rules will bring about the loss of coherence of a text. 2) Explicit information and implicit information. In view of the complexity of social life and relationships between entities, it is not possible for human beings to spell out in the text all the information relevant to the entities or states in the objective world. That is why natural language discourse, unlike formal discourse, is not fully EXPLICIT. Relationships between sentences or propositions may exist without being expressed. To put it another way, in a text, some details are explicit and some are implicit. A normal text is usually incomplete with some details, which we call the missing links and which can be inferred from the explicit details. The inadequacy or overabundance of details will inevitably result in the lack of coherence. 3) The sentential information structure in texts. Human beings are inclined to acquire new information from given information. Thus in order to express the continuity of a discourse, each sentence will in principle express this relation between OLD and NEW information, viz as topic and comment respectively. Reversing the order of given and new information will cause the lack of coherence.

Macro-structure refers to the general proposition of a text which can be expressed by a sentence. The macro-structure determines the GLOBAL or overall coherence of a discourse and are themselves determined by the linear coherence of sequences (van Dijk, 1977: 95). The macro-structure of a text can be shaped by condensing all its propositions with the help of four macro-rules. Macro-rules are in fact rules of information reduction. At first, van Dijk (1977: 144) insisted on four macro-rules, namely the deletion rule of accidental information, the deletion rule of inductively recoverable information, the generalization rule, and the construction rule. But later on, he (1981: 203) reduced these rules to three: the Deletion Rule, the Generalization Rule and the Construction Rule.

2.4.1.1.2 Coherence embodied as a unified rhetorical structure

A text is a unified whole composed of several components, each of which performs its specific function. The analysis of a text usually proceeds from the bigger rhetorical components to the smallest rhetorical structures, so that the base functional structure can be revealed (Mann ＆ Thompson, 1992). This rhetorical structure so shaped reflects the subjective rhetorical configuration of the author. Through an analysis of Zero Population Growth Fund Raising text, Mann ＆ Thompson identifies fourteen rhetorical relations between the modules of texts: background, means, motivation, proof, elaboration, comparison and so on. The modules that embody these rhetorical functions are all subsidiary units, modifying the bigger units. According to Mann, a text is coherent if a unified rhetorical structure can be shaped after an analysis; otherwise it is not coherent.

2.4.1.1.3 Coherence as the relevance in form, meaning and pragmatic force

According to Reinhart (1980), coherence comprises three elements: connectedness, consistency, and relevance. By connectedness he means that the sentences in a text are interconnected with each other in semantics and grammar, for example, the sentences have the same reference and they are usually connected with connectives. Consistency refers to the fact that there is no contradiction between the propositions expressed by these sentences and they are true to a certain extent. By relevance he means that a text should be related to the context, the sentences in a text should be related to each other, and the sentences should all be related to the general topic of the text.

2.4.1.1.4 Coherence as a reflection of common sense and experience

Some linguists study text from the perspective of psychology, and they hold that understanding of words and sentences, inference making about the text, etc. cannot be divorced from the people's existing knowledge. To them, the meaning of a text is the function between the linguistic input and the knowledge thus activated (Medlin ＆ Ross, 1992: 335). To put it another way, to understand a text, one must superpose his existing knowledge over the information of the text to check whether the information in the text conforms to the common sense. If the information conforms to the common sense, the text is coherent. If not, it is incoherent.

2.4.1.1.5 Coherence as pertinence in pragmatic force

Spoken communication is characterized by turn taking, and its basic structure is adjacency pair. Generally speaking, the speech act in the previous turn prefigures the speech act in the next turn. For example, initiation-response, question-answer, request-reply/refusal, are all adjacency pairs (Stubbs, 1983: 131; Mey, 1993: 242). Tsui (1991) generalizes the condition of coherence between the two turns into coherence principle: for one utterance (B) to be coherent with another utterance (A), B either has to meet the illocutionary force of A or has to be relevant to the A's pragmatic presupposition.

2.4.1.1.6 Coherence as the result of inference

Grice (1975) argues that human beings conform to the rule of cooperation in communication. When a reader is confronted with a text, he believes that every sentence in it is necessary, true and relevant for the completion of the communicative purpose. Thus on the one hand, the writer must make some guesses at the reader's knowledge and inferential ability when composing a text; on the other hand, the readers endeavor to revivify the writer's intention. The degree of revivification determines the coherence of the text.

2.4.1.1.7 Coherence as the result of participation, negotiation and cooperation

Conversational texts are the result of mutual interaction of the two parties. Interlocutors usually take an active part in producing and developing the text, and the interpretation is the result of their negotiation. The process of participation, negotiation and cooperation is the process of text coherence (Gumperz, 1982).

2.4.1.1.8 Summary

These seven different definitions of coherence fall into two kinds (Zhang, 1998). The first kind regards coherence as a feature belonging to the entity of text (1, 2, 3). The linguists holding this viewpoint are mainly engaged in the study of text the product. The second, on the contrary, holds that coherence is a rule governing communication and text processing (4, 5, 6, 7). The linguists in favor of this viewpoint regard text as a dynamic process of communication, and concentrate their research on the production and interpretation of text. Nevertheless, both of these points of view attach great importance to the fluency in the transmission of information and take it as a criterion to judge whether a text is coherent or not. In other words, all the linguists studying coherence emphasize the crucial role of informativity, which testifies that informativity is one of the enabling conditions of a text.2.4.1.2 Semantic compatibility

By semantic compatibility is meant that linguistic units in a text should be compatible with each other in the meaning. It is usually reflected in three aspects: semantic compatibility among words, semantic compatibility among sentences and semantic compatibility among paragraphs.

Semantic compatibility among words demands that the writer of a text not come out with the phrases like “enjoy free medicare”, “frenzied aggression” (a mistake typical of Chinese EFL learners), because these words are not compatible in the meaning. For example, in the second phrase, “aggression” is an act of intrusion into a country well schemed in advance by another country, while “frenzied” means “affected with or marked by temporary strong emotion; frantic”; therefore “aggression” is incompatible with “frenzied” in meaning. In Sinclair's terms, the words that compose a phrase should contain the same semantic prosody. Semantic prosody is a term coined by Sinclair for corpus linguistics, analogous to Firth's phonological prosody. While phonological prosody refers to the supra-segmental coloring of neighboring phonemes (Firth, 1957), semantic prosody refers to the common semantic elements shared by several words in a phrase (Sinclair, 1991; Partington, 1998: 68).

Semantic compatibility among sentences demands that special attention should be attached to the fluency of meaning between different sentences when a writer composes a text. Generally speaking, the writer is not expected to make too big a leap in the meaning between two sentences. It is widely accepted in the circle of Chinese literature that a text should be dotted with some leaps to be held in love and admiration for long. However, too much dependence upon leaps for the desired effects is likely to bring about incomprehensible texts.

Semantic compatibility between paragraphs demands that paragraphs should be linked to each other by means of certain logic relations to form a unified whole, such as succession, parallel, cause-effect, concession, transition and so on. Otherwise the text will fall into pieces and be rendered as unreadable.2.4.1.3 Associability

Associability refers to the requirement that a perfect text should activate the reader's full association in two aspects: associability within text and associability without text. Associability within text can be further divided into lexical associability and grammatical associability.

In fact, Saussure's distinction of syntagmatic and associative relations (Saussure, 1980) is a pioneering study of lexical associability. When a linguistic item is replaceable by other linguistic items without changing the overall structure of the phrase, this linguistic item is said to be in associative relation with them. When a linguistic item is likely to combine with other linguistic items to form a longer phrase or sentence, they are said to be in syntagmatic relation. In the phrase “a new house”, the onset word “a” is in associative relation with “the”, “this”, “that” and so on; and the central word is in associative relation with “old”, “beautiful”, and so on. And the three lexical items “a”, “new” and “house” are in syntagmatic relation.

Grammatical associability demands that the tenses, voices and mood to be used in the text should be consistent with the readers’ expectations. For instance, a popular science reading material should be in the simple present tense, passive voice, indicative mood and abundant in longer sentences; while a story, in stark contrast, is usually in the past tense, active voice and full of shorter sentences.

Associability without text demands that a text should enable the readers to bring to mind the happenings in the outside world. In other words, the text should act as a bridge between the readers and the objective world.2.4.1.4 Cognitivity

Cognitivity demands that a successful text should comply with the law of human cognition in that it must be understandable to the readers and win their identification. A text indeed has to be creative in wording, for example, the coinage of new expressions and the adoption of poetic licenses. However, these creations should be conducted within the limits of human cognition. Although the saying that “the composing of a text will not end without startling words” was held as a tenet among some ancient writers, their works can still be correctly interpreted by their descendants. This alone is enough to testify the fact that they had taken into consideration the readers’ cognition when composing a text. This also explains why their works can be passed onto the present generation.

As the story goes, Bai Juyi often read his new poem to the illiterate grannies and took their understanding as the criterion to judge the quality of the poem. If the grannies failed to understand it, Bai would rewrite it time and again until it could easily get across to them. This reflects what a high store ancient writers set by the readers’ cognition. This also highlights the reason underlying the fact that no matter how novel and creative the metaphysical poetry founded and represented by John Donne during the Restoration Period in Britain and the wisdom-inspiring poetry invented by ancient Chinese such as Su Shi are in form, they are still well within the limits of human cognition. Were this not the case, their works would be totally beyond recognition now.The Metaphysical School is a name given to a school of English poets

试读结束[说明：试读内容隐藏了图片]

下载完整电子书