QUALITATIVE DISCOURSE ANALYSIS
In the conclusion to a now classic qualitative analysis of a newspaper report
about racial conflict in Southern Africa, Tony Trew, one of the founders of ‘critical
linguistics,’ stated that the text ‘is so far from the truth that only a powerful grip on
the press and information and the diligence of the media in resolving the flood of
anomalies which they report are adequate to preserve the pretence that the press
is truthful’ (Trew 1979: 106). The quotation might give the impression that critical
linguistics simplistically checks whether a verbal account of political and other
events is ‘true’ or not. While this is not the case, analyses in the ‘critical linguistic’
tradition will often lapse into direct contradiction of the media’s picture of reality,
fueled by a strong political and educational commitment.
Critical linguistics developed in the 1970s as an influential early school of
discourse analysis. The approach was able to document a close relationship
between the linguistic details of media texts and the production of ideology and,
by implication, to substantiate that media ideology contributes to the
reproduction of a social order founded on inequality and oppression. Its
epistemology was a somewhat heavy-handed linguistic constructivism according
to which words construct a mental grid. Words are conventional, but constitutive
of the reality they designate, so that, for instance, newspapers inevitably
construct the social states and events they describe.
Critical linguistics went on to claim that not just lexical, but also syntactic
choices have semantic and hence ideological implications, as exemplified below.
Relying on Saussure’s notion of the arbitrary sign, the point is not that linguistic
choices can be made by journalists at will, but instead that these choices
articulate and enact social power in practice. Since the formation of public opinion
in capitalist societies is controlled by those with economic and organizational
power over the mass media, who will see to it that representations do not
jeopardize their interests and privileges, public communication has a pervasive
ideological thrust that renders current social arrangements natural, even
inevitable, while discrediting alternatives as contrary to common sense. As
summed up by Fowler (1985), the task of critical linguists is to expose such
ideology in the best interest of the majority: ‘Of course from my point of view this
has to be regarded as an undesirable situation and one that a critical linguist ought
to expose’ (p. 67).
Dissolving seemingly innocent linguistic features, critical textual analyses
can debunk ‘warped versions of reality’ (p. 68). Building on Halliday’s functional
grammar and social semiotic (Halliday 1973, 1978), numerous publications from
the late 1970s and into the 1980s (e.g., Fowler et al. 1979; Hodge and Kress 1988)
identified linguistic features that ‘will probably repay close examination,’
according to Fowler (1985: 68). He offered the following ‘checklist’ (with
examples in parentheses referring to the opening extract of Review of the Year
1998):
1. Lexical processes arise from the particular vocabulary, including
metaphors, which a media text mobilizes concerning a given area of
experience, such as scientific vocabulary in a cosmetics ad, or
management jargon in political coverage (the ‘launch’ of digital channels
in the BBC program refers simultaneously to actual, future channels and
to the program’s own format).
2. Transitivity, a relation obtaining between verbs and nouns, adds up to a
construction and configuration of the participants and processes in a text
(the review not only ‘launches’ new channels, but it ‘embraces’ the future,
and anticipates what, very soon, ‘we’ll all be able to watch’).
3. Syntactic transformations, particularly those labeled ‘passivization’ and
‘nominalization,’ can be considered ideologically problematic because
they may obscure agency – who did what to whom – or change the
participants’ relative prominence (‘we’re told’ that a digital cornucopia
will soon be available, but not who tells us, or with what justification).
4. Modality, in the form especially of modal verbs and adverbs, serves to
express speakers’ and other senders’ assessments of, and attitudes
towards, the events rendered by the sentences constituting the text (the
reference to ‘possibly’ thousands of channels is partly a qualification of
the prediction itself, but also partly an expression of skepticism toward
everything ‘we’re told’).
5. Speech acts and turn-taking refer to the fact that speech is a coordinated
form of action, in which speakers take turns in complex ways. One insight
of discourse analysis generally is that there is no one-to-one relation
between sentence types and communicative functions. Instead, analysts
should ask which speech act is accomplished by the utterance of a
sentence, and how both the single act and the structure of turn-taking
may build positions of power into the communicative context (e.g., a
classroom or a medical consultation). (While the news review is not
dialogic in the sense of talk or discussion programs, already the opening
‘Welcome,’ itself a conventional act of greeting, invites the viewer to
attend to the ‘turn’ of this program).
6. Implicature, comprising a variety of communicative features, is perhaps
best explained as the ability of most language users to ‘read between the
lines,’ to infer what is ‘really’ meant. It is this ability which is made
systematic and procedural in (critical) linguistics, particularly in attempts
to recover ‘latent’ meanings through a ‘hermeneutics of suspicion’ (the
reference to The Men’s Channel as being not for, but about, men implies
the existence of monogenetic pornographic channels, and relies on
viewers’ knowledge of this to poke gentle fun at men’s ‘follies’).
7. Address and personal reference, finally, allow linguistic analysis to
consider how stylistic choices anticipate an audience with a particular
social or educational status; how naming conventions affect the degree
of formality; and how personal pronouns assert a particular
‘interpersonal’ relationship with an audience (in addition to the inclusive
and exclusive we, ‘you understand’ is an explicit appeal to both female
and male viewers to accept the terminological joke about The Men’s
Channel).
Although this checklist has found its way into later discourse studies of
media, it is not always evident that its elements represent productive analytical
categories. Transitivity, for example, requires a very time-consuming scrutiny of
textual detail to establish an ideological pattern (e.g., Fowler 1985), which might
be predictable from a more cursory examination. Furthermore, it is doubtful
whether claims about the ‘mystifying’ effects of syntactic transformations are
warranted, being based on assumptions about the ‘non-recoverability’ of the
deleted linguistic component (Trew 1979). It seems equally probable that average
mediausers will have no difficulty inferring who did what to whom, despite passive
sentences or nominalizations, on the basis of their background knowledge of the
world, their familiarity with a current news agenda, and their general
communicative competence. Interestingly, an empirical study of readers’
reception of the newspaper articles previously analyzed by Trew found that the
readers’ views of the events reported in the articles were determined less by their
‘syntactically constructed’ ideology than by the readers’ social identities and life
histories (Sigman and Fry 1985).
In sum, critical linguistics made an important departure, both beyond
content analyses of manifest meaning, and beyond formal and poetic traditions of
textual analysis, but remained closely focused on linguistic indicators. Other
forms of discourse analysis have gone some way toward placing such indicators
in their communicative and social contexts.
