Annual Review of Applied Linguistics (1995) 15, 23-41. Printed in the USA.

Copyright © 1995 Cambridge University Press 0267-1905/95 $9.00 + .10

LANGUAGE AND THE MEDIA*

Allan Bell

INTRODUCTION

Media language has always attracted the attention of linguists, particularly

applied linguists and sociolinguists. There are four practical and principled
reasons for this interest. First, the media provide an easily accessible source of
language data for research and teaching purposes. Second, the media are impor-
tant linguistic institutions. Their output makes up a large proportion of the
language that people hear and read every day. Media usage reflects and shapes
both language use and attitudes in a speech community. For second language
learners, the media may function as the primary—or even the sole—source of
native-speaker models. Third, the ways in which the media use language are
interesting linguistically in their own right; these include how different dialects
and languages are used in advertising, how tabloid newspapers use language in a
projection of their assumed readers' speech, or how radio personalities use
language—and only language—to construct their own images and their relation-
ships to an unseen, unknown audience. Fourth, the media are important social
institutions. They are crucial presenters of culture, politics, and social life,
shaping as well as reflecting how these are formed and expressed. Media
'discourse' is important both for what it reveals about a society and for what it
contributes to the character of society.

Linguistic research on the media has always emphasized this last concern,
focusing on points at which issues of ideology and power are closest to the
surface. It has analyzed the macro level of discourse structures rather than
phonological or syntactic structures, concentrating mainly on the 'factual' genres
(particularly news) rather than fictional or advertising content (an emphasis shared
by sociological and communications research on media). This review will reflect
that weighting of research interest, concentrating on analyses of 'factual' media
discourse, but it will also touch briefly on other areas of interest to applied

23
24 ALLAN BELL

linguists and sociolinguists. As in most subfields of linguistics, work in and on

English predominates and is (unfortunately) reproduced in the present review
article.

RESEARCH ON MEDIA DISCOURSE

In the past few years, the study of media language and discourse has
gained a coherence and focus it previously lacked (a deficiency shown in the
rather fragmented scene described in a previous ARAL overview by Geis [1987]).
Pioneering analyses of media discourse conducted in the 'Critical Linguistics'
framework (e.g., Fowler, et al. 1979, Kress and Hodge 1979) had been stimulat-
ing but less dian satisfactory. Advances were made, however, when van Dijk
began to apply his theory and methodology of discourse analysis to media texts
(1985; 1988a; 1988b). Subsequently, the first general textbooks in English on
media language appeared, by Fowler and Bell, in 1991 (cf. Burger 1984 for
German). Published almost simultaneously, these two books are core texts for
the increasing number of courses in media language or discourse being taught in
European universities. A British Open University reader edited by Graddol and
Boyd-Barrett followed in 1994, bringing together new and existing papers in the
field. To these there is to be added a forthcoming book by Fairclough, whose
important contributions on media discourse have until now been scattered through
a number of publications.

The predominant research on media discourse in recent years falls under

the umbrella of 'Critical Discourse Analysis' (CDA).1 It represents an outgrowth
of the work of the British and Australian pioneers of Critical Linguistics, particu-
larly Fowler and Kress, in convergence with the approaches of the British
discourse analyst Fairclough and the Dutch text linguist van Dijk. All of these
researchers continue to carry out significant work on the language of media.
CDA is increasingly the standard approach to media texts within European
linguistics, although the paradigm is less familiar in the United States. The major
research reviewed below either adopts explicitly some form of CDA or is conge-
nial to it where its concerns overlap with other perspectives (e.g., Bell 1991b;
1994). Boyd-Barrett (1994) provides a clear summary and even-handed assess-
ment of three of these approaches (those of van Dijk, Fowler, and Bell).

CDA is more a shared perspective than a school or a methodology. It

has an explicit sociopolitical agenda, a concern to discover and bear witness to
unequal relations of power which underlie ways of talking in a society, and it
aims, in particular, to reveal the role of discourse in reproducing or challenging
sociopolitical dominance. The new journal, Discourse and Society (see Annotated
Bibliography), is a specific forum for CDA and related work. The media are a
particularly important subject of CDA analysis because of their manifestly pivotal
role as discourse-bearing institutions.
 LANGUAGE AND THE MEDIA 25

Parallel to the growth of CDA, there has been a convergence between the
methods and interests of linguistic discourse analysis and those of European
critical sociopolitical theory, literary criticism, and cultural studies. Both strands
share an interest in media texts as manifestation of, and contributor to, sociopolit-
ical structures and trends. The convergence is reflected in thematic issues of
communications journals, such as those edited by Mancini (1988; European
Journal of Communication) and Scannell (1990; Media, Culture and Society).

1. Van Diik

The most comprehensive work on media discourse to date has been

published by Teun van Dijk (1988a; 1988b; 1991). Van Dijk's background is in
European text linguistics, and he is a leading theorist of discourse analysis. His
framework aims to integrate the production and interpretation of discourse as well
as its textual analysis. He puts forward his approach to news discourse as a
"new, interdisciplinary theory of news in the press" (1988b:vii). Following a
number of earlier presentations, News as Discourse (1988b) appeared as a
primary theoretical contribution to the analysis of news stories. It is supple-
mented by a volume of case studies, News Analysis (1988a). These are drawn
mainly from a massive sample of international news reporting and a study of
racism in the European press. His 1991 book, Racism in the Press, analyzes the
reporting of ethnicity in a large sample of stories from British newspapers.

Van Dijk's main contribution is a framework for analyzing the discourse

structure of news stories. The thematic structure of a story consists of its topics
and their organization within the story, providing a broad semantic structure.
Closely parallel is the syntactic structure, which van Dijk terms 'news schemata.'
Schemata consist of a set of characteristic categories, organized by rules. The
structure is derived from the discourse by 'macrorules.' These come in three
main kinds—deletions, generalizations, and 'constructions' (which summarize a
series of actions)—and reduce the information in a discourse to its gist.

News schemata are thus a syntax of news stories, the formal categories
into which news can be analyzed, and their relations to each other. These can be
tree-diagrammed to show the discourse structure of the story. The categories
include summary (headline and lead), main events, background, and conse-
quences. Some categories, such as background or lead, are used by journalists to
organize their product; others, such as headline, are also known to news audi-
ences. Van Dijk's approach to comprehension stresses the psychological reality
of some schemata (or 'scripts') which news consumers bring to understanding
news.

Van Dijk's analysis of newspaper stories in many languages (1988a)

found few significant cross-linguistic differences in news discourse structure. In a
comparison of different treatments of the assassination of Lebanese President-elect
26 ALLAN BELL

Gemayel in 1975, he found stories in Spanish, Chinese, and Swedish all followed
a pattern similar to English-language news. There were some differences between
papers in the 'First' and 'Third' Worlds, but the greatest differences were
between 'quality' and 'popular' papers within certain countries, for instance,
within West Germany and the United Kingdom.

Van Dijk's (1991) book on British press coverage of ethnic relations

focuses on quantitative analyses of recurring features and structures rather than
qualitative application of van Dijk's discourse analysis framework. Analysis of
the structure and function of headlines shows that majority-group actors have a
more prominent position than do minority figures. Minority actors are associated
more with negative predicates, and they tend to get first position in a headline
when they are agents of negative actions. Further analyses show that minorities
are quoted as sources much less often than are majority members, and minority
groups are labelled with negative terms of reference.

2. Bell

The Language of News Media (Bell 1991b) covers broadcast media as

well as the press, with a concentration on news. In addition to reporting the
author's specific research findings and methods, the book aims to introduce a
range of data and approaches on media language as well as relevant areas of mass
communication research, such as how news is produced and framed. The book's
examples include materials gained by 'observant participation'—drawing on
stories written or edited by me and other journalist colleagues. Three themes run
throughout the volume: the importance of the processes which produce media
language, the notion of the news story, and the role of the media audience.2

News discourse is the product of multiple hands, and the processes by

which it is molded and modified are both crucial and enlightening for an under-
standing of the eventual news text, its form, and its content. As well as taking
material for their stories from spoken sources such as interviews and speeches,
journalists draw on a range of written texts and embed whole chunks of prefabri-
cated text into their own stories—from earlier stories, news agency copy, press
releases, reports, and the like (cf. Fairclough's concept of intertextuality below).
Once the story leaves the journalist's hands, it then passes through multiple
editing stages before being eventually published, with proven scope for inaccu-
racy expected to creep in as it moves along the editorial chain. The editing
process by which copy editors cut, clarify, and rewrite stories is largely driven by
the desire to maximize 'news value.'

The idea of the story is central to news. Journalists do not write articles;
they write stories—with structure, order, viewpoint, and values. Bell (1991b)
examines how news stories differ from other kinds of narratives, beginning with a
summary 'lead' sentence and eschewing chronological order. News values drive
 LANGUAGE AND THE MEDIA 27

the way news stories are structured, and this may lead to inaccurate reporting of
source information (see also Bell 1994). The lead sentence embodies the charac-
ter of the story. News discourse is analyzed in terms developed from analysis of
personal narratives and van Dijk's framework: summary (headline and lead),
attribution, events, actors, setting (time, place), followup, commentary, and
background. The analysis examines a number of discourse features:

1. How headlines and lead paragraphs are structured,

2. Which people act and speak in the news and how they are described,
3. How time and place are expressed,
4. How the words of news makers are reported,
5. How much of news is talk about talk.

The importance of the audience is investigated in relation to both linguis-

tic styles and discourse comprehension. Audiences bring the power of their own
choices, understandings, and preconceptions to media reception. Audiences for
news media also fulfill multiple roles in relation to the media and their content—
they may be addressees, auditors, or overhearers. Structured variation of a
number of sociolinguistic variables shows how media accommodate their news
styles to different kinds of audiences. By contrast, advertisers take the linguistic
initiative by adopting accents foreign to their audience but with attractive
associations (see also Bell 1992). To their understanding of media content,
audiences bring their own pre-existing mental scripts through which they interpret
(and 'misinterpret') the news that they receive.

3. Fowler

Roger Fowler's Language in the News: Discourse and Ideology in the

Press (1991) expounds the framework of Critical Linguistics and applies it to
specific news texts. Critical Linguistics was developed by Fowler and his
colleagues in the 1970s, building on the functional grammar originated by
Halliday. Original versions of Critical Linguistics were open to the charge of
both linguistic and social inadequacy; more recently, critical linguists have
themselves critiqued some of their own earlier stances and from this has evolved
the Critical Discourse Analysis approach outlined above.

Fowler's approach is openly political. A strong impetus for his book lay
in the political changes involved in 'Thatcherism' in the United Kingdom in the
1980s and the part which media discourse played in those shifts. He regards the
forms of language as not neutral, so that "any aspect of linguistic structure...can
carry ideological significance" (67). The tools with which he approaches media
discourse are drawn from functional linguistics and involve the analysis of the
transitivity of sentences (including the role and nature of participants), the use of
passives and nominalizations (which can mystify relations by omitting agents, or
28 ALLAN BELL

reify events or processes by naming them), and the impact of modality (which
indicates the stance of the speaker towards what is said).

Fowler's analyses concentrate on vocabulary rather than linguistic or

discourse structure. He is concerned with how groups and individuals are
labelled in the media—women, young people, strikers, hospital patients. He
identifies 'overlexicalization,' in which society—including the media—applies an
excess of labels to a particular kind of person. The surfeit of terms is indicative
of stereotyping, distancing, and downgrading the persons and the group with
which the terms are associated. Fowler's case studies include the 1986 United
States bombing of Libya, which was launched from bases in the UK, and cover-
age that labelled hospital patients as 'cases' and made subsequent reference with
the impersonal it rather than a personal pronoun.

The 1988-89 affair of salmonella contamination of food in the UK is

Fowler's main hands-on analysis, occupying over a quarter of the book. He
argues that the salmonella issue was an instance of media-generated hysteria,
involving massive coverage, generalization to a host of environmental issues, and
highly emotive vocabulary. The issue was, in large measure, caused by the
nature and amount of media coverage; that is, the supposed crisis was itself
created and constituted by the discourse, independent of the facts of the situation.
Fowler analyzes the sets of vocabulary used in 300 stories clipped over the three
months of the affair, looking at the metaphors used (particular war imagery) and
at the rhetoric of quantification (cf. Potter, et al. 1991). He examines how the
affair was also labelled with a range of lexical alternates such as crisis, contro-
versy, and outbreak. Such formulaic patterns played a basic role in making up
the crisis. His conclusion from the increasing personalization of media coverage
as the affair developed is that responsibility in the crisis was being transferred
away from the government or regulatory institutions and on to individuals
(particularly 'housewives').

4. Fairclough

Like the critical linguists, Norman Fairclough's linguistic approach draws

on Halliday's functional framework, but he also brings a knowledge of recent
social theory which is rarely encountered in linguistics (or even socio- or applied
linguistics), drawing particularly on the French social theorist Foucault. His
framework has been developed independent of a concentration on media language,
although it is no accident that many of his example analyses are of media texts
since the media—whose business is discourse—play a crucial role in shaping and
purveying discourses. His two principal books, Language and Power (1989) and
Discourse and Social Change (1992), and a book on Language in the Media
(forthcoming), develop his framework. He ranges more widely across the media
than the previous three authors, covering a variety of 'non-fictional' broadcast
genres as well as news.
 LANGUAGE AND THE MEDIA 29

Fairclough proposes three dimensions to discourse analysis:

1. Text analysis concentrates on the actual textual language, examining

vocabulary, syntax and cohesion, and the macro level of text structure. It
also highlights 'interpersonal' elements in the text—both personal identity
and relations between persons.
2. Analysis of discourse examines how the text is produced and interpreted,
and also how it is 'distributed,' with the media as a case of complex
distribution of texts from other sources. This level of analysis also looks
at wider issues of 'orders of discourse'—the characteristic discourse
practices of institutions such as the media.
3. Analysis of social practice focuses on discourse in society, institutions,
and culture, and especially on the relation of discourse to power and
ideology.

Fairclough emphasizes several trends towards change in the discourses of

contemporary society, particularly in the media. 'Commodification' or 'market-
ization' encroached on many areas of life in Western democracies in the 1980s so
that many aspects of life and institutions are increasingly thought of, and follow
from, commercial models. Fairclough examines the discourse of Thatcherism in
the United Kingdom, making the argument that the ideological conflicts which
surrounded it were in large part fought over the discourse itself. Another aspect
of marketization is the increasing use of promotional language transferred from
advertising into other domains. Thus, one genre of media discourse colonizes
other areas of discourse in society (Fairclough 1993).

A second trend is towards the 'democratization' or 'conversationalization'

of discourse in the media as in other institutions—a shift to increased informality
of language. Fairclough analyzes certain political broadcasts in these terms.
Panel interviews are starting to be conducted in ways reflecting television chat
shows, which are of course themselves modelled on (a projection of) everyday
conversation. Fairclough notes the ambivalence of these processes—the positive
effect of more open access to previously closed domains, but also the potential for
pseudo-conversation to mask even more effectively the exercise of power.

These changes are just two examples of what Fairclough (following

Bakhtin) calls 'intertextuality'—the mixing of different language genres or styles.
In addition to the manifest case of the cut-and-paste origins of news text, this
represents more abstractly the combination of previously different ways of
talking. Fairclough analyzes a story from the (British) Sun newspaper which
mixes its own populist, conversation-like style with the bureaucratic language of
the government document on which it is reporting. This is part of the translation
of public life into hitherto private terms, in which the media play a central role.
30 ALLAN BELL

TALK ON THE AIR

Where the press is limited to printed-language genres, broadcasting offers

the opportunity to examine a range of spoken genres, including interviews,
telephone conversations, and assorted kinds of monologues. Research on such
genres in broadcast media has mainly used the methods of Conversational
Analysis (CA) developed from the work of the sociologist Harvey Sacks, particu-
larly by Schegloff and Jefferson. CA's interest is in describing the particulars of
how conversations are structured, especially their openings, closings, and turn-
takings. Since the mid-1980s, a number of researchers have used CA to analyze
broadcast conversations such as interviews, chat shows, and phone-ins,
particularly comparing the broadcast genre with typical face-to-face conversation
(for work in German, see Burger 1991). Researchers have also drawn on
Goffman's approaches to the 'interaction order,' the presentation of self, and the
construct of 'footing' in talk. (Much of Goffman's later work was focused on
language, including 'radio talk' [1981]).

1. Interviews

Interviews that are broadcast on radio and television have been the focus
of much of this research, largely owing to the sociopolitical salience of the
content and participants in political interviews. As Clayman states: "It is now
apparent that the interview...is in fact a strongly institutionalized genre of
discourse that exerts a pervasive influence on the conduct of journalists and public
figures, and on the manner in which they form their talk with one another" (1991:
48).

The interview is the main technique by which all journalists get informa-
tion for stories they write. What broadcast technology does is turn the one-to-one
private speech event of the interview into a genre which is openly showcased for
public consumption. It thus develops its own structure and norms and becomes
institutionalized (Heritage and Greatbatch 1991). Such interviews have been
researched since the early 1980s (e.g., Heritage 1985, Jucker 1986). Clayman,
working in the United States, and Greatbatch, in the United Kingdom, have both
studied how questions and answers are structured within news interviews, finding
systematic differences from ordinary conversational practice (e.g., Greatbatch
1988). In particular, interviews work under the institutional constraint that
interviewers ask questions and interviewees answer them. When this is departed
from—for instance, when an interviewee starts asking questions instead of
answering them—the interviewer will sanction the diversion and repair the flow of
the interview to get it back on track.

Politicians are publicly perceived as avoiding straight answers to straight

questions. Clayman (1993) and Harris (1991) have examined the ways in which
politicians evade direct questions. Harris found that nearly half of politicians'
 LANGUAGE AND THE MEDIA 31

answers to direct yes/no questions were indirect. Some interviews also involve
multiple participants, usually brought together specifically because they represent
opposing viewpoints on an issue. Greatbatch (1992) examines how disagreements
between participants in an interview panel are managed. In everyday conversa-
tion, disagreement is routinely discouraged, but in these interviews it is expected.
However, the disagreement is mitigated by being largely mediated through the
interviewer rather than addressed directly to the opposing party. Interviewers
also perform the task of providing subjects with exits from escalating disagree-
ments.

dayman (1992) has also investigated how television news interviewers

alter their footing (in Goffman's terms) to achieve the appearance of neutrality in
their questioning of an interview subject. Typically, an interviewer will attribute
opposing views to absent third parties rather than presenting them to the subject
as the interviewer's own position on a disputed matter.

2. Chat and self-presentation

Other forms of broadcast conversations include what Tolson has termed

'chat,' which is "a form of studio talk which can be found in all types of
interviews, panel discussions, game shows, and human interest programmes"
(1991:179). Chat is an institutionalized proxy for ordinary conversation, often
intended to display the verbal skills of the participants. Tolson analyzes a
standard television chat show (Wogan on the BBC), and then turns to analyze the
version presented by Dame Edna Everidge (comedian Barry Humphries), which is
simultaneously exemplar and parody of the genre. Phone-ins are another form of
on-air conversation, analyzed for example in Cameron and Hills (1990), Scannell
(1991), and Liddicoat, et al. (1992).

A majority of radio talk is monologue, not dialogue, and presenters rely

largely on their use of language to project their on-air identities and their relation-
ship with listeners. Brand and Scannell (1991) analyze the skill and work which
one DJ, Tony Blackburn, brings to forming his public persona on BBC Radio 1.
Aside from his baseline DJ mode, Blackburn routinely employs several changes
of voice—authoritative, empathetic (initiating address in phone calls), and several
camp parody voices. On air, he 'talks himself into being,' creating a self-aware
identity of what Brand and Scannell call 'the playful self.'

OTHER THEMES

1. Micro-linguistic studies

Language structure below the level of the sentence—syntax and

phonology—has received considerably less attention than the discourse level of
media language. The four discourse analysis frameworks outlined above do pay
32 ALLAN BELL

some attention to syntactic analysis, particularly those structures emphasized in a

functional grammar: transitivity, modality, nominalization, passivization, agent
deletion. But the focus of these analyses is their sociopolitical significance rather
than a concentration on the nature of the language structure itself.

Some research has focused more closely on how linguistic features mark
social identity or respond to the differentiated characteristics of media audiences.
Jucker (1992) found that the absence of determiners in referring expressions in
the British daily press is highly sensitive to the social status of newspapers'
readerships. There is also strong evidence that New Zealand radio newsreaders
accommodate to the social status of their audiences on several phonological
variables (Bell 1991a). In addition, the use of a distinctive phonological variant
by Hebrew pop singers signals a complex ambivalence in their affiliation to ethnic
groups in Israel (Yaeger-Dror 1993).

Change in media language and cross-language differences represent two

other areas receiving interest in a number of studies. The introduction to
Scannell (1991) notes the transition in the BBC's language use from written to
spoken models and from authoritarian to populist use (cf. Fairclough's notion of
conversationalization). The mixing of written and spoken linguistic forms in
Norwegian radio language is the focus of a pilot study by Vagle (1991); this
study is also a precursor to comprehensive historical research on the development
of broadcast Norwegian. Interesting cross-cultural patterns in media language
emerge in a comparison of illocutionary acts between American English news-
paper editorials and Egyptian editorials in English and Arabic (Reynolds 1993).
Non-native use of the pluperfect tense in Nepali English is shown to be derived
from tense usage in Nepali signalling the remoteness of subsidiary events from
the main events of a news story (Hartford 1993).

2. Advertising language

Next to the 'factual' genres, advertising has been the other main genre
researched by linguists. It is of interest for both linguistic and sociopolitical
reasons because of the creative ways in which advertisers use language and the
part that language plays in the persuasive intent of advertisements. Studies have
continued since Leech's pioneering work 30 years ago. Recent works (e.g.,
Cook 1992, Myers 1994, Tanaka 1994) rightly give a place to the literary-like
creativity of advertising language in its use of metaphors, puns, alliteration, and
many quasi-poetical devices. Local or external dialects are used in some adver-
tisements for particular associations in countries such as Switzerland (Lee 1991)
and New Zealand (Bell 1992). More extremely, English is widely used in
advertisements in non-English speaking areas such as Japan, Hindi-speaking
India, and Continental Europe (Bhatia 1992).
 LANGUAGE AND THE MEDIA 33

3. Language maintenance and shift

Finally, there is the issue of the media's role in language maintenance

and shift. Many languages world-wide are endangered, and the media often
appear as instruments by which powerful languages (particularly English) en-
croach on vulnerable ones. But conversely, the media are also seen as potential
means of language maintenance or revival, for instance with the advent of the
Welsh language television channel S4C and the campaigns for Maori-language
broadcasting in New Zealand. The role of media in language maintenance is not
easy to research because of the difficulty of isolating media from other effects,
but the Celtic languages in particular have received some attention (e.g., Browne
1992, Cormack 1993).

CONCLUSION

Research on media language and discourse tends to be inherently applied.

It must always work on actual texts, and most practitioners bring to it social or
political concerns and a desire to see others make use of their frameworks. All
four of the approaches to media discourse outlined above are intended for
application in understanding the nature and sociopolitical role of media discourse.
Fairclough and Fowler are both explicit in urging others to apply their
approaches, recognizing as well that such an undertaking requires appropriate
training.

The approaches outlined are intended to be applied, but there is a caveat:

grasping and using these techniques demands time and commitment. The frame-
works themselves can be difficult to understand (although all have been taught to
students). The terminologies used and the explanations of categories and methods
are not always clearly presented. In addition, the analyses are time-consuming if
undertaken at all comprehensively. The originators are aware of this (e.g., van
Dijk 1991:10), and they themselves tend to do full-scale analyses on only a few
texts, alongside more piecemeal, specific analyses on larger samples.

With its emphasis on vocabulary, Fowler's approach is probably the most

accessible. The amount and complexity of apparatus proposed by Bell, van Dijk,
and Fairclough can be daunting to students or lay people (and sometimes to
specialists). The writings of van Dijk and Fairclough are also often demanding to
understand, but close engagement with their work is rewarding for the percep-
tions they yield. The complexity of the frameworks proposed is in fact a true
reflection of the real complexity of texts such as news stories, which sometimes
appear deceptively simple. Media research also offers many instances of the
danger of basing conclusions on inadequate language analysis.

Writing from the perspective of a media scholar, Boyd-Barrett observes

that linguists have not often applied their analytical skills to media texts, "but
34 ALLAN BELL

when they do, the dividends are rewarding" (1994:38). The language of news
media is prominent and pervasive in society, and it is important to understand
how that language works, how it affects our perceptions of others and ourselves,
how it is produced, and how it is shaped by social structures and forces. What is
media discourse like? What can it tell us about media? What can it tell us about
language? These questions, addressed in the research reviewed here, are signifi-
cant for applied linguistics.

NOTES

* I would like to thank the Centre for Language and Communication Research in
the School of English, Communication and Philosophy, University of Wales
College of Cardiff, for their support and hospitality to me during my residency
there as a Visiting Research Fellow in 1994.

1. For a more detailed exposition of Critical Discourse Analysis, see Kress's

overview in ARAL 1990 (Kress 1991); the entire issue of Discourse and Society
(4.2, 1993), especially van Dijk's general concluding paper (van Dijk 1993); and
the second edition of Hodge and Kress's (1993) foundational text.

2. To reduce the difficulty of trying to provide an even-handed assessment of my

own work alongside that of others, I quote directly the (happily positive) evalua-
tion of Bell (1991b) in Boyd-Barrett's survey of approaches to media discourse
(1994:34): "Bell is an unusual combination of linguist and journalist.... His
thinking has a grass-roots feel to it, reflected in a less abstract, technical style.
He is excellent on the integration of issues of text structure, production and
audience."

ANNOTATED BIBLIOGRAPHY

Bell, A. 1991b. The language of news media. Oxford: Basil Blackwell.

One of two general texts on media language now available in English (the
other is Fowler 1991), this book emphasizes three themes: the importance
of the processes which produce media language, the notion of the news
story, and the role of the media audience. The book examines media
language as the product of multiple hands, and it looks at the unique
structure of the news story, particularly in its avoidance of chronological
order. The role of the audience is investigated in relation to both
 LANGUAGE AND THE MEDIA 35

linguistic styles (cf. Bell 1991a) and discourse comprehension (cf. Bell
1994).

Biere, B. U. and H. Henne (eds.) 1993. Sprache in den Medien nach 1945.
[Language in the media since 1945.] Tubingen: Niemeyer.

There is a significant German literature on media language which is

scarcely known in English, including the work of Burger (e.g., 1991),
whose general text on media language (1984) predated comparable
publications in English. Biere and Henne's recent collection focuses on
the development of media genres over the past 50 years. There is an
introductory overview (by Schwitalla), followed by papers on radio
language (Fluck, Reinke), television (Holly and Piischel), readers' letters
(Fix), and media pedagogy (Hebel). Biere examines the processes by
which the texts of press stories are constituted from source documents,
and Burkhardt investigates the influence of the media on parliamentary
discourse.

Discourse and Society Journal [Quarterly publication since 1990.]

This is the main outlet for media discourse analysis. Founded in 1990 by
van Dijk, Discourse and Society has an explicitly sociopolitical and
'critical' agenda. The editorial statement carried in each issue indicates
its character: "Discourse and Society...favours contributions that pay
attention to the detailed analysis of social and political relations of power,
dominance and inequality, and to the role of discourse in their legitima-
tion and reproduction in society, for instance in the domains of gender,
race, ethnicity, class or world region." Some 40 percent of the papers
published deal in whole or in part with media discourse. Many of these
are referenced in the general bibliography below (see in particular Bell
1994, Meyers 1994, Nir and Roeh 1992, O'Donnell 1994, Tulloch and
Chapman 1992, Wodak 1991). Other leading journals in which papers
on media language and discourse can commonly be found are Discourse
Processes; European Journal of Communication; Media, Culture and
Society; Text; and World Englishes.

Fairclough, N. 1989. Language and power. London: Longman.

1992. Discourse and social change. Cambridge, UK: Polity Press.
Forthcoming. Language in the media. London: Edward Arnold.

Fairclough's approach to language, and his frequent use of media texts in

the first two references, is outlined in the main article above. His work
combines sophisticated linguistic and social analyses. (Fairclough [1993]
includes a brief exposition of his approach and glossary of its termino-
logy.) His forthcoming monograph on media language covers general
36 ALLAN BELL

material about the study of media discourse, the nature of mass communi-
cation, and his own approach to critical discourse analysis of the media.
It expounds in detail some of his characteristic emphases on media
language: its 'intertextual' nature, especially the 'conversationalization' of
much media discourse; the identities and relationships which mass
communication sets up for participants (including the audience); and the
ways these and other features are represented in the linguistic structure.

Fowler, R. 1991. Language in the news: Discourse and ideology in the press.
London: Routledge.

Fowler expounds the framework of Critical Linguistics and applies it to

specific media texts. His analysis of language structures uses the tools of
functional linguistics, including analyses of the transitivity of sentences,
the use of passives and nominalizations, and the impact of modality.
Much of his focus is on vocabulary, including a concern with how groups
and individuals are labelled in the media. His main case study is the
1988-1989 affair of salmonella contamination of food in the UK, which
he argues was an instance of media-generated hysteria.

Graddol, D. and O. Boyd-Barrett (eds.) 1994. Media texts: Authors and readers.
Clevedon, UK: Multilingual Matters and Open University.

This British Open University reader is an eclectic mix of new papers and
excerpts from recent and classic existing work, with varying pertinence to
the discourse of media. The new work includes a perceptive and lucid
summary by Boyd-Barrett (a media sociologist) of three current
approaches to media discourse (van Dijk, Fowler, and Bell). Graddol
contributes an insightful chapter on the relationship of the visual channel
to the verbal script in television news, an issue also dealt with by Ulrike
Meinhof. The remainder of the 18 chapters include excerpts from
Halliday, Hasan, Fowler, Bell, Barthes, Hall, Fiske, and Moores.

Hodge, R. and G. Kress. 1993. Language as ideology, 2nd ed. London:

Routledge.

The first edition of this work (Kress and Hodge 1979) was one of two
foundation texts of Critical Linguistics (along with Fowler, et al. 1979),
an approach which had a focus on media language from the beginning.
This revised edition reprints the original text intact, adding a preface
which disarmingly acknowledges the criticisms and shortcomings of the
original but re-affirms the general approach which has now grown into
Critical Discourse Analysis. A long new final chapter updates the
framework and illustrates it with analyses of media coverage of the 1991
'Gulf War.'
 LANGUAGE AND THE MEDIA 37

Jucker, A. 1992. Social stylistics: Syntactic variation in British newspapers.

Berlin: Mouton de Gruyter.

This is a major study of stylistic differences in the British national press,

based on close syntactic analysis of a large corpus of noun phrases.
Jucker found that the complexity of NPs serves as an indicator of style,
with 'down-market' newspapers using a distinctly higher percentage of
names and pronouns as NPs. The most socially sensitive aspect of the
NP is what Bell (e.g., 1991b) has analyzed as 'determiner deletion'—the
characteristic media usage by which expressions such as the Prime
Minister Margaret Thatcher become Prime Minister Margaret Thatcher.
This is a socially diagnostic variable, with the degree of deletion directly
proportionate to the social status of the newspaper's readership, in accord
with the principle of audience design.

Potter, J., M. Wetherell and A. Chitty. 1991. Quantification rhetoric—Cancer on

television. Discourse and Society. 2.333-365.

The way in which the news media use 'facts and figures' as part of their
rhetorical strategies has received some attention from media discourse
analysts (e.g., Bell 1991b, van Dijk 1988a; 1988b). Potter, et al. present
an analysis of how quantification is used to buttress argumentation in
television coverage of the efficacy of cancer treatments. Other recent
works which include analysis of quantification rhetoric are Lupton (1993)
and Rae and Drury (1993).

Scannell, P. (ed.) 1990. Texts and audiences. London: Sage. [Special issue of
Media, Culture and Society. 12.1]

This is one of a number of collections which represent the convergence of

semiotic, mass communication, and more linguistic approaches to media
discourse. The audience-oriented contributions (Moores, Morley and
Silverstone, and Jensen) stress the multiple meanings which can be taken
from broadcast content by different audiences. Corner, Richardson, and
Fenton relate the discourse analysis of documentaries on nuclear energy
to audience understandings of them.

Scannell, P. (ed.) 1991. Broadcast talk. London: Sage.

This admirable collection focuses on monologue and dialogue on (mainly)

British radio and television. Three papers cover the interview as a
broadcast genre (Corner, Clayman, and Harris), primarily using the
techniques of conversational analysis to examine its structure. Tolson
focuses on televised 'chat,' and Hutchby examines the 'organization of
talk on talk radio.' One broadcaster's creation of his on-air identity
38 ALLAN BELL

through discourse is analyzed in a fascinating study by Brand and

Scannell. Montgomery offers a meticulous and perceptive examination of
a daily program segment on BBC radio.

Van Dijk, T. A. 1988a. News analysis: Case studies of international and national
news in the press. Hillsdale, NJ: Lawrence Erlbaum.
1988b. News as discourse. Hillsdale, NJ: Lawrence Erlbaum.
1991. Racism and the press. London: Routledge.

Van Dijk's research on news is the most comprehensive published work

on media discourse. The two earlier books develop the structural
analysis of news discourse and cognitive approaches to news comprehen-
sion. News as Discourse is his main theoretical and methodological
statement, illustrated by international case studies in News Analysis. Van
Dijk has increasingly concentrated on the sociopolitical dimensions of
media discourse. Racism in the Press analyzes the reporting of ethnicity,
but uses mainly content analysis and other quantitative measures rather
than discourse analysis.

UNANNOTATED BIBLIOGRAPHY
Almeida, E. P. 1992. A category system for the analysis of factuality in
newspaper discourse. Text. 12.233-262.
Bell, A. 1991a. Audience accommodation in the mass media. In H. Giles, N.
Coupland and J. Coupland (eds.) Contexts of accommodation-
Developments in applied sociolinguistics. Cambridge, UK: Cambridge
University Press. 69-102.
1992. Hit and miss: Referee design in the dialects of New Zealand
television advertisements. Language and Communication. 12.327-340.
1994. Climate of opinion: Public and media discourse on the global
environment. Discourse and Society. 5.33-63.
Bhatia, T. K. 1992. Discourse functions and pragmatics of mixing: Advertising
across cultures. World Englishes. 11.195-215.
Boyd-Barrett, O. 1994. Language and media: A question of convergence. In D.
Graddol and 0. Boyd-Barrett (eds.) Media texts: Authors and readers.
Clevedon, UK: Multilingual Matters and Open University. 22-39.
Brand, G. and P. Scannell. 1991. Talk, identity and performance: The Tony
Blackburn Show. In P. Scannell (ed.) Broadcast talk. London: Sage.
201-226.
Browne, D. R. 1992. Raidio na Gaeltachta: Reviver, preserver or swan song of
the Irish language? European Journal of Communication. 7.415-433.
 LANGUAGE AND THE MEDIA 39

Burger, H. 1984. Sprache der Massenmedien. [Language of the mass media.]

Berlin: Walter de Gruyter.
1991. Das Gesprdch in den Massenmedien. [Conversation in the mass
media.] Berlin: Walter de Gruyter.
Cameron, D. and D. Hills. 1990. "Listening in": Negotiating relationships
between listeners and presenters on radio phone-in programmes. In G.
McGregor and R. S. White (eds.) Reception and response. London:
Routledge. 52-68.
Clayman, S. E. 1990. From talk to text: Newspaper accounts of reporter-source
interactions. Media, Culture and Society. 12.79-103.
1991. News interview openings: Aspects of sequential
organization. In P. Scannell (ed.) Broadcast talk. London: Sage. 48-75.
1992. Footing in the achievement of neutrality: The case of
news-interview discourse. In P. Drew and J. Heritage (eds.) Talk at
work: Interaction in institutional settings. Cambridge, UK: Cambridge
University Press. 163-198.
1993. Reformulating the question: A device for answering/not
answering questions in news interviews and press conferences. Text.
13.159-188.
Cook, G. 1992. The discourse of advertising. London: Routledge.
Cormack, M. 1993. Problems of minority language broadcasting: Gaelic in
Scotland. European Journal of Communication. 8.101-117.
Corner, J., K. Richardson and N. Fenton. 1990. Textualizing risk: TV discourse
and the issue of nuclear energy. Media, Culture and Society. 12.105-124.
Duszak, A. 1991. Schematic and topical categories in news story reconstruction.
Text. 11.503-522.
Fairclough, N. 1993. Critical discourse analysis and the marketization of public
discourse: The universities. Discourse and Society. 4.133-168.
Fasold, R., H. Yamada, D. Robinson and S. Barish. 1990. The language-
planning effect of newspaper editorial policy: Gender differences in The
Washington Post. Language in Society. 19.521-539.
Fowler, R., R. Hodge, G. Kress and A. Trew. 1979. Language and control.
London: Routledge and Kegan Paul.
Geis, M. 1987. Language and media. In R. B. Kaplan, et al. (eds.) Annual
Review of Applied Linguistics, 1986. New York: Cambridge University
Press. 64-73.
Goffman, E. 1981. Forms of talk. Philadelphia: University of Pennsylvania Press.
Greatbatch, D. 1988. A turn-taking system for British news interviews. Language
in Society. 17.401-430.
1992. On the management of disagreement between news
interviewees. In P. Drew and J. Heritage (eds.) Talk at work: Interaction
in institutional settings. Cambridge, UK: Cambridge University Press.
268-301.
Harris, S. 1991. Evasive action: How politicians respond to questions in political
interviews. In P. Scannell (ed.) Broadcast talk. London: Sage. 76-99.
40 ALLAN BELL

Hartford, B. A. S. 1993. Tense and aspect in the news discourse of Nepali

English. World Englishes. 12.1-13.
Heritage, J. 1985. Analysing news interviews: Aspects of the production of talk
for an overhearing audience. In T. A. van Dijk (ed.) Discourse and
dialogue. London: Academic Press. 95-119. [Handbook of discourse
analysis, Volume 3.]
and D. Greatbatch. 1991. On the institutional character of
institutional talk: The case of news interviews. In D. Boden and D. H.
Zimmerman (eds.) Talk and social structure: Studies in ethnomethodology
and conversation analysis. Cambridge, UK: Polity Press. 93-137.
Hjarvard, S. 1994. TV news: From discrete items to continuous narrative? The
social meaning of changing temporal structures. Cultural Studies.
8.306-320.
Jucker, A. H. 1986. News interviews: A pragmalinguistic analysis. Amsterdam:
John Benjamins. [Pragmatics and Beyond. 7.4].
Kress, G. 1991. Critical discourse analysis. In W. Grabe, et al. (eds.) Annual
Review of Applied Linguistics, 11. New York: Cambridge University
Press. 84-99.
and R. Hodge. 1979. Language as ideology. London: Routledge and
Kegan Paul.
Lee, D. 1991. The voices of Swiss television commercials. Multilingua.
10.295-323.
Liddicoat, A., A. Brown, S. Dopke and K. Love. 1992. The effect of the
institution: Openings in talkback radio. Text. 12.541-562.
Lupton, D. 1993. AIDS risk and heterosexuality in the Australian press.
Discourse and Society. 4.307-328.
Mancini, P. (ed.) 1988. The analysis of news texts. London: Sage. [Special issue
of European Journal of Communication. 3.2.]
Meyers, M. 1994. Defining homosexuality: News coverage of the "repeal the
ban" controversy. Discourse and Society. 5.321-344.
Montgomery, M. 1991. Our Tune: A study of a discourse genre. In P. Scannell
(ed.) Broadcast talk. London: Sage. 138-177.
Myers, G. 1994. Words in ads. London: Edward Arnold.
Nir, R. and I. Roeh. 1992. Intifada coverage in the Israeli press: Popular and
quality papers assume a rhetoric of conformity. Discourse and Society.
3.47-60.
O'Donnell, H. 1994. Mapping the mythical: A geopolitics of national sporting
stereotypes. Discourse and Society. 5.345-380.
Pietila, V. 1992. Beyond the news story: News as discursive composition.
European Journal of Communication. 7.37-67.
Rae, J. and J. Drury. 1993. Reification and evidence in rhetoric on economic
recession: Some methods used in the UK press, final quarter 1990.
Discourse and Society. 4.329-356.
Reynolds, D. W. 1993. Illocutionary acts across languages: Editorializing in
Egyptian English. World Englishes. 12.35-46.
 LANGUAGE AND THE MEDIA 41

Richardson, K. 1994. Audience research—Beyond comprehension. Paper

presented to UK Sociolinguistics Symposium 10, Lancaster University.
Lancaster, UK, 1994.
Roeh, I. and R. Nir. 1990. Speech presentation in the Israel radio news:
Ideological constraints and rhetorical strategies. Text. 10.225-244.
Tanaka, K. 1994. Advertising language: A pragmatic approach to advertisements
in Britain and Japan. London: Routledge.
Tolson, A. 1991. Televised chat and the synthetic personality. In P. Scannell
(ed.) Broadcast talk. London: Sage. 178-200.
Tulloch, J. and S. Chapman. 1992. Experts in crisis: The framing of radio debate
about the risk of AIDS to heterosexuals. Discourse and Society.
3.437-467.
Vagle, W. 1991. Radio language—Spoken or written? International Journal of
Applied Linguistics. 1.118-131.
Van Dijk, T. A. 1985. Structures of news in the press. In T. A. van Dijk (ed.)
Discourse and communication: New approaches to the analysis of mass
media discourse and communication. Berlin: de Gruyter. 69-93.
1993. Principles of critical discourse analysis. Discourse and
Society. 4.249-283.
Van Leeuwen, T. 1994. Grammar, identity and the press. Paper presented to UK
Sociolinguistics Symposium 10, Lancaster University. Lancaster, UK,
1994.
Wang, S. 1993. The New York Times' and Renmin Ribao's news coverage of the
1991 Soviet coup: A case study of international news discourse. Text.
13.559-598.
Wodak, R. 1991. Turning the tables: Antisemitic discourse in post-war Austria.
Discourse and Society. 2.65-83.
Yaeger-Dror, M. 1993. Linguistic analysis of dialect "correction" and its
interaction with cognitive salience. Language Variation and Change.
5.189-224.