留學生新聞媒體學專業(yè)碩士dissertation定制：新聞語言

Chapter One Introduction
1.1 Background
In modern times, with the development of Internet, western mass media, such as newspaper, English broadcasts etc. can be easily accessed. Media discourse has consequently become a major source of people’s information, knowledge and attitudes. As an important channel of information transmission, media discourse, news report in particular, is usually presented as the unbiased recording of ‘hard 留學生新聞媒體學專業(yè)碩士dissertation定制facts’ and newspaper reporters always claim that what they spread to the public is objective and neutral. This is just one side of the coin, and news is in no way like what is mentioned above. Inevitably, this type of biased report will insinuatingly impose greater influence on people’s life and mind.
News discourse has inspired critical studies in many disciplines. Linguistics, semiotics, pragmatics and discourse analysis have produced a number of critical studies of news reporting or other programs. Since its emergence in the late 1970s, with its counterparts in “critical” developments in sociolinguistics, psychology and the social sciences, critical linguistics, or otherwise known as critical discourse analysis has developed rapidly in the past three decades or so. Critical discourse analysts regard language as a form of ideology and view linguistic activity as a social practice. They attempt to discover the ideology and power relations concealed in news discourse and the counteraction of discourse in the formation of ideology by means of not only analyzing texts or probing how texts are produced and interpreted, but also investigating the relationship between texts, processes of text production and social conditions as well as examining the political, economic and cultural contexts. The overriding objective of CDA is to reveal the discursive sources of power, dominance, inequality, and bias and how these sources are initiated, maintained, reproduced, and transformed within specific social, economic, political and historical contexts (van Dijk, 1988).
According to Fowler (1991), news is a practice, a product of social and political world on which it reports. News discourse, especially those issued by public service agency is always and inevitably influenced by the domain ideologies. News can influence and control people’s cognitions directly and indirectly and thus play a significant role in forming people’s ideologies and attitudes. And this is the very reason why news reports are the most salient type of discourses which CDA is applied to (Fowler, 1991) and are chosen as the research object here.
One year has elapsed since the eruption of Lhasa Riots, many stories and comment on it have appeared in media, some of these reports and comments are even contradictory to each other. Until recently, despite the extensive coverage of the Lhasa Riots home and abroad, no analysis of the nature and the quality has been conducted, let alone the analysis of the coverage of them under the theoretical framework of CDA. Therefore, the present thesis is of some significance and will make humble contribution to CDA.#p#分頁標題#e#
1.2 Purpose and Significance of the Study
The thesis is mainly concerned with the function of language as social practice and examines the seemingly unclear and hidden ideological effects in language use in news discourse. It studies the relationship between language and ideology through a close examination of the ideational function and textual metafunctions of language. It is expected to testify and universalize the validity of the assumption that news discourse is value-laden and ideologically invested with its theoretical framework and analytical categories based on Fairclough’s approach of critical discourse analysis, which find its root in the systemic-functional grammar of Halliday, and with a comparative analysis of news discourse from New York Times and People’s Daily.
Through the revelation of how linguistic features and structures are exploited to cover up ideologies, the author attempts to shed light on the aspects of language loaded with ideology, which is taken for granted or even ignored by the reader and the language learners, and explores the ways and causes that such beliefs or ideas of particular communities are disseminated. It is hoped that the present study contribute to improve the critical reading ability of readers and raise their awareness of ideology hidden in the news discourse.
1.3 Research Questions
In order to fulfill the above purposes, the following research questions will be addressed in this thesis:
1. How are ideologies of the news reporters with different interest embedded in the news discourse?
2. How does language act on the reader’s ideology?
3. How does language use relate to ideology in certain institutional and social context?
1.4 Organization of the Thesis
The present thesis consists of five parts. Chapter one is an introduction to the research background, the goal, the significance and organization of the thesis. Chapter Two manifests some key concepts related to the present study, and provides a brief review of the development of CDA as well as the contributions made to critical studies of news discourse by prominent figures both broad and at home. Chapter Three is an elaboration of the theoretical framework and methodology of the present study, primarily drawing upon Halliday’s systemic-functional grammar and Fairclough’s three-dimensional model. Chapter Four is a detailed analysis of news samples with the analytical tools developed in the previous chapter at three stages: description, interpretation and explanation in accordance with the model proposed by the Fairclough. Differences between these two newspapers are identified and a reasonable explanation of differences is also presented. Finally, the conclusion part summarizes major findings of this study, puts forward theoretical and practical implications derived from the research findings, and points out the limitations of the present study as well as areas for further research.
Chapter Two Literature Review
This chapter briefly surveys the history of CDA, summarizes three different approaches to the critical discourse study, illustrate some key concepts in this field and the rationale for choosing news discourse as the research object, and review the previous studies of news discourse from a critical perspective. #p#分頁標題#e#
2.1 Birth and Development of CDA
2.1.1Birth and development of CDA abroad
CDA represents an outgrowth of the work of British and Australian pioneers of critical linguistics (CL), particularly Fowler and Kress. Nowadays, CDA is also termed as CL(critical linguistics) or CLS (critical language study) in different works, but the term “CDA” seems to have been favored recently and is being used to denote the theories formerly identified as CL.
The term of critical linguistics was first used in its current sense in 1979 in a book named Language and Control by Fowler, Hodge, Kress and Trew, a group of colleagues at the time working at the University of East Anglia, Norwich. In this book, Roger Fowler and his colleagues criticize the dominant formal paradigms of the 1960s and 1970s, which see language usage completely in a de-contextualized way. It argues for a kind of critical linguistics capable of exposing distortion and mystification in language and introduces some fundamental tools for the analysis of language, ideology and power. In the concluding chapter of this book, Fowler and Kress poses the concept of “critical linguistics” and gives some explanations of its theoretical sources of systemic functional grammar and the analytical tools. From the very beginning, critical linguistics is employed as an instrument in the sense that ”linguistic forms allow significance to be conveyed and distorted”(Fowler, Hodge, Kress&Trew, 1979, p.6)
Later, Fowler (1991: 67) holds that “critical linguistics seeks, by studying the minute details of linguistic structure in the light of the social and historical situation of the text, to display to consciousness the patterns of belief and value which are encoded in the language and which are below the threshold of note for anyone who accepts the discourse as natural.”
Ten years after the publication of Language and Control (1979), another scholar Norman Fairclough publishes a book entitled Language and Power, which is considered to be a landmark of the beginning of CDA. At that time, Fairclough’s understanding of the role of discourse is far more complex than that proposed by the previous critical linguists for he believes that discourse is both socially constituted and socially constitutive, and that discourse is not only a product or reflection of social processes, but also a contributor towards the production of these processes. In Language and Power, he refers the critical approach to discourse as “critical discourse analysis”, thus, replacing the word ‘linguistics’ with the term ‘discourse analysis’. This replacement has great significance as it makes connections with sociological and philosophical theories of ‘discourse’ which are broader-ranging than what the term ‘linguistics’ implies.
In the past thirty years, many other experts in this field have made seminal contribution to the development of CDA and established it internationally as an interdisciplinary approach to the study of discourse. #p#分頁標題#e#
CDA deals with examining and analyzing spoken and written texts to reveal the discursive sources of power, dominance and inequality and explaining how these sources are initiated, maintained, reproduced, and transformed within specific social, economic, political, and historical contexts (van Dijk, 1988). CDA has become a field which aims at demystifying naturalized discourse and showing that grammatical and semantic forms have close connections with ideology and power. It regards language as “social practice” and provides theories and methods for empirical studies of the relationship between discourse and social and cultural developments in different domains.
According to Wodak, CDA as a network of scholars did not emerge until the early 1990s. Since then, it has evolved very quickly and attracts considerable attention in Great Britain and some other European countries.
2.1.2 Birth and development of CDA in China
Compared with CDA research abroad, the CDA studies in China start much later. It was not until middle 1990s that scholars in China, for instance, Chen Zhongzhu, XinBin etc. started to pay attention to the critical perspective on discourse analysis. They introduce the history, philosophical and theoretical foundations of CDA, make investigation of some key terms and apply the critical approach to the http://www.mythingswp7.com/dissertation_writing/Journalism_and_Mass_Communica/practical analysis of some news samples.
Chen Zhongzhu is a pioneer in the field of CDA in China. It is he who first introduces the concept of CDA to Chin among many scholars. He discusses the five development periods of CDA in great details. According to Chen Zhongzhu (1995), CDA has experienced several periods of development since the late 1970s: the sprouting period (1976-79), the stagnation period (1979-85), the self-reflection period (1985-87) and the new development period (from1988 up to the present). During the new development period, CDA has established its international fame. And in his article: Text and Ideology: Critical Discourse Analysis, he employs the analytical tools, for instance, transitivity, modality, speech reporting and thematic choice, which are based on the three meta-functions proposed by Halliday to analyze two news reports about coal miners’ strikes in 1985 from two British newspapers. He attempts to shed some light on the concept that language is both socially constitutive and socially constituted and encourages the development of the critical approach to the analysis of public discourses in China.
Another prominent scholar in this field domestically is Xin Bin, the most productive and comprehensive research among the Chinese scholars in this field. He writes a number of papers in which not only the theoretical origin and development of CL are introduced, but also the application to the real discourses is also illustrated. Just like Fairclough, he imposes special emphasis on intertextuality, and makes a case study of two news reports through analyzing modes of speech reporting from the angle of news sources, reporting verbs, direct and indirect speeches in 1998. In 2000, he published his dissertation Intertextuality from a Critical Perspective in order to “acertain how intertextuality contributes to the evolutions of the ideological meaning of a text and to show that text or discourse, far from being merely a reflection of a stable social structure, is itself a site of , and a stake in , ideological struggle ” (Xin Bin, 2000:ix-x) Xin (1997) also points out that it is of great importance to apply theories and methodologies of CDA into English classroom to enhance the critical language awareness of foreign language learners in China. #p#分頁標題#e#
In summary, the study on CDA in China starts much later than that abroad and it is not as prosperous as that in the same field in the western countries. Though many articles have been published, most of them are mainly limited on the theoretical introduction, the development of this field abroad and its application in analyzing the ideology in western media and few creative works in this field have emerged. Thus, a large amount of efforts are needed in order to make breakthrough in this field.
2.1.3 Summary
Though CDA has been in existence more than three decades, it has not had a well-defined method. van Dijk acknowledges that CDA does not have a unitary theoretical framework or methodology because it is best viewed as a shared perspective encompassing a range of approaches rather than as just one school” (Bell & Garret, 1998: 7). Different schools of CDA and various approaches to CDA thus have developed, with their own strength and weakness. The most representative approaches are the three-dimension model proposed Fairclough, the socio-cognitive approach by van Dijk and the discourse-historical approach by Ruth Wodak. All these approaches will be discussed in detail in the following sections and their strength and weakness will also be examined.
2.2 Different Approaches to CDA
CDA adopts a critical approach to study connections which may be hidden from people, such as the connection between language, power and ideology. It critically deals with the norms and expectations of particular discourse communities and raises social, economic and political issues. CDA studies social problems and the common domain and topics of CDA incorporate political discourse, ideology, racism, economic discourse, advertisement, promotional culture, gender, education and literacy.
Critical discourse analysis is a family of approaches instead of a single theory, and it does not have a unitary theoretical framework.
Van Dijk believes that critical discourse analysis is not so much a direction, school, or specialization. Rather, it aims at offering a different perspective of theorizing, analysis and application throughout the whole field. It is generally agreed that CDA cannot be classified as a single method but is rather views as an approach, which consists of different perspectives and different methods for the studying the relationship between the use of language and social context.
There have been many attempts to systematize CDA. The most notable work has been conducted by Fairclough (1992a, 1995a, 1998, 2000,2001,2003), CHouliaraki and Fairclough (1999), van Dijk (1993, 1997, 2001, ), RuthWodak (1996,2001), Scollon(1998,1999,2001,2004), Gee (1999,2005) and van Leeuwen (1993,199,1996).
Although they vary considerably in technical specification, they share a common strategy. According to Luke:
CDA involves a principled and transparent shunting back and forth between the microanalysis of texts using varied tools of linguistics, semiotic, and literary analysis and the macroanalysis of social formations, institutions, and power relations that theses texts index and construct (Luke 2002, p.100) #p#分頁標題#e#
The following sections will give a short overview of three of more representative and better established theoretical approaches to CDA in terms of their key features: namely the three-dimensional model proposed by the British critical linguist Norman Faiclough, the socio-cognitive approach proposed by Teun A. van Dijk, and the discourse-historical approach mainly proposed by Ruth Wodak, a Vienna-based professor of applied linguist. “Their approaches differ in various ways. They share a commitment to going beyond linguistic description to attempt explanation, to showing how social inequalities are reflected and created in language, and to finding ways through their work to change the conditions of the inequality that their work uncovers.”(Pennycook, 1994:121)
2.2.1Van Dijk and his socio-cognitive approach
Teun A. van Dijk has long been a leading theorist and advocate of discourse analysis as an interdisciplinary approach to the analysis of texts in social context, and he is also the one who is often mentioned in critical studies of media discourse.
Van Dijk proposes that CDA can be conducted in, and combined with any approach and subdiscipline in the humanities and the social sciences with its focus on social problems, and especially on the role of discourse in the production and reproduction of power abuse or domination. (2001:96). His socio-cognitive approach to analyzing discourse consists of three parts: discourse analysis which is primarily text based, social analysis pertaining to examining the overall social structures and cognitive analysis mediating between society and discourse.
In the 1980s, van Dijk presented a new interdisciplinary theory about news in the press, which he applied to describe and investigate the reproduction ethnic prejudices in the media, the abuse of power and reproduction of inequality through ideology. Most of his empirical studies are conducted on the basis of an extensive empirical data collection and text corpus and aim to display how social structures are enacted, instituted, legitimated, confirmed or challenged by text and talk.
However, what is the most distinguishable feature of van Dijk’s approach is his concern over cognitive analysis, an element of the discourse-cognition-society triangle (van Dijk, 2001, P98). By this triangle, van Dijk, claims to conduct analysis from discourse perspective, cognitive perspective and social perspective. Discourse analysis is primarily text based (lexicon, local semantics, topics, schematic structure, etc.) social analysis pertains to examining the overall societal structures, the context, and cognition-social cognition and personal cognition is a meditation between society and discourse
Generally speaking, van Dijk conducts analysis at two levels, namely: microstructure and macrostructure. Analysis at the level of microstructure mainly focuses on the semantic relations, between propositions, syntactic, lexical rhetorical elements that provide coherence and rhetorical elements such as quotations, direct or indirect reporting that give factuality to the news reports. Whereas, he analysis at the macrostructure level pertains to the thematic/topic structure of the news stories and their overall schema.#p#分頁標題#e#
Comparatively, Van Dijk shows greater interest in macro socio-psychological research and makes great achievements in this area. He succeeds in introducing the concept of context models, understood as mental representations of the structures of the communicative situation that are discursively relevant for a participant. In addition, he displays three forms of social representations relevant to the understanding of discourse, for instance, socio-psychological understanding and ideology, but he rarely mentions the language itself. However, this theory also has some limitation. It gives overloaded attention to the social relations and how theses relations are accomplished through routine practices. Therefore, van Dijk rarely mentions the language itself and underestimates the role of the interpersonal function language and the intertextual relations.
2.2.2Ruth Wodak and the discourse-historical approach
Ruth Wodak is a linguistically oriented CDA scholar, and she sees discourse from a discourse-historical perspective. Wodak, together with her colleagues, has developed the new approach to discourse analysis primarily based on sociolinguistics in the Bernsteinian tradition and on the ideas of the Frankfurt school, and this approach widely applied in CDA is closely associated with the Vienna school around Wodak.
In this approach, the connection between fields of action, genres, discourses and texts is described and modeled, and context is understood mainly historically. The discourse-historical approach is both hermeneutic and interpretative, with some influence from cognitive science (Wodak, 1996). This approach concerns linguistic and iconic characteristics of discourse, concentrating on collective symbols (topoi) which possess important cohesive functions in texts. This approach links the language changes to the changing social-historical background by fitting linguistic theories into argumentation theory.
The key element of this approach is its attempt to integrate systematically all available background information in the analysis and interpretation of discourse. The relationship between discourse and social power is the center of the Vienna school interest, from this perspective, language “manifests social processes and interaction and constitutes those processes as well (Wodak & Ludwig, 1999: 12).
In their studies of racist and discriminatory discourse Reisigl, M.& Wodak, R (2001) developed a four-step strategy of analysis: firstly, establishing the specific contents or topics of a specific discourse with racist, anti-semitic, nationalist or ethnicist ingredients; secondly, investigating the discourse strategies (including argumentation strategies); thirdly, examining the linguistic means, and finally the specific, context-dependent linguistic realizations of the discriminatory stereotypes . Three dimensions are central to the discourse-historical approach: the content of the data, the discursive strategies employed, and the linguistic realization of these contents and strategies, and the discursive strategies which are concentrated on in the discourse-historical approach are as follows: referential strategy or strategy of nomination, strategies of predication, strategies of argumentation, strategies of perspectivation, Strategies of intensification and mitigation.#p#分頁標題#e#
The general principles of discourse-historical approach can be summarized as follows: (Wodak,2001):
(1) Setting and context should be recorded as accurately as possible, since discourses can only be described, understood and interpreted in their specific context.
(2) The context of an utterance must be confronted with historical events and fact
as well as presented reports (intertextuality).Texts must be interpreted by other subject specialists (sociology, history, psychology). All stages imply an interdisciplinary approach as an important characteristic of the discourse-historical method.
(3 )Texts must be described as precisely as possible at all linguistic levels.
Combining historical and quantitative analysis on the level of discourse and social practice with qualitative analysis on the textual level, the Vienna School has dealt with racist media discourse and postwar anti-semitism. Wodak, in particular, has carried out research in various institutional settings such as courts, schools, and hospitals, and on a variety of social issues such as sexism, racism, and anti-Semitism. But this approach is also not perfect and many critiques are imposed on it, which are mostly targeted against the complexity of the use of this approach in empirical studies.
2.2.3 Fairclough and his three-dimensional model
The third principal approach in CDA to be mentioned in this thesis is Fairclough’s three-dimensional model, which is central to CDA over the past ten years or so. Fairclough’s idea on CDA primarily developed from Halliday’s systemic-functional Grammar, and he especially concentrates on the three metafunctions of language: the ideational function of language constitutes system of knowledge; the interpersonal function creates social subjects or identity and the relationship between them; the textual function creates discourse. (Faiclough,1995).
In his 1989 book, Language and Power, Fairclough developed a model of discourse analysis. He divided the discourse into three dimensions-texts, interaction and contexts, corresponding to which ,three stages of discourse analysis are distinguished；description, interpretation and explanation. Later, the model was revised and applied to critical discourse analysis.
For Fairclough, CDA is mainly concerned with the investigation of the relation between two assumptions about language use: that language use is both socially shaped and socially shaping. The three components mentioned above: description, interpretation, and explanation are the basis of analysis of Fairclough. Linguistic properties of texts are described (text analysis), the relationship between the productive and interpretative processes of discursive practices and the texts is interpreted, and the relationship between discursive practice and social practice is explained (Fairclough, 1995a). In doing this, Fairclough attempts to establish a systematic method to explore the relationship between text and its social context. The dimensions on which the method is based on are as follows(this figure will be elaborated in detail in section 3.2):#p#分頁標題#e#

Process of production
Text Description (text analysis)

Process of interpretation Interpretation (processing analysis)
Discourse practice
Explanation (social analysis)
Sociocultural practice
(situational; institutional; social)

Dimensions of discourse Dimensions of discourse analysis
A three-dimensional view of discourse analysis (Fairclough, 1995)

A distinctive feature of this approach is that the link between texts and society is mediated by discourse practices. Since him, the term “mediation” has been widely applied to describe the relation between language use and society, which means that the relationship between language and society is not direct, and language or discourse is just the medium to bring out values, beliefs, conventions and norms of society.
Besides, what merits the attention of this approach is the emphasis on intetextuality analysis. 1n his 1995 book Media Discourse, Fairclough puts forward a three-dimension framework for the intertexuality analysis in media discourse.
For Fairclough, intertextuality refers to “the property texts have of being full of snatches of other texts, and which may be explicitly demarcated or emerged in, and which the text may assimilate contradict, ironically echo, and so forth” (Fairclough, 1989a), and in the view of Fairclough, intertextual analysis draws attention to the dependence of texts upon society and history in the form of resources made available. In addition, intertextual analysis also draws attention to how texts may transform social and historical resources and how discourses are mixed in texts.
2.2.4 Summary of different approaches to CDA
Although these approaches differ in various ways, they “share a commitment to going beyond linguistic description to attempting explanation, to showing how social inequalities are reflected and created in language, and to finding ways through their work to change the conditions of the inequality that their work uncovers” (Pennycook, 1994: 121). Despite the differences in approaches, Fairclough and Wodak summarize the general principles of CDA, which could show some highlight for the further study.
1.Language is a social practice through which the world is represented
Discourse/language use as a form of social practice in itself not only represents and signifies other social practices but it also constitutes other social peactices sucha s the exercise of power, domination, prejudice, resistance and so forth.
1. Texts acquire their meanings by the dialectical relationship between texts and the social subjects: writers and readers, who always operate with various degrees of choice and access to texts and means of interpretation.
2. Linguistic features and structures are not arbitrary. They are purposeful whether or not the choices are conscious or unconscious.
3. Power relations are produced, exercised, and reproduced through discourse.
4. All speakers and writers operate from specific discursive practices originating in special interests and ims which involve inclusions and exclusions.#p#分頁標題#e#
5. `discourse is historical in the sense that texts acquire their meanings by being situated I specific social, cultural and ideological contexts, and time and space
6. CDA does not solely interpret texts, but also explains them.
2.3 Key Concepts in CDA
The critical approach to discourse analysis is an application of linguistic analysis on the basis of systemic-functional linguistics developed by Halliday. This study aims at examining the media language and discourse, which is actually a site of ideological struggle instead of being merely a reflection of facts. Briefly speaking, CDA aims at making clear the usually hidden connections between discourse and its underlying ideology. Two terms are essential to the understanding of CDA, namely, “discourse” and “ideology”.
2.3.1 Discourse
Discourse can be defined in many ways by different researchers and in different academic cultures. In the German and central European context, a distinction is made between “text” and “discourse” relating to the tradition in text linguistics; in the English speaking world, “discourse” is often used both for written and spoken texts (see Schiffrin 1992); other researchers distinguish between different levels of abstractness: Lemke (1995) adheres to a more Foucauldian approach and defines “text” as the concrete realization of abstract forms of knowledge (“discourse”),
Even in the field of critical discourse analysis, there are a number of definitions available for the term “discourse”. Fairclough, van Dijk, Kress all define discourse in different ways. In his book Critical Language Awareness, Fairclough defines “discourse” as “language use”, which “shapes and is shaped by society and is invested with ideologies”(Fairclough, 1992:p:8 ) later, in his 1995 book, Fairclough again claims that “discourse is use of language seen as a form of social practice, and discourse analysis is analysis of how texts work within sociocultural practice” describing discourse as social practice implies a dialectical relationship between a particular discursive event and the situations, institutions and social structures which frame the discourse. That also implies that discourse is socially constitutive as well as socially conditioned.
In this thesis, the author will follow Fairclough’s view of discourse as the working definition to analyze news reports.
One thing need to be mentioned here, discourse can manipulate readers. Readers can be positioned by texts in that they are drawing upon a particular discourse without necessarily realizing that this discourse sets limits on their reading. (O’Halloran 2003:13)
2.3.2 Ideology
Ideology is also very hard to define and different scholars understand ideology differently. Thompson (1990) points out that the concept of ideology first appeared in the late 18th century in France and it thus has been in existence for more than two centuries. Ever since its birth, the term has been given a range of functions and meanings.#p#分頁標題#e#
Ideology is closely related to language. Thompson considers ideology as “social forms and processes within which and by means of which, symbolic forms circulate in the social world”(Wodak, 2001:10). Simpson (1993:5)defines ideology as “taken-for granted assumptions, beliefs and value-systems which are shared collectively by social groups.”
Fowler (1991: 92) refers to ideology as “the sum of ways in which people both live and represent to themselves their relationships to the conditions of their existence.” Van Dijk (1998a: 24) argues that ideologies are the knowledge, attitudes and beliefs which are shared by the members of groups, organizations, or other social collectives of people and According to Fairclough and Wodak, ideologies are cognitive representations of practices formed from particular perspectives in the interest of maintaining unequal power relations and dominance. (1997), which are considered to be embedded in linguistic structures. Despite the superficial divergence, these diverse views of ideology share one common ground and they all hold that an ideology is a fairly abstract system of evaluative beliefs, typically shared by a social group, which underlies the attitudes of a group.
Ideology is closely related to language and ideology can be constructed through news discourse at deeper levels. News discourse can affect what we think. Though people have their own social or political opinions, news discourse controls the whole social knowledge and ideology. Language is in no sense a transparent medium, it carries much ideology, loaded with value concept, historical concept and political attitudes to the information it conveys, and exerts influence beyond people’s awareness. The political tendency and ideology, particularly those in reports concerning hot events are often covered by the seemingly objective narration. People that lack background knowledge or political sense are inclined to be influence unconsciously by the ideologies hidden in language. Therefore, it is necessary to guide people from the perspective of critical discourse analysis in regarding news discourse.
2.4 Rationale for News Discourse Analysis
Newspaper is playing an increasingly significant role in people’s daily life. For some people, reading newspapers has become an indispensible part of their life. In addition, the information conveyed by the channel of newspaper imposes a large impact on the mass population. In short, news discourse research is of great necessity and importance.
CDA is particularly interested in media discourse such as newspaper discourse and a quantity of research has been conducted in this aspect. Norman Fairclough, Fowler and van Dijk have made remarkable contribution to this filed.
Discourse analysts have given a multiplicity of reasons for choosing media discourses as the object of study(Bell &Garrett, 1998, p.3;van Dijk, 1985(b), p.27; Fowler, 1991b, p.120)
Bell summarizes four main reasons for taking interest in the studies of media discourse.#p#分頁標題#e#
Firstly, media are a rich source of readily accessible data for research and teaching. Secondly, media usage influences and represents people’s use of and attitudes towards language in a speech community. Thirdly, media use can tell us a great deal about social meanings and stereotypes projected through language and communication: for example in the use of foreign languages in advertisements. Fourthly, the media reflect and influence the formation and expression of culture, politics and social life.
(Bell & Garrett, 1998: p.3)
It can be seen from the previous quotation that is an important source of information, pleasure and enjoyment, and it is also a sharper of perception, behavior and institutions. Media discourses are also widely used in many research projects because media data is “good in quality, adequate in quantity, comparatively easy to access and not modified by an observer effect” (Bell, 1991, p.9).
Yu Yanmin also points out that when people are not informed on certain issues, they are more vunerable to the influence and manipulation of what they read in newspaper and what they watch on television. Media have great influential and manipulating power, they are able to transmit messages to a diversity of audiences at an enormous size, representing, influencing and changing people’s understanding and attitudes towards social realities and relationships.
In the words of Leeuen, media studies attempt to show that the seemingly factual media products were not transparent windows on the world or ‘slice of life’ but ‘constructed realities’ with an ideological dimension (1997: p.214).
2.5 News Discourse and CDA
Throughout the development of CDA, news discourse has maintained the position of an important research object and a rich source of research data. Almost all established critical discourse analysts, like Fowler, Fairclough, van Dijk, Bell, etc. have devoted much attention to news discourse. Its practitioners have accumulated invaluable analytical modes and methods on news discourse analysis (Fowler, 1991; van Dijk, 1993, 1995; Fairclough, 1989, 1995).
According to Fowler (1991: 25), “news representation in the press as in all other kinds of media and discourse, is a constructive practice. Events and ideas are not communicated neutrally, in their natural structure, as it were.”
In his view, language use is never neutral. Any form of language choice incorporates ideological meanings. After proposing the concept of CL in Language and Control (1979), Fowler expounds the framework of CDA and applies it to specific news texts in is 1991 book “Language in the News: Discourse and Ideology in the Press”. Fowler holds that CDA is ‘instrumental linguistics’ and his analysis of media language draws upon functional linguistics. He employs the theoretical framework of CDA in the actual news texts by analyzing its transitivity, transformation, classification, etc. and his analysis concentrates on vocabulary rather than the macrostructure of the news reports. He carries out case studies of major news stories of the United States’ bombing of Libya, and coverage of the conversations between patients and doctors in the press. After strenuous study of the news discourse, he reaches a conclusion that some crisis is created and constituted by media discourse.#p#分頁標題#e#
British linguist Norman Fairclough is a central figure who is often referred to and quoted in critical studies of news discourse. Fairclough (1992a: 110) suggests that “news media can be regarded as effecting the ideological work of transmitting the voices of power in a disguised and covert form.” He elaborates the relationship between language and ideology, and in this way, he gives a general picture of the place of language in society.
He writes extensively on news discourse, contributes remarkably to media studies by imposing special emphasis on intertextuality and suggests that a systematic study of news discourse should be made. He puts forwards the recently widely applied three-dimensional model which deals with language, discourse and power in his 1989 book Language and Power. Later, Faircough (1998a) applies his three-dimensional framework to a case study on the discourse of Thatcherism, imposing concentration on the examination of institutional and social contexts rather than linguistic features. Since then, his three-dimensional model (including the later revised version) has been adopted as theoretical framework in many studies and in the present study as well. He also analyzes a news report from the British national newspaper, Sun, which mixes its own conversation-like style with the bureaucratic language of the official document produced by a parliament committee.
Van Dijk is another major contributor to the advancement of critical news analysis. Though his early works mainly focus on literary studies and text grammar, since the late 1980s, he turns to media discourse and publishes many works on it. His interest is in ‘developing a theoretical model that will explain cognitive discourse processing mechanisms’ (Wodak and van Dijk: 2000).
In the 1970s, he was one of the most important scholars who insisted that textual theories and media discourse analysis should take account of the context of texts.
In the 1980s, van Dijk (1988a, 1988b) presented a new interdisciplinary theory about news in the press and employed it to the description and exploration of the reproduction of ethnic prejudices in the media, the abuse of power as well as the reproduction of inequality through ideology.
What is most distinguishable of van Dijk main contribution to critical news analysis is his cognitive model for CDA. In order to uncover ideologies generated in discourse, van Dijk (2001: p.97) develops a socio-cognitive model for CDA and resorts to discourse analysis of the text, cognitive analysis and social analysis. Discourse analysis is primarily text based, while social analysis pertains to examining the overall social structures, the context. It is socio-cognition—social cognition and personal cognition—that mediates between society and discourse. Through this model, he attempts to concentrate on the interpretation of the discourse in a psychological perspective.
Recently, van Dijk switches his attention to the relations between power, language and ideology in news discourse. He attempts to expound on the idea that the news reports themselves need to be investigated in a dynamic process.#p#分頁標題#e#
Chapter Three Theoretical Framework and Methodology
In this chapter, the theoretical framework and the analytical tools for the present study are presented first, followed by the illustrations of the data collection and data analysis procedures.
Van Dijk (2001a) points out that CDA is not a specific direction of research, and it does not have a unitary theoretical framework. Just as mentioned in section 2.2, different scholars provide various perspective of theorizing, analysis and application, among which the three-dimensional model of Fairclough, the socio-cognitive approach of van Dijk and the discourse-historical approach of Ruth Wodak are more representative and better established. The three-dimensional model of Fairclough is the best received and most widely adopted approach, and it also serves as the theoretical framework here.
Since its emergence as an analytic approach, CDA has been indebted much to Halliday’s systemic-functional grammar. SFG serves as the main source and a useful analytic tool for CDA practitioners by offering clear and rigorous linguistic categories for analyzing the relationship between discourse and social meaning (Chouliaraki & Fairclough, 1999). SFG is a lexico-grammatical theory formulated on the basis of the notion of choice. It constructs language as a system of meaning which is realized mainly through the ideational, interpersonal and textual functions embodied in the grammatical choices made 留學生新聞媒體學專業(yè)碩士dissertation定制by language users.
The present study adopts simultaneously Fairclough’s (1995a) three-dimensional model as the overall analytical framework and Halliday’s (1994) theory of metafunctions as specific analytical tools. Based on Fairclough’s three-dimensional model, the present analysis will be conducted at three stages: description, interpretation and explanation; or in other words, at three different corresponding levels: the textual level, the discursive level and the socio-cultural level. It is hoped that this three-level analysis of language use might provide new insights into discerning the dominant ideologies embedded in the news reports represented by different groups of Chinese and American media.
3.1 SFG and Halliday
3.1.1 Systemic-functional grammar
Halliday once emphasize “a discourse analysis that is not based on grammar is not an analysis at all, but simply a running commentary on a text”(Halliday, 2000,p47). This quotation implies the paramount importance of choosing a holistic grammatical system to be draw upon in the critical discourse analysis. Following the line of Fairclough, this thesis draws upon SFG as underpinning in the description of the discourses.
Systemic functional grammar is a linguistic theory with related analytical methods particularly associated with Michael Halliday (Halliday 1978,1994). According to Fowler (1979), Hodge and Kress (1988) and Van Leeuwen (2001), it is distinct from the traditional grammar and it is mainly concerned with the relationship between language and other elements and aspects of social life, and its approach to the linguistic analysis of texts is always oriented to the social character of discourse.#p#分頁標題#e#
Systemic-functional linguistics views language as an instrument of social interaction, and it maintains that language should not be studied in isolation from the social context. It holds the belief that there is bilateral relationship between society and language, in other words, language is not only applied to construct reality, it is also socially formed.. it is believed that language is creating and being created by identity, social relations and system of knowledge and beliefs. (Halliday, 1994)
In the view of Xin Bin (1996b), SFG possesses a number of merits which makes it widely applied in CDA .Firstly, it is a semantically driven grammar, which seeks to consider and identify the role of various linguistic items in any text in terms of their functions in conveying meaning; secondly, Halliday’s approach is socio-semiotic. He defines language as meaning potential. Thirdly, the meaning potential is the linguistic realization of the behavior potential. Besides, SFG shares the same view with CDA that no text can be free of context, so no text is free of ideology.CDA is ideology-oriented, and SFG just provide a systemic grammatic system to identify ideology in text.
3.1.2 Three metafuncitons of language by Halliday
Michael Halliday is the foremost exponent of SFL and the basis of his theory is the notion of function. According to him, language is functional and all language performs three metafunctions simultaneously, which he terms as: the ideational function, the interpersonal function and the textual function. Each of them deals with one aspect of the world and is concerned with a different mode of meaning of clauses.
The ideational function is about the natural worlds in the broadest sense. Language serves for the expression of “content”, that is, the speaker’s experience, including the inner world of consciousness. It is also concerned with clauses as presentation. The ideational function, which is about the expression of content is crucial to the theory of critical linguistics.
Ideational function is the one concerning all kinds of human experience consisting of “goings-on”-happening, doing, sensing, meaning, being and becoming. In this sense, clause not only represents a mode of action, but also reflects the endless variation and flow of event.
The ideational metafunction is mainly realized through transitivity system, which “construes the world of experience into a manageable set of process types.”(Halliday, 1994, p.106) The ideational metafunction is also realized by vocabulary. Vocabulary reflects human’s cognition and judgment of the world. Apparently, speakers choose different words to describe objects, and display their attitude and opinion. Therefore, after a careful analysis of the vocabulary classification, the preference and attitude of the language user can be exposed.
The interpersonal function is about the social world. Language serves to establish and maintain social relations. It is concerned with clauses as exchanges.#p#分頁標題#e#
The textual function is about the verbal world. Language serves to provide for making links with itself and with the features of the situation, especially the flow of information in a text. It is concerned with clauses as message.
3.2 Three-dimensional Model of Fairclough
Viewing language or discourse as a social practice, Fairclough suggests that critical analysts should not only analyze the texts, the processes of text production and interpretation, but also investigate the relationship between texts, processes and their social conditions. Based on such assumptions, Fairclough (1989a; 1992a; 2003) develops his three-dimensional model of CDA.
Fairclough’s three-dimensional model is based on his three-dimensional conception of discourse as language in use. Every instance of language is viewed as a communicative event consisting of three dimensions: it is a text (speech, writing, visual images or a combination of these); it is a discursive practice which involves the production and consumption of texts; and it is a social practice. Corresponding to three dimensions of discourse, he proscribes three dimensions or stages of CDA (1989a: 26):
Description is the stage which is concerned with the formal properties of the text.
Interpretation is concerned with the relationship between text and interaction – with seeing the text as a product of a process of production, and as a resource in the process of interpretation.
Explanation is concerned with the relationship between interaction and social context – with the social determination of the processes of production and interpretation, and their social effects.
Fairclough sees all three stages of analysis as interconnected, each serving a particular purpose and all combined aiming at uncovering the underlying ideologies embedded in discourse. Figure 3-1 shows the interrelationship between the dimensions of discourse and the analytic levels or stages.

Process of production
Text Description (text analysis)

Process of interpretation Interpretation (processing analysis)
Discourse practice
Explanation (social analysis)
Sociocultural practice
(situational; institutional; social)

Dimensions of discourse Dimensions of discourse analysis
Figure 3-1 A three-dimensional view of discourse analysis (Fairclough, 1995)
The embedding of the boxes emphasizes the interdependence of these three dimensions and the intricate moving backwards and forwards between different types of analysis. This three-dimensional image enables one to understand that an analytic move to examine a single box necessarily breaks the interdependence between the boxes and requires subsequent moves which re-insert that box into interconnected place (Janks, 1997: 27). A special feature of this approach is that the link between text and sociocultural practice is mediated by discourse practice. The production or interpretation of a text depends upon the nature of the sociocultural practice which the discourse is a part of. The nature of the discourse practice of text production shapes the text, and leaves traces in surface features of the text and meanwhile the nature of the discourse practice of text interpretation determines how the surface features of a text will be interpreted. Therefore, Fairclough’s three-dimensional model which coherently conflates and connects the macro-social interpretation with micro-linguistic analysis, provides us with a synthesis of social theoretical concepts within a critical analytical framework. #p#分頁標題#e#
3.3Analytical Tools
Actually, both the theory and the methodology of CDA are eclectic, theories and methods are integrated so that it is adequate in understanding and explaining the object under investigation (Wodak, 2000). CDA has the nature of being interdisplinary and there has been no accepted list of analytic categories for CDA yet. Different CDA practitioners prefer to varying analytical tools, such as transitivity, modality, nominalization, transformation, etc. It is not necessary and not possible to employ all of them when approaching certain texts.
The applied tools applied in this study are selectively borrowed from Fowler, van Dijk, and Fairclough, and conceptually organized according to the meta-functions of language proposed by Halliday (1985, 1994).
Based on the specific research questions and corresponding to the three meta-functions of language by Halliday, the following analytic categories are chosen for the present study, namely, lexical classification, transitivity, and intertextuality. Starting from the ideational function of language,
Since representation corresponds to Halliday’s ideational function, lexical classification and transitivity which mainly deal with exploring the ideational function of grammar are chosen for the analysis at the textual level. Analysis at the discursive level will focus on intertextuality, while at the final level, the impact of the social background as well as the institutional background is illustrated.
3.3.1 Lexical classification
Lexis or vocabulary, a most basic component of language, is regarded as one of the major determinants of ideational structure in Halliday’s (1994) systemic-functional grammar. It is regarded as a representation of the world for a culture, while the world as perceived according to the ideological needs of culture.
The terms “classification’ “categorization” and “lexical structure” are interchangably used in various works of critical discourse analysis, referring to word choice, the basic level of language function.
In the view of Kress and Hodge (1993, p.62) “classification is at the basis of language and thought. Without acts of classification no one could relate concepts or words to new concepts or message because words and concepts only exist through classifications”.
For a period of time, it has been employed by critical scholars as an essential tool for discourse analysis. Fowler (et al.,1979,1991) imposes significant on the lexical analysis of news discourse. He refers the terms of classification to word choice, the basic level of language function. According to Fowler, vocabulary is an internal part of reproduction of ideology or proposition. In a similar vein with Fowler, van Dijk (1988: 177) views lexical choice as an eminent aspect of news discourse in which hidden opinions or ideologies may surface. Word choice is not made at random, instead, words which may be chosen generally or conceptually express values or norms, and that therefore are used to express a value judgment (van Dijk, 1995: 20)#p#分頁標題#e#
Fairclough also imposes great significance on the analysis of lexical classification. According to him (1989a: 116), “a text’s choice of wording depends on, and helps create, social relationships between participants”. And he also claims that “the classification scheme constitutes a particular way of dividing up some aspect of reality which is built upon a particular ideological representation of that reality.” Therefore, lexical classification serves as important cognitive ability for human being, and endows the exterior world with order.
Van Dijk makes remarkable contribution to the lexical choice in news discourse by proposing the notion of “ideological square” which mainly aims at establishing a positive ingroup image and also a negative outgroup image. In short, people with different interests have different ways of classification with regard to the same events. Therefore to examine the lexical classification about the participants or the event is of much significance and offers to reveal the reporters’ attitude and the hidden ideology a crucial clue to the hidden ideologies in the text and the related political positions.
3.3.2 Transitivity
Transitivity is a fundamental and powerful semantic concept in the theory of Halliday. It is the main element of Halliday’s category of the ideational function and an essential tool in the analysis of representation. “The term of transitivity is the systemic linguistic term for exploring the ideational function of grammar at the level of the clause” (Fairclough, 1992a: 177). It shows how meaning is represented in the clause, how producers encode their mental picture of reality and how they account for their experience of the world around them and inside them. Since transitivity makes options possible, the choice we make or the choice made by discourse indicates our point of view and is ideologicaly significant( Fowler R.,1991a).
Halliday (1994:107) identifies the way in which transitivity carries out the ideational function as processes. According to Halliday, the transitivity system is concerned with the transmission of ideas and construes the world of experience into a manageable set of process types ( Teo P. 2000,11(1):p.7-49 ). According to what they represent, for instance, whether they represent actions, events, states of mind or simply states of being, Halliday identifies three main types of processes which include material process, mental process and relational process, and three subsidiary types of processes which consist of behavioral process, verbal process and existential process. All these six types of processes are potentially composed of three components: the process itself, which is expressed by the verb phrase in the clause, participants in the process which is typically realized by noun phrases in the clause and circumstances associated with the process which are normally expressed by adverbial and prepositional phrases (time and location).
In the words of Halliday, “material processes are processes of doing,” for instance, run, read or jump, the processes of action and physical doing, as well as resign, a process of abstract doing. Material processes can have two participants: one is the Actor, an obligatory element which serves as the “doer” or “performer” of the process and is realized by the animate or the inanimate; and the other is Goal, an optional element which assumes the role of one affected by the process carried out by the Actor. (Halliday,1994). Typically in the media discourse, the Actor in the material processes are sometimes deliberately omitted or exhibited as non-human Actor, therefore it becomes unclear who should be responsible for a certain event.#p#分頁標題#e#
Example one:
Several people(Actor ) clashed with and stoned the local police(Goal) around the Ramogia(Circumstance) Monastery in downtown Lhasa.
( People’s Daily, March 15(a),2008) )
Example two:
Several clothing shops, restaurants, and mobile phone stores(Goal ) were looted. ( People’s Daily, March 15(a),2008) )
Mental processes are “processes of sensing,” perception, cognition and affection, such as see, think or feel (Halliday, 1994, p.112). They are associated with human emotions and often convey opinions, feelings and thoughts. According to Simpson, mental process may be more delicately defined as perception process (seeing, hearing, etc.), affection process (liking, hating, etc.) and process of cognition (thinking, understanding, etc) (Simpson, 1993: 91). This type of processes can also have two participants: Sensor, a “conscious being” that the conscious being that is perceiving, reacting or thinking, and the Phenomenon, the one/ting that is sensed or felt. Mental processes often reflect the Sensor’s opinion or attitude towards what has happened. Some examples are illustrated as follows:
Example one:
I (Senser)saw the ground was covered with rocks weighing one or two kilograms and a cab was burnt down(phenomenon).
( People’s Daily, March15(a),2008)
Example two
I (Senser) heard a soldier shouting, ‘Please go home and stopfighting!’ (Phenomenon)
( New York Times, March 15,2008)
Relational processes are processes of “being”, which operates with three main types: intensive, circumastantial and possessive. And each of these comes in two distinct modes: attributive and identifying. (Halliday, 1994: P.119) The attributive processes can have two participants, Carrier and Attribute. Different from the attributive processes, the identifying processes can have Identififer, the Identified, Token and Value. The identifying relational processes are reversible since both the Identified and Identifier refer to the same world entity. Identification is a matter of relating a specific realization to a more general category.
Example one:
The chaotic scene (Identified) was the latest, and most violent, confrontation (Identifier) in a series of protests that began on Monday(Circumstance).
(New York Times, March 15, 2008)
Example two:
But Chinese leaders(Carrier) may be more reluctant(Attribute) to order such heavy-handed tactics as Beijing prepares for the Olympics
(New York Times, March 15,2008)
On the borderline of material processes and mental processes is the category of behavioral process, a process of behaving, such as smiling or dreaming. Though it is grammatically more like a process of ‘doing’, yet it is more of human physiological and psychological behavior.(Halliday,1994:139). There is only one participant in this process, which is the Behaver.
Example one:
The monks (Behaver) are still protesting.#p#分頁標題#e#
(New York Times, March 15,2008)
Example two:
The ethnic friction evident in Friday’s violence(Behaver) has long simmered just below the surface in Lhasa.
(New York Times, March 15,2008)
Verbal processes are processes of “saying”, such as tell, say, or accuse. This category of verbal process is on the borderline of mental processes and relational processes and is symbolic relationships constructed in human consciousness and enacted in the form of language like saying and meaning (Halliday, 1985:107). A verbal process may consist of three participants: the Sayer which refers to anything that puts out a signal, the Receiver that is the addressee to whom the verbalization is addressed, and the Verbiage, the verbalization itself or the content of the massage (Halliday, 1994: 115).
Example one:
The mobs assaulted passersby, sparing no women or children(verbiage), witnesses (sayer)said
(People’s Daily, March15(a),2008)
Example two:
A Tibetan government official (sayer)told Xinhua(receiver) that there had been enough evidence to prove that the sabotage in Lhasa was "organized, premeditated and masterminded" by the Dalai clique(verbiage).
(People’s Daily, March15(a),2008)
Finally, the Existential processes are processes of “existing”, as there is a tree. This type of processes is between relational and material processes and represents something exists or happens. Typically, existential processes are realized by there-be construction. Sometimes, this category of processes can also be realized by other verbs, for instance, remain, occur, follow, emerge, flourish, prevail etc. This type of processes has only one participant, which is the Existent., an event, an object or a person. This type of processes is rare in media discourses.
Example one:
On Friday, different accounts(existent) emerged (process verb)about how the Chinese military police in Lhasa handled the demonstrations.
( New York Times, March 15,2008)
Example two:
There is police and military (existent)everwhere(circumstance).
( New York Times, March 15,2008)
Our most powerful impression of experience is that it consists of ‘goings-on’-doing, happening, sensing, meaning, and being and becoming (Halliday 1994:106). The following table is a summary of the types of processes Halliday identified in the grammar, as well as the general category meaning and the principal participant functions that are associated with each other.
Process type Category meaning Participants
Material:
Action
event ‘doing’
‘doing’
‘happening’ Actor, Goal
Behavioral ‘behaving’ Behaver

Mental :
Perception
Affection
Cognition ‘sensing’
‘seeing’
‘feeling’
‘thinking’ Senser
Phenomenon

Verbal ‘saying’ Sayer, Target#p#分頁標題#e#
Relational:
Attribution
identification ‘being’
‘attributing’
‘identifying’
Carrier, Attribute
Identified, Idenfifier
Existential ‘existing’ Existent
3.3.3 Intertextuality and discourse representation
Intertextuality is considered as one of the most striking characteristics of news discourse. News discourses are always profuse with snatches of other texts, which may be explicitly demarcated or merged in the text.
It is widely believed that it is Kristeva(1986) who coined the term intertextuality on the basis of Bakhtin’s work. Kristeva (1981, p. 36) points out that a given text is “a permutation of texts, an intertextuality: in the space of a given text, several utterances, taken from other texts, intersect and neutralize one other”. But it is Fairclough who introduces the concept of intertextuality into CDA for the analysis of discursive practice
Scholars in different fields consider the notion of ‘intertextuality’ from different perspectives and for various purposes. Some scholars are from semiotics, with Kristeva(1981), Frow (1986),Smith (2000), Chandler (2005) etc. as the leading figures, and they concentrate on the complex and heterogenous nature of literary works by appropriating the concept of intertextualiy. Other scholars are from the field of discourse analysis and critical discourse analysis, with Fairclough (1992,1995) Scollon (2004),Bazerman(1993,2004),Lemke (1983,1985,1988a,b, 1995a,b),White (2002) etc. as their key figures and their major concern is with non-literary works. According to Fairclough(1992, 1995), for many scholars, intertextuality is perceived not only as a form through which texts are interrelated, but also as a social practice which involves particular socially regulated ways of producing and interpreting discourse.
Fairclough(1992a:84) defines intertextuality as “basically the property texts have of being full of snatches of other text, which may be explicitly demarcated or merged in, and which the text may assimilate, contradict, ironically echo, and so forth”. It implies that any text is a mixture of quotations and any text is the absorption and transformation of another.
The notion of intertextuality offers a perspective of both reading and writing texts as a way of looking at a text’s interactions with prior texts, writers, readers, and conventions. Thibault (1994,p1751) explains this perspective claiming that all texts, spoken and written, are constructed and have the meanings which text-users assign to them in and through their relations with other texts in some social forms.”
The present study will focus on discourse representation, a type of “manifest intertextuality”, which refers to the heterogeneous constitution of texts by which specific other texts are overtly drawn upon within a text and are marked by such explicit signs as quotation marks (Chouliaraki & Fairclough, 1999). Discourse representation, traditionally known as speech reporting, is a kind of manifest intertextuality. A framework for analysis of discourse representation is developed by Fairclough (1995a).#p#分頁標題#e#
In his 1995 book Media Discourse (Fairclough, N.1995b), Fairclough put forwards a three-dimension framework for analyzing intertextuality in media discourse, namely, the analysis of ‘discourse representation’, generic analysis of discourse types, and an analysis of discourses in texts. To Fairclough, ’discourse representation’ is a term used to refer to what is traditionally called ‘speech reporting;’ it is a form of intertextuality assigned to discourse practice (ChouliarakiL,Fairclough N. 1999); it is a form of intertextuality in which parts of specific texts are incorporated into a text and are usually, but not always, explicitly marked with such devices as quotation marks and reporting clauses(Fairclough,1992a,b, 1995b). Discourse representation accounts for a major part of what news is in media discourse. Fairclough replaces “discourse representation” with the more familiar “speech reporting” for the following reasons(a) writing, as well as speech, may be represented, and (b) rather than a transparent “report ”of what was said or written, there is always a decision to interpret and represent it in one way rather than another (Fairclough, 1995:54).
Just like intertextuality is mainly realized by two modes of speech reporting traditionally: direct speech and indirect speech, based on Volosinov’s account. Fairclough distinguishes two main modes of discourse representation: Direct Discourse (DD) and Indirect Discourse (ID). DD is the copy of original discourse, and reporters make the quotation without changing the form and content of the original discourse; direct discourse is usually claimed as the most objective one, and it is believed that the reporter is faithfully rendering the form and content of the quoted speech without imposing his own.; ID is usually expressed in reporters’ own words and is a sort of paraphrase or interpretation of the original utterance. In this sense, it involves more efforts on the part of the reporters and it is inevitable that the attitude and the perspective of the reporter are incorporated. Besides, two other types are also distinguished, namely DD(s), and UNSIG. The former is treated by him as a sub-type of DD, and is a category for cases of ‘slipping’ between modes, which is used to distance the reporter from the language highlighted and to indicate the reporter’s disassociation with the words， It allows an author to slip from narrative statement to interior portrayal which is difficult for readers to notice what has occurred, and makes readers have little choice but to trust the views of the narrator (Leech and Short，1981), Slipping is the most subtle way of manipulating readers’ points of view. The UNSIG is a category for cases where what is clearly secondary discourse appears in primary discourse without being explicitly marked as represented discourse. In short, news discourse is by no means completed by piling up the previous discourse at random, but by a careful compilation of the news reporter. Whatever mode is employed, it is inevitable for the reporter to impose his voice on the original utterance and complete objectivity is impossible.#p#分頁標題#e#
In addition to the modes of discourse representation, news sources should also be addressed and investigated. News reports rely heavily on other People’sverbal descriptions of events. The quality of the news sources counts because it directly affects the value of the news reports produced. Meanwhile, the preference to different sources and voices implies diverse attitudes towards each source. The press, in general, quotes those whose views do not contradict the points of view of the newspapers (van Dijk, 1993). An analysis of news sources plays an essential role in ideological analysis since it indicates whose voice is introduced and supported by the press.
According to Xin Bin (1998, 2006), in reporting others’ comments or opinions, reporters generally resort to three ways to introduce the news sources: specific source, semi-identified source and unidentified source.
As for the specific source, the information source is clearly and fully identified. For instance, the full name, official position, occupation of the information provider, institution or any other information related to the event are specified in great detail.
For instance:
Lhazom Zhoigar, vice-chairwoman of the regional Chinese People's Political Consultative Conference, said, "I am a witness of old and new Tibet. Before the peaceful liberation in 1959, poor Tibetans lived worse than beasts of burden. The new Tibet, especially since the national reform and opening up in 1978, experienced rapid development in all fields of politics, economy and culture”.
(People’sDaily, March 15(b), 2008)
As for the semi-identified source, it exists because some information provider is reluctant to expose their identity, and they might provide speaker more freely and comprehensively without the exposure of their identity. Thus, the reporters only drop a hint of the news source by using the names of institutions, groups, etc. and do not specify the news source in detail.
The Tibetan higher People’s court, the regional People’s procuratorate and the regional public security bureau said in the notice the rioters had set fire to schools, hospitals, children entertainment centers, shops and civilian houses, confronted government departments and assaulted law enforcement officers.
(People’sDaily, March 15(b), 2008)
The reporters sometimes also use some vague words or usual expression to introduce the source, which are unidentified one. In one case, they might be unclear about the exact source of the news, in another, they might consider it insignificant, or want to conceal it on purpose but still to show its objectivity and accuracy. If the information is incorrect, the reporter can have a good excuse to shift responsibility.
Sources told Xinhua that policemen were ordered not to use force against the attacker. But they were forced to use a limited amount of tear gas and fired warning shots to disperse the desperate crowds.
(People’sDaily, March 19,2008)#p#分頁標題#e#
3.4 Data Collection and Data Analysis
3.4.1 Data collection
Ten English new reports Data are chosen as the research object for the present stud, five of which are extracted from People’s Daily (PD) of China and the rest from New York Times (NYT) in the U.S. The newspaper selections are all downloaded from the Internet sites of the publication, which mirror the contents of the hard copy versions.
People’s Daily is chosen mainly because of its comprehensive coverage and large publication. It has been the most authoritative and influential newspaper in China since its launching in June 1948. It is an official government newspaper representing the interests of the Communist Party, the leading party in China and covers latest government policies and resolutions and other major events both in China and abroad.
New York Times is chosen out of similar reasons. It is the most influential, formal and authoritative newspaper of the United States. It enjoys great reputation worldwide and “probably exhibits fragments of an overall American ideological perspective on news events and the world” (van Dijk, 1988: 21). It is one of the leading newspapers in the U.S. with regard to the coverage of international news and views, attracting readers from every state and around the world. According to Golan (2006), New York Times serves an inter-media agenda-setting function for other news sources, in particular with regard to the coverage of international events and issues.
The texts selected are all closely related to the March Lhasa Riots. In the process of corpus construction, first of all, all the relevant news reports are retrieved from the archives of websites of these two newspapers: http://english.peopledaily.com.cn and http://www.nytimes.com. After a careful corpus design and compilation, ten news samples are selected divided into two groups: the reports from China are labeled as Group A, and those from the U.S. as Group B (cf. Appendix I).
3.4.2 Data analysis
Both quantitative and qualitative analysis will be conducted to in the present study, in order to make the analysis more reliable and systematic. The quantitative study mainly serves as a basis and complement to the qualitative study which is concerned with confirming the linguistic phenomena under analysis.
The quantitative method is adopted in the analysis of some corresponding linguistic items or patterns in different news texts to probe into their differences and similarities in terms of frequency, distribution and usage. Ten sample texts are tagged and annotated manually, and then raw frequencies of some distinguished linguistic features are presented, such as types of processes, direct and indirect speech as well as news sources.
The qualitative research is mainly concerned with observing how the event is reported and how ideologies function in representation processes. It proceeds clause by clause, detecting all salient features of news samples. The empirical interpretation also has to make use of theories mentioned in previous chapters to explain meanings resulting from the qualitative analysis. The results of analysis are illustrated with some specific and typical examples drawn from the news samples to make the analysis clear and convincing. #p#分頁標題#e#
Chapter Four Data Analysis
4.0 Background of Lhasa Riots
China achieves great glories as well as experiences great hardships in 2008. Most people would never forget a day in history, March 14, 2008, a “black Friday” for most Chinese, especially those in the Tibetan residence, as on this very day, fully prepared and well-designed riots erupted in Lhasa, the capital of Tibetan autonomous region , which began at about 10 a.m. and largely quieted down by 3:30 p.m..
Tibet has undergone some unrest in history, but no one can be compared with this one except the one in 1959. From the morning on, a handful of lawless people, chanting separatist words in downtown Lhasa initiated a rampage, beating people, smashing businesses, looting and burning. Many innocent civilians were burned or stabbed to death in the riots; cars were damaged or burned; dozens of public security officers and scores of armed police were injured, a number of public facilities and even residential houses were destroyed. The life a of the Lhasa residents and stability of the society were severely threatened and significant property damages were caused.
Confronted with the riots, Party and government officials of the Tibet Autonomous Region made timely reaction to control the situation and ensure the security, and China's public security and armed police have exerted the highest restraint. “We fired no gunshots," Qiangba Puncog, chairman of the Tibet Autonomous Regional Government, told the press in Beijing on March 15th when asked about the unrest in Lhasa on the previous day. But according to a news report issued by new York times, Violence in Tibet as Monks Clash With the Police，situations were different, in which the Riots which caused profound damage was originated from a peaceful protest, and the situation only escalated after some Chinese police shot the namely protester. The White House called on China to “respect Tibetan culture” and issued a renewed call for dialogue between Beijing and the Dalai Lama, the exiled spiritual leader of Tibetan Buddhism.
That article even linked the present riots with China’s Human right and the uprising which aimed at separating from Tibet from China in 1959 and displayed sympathy for the rioters and displayed condemn on some Chinese Policy. In short, the news reports from the New York Times are quite different from those issued in China. And these different reports provide us with authentic materials and an opportunity to conduct a comparative study on the ideological differences in news reporting between People’s Daily and New York Time.
The following sections present the detailed analysis at textual, discursive and socio-cultural level. First a textual analysis in detail is presented with the instrument of lexical classification and transitivity.
4.1 Lexical Classification Analysis
More specifically, the following 10 news discourse will be analyzed from three aspects: firstly, the coverage concerning the events which includes the naming of the event, the verbs employed to describe the event will be analyzed; secondly, the media coverage of the performers of the event will be analyzed and finally, the media coverage of the inflicted people and organization will also be analyzed. By examining the way that the events are presented, we attempt to have a relatively thorough idea of the events and to have a better understanding of how different viewpoints or ideological positions are constructed through linguistic forms.#p#分頁標題#e#
Referring to the event of the Lhasa riots, different lexical choices are made by the two groups of reporters of PD and NYT. There are striking differences in the naming and the description of the same event. Different nouns, adjectives and verbs are employed by these two groups, which nonetheless represent their distinct attitudes and positions.
Table 4.1Media coverage concerning the event
People’s Daily New York Times

riot
The violent crime
a bloody violence
"separatist sabotage"
separatist plot
activities of splitting the nation
evil acts
a political conspiracy plotted by the Dalai clique to split Tibet,
violent array of sabotage
organized and premeditated" by the Dalai clique.
organized, premeditated and masterminded" by the Dalai clique.
sabotage the peaceful and harmonious life,
disrupted the public order and jeopardized people's lives and property
confronted government departments and assaulted law enforcement officers,
violated China's criminal law
their criminal activities
undermined the religious order and peace in Lhasa
destroy the order of the Tibetan Buddhism
poses severe threat to life, stability
aroused huge indignation among the people
undermine the normal life and harmony Friday’s unrest
the protests
The tumult
bloody clashes
The protests in Lhasa
most violent, confrontation in a series of protests
major challenge to the ruling Communist Party
most serious and prolonged demonstrations
demonstration
the unexpected demonstrations
peaceful protests against religious restrictions by Chinese authorities.
a political and public relations challenge for the ruling Communist Party as Beijing prepares to play host to the Olympic Games in August.
largely ignored in the Chinese news media
a blow to President Hu Jintao
undercuts a theme regularly promoted by China’s propaganda officials, that Tibetans are a happy minority group

The different lexical structure employed in naming and describing the Lhasa riots contains important information and reflects the news reporters’ attitude. Language is never transparent and neutral, instead, it is loaded ideology, and new reports with its large circulation and great influence can somehow affect or even manipulate people’s opinions of the facts.
The Chinese sides employs “riots” which is of great negative connotation to describe what happened in Lhasa on March 14th, 2008 and the following up events. They consider the event as “evil acts” and “violent array of sabotage”, they believe they are “activities of separatist plot” and a “political conspiracy plotted by the Dalai clique to split Tibet”. All of these words, such as “evil” “plot” “conspiracy” clearly manifest the stance of the Chinese news reporters. They are supporters of the Chinese government, and hold quite negative attitude to the rioters. Indeed, during the horrible day, people lives were deprived, damages were conducted and the normal life and harmony of the Tibetan residents were ruined. It is reasonable for the Chinese reporters to depict the reality as objective as possible, but at the same time, under the influence of their culture and value, to depict what happened in this way.#p#分頁標題#e#
As for the American side, they employ much gentler words to describe the event. Though they employ “bloody clashes”, “violence” occasionally in order to emphasize the seriousness of the events, they use “Friday’s unrest”, “protest in Lhasa”, “demonstration” more frequently. And they report that initially, they are “peaceful protests against religious restrictions by Chinese authorities” The choice of the latter words severely distort the nature of what happened in Lhasa that day, and reflects American’s attitudes to the riots and the rioters: they share some sympathy with them. Besides, by describing the riots as initially “peaceful protests against religious restrictions” by Chinese authorities, an illusion of dictator China is created, in which people have no freedom of religion and have terrible human rights.
For a very long period of time, the United States shares different viewpoints with the Chinese government on many aspects: the Tibet Issue, the Tai Wan issue and the Falun Gong spiritual movement. In their eyes of the ordinary America, China is notorious for its abuse of human rights and suppression of dissent, and on the occasion of the Tibet riots, they underlying value is again reflected through the word choice they make.
Table 4.2 Media coverage concerning Dalai Lama and the rioters
People’s Daily New York Times
law breaker
lawless people
violent saboteurs
rioter
a handful of Buddhist monks
reactionary force
the mobs
the insane people
instigated and planned the riot in Lhasa
premeditated the riot angry Tibetan crowds
Tibetan protesters
angry mobs
disobedient monks.
protest the rough treatment of monks
Groups that promote Tibetan independence
are still protesting

Concerning the performers of the Lhasa riots, different expressions are also employed. According to the Chinese side, the performers of the event are “lawless people”, “insane people” “reactionary force”. These words illustrate the seriousness and nature of their behavior, which is not in the mainline with the Chinese government and might affect the construction of a harmonious and lawful Tibet. But the American media describe these people as “angry” and “disobedient”. People become angry when their benefit is threaten or in bad terms with others, thus the word “angry” depicts a tense relation between the Tibetans and the Chinese government, while the word “disobedient” undermines the bad consequences of the rioter’s actions.
Table 4.3Media coverage concerning the police and the Chinese Government
People’s Daily New York Times

have rescued more than 580 people
ordered not to use force against the attacker
on close lookout for comeback of violence.
rescue residents under attack. #p#分頁標題#e#
were stoned
taken necessary measures to maintain stability
condemned the "separatist sabotage" firing into crowds of protesters
fled after an initial skirmish
killing at least two people
pouring in to stifle new protests
suppressed a rebellion there with lethal force
grappling with growing criticism of China’s domestic human rights record
denied that they had fired on protesters
defended their response to the violence in Lhasa
failed to earn the respect of the people there.
smother separatism by sparking economic development
tightly controlled its coverage to focus on Tibetans burning
acutely worried about an international reacted
tried to soften its image on Tibet
infuriated many Tibetans by trying to monopolize the most sacred rituals of Tibetan Buddhism
cannot win the hearts of Tibetans
Expressions in NYT concerning the Chinese police and the Chinese government, such as “denied that they had fired on protesters”, “suppressed a rebellion there with lethal force”, “stifle new protests”, “tightly controlled its coverage to focus on Tibetans burning” provide readers with an illusion that people in China are deprived of freedom and are put in tight control of an arbitrary government. These expressions contain negative connotations that Chinese government is a dictatorial government and Chinese people are living miserably under political pressure. The Tibetans have no right to express themselves and make their own decisions. They are in such an inferior position that being obedient is the only way to survive. The western reporters also depict China government as　“failed to earn the respect of the people there and ”infuriated many Tibetans” to depict a negative picture of relationship between Chinese government and its people.
While on the Chinese side, another different picture was shown. It is reported that policemen have" rescued more than 580 people” instead of “killing at least two people”; they are “ordered not to use force against the attacker” even when their lives are under threat instead of “firing into crowds of protesters" and “killing at least two people”. The reports from the Chinese side contribute to a positive image of the Chinese police-the public servants of Chinese government, in contrast with that depicted in the New York Times.
The detailed study of the lexical classification in the collected news reports explicitly display that China, and the U.S. do not report the Lhasa Riots in the same way. Towards the event itself, and the people inflicted, including both the initiators and the harmed, they employ different words and thus show their different attitudes and emotions. Actually, lexical classification is the representation of the world for a social group, and the lexical structures reporters use are reflections of their ideologies. From the analysis of the lexical classification, a conclusion can be reached that the Chinese side makes strenuous efforts to ensure the safety of the Lhasa residents and the social stability, while the U.S. attempts to impose some negative image on it.#p#分頁標題#e#
4.2 Transitivity Analysis
Transitivity is a fundamental and powerful semantic concept in the theory of Halliday. It is the main element of Halliday’s category of the ideational function and an essential tool in the analysis of representation. The central idea of Halliday’s transitivity refers to how meaning is represented in the clause; the way the clause is used to analyze events and situations as being of certain types. Transitivity enables the same events to be analyzed in different ways and therefore attracts great interest in newspaper analysis
Halliday (1994:107) identifies the way in which transitivity carries out the ideational function as processes. Halliday identifies three main types of processes: material process, mental process and relational process, and three subsidiary types of processes: behavioral process, verbal process and existential process. All these six types of processes are potentially composed of three components: the process itself, participants and circumstances
Transitivity analysis operates on the basis of identifying types of process, the role of participants in the processes and also the associated circumstances. Since there is always a choice between different grammatical process and participant types, the selection can be ideologically significant. Which type of process is employed or how to arrange the location of the participants is mainly determined by ideology.
Table 4.4 Distribution of process types in news samples
Material Mental Relational Behavioral Verbal Existential Total
PD 64(70%) 6(6.5%) 9(9.8%) 0(0%) 12(13.1%) 0(0%) 91(100%
NYT 141
(70.1%) 18
(8.9%) 16
(7.9%) 3
(1.4%) 20
(9.9%) 3
(1.5%) 201
(100%)
The table above illustrates that material processes have the largest distribution in the news discourse from both sides. “Material processes are processes of ‘doing’. They express the notion that some entity ‘does’ something—which may be done ‘to’ some other entity” (Halliday, 1994: 110). In the news discourse, material processes have the largest distribution, which contributes to the objectiveness and validity.
The ‘doer’ of the action is labeled as Actor which is obligatory in any material process and at the same times can be omitted in the clause. Second participant the action may affect or ‘being done to’ is called Goal and the following are some material examples.
Table 4.4Material processes examples
Actors Process
The regional government
reinforced57- protection for schools, hospitals and gas stations,
took 55-emergency measures to rescue56- residents under attack,
required the government agencies and businesses to ensure safety of their employees.
Local government informed the citizens of the sabotage through TV
imposed traffic control on the main streets in Lhasa Friday night
The police didn't chase them#p#分頁標題#e#
As mentioned above, Actor is always being represented as powerful and responsible for the action and the Goal less powerful and more disadvantaged. In the material processes, the actors are always the local government or the police that serve the local government, therefore displace a clear sense of responsibility of them and this structure is chosen to place emphasis on the agent of the action, implying clear responsibility, and subtly offering readers a favorable picture of the local government and the police there. The material processes here emphasize the quick reaction and reasonable handle of the regional government, and illustrates their concern for the people and property in the local area. The Lhasa riot erupts not because of a tense relation ship between the government and the local residents, at least not from the side of the government. In confrontation of the riot, what the government attempted to do is to protect public facilities, keep the people well informed and ensure the safety of the innocent people. The police refuse to resort to life-threatening weapons even when their own lives are under threat. This helps to create a positive image of China and it is in complete accordance with the principle advocated in China that the interest of the people comes first.
Table 4.5 Examples of material processes
Actors Process
Beijing is facing the most serious and prolonged demonstrations in Tibet since the late 1980s
It(Beijing) suppressed a rebellion there with lethal force, that left scores, and possibly hundreds, of ethnic Tibetans dead
military police officers were blocking many city streets.
It is evident that Actors in NYT are mainly associated with negative processes. Beijing, the capital and the embodiment of China, is in crisis and the military police were blocking many city street. Besides, the purpose of western reporters employ Beijing, or the police officers as Actor is to imply the clear responsibility of them for the deed and implicitly blame Chinese government for its abuse of right and the terrible treatment of the Tibetans，thus sending a clear message to readers that the Tibetans in China receive bad treatment and enjoy little human right, and at the same time that the Chinese government lack the competence to handle the Lhasa riot.
4.3Intertextuality Analysis
4.3.1Analysis of the discourse representation mode
As mentioned in 3.3.3, Fairclough (1995a) identifies DD (Direct Discourse), ID (Indirect Discourse), DD(S) ( a category for cases of ‘slipping between modes ID and DD) and UNSIG. Among the four modes, UNSIG could only be identified by checking the discourse at issue against the discourse it has possibly drawn upon. Its identification is impractical in the present study. The criteria for identifying modes of discourse representation mainly depend on the discussion in section 3.3.3. In both Chinese and American news samples, all three modes of discourse representation are employed. The following table shows the incidence of modes of discourse representation in the news reports.#p#分頁標題#e#
Table 4.6 Modes of discourse representation
DD DD(S) ID Total
PD 18(37.5%) 5(10.41%) 25(52.03%) 48
NYT 33(34.02%) 14(14.43%) 50(51.55%) 97
Notes: DD=Direct Discourse DD(S)=Sub-type Direct Discourse
ID=Indirect Discourse
It can be concluded from the table that in the reports of both Chinese media and the American media, all three modes of discourse representation are employed to provide information as well as the standpoints. The two newspapers share some common characteristic in the low distribution of direct discourse slipping, and show subtle difference in the employment of the other two modes. The people’s daily has a slightly higher frequency in the mode of direct discourse while the New York Times has more indirect discourse while the voice of the government or common people is transmitted.
The employment of DD may serve many purposes. Since the direct quote is exactly what the news source presents, DD could achieve ostensible truthfulness, and leave an impression on the readers that the event is represented objectively and neutrally. It is for the objectiveness and neutrality that the people’s daily employs the mode of DD, and a close look at the sources can reveal that most of the second discourse is quoted from the government, or government officers, who are the representatives of the voice of the masses and inclined to give the facts of the event rather than some biased opinions.
The following are two examples extracted from PD:
"We have sufficient evidence proving that the Lhasa riot was instigated, planned and organized by the Dalai clique," Ragdi said.
(PD, March 18, 2008)
Lhazom Zhoigar, vice-chairwoman of the regional Chinese People's Political Consultative Conference, said, "I am a witness of old and new Tibet. Before the peaceful liberation in 1959, poor Tibetans lived worse than beasts of burden. The new Tibet, especially since the national reform and opening up in 1978, experienced rapid development in all fields of politics, economy and culture, she said.
（PD, March 16,2008）
Though New York Times also employs the mode of DD, it is in slightly lower incidence and some of second discourse is from organizations or leading figures that displays unfavorable opinions on China, for instance:
“Tibet is not free! Tibet is not free!” yelled one young Buddhist monk, who then started crying.
（ NYT, March 27, 2008）
“This wasn’t organized, but it was very clear that they wanted the Chinese out,” said the American woman who witnessed the riots and asked not to be identified for fear of reprisals.
（NYT, March 24, 2008）
A close look at the second discourse represented in the mode of DD in NYT reveals that most of them are expressions of personal judgments and attitudes rather than stating facts and are concerned with doubt。 Though Reporters in NYT attempt to achieve an air of objectivity by quoting the statements directly, what they quote might depict a negative image of China and mislead the readers, for instance:#p#分頁標題#e#
“What we see right now, what is happening in Tibet, blows the whole propaganda strategy in Tibet wide open,” said Lhadon Tethong, an official with the New York-based advocacy group Students for a Free Tibet.
(NYT, March 16,2008)
Why did the unrest take off?” asked Liu Junning, a liberal political scientist in Beijing. “I think it has something to do with the long-term policy failure of the central authorities. They failed to earn the respect of the people there.”
(NYT, March18, 2008)
ID is inherently ambivalent in voice, which in turn could blur the boundary between the reported and reporting speech. According to Fairclough (1995b), mass media are inclined to employ indirect speech, which is well illustrated by the frequent use of ID mode by the two newspapers here.
In his account of media discourse, Fairclough (1995, p.81) argues that “one feature of indirect speech is that although it is expected to be accurate about the propositional content of what was said, it is ambivalent about the actual words that were used-it may simply reproduce them, or it may transform and translate them into discourses which fit more easily with the reporter’s voice.”The choice of indirect speech depends upon reporter’s personal understanding of the original verbiage, combined with his stance and the communicative purpose. Therefore, the reporting of others’ words is not a neutral account of what has been said and the indirect speech is highly interpretative.
Although the two newspapers both employ indirect discourse as the major means of representation, differences emerge after a careful look of the sources of representation. PD mainly takes Chinese officials, Chinese and even some foreigner as the news sources, while the NYT employs western agencies and Tibet-free advocate as the sources. Sometimes, it is difficult to distinguish the true intention of the speakers and that of the newspapers.
The Dalai Lama released a statement on Friday calling on both sides to avoid violence and appealing to China’s leaders to “address the long simmering resentment of the Tibetan people through dialogue with the Tibetan people.
(NYT,March15, 2008)
When police officers began beating the monks, Tibetans rioted in the Barkhor area, the advocate said. （NYT,March15, 2008）
As for the type DDs, the sub-type DD, it remains the least in the distribution of discourse representation modes in the sample reports from both sides. It allows an author to slip from narrative statement to interior portrayal which is difficult for readers to notice what has occurred, and makes readers have little choice but to trust the views of the narrator (Leech and Short，1981). Thus the most subtle manipulate readers’ understanding and their points of view.
After a careful examination of the content of DDs in PD, it is shown that “separatist sabotage”, and “organized and premeditated” are more frequently employed, which uncover the nature of the Lhasa riots and the harm that the event causes. In this way, the standpoint of the Chinese government toward the event itself is implied.#p#分頁標題#e#
However, in the NYT, quotations are made from different sources. For instance,
A spokesman for the Dalai Lama called China’s accusations “absolutely baseless.”
(NYT,March16,2008 ）
The White House called on China to “respect Tibetan culture” and issued a renewed call for dialogue between Beijing and the Dalai Lama, the exiled spiritual leader of Tibetan Buddhism.
NYT ,March16, 2008)
It demonstrates that the American press attempts to belittle the severe harm caused by the Lhasa riots, and display their sympathy for the so named Dalai Lama. To some extent, the NTY even depict China as a dictator country in which the freedom of religion is controlled and different cultures are undermined.
In general, no discourse representation is absolutely objective or neutral. No matter which mode is employed, an imposition of reporter’s voice on the original voice is inevitable. Readers have little idea of whether the original utterances are reproduced or not. It is impossible to distinguish the view of the reporters from that of the press itself. Through selecting what and how to report the event, news reporters give subtle cues to their own stances and values.
4.3.2News sources
News reports rely heavily on other people’s verbal descriptions of events. The quality of the news sources counts because it directly affects the value of the news reports produced. Meanwhile, the preference to different sources and voices implies diverse attitudes towards each source. The press, in general, quotes those whose views do not contradict the points of view of the newspapers (van Dijk, 1993). Xin Bin (2000:171) also claims that it is up to the reporter, and ultimately the newspaper he or she works for, to decide whose voice is to be heard on what issue and such decisions are usually highly significant. Therefore, the sources of information need to be investigated.
Table 4.7 Distribution of news sources
Specific Semi-identified unidentified total
PD 28（62.3%） 15(33.3%) 2(4.4%) 45(100%)
NYT 28（62.3%） 15(33.3%) 2(4.4%) 45(100%)
The above table is a clear illustration of the news sources for the different speech. Both sides employs information with unidentified sources at the lowest level, (being4.4% in PD and 6.0% in NYT respectively), and they employ specified and semi-specified sources more frequently.，which is in accordance with the newspaper’s goal to achieve objectivity. However, PD and NYT display difference 留學生新聞媒體學專業(yè)碩士dissertation定制to some extent. Reporters from the Chinese side employ more use of news with specified sources, for instance：Lhazom Zhoigar, vice-chairwoman of the regional Chinese People's Political Consultative Conference, said; Dawa Toinzhub, president of Lhasa-based Dashi Group, said; said Qiangba Puncog, chairman of the Tibet Autonomous Regional Government, on the sidelines of China's annual parliament session.(PD, March 19,2008 ).#p#分頁標題#e#
All this information provides a clear source of the secondary discourse, hold the very specific person for responsibility, and make the reports more convincing and trustworthy and achieve the objectivity and credibility.
On the other side, the American reporters rarely mention the specific sources of information, and when it does, it is to the disadvantage of China.
“The party must surely know these monks are not going to change their minds” about the Dalai Lama, said Tsering Wangdu Shakya, a Tibet expert at the University of British Columbia. “So the whole point of the meetings is to intimidate the monks.” (NYT, March 18, 2008)
The American side employs reported speech from semi-specified sources most frequently. In their reports are most frequently are: Witnesses said, a spokeman for DalaiLama said, a Tibet analyst said, a Tibet advocate said, in which no one bear the responsibility. They are also attempting to achieve the objectivity. But a number of reported speeches come from people’s complaints or criticisms, which might subtly create a negative image of China to the world.
4.4 Explanation
The explanation of social and institutional context will also enable us to figure out why different newspapers present readers with remarkably different news reports.
4.4.1 Institutional practice
The content and the tones of the news reports are determined by a country’s institutional practice to a large extent. Therefore, to explore how the Chinese and American media is institutionalized is of great help and importance in the explanation of divergence on the news discourse on the Lhasa Riots.
Generally speaking, the American mass media has been commercialized. According to Hiebert (1999, P. 12), the American mass media, which are owned now primarily by large, public held corporations, have been oriented to the bottom line of money. In terms of censorship, Bakhtin (1993) claims that the American institutional press is largely free and unregulated and is generally subject less to government’s whim than to its own lapses of judgment. In theory, American media (here with NYT as its representative) is quite different from that of China, as in China the media is mainly government–controlled. In fact, it is impossible for the mass media to be free from the influence of the government. That is why the tones and mainstream ideology of the U.S. news reporting can still be unanimous and be in accord with the government most of the time. As American government has an inclination to depict a negative image of China. Tragedies and disasters in China are inclined to be emphasized.
In China, the mass media is controlled by government, and the voice of the officials can often be heard. Xinhua news agency as the official one, from which PD mainly obtains the up-to-date information, displays the voice of the government. What they attempt to do in the Lhasa Riot is to give a factual statement of the event and establish a positive image of the authority and maintain the stability of the society. That explains positive linguistic description of the policemen in the riot. #p#分頁標題#e#
4.4.2 Cultural and political context
China and the U.S. are culturally and politically different. From the perspective of culture, American society is mainly a religious one and most of the Americans are either Christian or other religion-supporter. Though Christianity is not longer as powerful as it was in the middle age, it still plays a significant role in modern America. Since the first settlers came to the American continent, they have viewed themselves as the most noble nationality who were sent by God to govern the world and to spread the values of ‘democracy’ and ‘freedom’ to those unenlightened nations. In their eyes, in Tibet, a remote place in China under the rule of the communist party, democracy and freedom, especially freedom of religion rarely exist. This cultural context help to explain why in the news discourse of NYT, some expressions that display sympathy for the violent Tibetans rioter exist.
As for China, it receives influence from the traditional Confucianism. Though some religious people also can be found here, they are of the minority. Most Chinese are religion-free. Under the influence of Confucianism and what they attempt to achieve is a concept of harmony. Therefore, in face of the Lhasa riot, they reject to employ fatal weapons, which are factually reported by the Chinese media.
The political system underlying both countries also creates a wide gap. The U.S. is a capitalist country, and two leading parties, the Republicans and the Democrats take control of the government alternatively. “Though they might differ in certain domestic policies to some extent, they are united in the cause of anti-communism in foreign policy and export of American capitalist’s values”. (Pan,2003, p.315). There has existed a sense of deeply rooted anti-communism in the American society. According to pan (2003), it needs great courage and tactics to say something good about the communist in the American society. This is proved by the news discourse samples collected, in which no positive image of the Chinese authority or the police is created. What impresses the readers is that they “suppress” the riots, “firing into the angry crowd”, “China security wavered”, all of which depict an image associated with “oppression, totalitarianism”. China, as a socialist country, has the communist party as its leading party, which believes in Marxism and the significance of the unity of all ethnic groups. Actually, just a Tibetan said (PD, March 16, 2008), "I am a witness of old and new Tibet. Before the peaceful liberation in 1959, poor Tibetans lived worse than beasts of burden. The new Tibet, especially since the national reform and opening up in 1978, experienced rapid development in all fields of politics, economy and culture”. This helps to create a positive image of the leading party in China, which is quite different from that the American media depict.
4.4.3 Historical background
Historically, the US and China hold different stances on the Tibet Issue which hence influence the discourse representation of reports on Lhasa Riots. U.S. will give its full attention to anything related to Tibet. Ever since 1940s, the U.S has devoted their attention to Tibet, and encouraged the independence of it. The armed rebellion staged by the Dalai clique to split the country in 1959 was partly supported by U.S. Ever since the Dalai clique went into exile, U.S has never ceased their instigation and training of the Dalai clique to support their split and sabotage activities. It is shown that Voice of America has started a Tibetan channel ever since 1991, in which year the Dalai Lama in exile was met privately by the present Bush; and it is reported that VOA assisted the spread of the information for the 3.14 riots for the Tibet splitists. Many organizations for Tibetan independence are established in the U.S. and ICT, a leading organization by the Dalai clique has its headquarters established in Washington D.C .Information from other sources also shows that Americans hailed the 14th Dalai Lama -- chief representative of the theocratic feudal serfdom and the Tibetan serf-owners -- as a "guardian of human rights," "peace envoy," and "spiritual leader," and U.S has been providing enormous financial support for the Dalai Clique. All of these events happened in history and recently clearly displayed the stance of U.S. on the issue of Tibet. T. Many westerners, including some Americans still use the Dalai Lama as a tool to attack Chinese government and Communist Party. Though some of them acknowledged Tibet as an integral part of China, meanwhile they also support Dalai Lama’s secessionist activities. They might lose no opportunity in using Lhasa Riots against China. What happened in history inevitably has a profound influence upon the present American citizens, no exception to the news reporters who explicitly or implicitly display their ideologies in the news discourse.#p#分頁標題#e#
On the other side, China is, and will always, stick to the policy of national unity and territorial integrity, and will make no concession to the Tibet independence force on the matter of sovereignty and principle. The year 2009 marks the 50th anniversary of the end of rebellion and democratic reform in Tibet. During these years, Tibet has witnessed profound changes and experienced rapid development in all fields of politics, economy and culture, traditional Tibetan culture and heritage is protected, freedom in religious belief is promoted, and the human rights are ensured. An increasing number of tourists in Tibet can prove that. Without the abolishment of the feudal serfdom, without the correct lead of the Chinese Party, all the above can not be achieved.
In summary, different representations of the same issue in different newspapers are mainly the different due to ideologies out of which opinions are formed and judgments made.
Chapter Five Conclusion
Based on Halliday’s systemic-functional grammar and Fairclough’s three-dimensional model of discourse analysis, the present thesis has conducted a comparative critical analysis of ten news reports from People’s Daily and New York Times to illustrate how the same issue is represented differently in these two newspapers..
The analysis leads to a conclusion that obvious difference exists between the discourse representations of the news reports from these two sides. Although news reporters attempt to be impartial, serious and objective, the seemingly neutral lexical and structural patterns conceal the naturalized ideological assumptions and implications.
5.1 Major Findings
The results of the present study are in accordance with those of the former critical analysts, indicating that news discourse is value-loaded rather than value free. They “constructed realities” with an ideological dimension which is embedded in the discourse representation. Just as Fairclough states that ideology works best when it is least apparent. Thus the news discourse, with concealed ideology will inevitably influence the understanding of the readers.
Linguistic strategies are employed by newspapers to represent event in a manner congruent with the socio-ideological context. The relevant underlying social and intuitional contexts do exert significant influence on the modes of the event being represented. It is revealed that language and discourse are essentially socially constructed and are products of values, beliefs or ideologies which are, in turn, reinforced and maintained through language and discourse.
5.2 Implication
With the findings and insights drawn from the previous sections, it is believed that the study has some enlightening implications.
Firstly, in doing CDA research, especially those related with news discourse, a comparative study is an effective way to reveal the hidden ideology encoded in the linguistic features. It is very difficult for people to perceive or identify the ideological stance embedded in news texts from unitary sources. It will be helpful to have the news texts from different sources in terms of ideological standpoints available when analyze news discourse from the perspective of CDA.#p#分頁標題#e#
Secondly, nowadays, mass media develop very quickly and they become a major source of people’s information. In face of the large pour of news articles, it is necessary for the readers to hold a critical attitude towards what they read in order to have a better understanding of what has happened.
Thirdly, traditionally, in the process of language learning and teaching, the meanings of words and sentence structures are stressed, while the macro social and historical contexts are mainly neglected. Actually, language is both socially shaping and socially shaped. Therefore, it is necessary to attach emphasis to the social and historical background and the critical awareness should be raised.
5.3 Limitation
Though the present study provides some invaluable implications, some limitationst do exist and there is much room for further study.
Firstly, this study is mainly a qualitative one, it is just a study based on a detailed analysis of a news samples with limited size, thus the conclusions reached here are tentative and open. They are just a tip of iceberg and might not be applied to the discourse of other cases. Though some quantitative work is also conducted here, the accuracy of the data can not reach the level of being one hundred percent precise as all the work is done manually.
Secondly, only a limited number of analytical tools are employed, such as the transitivity, intertextuality, etc. though they are also important tools for critical study, they could not replace other tools used in the field of CDA and this also undermines the validity of the study.
Thirdly, due to the limit in knowledge scope and cultural background, it is difficult to be absolutely objective in presenting the descriptions and interpretations of the ideological assumptions. Consequently, the validity and accuracy of the analysis might be reduced.

Bibliography
Bell, A. (1991). The Language of News Media. Cambridge, MA: Blackwell.
Bell, A. (1995). Language and the Media. Cambridge: Cambridge University Press.
Bell, A. & Garrett, P. (1998). Approaches to Media Discourse. Oxford: Blackwell.
Bloor, T. & Bloor, M. (2001). The Functional Analysis of English: A Hallidayan Approach. Beijing: Foreign Language Teaching and Research Press.
Brown, G. & Yule, G. (1983). Discourse Analysis. Cambridge: Cambridge University Press.
Cameron, D. (1993). Style policy and style politics: A neglected aspect of the language of the news. Media, Culture & Society, 15: 315-33.
Chouliaraki, L. & Fairclough, N. (1999). Discourse in Late Modernity: Rethinking Critical Discourse Analysis. Edinburgh: Edinburgh University Press.
de Beaugrande, R. & Dressler, W. (1981). Introduction to Text Linguistics. London: Longman.
Fairclough, N. (1989a). Language and Power. London: Longman.
Fairclough, N. (1989b). Language and ideology. English Language Research Journal, 3: 9-28.
Fairclough, N. (1992a). Discourse and Social Change. Cambridge: Polity Press.#p#分頁標題#e#
Fairclough, N. (1992b). Critical Language Awareness. London: Longman.
Fairclough, N. (1993). Critical discourse analysis and the marketization of public discourse. Discourse & Society 4(2): 133-168.
Fairclough, N. (1995a). Critical Discourse Analysis: the Critical Study of Language. London: Longman.
Fairclough, N. (1995b). Media Discourse. London: Edward Arnold.
Fairclough, N. (1998). Political news in the media. In Bell, A. & Garrett, P. (eds.), Approaches to Media Discourse, pp. 142-162. Oxford: Blackwell.
Fairclough, N. (2001). Critical discourse analysis as a method in social scientific research. In Wodak, R. & Meyer, M. (eds.), Methods of Discourse Analysis, pp. 121-138. London: Sage.
Fairclough, N. (2003). Analyzing Discourse: Text Analysis for Social Research. London: Routledge.
Fang, Y. J. (1994). ‘Riots’ and demonstrations in the Chinese press: A case study of language and ideology. Discourse & Society, 5(4): 463-481.
Fang, Y. J. (2001). Reporting the same event? A critical analysis of Chinese print news media texts. Discourse & Society 12(5): 585-613.
Fowler, R., Hodge, B., Kress, G. & Trew, T. (1979). Language and Control. London: Routledge.
Fowler, R. (1987). Notes on Critical Linguistics. In Steele, R. & Threadgold, T. (eds.), Language Topics: Essay in Honor of Michael Halliday. Amsterdam Philadelphia: John Benjamins.
Fowler, R. (1991a). Language in the News: Discourse and Ideology in the Press. London: Routledge.
Gee, P. (2000). Introduction to Discourse Analysis: Theory and Method. Beijing: Foreign Language Teaching and Research Press.
Geis, M. L. (1987). The language of Politics. New York: Springer-Verlag.
Golan, G. (2006). Inter-media agenda setting and global news coverage. Journalism Studies, 7(2): 323-333.
Hall, S., Critcher, C., Jefferson, T., Clarke, J. & Roberts, B. (1978). Policing the Crisis: Mugging, the State, and Law and Order. London: Macmillan.
Hall, S. (1980). Culture, Media, Language. London: Hutchinson.
Halliday, M. A. K. (1978). Language as Social Semiotics: The Social Interpretation of Language and Meaning. London: Edward Arnold.
Halliday, M. A. K. (1994). An Introduction to Functional Grammar. London: Edward Arnold.
Harvey, D. (1996). Justice, Nature and the Geography of Difference. Oxford: Blackwell.
Herman, E. S. & Chomsky, N. (2002). Manufacturing Consent: the Political Economy of the Mass Media. New York: Pantheon Books.
Hodge, R. & Kress, G. (1988). Social Semiotics. Cambridge: Cornell University Press.
Hodge, R. & Kress, G. (1993). Language as Ideology. London: Routledge.
Janks, H. (1997). Critical discourse analysis as a research tool. Discourse: Studies in the Cultural Politics of Education 18(3): 329-342.
Johnstone, B. (2002). Discourse Analysis. Oxford: Blackwell.
Kress, G. (1985). Linguistics Processes in Sociocultural Practice. Oxford: Oxford University Press.#p#分頁標題#e#
Kress, G. (1985). Ideological structure in discourse. In van Dijk, T. A. (ed.), Handbook of Discourse Analysis, (Vol. 4), pp. 27-42. New York: Academic Press.
Kress, G. (1986). Language in the Media: the Construction of the Domains of Public and Private. Media, Culture and Society, (8): 395-419.
Kress, G. & van Leeuwen, T. (1996). Narrative representations: designing social action. In Kress, G. & van Leeuwen, T. (eds.), Reading Images: the Grammar of Visual Design, pp. 43-78. London: Routledge.
Kress, G & van Leeuwen, T. (1996). Reading Images: the Grammar of Visual Design. London: Routledge.
Leech, G. & Short, M. (1981). Style in Fiction. London: Longman.
Luke, A. (1997). Theory and practice in critical science discourse. In Saha, L (ed.)
Pennycook, A. (1994). Incommensurable discourse. Applied Linguistics, (5): 119-138.
Pennycook, A. (2001). Critical Applied Linguistics. Mahwah, NJ: Lawrence Erlbaum.
Schiffrin, D. (1994). Approaches to Discourse. Oxford: Basil Blackwell.
Schiffrin, D., Tannen, D. & Hamilton, H. E. (2001). Handbook of Discourse Analysis. Oxford: Blackwell.
Simpson, P. (1993). Language, Ideology and Point of View. London: Routledge.
Skyes, M. (1985). Discrimination in discourse. In van Dijk, T. A. (ed.), Handbook of Discourse Analysis, (Vol. 4), pp. 83-101. New York: Academic Press.
Stubbs, M. (1983). Discourse Analysis:The Sociolinguistic Analysis of Natural Language. Oxford: Blackwell.
Teo, P. (2000). Racism in the news: A critical discourse analysis of news reporting in two Australian newspapers. Discourse & Society, 11(1): 7-49.
Thompson, J. B. (1990). Ideology and Modern Culture. Cambridge: Polity Press.
Thompson, G. (2000). Introducing Functional Grammar. Beijing: Foreign Language Teaching and Research Press.
Toolan, M. (2001). Narrative: A Critical Linguistic Introduction. London: Routledge.
Toolan, M. (2002). Critical Discourse Analysis: Critical Concepts in Linguistics, 4 vols. London: Routledge.
van Dijk, T. A. (1977). Text and Context: Explorations in the Semantics and Pragmatics of Discourse. London: Longman.
van Dijk, T. A. (1983). Discourse Analysis and the Structure of News. Journal of Communication, 33(2): 20-43.
van Dijk, T. A. (1985b). Handbook of Discourse Analysis, 4 vols. New York: Academic Press.
van Dijk, T. A. (1988a). News as Discourse. Hillsdale, NJ: Lawrence Erlbaum Associates.
van Dijk, T. A. (1988b). News Analysis: Case Studies of International and National News in the Press. Hillsdale, NJ: Lawrence Erlbaum Associates.
van Dijk, T. A. (1991). Racism and the Press. London: Routledge.
van Dijk, T. A. (1993). Elite Discourse and Racism. London: Sage.
van Dijk, T. A. (1993). Principles of Critical Discourse Analysis. Discourse & Society, 4(2):249-283.
van Dijk, T. A. (1995). Discourse analysis as ideology analysis. In Schaffner, C & Weden, A. L. (eds.), Language and Peace. Dartmouth: Aldershot.#p#分頁標題#e#
van Dijk, T. A. (1997). Discourse as interaction in society. In van Dijk, T. A. (ed.), Discourse as Social Interaction – Discourse Studies: A Multidisciplinary Introduction. (Vol. 2). London: Sage Publications.
van Dijk, T. A. (1998). Opinions and ideologies in the press. In Bell, A. & Garrett, P. (eds.), Approaches to Media Discourse. Oxford: Blackwell.
van Dijk, T. A. (2001a). Critical discourse analysis. In Schiffrin, D., Tannern, D. & Hamilton, H. E. (eds.), Handbook of Discourse Analysis, pp. 352-371. London: Blackwell.
van Dijk, T. A. (2001b). Multidisciplinary CDA: a plea for diversity. In Wodak, R. & Meyer, M. (eds.), Methods of Critical Discourse Analysis, pp. 95-120. London: Sage.
van Leeuwen, T. (1993). Genre and field in discourse analysis. Discourse & Society 4(2): 193-223.
van Leewuen, T. (1997). Media in education. In Wodak, R & Corson, D. (eds.), Encyclopedia of Language and Education. Netherland: Kluwer Academic Publishers.
Weiss, G. & Wodak, R. (2003). Critical Discourse Analysis: Theory and Interdisciplinarity. New York: Palgrave Macmillan.
Wodak, R. (1995). Critical linguistics and critical discourse analysis. In Verschueren, J, Ostman J. O. & Blommaert, J (eds.), Handbook of Pragmatics. Amserdam: Benjamins.
Wodak, R. (1997). Critical discourse analysis and the study of doctor-patient interaction. In Gunnarsson, B. L, Linell, P. & Nordberg, B. (eds.), The Construction of Professional Discourse. London: Longman.
Wodak, R. & Meyer, M. (2001). Methods of Critical Discourse Analysis. London: Sage.
Wodak R. The Discourse-historical Approach. In Wodak R, MeyerM. Approac hes to Critical Discourse Analysis. ]London:Sage,2001,63-94
Wodak.R.&Chilton, P. (Eds).(2005).A new agenda in critical discourse analysis: theory, methodology and interdisplinary. Philadephis: John Benjamins.
陳中竺,留學生新聞媒體學專業(yè)碩士dissertation定制 (1995a), 批評語言學述評,《外語教學與研究》，(1): 21-27.
陳中竺, (1995b), 語篇與意識形態(tài)：批評性語篇分析,《外國語》，(3):42-45.
苗興偉,(2004), 話語轉向時代的語篇分析:《中國海洋大學學報》 (6)
苗興偉,(2006),語篇分析的進展與前沿: 《外語學刊》 (1)
戴煒華、高軍, (2002), 批評語篇分析：理論評述和實例分析,《外國語》，(6): 42-48.
潘志高, (2003), China’s Image in the New York Times: Political, Historical and Cultural Factors, 河南：河南出版社.
彭宣維, (2000),《英漢語篇綜合對比》, 上海: 上海外語教育出版社.
王慶新、紀衛(wèi)寧, (2000), 批評式語篇分析及其應用,《山東外語教學》，(1): 29-33.
辛斌, (1996a), 語言、權力與意識形態(tài)：批評語言學,《現(xiàn)代外語》，(1): 21-26.
辛斌,(1996b) 語篇互文性的批評性分析, 蘇州：蘇州大學出版社,26#p#分頁標題#e#
辛斌, (1997), 英語語篇的批評分析芻議,《四川外國語學院學報》，(4): 43-49.
辛斌, (1998), 新聞語篇轉述引語的批評性分析,《外國教學與研究》，(2): 9-14.
辛斌, (2000a), 語篇互文性的語用分析,《外語研究》，(3): 14-16.
辛斌, (2000b), 批評語言學與英語新聞語篇的批評性分析,《外語教學》，(4): 44-48.
辛斌, (2000c),《語篇互文性的批評性分析》, 蘇州: 蘇州大學出版社.
辛斌, (2002), 批評性語篇分析方法論,《外國語》，(6): 34-41.
辛斌, (2003), 語言 語篇 權力,《外語學刊》，(4): 1-6.
辛斌, (2004), 批評語篇分析：問題與討論,《外國語》，(5): 64-69.
辛斌, (2006),《中國日報》和《紐約時報》中轉述方式和消息來源的比較分析,《外語與外語教學》，(3): 1-4.
徐立新, (1999), 語言反映意識形態(tài)---對四家報紙關于印度核試驗報道的語篇分析, 《外語教學》，(4): 12-16.
張榮建, (2005), 會話分析與批評會話分析,《四川外語學院學報》，(2): 47-52.

Appendix I

Group Headline time

People’s Daily(Group A) Dalai-backed Violence Scars Lhasa March 15(a),2008
Tibetan Authorities Demand Rioters Surrendering, Offer Leniency March 15(b), 2008
Tibetan Religious Leader, Locals Chide Lawless Riot in Lhasa March 16, 2008
Relevant Countries Urged to Respect Fact, Distinguish Right and Wrong on Lhasa Riot March 18,2008
Lhasa Riot Poses Severe Threat to Life, Stability March 19,2008

New York Times(Group B) Violence in Tibet as Monks Clash With the Police March 15,2008
Simmering Resentments Led to Tibetan Backlash March 18,2008
China Steps Up Its Accusations Against the Dalai Lama March 19,2008
As Tibet Erupted, China Security Forces Wavered March 24,2008
660 Held in Tibetan Uprising March 27,2008
Appendix II
Violence in Tibet as Monks Clash with the Police
Published: March 15, 2008 NYT
BEIJING — Violence erupted Friday morning in a busy market area of the Tibetan capital, Lhasa, as Buddhist monks and other ethnic Tibetans brawled with Chinese security forces in bloody clashes. Witnesses said angry Tibetan crowds burned shops, cars, military vehicles and at least one tourist bus. State media said at least 10 people died.
The chaotic scene was the latest, and most violent, confrontation in a series of protests that began on Monday and now represent a major challenge to the ruling Communist Party as it prepares to play host to the Olympic Games in August. By Saturday morning, Chinese armored vehicles were reportedly patrolling the center of the city.
Beijing is facing the most serious and prolonged demonstrations in Tibet since the late 1980s, when it suppressed a rebellion there with lethal force that left scores, and possibly hundreds, of ethnic Tibetans dead. The leadership is clearly alarmed that a wave of negative publicity could disrupt its elaborate plans for the Olympics and its hopes that the games will showcase its rising influence and prosperity rather than domestic turmoil.#p#分頁標題#e#
Thousands of Buddhists in neighboring India and Nepal took to the streets Friday in solidarity. Concerned that the protests might spread elsewhere in China, the authorities appeared to be moving the military police into other regions with large Tibetan populations.
Roughly 1,000 special police officers were deployed in the town of Bamei, in Sichuan Province, the site of a temple sacred to Tibetans, witnesses said telephone on Friday. Residents in Lhasa, reached by telephone, the authorities had placed much of the city under a curfew by Friday night while military police officers were blocking many city streets. One resident reported seeing armored vehicles in the center of the city.
The United States Embassy in Beijing warned American citizens on Friday not to travel to Lhasa. The embassy said it had “received firsthand reports from American citizens in the city who report gunfire and other indications of violence.”
In a meeting in Beijing on Friday, the United States ambassador to China, Clark Randt, urged Chinese officials to act with restraint, “and not resort to use of force in dealing with the protesters,” the State Department spokesman, Sean McCormack, told reporters.
The Chinese authorities blamed the Dalai Lama, the exiled spiritual leader of Tibet, for the violence and said the government would maintain stability in Lhasa. “The government of Tibet Autonomous Region said Friday there had been enough evidence to prove that the recent sabotage was ‘organized, premeditated and masterminded’ by the Dalai clique,” reported Xinhua, the Chinese government’s official news agency.
The Dalai Lama released a statement on Friday calling on both sides to avoid violence and appealing to China’s leaders to “address the long simmering resentment of the Tibetan people through dialogue with the Tibetan people.” A spokesman for the Dalai Lama called China’s accusations “absolutely baseless.”
The situation in Lhasa represents a complicated predicament for the Communist Party, which is now holding its annual meeting of the National People’s Congress in Beijing. Party leaders are already grappling with growing criticism of China’s domestic human rights record and its ties to Sudan, which the United States has accused of waging a genocidal campaign in its Darfur region.
In the past China has not hesitated to crush major protests in Tibet or to jail disobedient monks. President Hu Jintao, who is also the general secretary of the Communist Party, served as party boss in Tibet during a violent crackdown in 1989. His support for the bloody suppression of unrest that year earned him the good will of Deng Xiaoping, then the paramount leader, and led directly to his elevation to the Politburo Standing Committee and eventually to China’s top leadership posts.
But Chinese leaders may be more reluctant to order such heavy-handed tactics as Beijing prepares for the Olympics. On Friday, different accounts emerged about how the Chinese military police in Lhasa handled the demonstrations.#p#分頁標題#e#
Radio Free Asia, a nonprofit news agency financed by the United States government, quoted Tibetan witnesses who described police officers firing into crowds of protesters and killing at least two people in the city’s ancient Barkhor area. On Saturday morning, Radio Free Asia quoted witnesses who described seeing dead bodies around Lhasa.
Later on Saturday Xinhua reported (10-ID-M)10 deaths had been confirmed
A Chinese resident in Lhasa, reached by telephone, said (11-ID-M)stories were circulating among local Chinese that soldiers had been wounded and had not been allowed to fight back against Tibetans throwing rocks. Another Chinese man living near the Barkhor area said (12-ID-M)family members had told him that two soldiers died and that Tibetans were beating Chinese residents with iron rods.
Friday’s sharp escalation in violence, and the sense of dread described by several residents, came a day after China’s Foreign Ministry told reporters that the situation in Lhasa had stabilized. The protest started Monday when Buddhist monks began peaceful protests against religious restrictions by Chinese authorities. The police arrested 50 or 60 monks, but other protests followed Tuesday and Wednesday as monks in two different monasteries took to the streets.
The apparent epicenter of Friday’s protests was the Tromsikhang Market, a large, concrete structure built in the Barkhor area by the Chinese authorities in the early 1990s. “It’s chaos in the streets,” said a person who answered the telephone at a bread shop near the market.
What actually set off the violence is unclear, as accounts differed between Chinese and Tibetan residents. Monks from the Ramoche Temple, a short walk from the market, reportedly began to march in the Barkhor area. The Ramoche monks intended to protest the rough treatment of monks who had marched earlier in the week, according to a Tibetan rights advocate in the United States who has communicated with people in Lhasa.
When police officers began beating the monks, Tibetans rioted in the Barkhor area, the advocate said. Angry mobs set fire to a police car and a store owned by a Chinese shopkeeper, said the advocate, who refused to be publicly identified for fear of reprisals.
But a Chinese travel agent in Lhasa, reached by telephone, said Tibetans had instigated the violence and set fire to an empty tour bus parked outside the Ramoche Temple. Another Chinese resident described 50 or 60 young Tibetans burning stores owned by Chinese merchants as well as two fire trucks and two police cars.
“I saw someone who was dead and covered in a sheet,” (15-DD-M)the Chinese resident said in a telephone interview. “The Tromsikhang market was destroyed, except for the shops owned by Tibetans. I heard a soldier shouting, ‘Please go home and stop fighting!’ ”(16-DD-M)
News agencies also reported clashes between monks from Ramoche Temple and military police officers. “The monks are still protesting,” one witness told The Associated Press. “Police and army cars were burned. There are people crying. Hundreds of people, including monks and civilians, are in the protests.”#p#分頁標題#e#
Radio Free Asia reported that Tibetan protesters were waving traditional white scarves and shouting, “Free Tibet.” The agency said the riots began about 10 a.m. and had largely quieted down by 3:30 p.m., after the paramilitary police were mobilized.
Meanwhile, anxious tourists stranded in Lhasa posted worried comments on online forums for travelers. “The situation seems to be very nervous and paranoid up here,” wrote one person in broken and misspelled English in a chat room sponsored by the Lonely Planet tour guide. “There is police and military everwhere. Suddenly you would see policeman running and rushing somewhere”.
The ethnic friction evident in Friday’s violence has long simmered just below the surface in Lhasa. For more than two decades, a steady influx of Chinese migrants has transformed and stratified the city. A newer, Chinese section of Lhasa is focused along Beijing Road and is lined with shops and concrete buildings.
But the older Tibetan neighborhoods emanate from the Jokhang, the most sacred temple in Tibet, and the Potala Palace, the former residence of the Dalai Lama. Tibetans also have complained that Chinese merchants now control most of the tourist shops in the Barkhor area.
The protests in Lhasa coincided with the anniversary of a failed 1959 Tibetan uprising against Chinese rule that forced the Dalai Lama to flee to India. Groups that promote Tibetan independence have marked the anniversary with demonstrations around the world, including in India, where a group of advocates tried to march to Tibet.
But the unexpected demonstrations in Lhasa are the largest Tibetan protests against Chinese rule since 1989. Military police officers and soldiers are now reportedly surrounding the three monasteries that were at the center of the protests earlier this week. Two monks have reportedly tried to kill themselves, while pro-Tibetan groups say others have started a hunger strike.
In its travel warning, the United States Embassy in Beijing advised American tourists in Lhasa to “seek safe havens in hotels and other buildings and remain indoors.”
“All care should be taken to avoid unnecessary movement within the city until the situation is under control,” it said.
On Friday night, the Chinese resident living in the Barkhor area said his family was huddled at home with the lights turned off. He said he could hear shouting on the streets outside and feared for the safety of his family.
Appendix III
Dalai-Backed Violence Scars Lhasa
March 15,2008 People’s Daily
The outbreak of violence died down in Lhasa Friday night, after a tumultuous day that saw windows smashed, shops robbed, mosque burnt down and reportedly many casualties.

Witnesses said the unrest started around 1:10 p.m. on Friday, several people clashed with and stoned the local police around the Ramogia Monastery in downtown Lhasa.

Rioters began gathering around 2 p.m. around the Ramogia Monastery, and set fire to shops along two main streets in the capital, and around Jokhang Temple, Ramogia Monastery and Chomsigkang Market. At least five blazing spots were reported and dense smoke was seen blanketing the area.#p#分頁標題#e#

A number of shops, banks and hotels were burnt, causing blackouts and interruption of communications in some areas. Shops close to the Jokhang Temple and Ramogia Monastery were shut down.

A Tibetan government official told Xinhua that there had been enough evidence to prove that the sabotage in Lhasa was "organized, premeditated and masterminded" by the Dalai clique.

The violence, involving beating, smashing, looting and burning, has disrupted the public order and jeopardized people's lives and property, the official said.

Xinhua reporters in Lhasa saw many rioters were carrying backpacks filled with stones and bottles of inflammable liquids, some holding iron bars, wooden sticks and long knifes, a sign that the crowd came fully prepared and meant harm.

The mobs assaulted passersby, sparing no women or children, witnesses said
They hit at things along their path, smashing windows, automatic teller machines and traffic lights. Several clothing shops, restaurants, and mobile phone stores were looted. Bikes, motorcycles and cars were burnt down.

The vandals started burning the local Sifang Supermarket, Landun Shopping Mall and Wenzhou Mall around 3:00 p.m. Friday, causing more blazing spots. A Muslim mosque was also set on fire at around 8:30 p.m..

There were injuries reported in the violence and the wounded were sent to the hospital. People were also seen burnt by the attackers. But death toll is not yet available.

GOVERNMENT ACTIONS

留學生新聞媒體學專業(yè)碩士dissertation定制Sources told Xinhua that policemen were ordered not to use force against the attacker. But they were forced to use a limited amount of tear gas and fired warning shots to disperse the desperate crowds.

Xinhua reporters learnt that many policemen on duty were badly injured.

Police have not made any announcement of arrest, but an officer said the search for the vandals could be difficult as the mobs disguised themselves in plain clothes as ordinary citizens.

Around midnight, fire-fighters and policemen were cleaning the burning wreckages discarded on the Beijing Middle Road, one of the main streets in downtown Lhasa.

Police cordoned off a few downtown sections and are on close lookout for comeback of violence.

The regional government took emergency measures to rescue residents under attack, reinforced protection for schools, hospitals and gas stations, and required the government agencies and businesses to ensure safety of their employees.

Local government imposed traffic control on the main streets in Lhasa Friday night and it also informed the citizens of the sabotage through TV, calling for them to take precautions.

BLOGGER ACCOUNT

A blogger who called himself Han Jingshan, a Lhasa resident, recalled the sabotage in a post titled Four-hour Personal Experience of Lhasa's Riots.

The man drove a car onto the streets in the afternoon only to find flames with heavy smoke blanketing the area of the Ramogia Monastery and ambulances whistling by, according to the post.#p#分頁標題#e#

"Arriving at the road entrance to the Ramogia Monastery, I saw the ground was covered with rocks weighing one or two kilograms and a cab was burnt down," he wrote.
"I saw a dozen of mobs, mainly young people in their twenties, were burning cars in front of the Baiyi Supermarket with more than 200 people standing by and watching," he said.
"At 17:56, a police car arrived and the mobs ran away. The police didn't chase them as another two cabs were on fire at the New Jiangsu Avenue some 300 meters ahead."

"At this time, a Han Chinese women, whose face was bleeding, ran by me and later a cab with broken windows passed by," he wrote.

"The history of Tibet and even of China will remember the day of March 14 in 2008 forever," the writer said.
Appendix IV Major Abbreviation Used in the Thesis
CDA-Critical discourse analysis
CL-Critical linguistics
SFG-systemic-functional grammar
PD-people’s daily
NYT-New York Times