9

Critical Discourse Analysis

The norms and values which underlie texts are often ‘out of sight’ rather than overtly stated.
As Hyland (2005b: 4) observes, acts of meaning making (and in turn discourse) are ‘always
engaged in that they realize the interests, the positions, the perspectives and the values
of those who enact them’. The aim of a critical approach to discourse analysis is to help
reveal some of these hidden and ‘often out of sight’ values, positions and perspectives. As
Rogers (2004: 6) puts it, discourses ‘are always socially, politically, racially and economically
loaded’. Critical discourse analysis examines the use of discourse in relation to social and
cultural issues such as race, politics, gender and identity and asks why the discourse is used
in a particular way and what the implications are of this kind of use.

Critical discourse analysis explores the connections between the use of language and the social and
political contexts in which it occurs. It explores issues such as gender, ethnicity, cultural difference, ide-
ology and identity and how these are both constructed and reflected in texts. It also investigates ways
in which language constructs and is constructed by social relationships. A critical analysis may include
a detailed textual analysis and move from there to an explanation and interpretation of the analysis. It
might proceed from there to deconstruct and challenge the text(s) being examined. This may include
tracing underlying ideologies from the linguistic features of a text, unpacking particular biases and
ideological presuppositions underlying the text, and relating the text to other texts and to people’s
experiences and beliefs.

Critical discourse analysis starts with the assumption that language use is always social
and that discourse both ‘reflects and constructs the social world’ (Rogers 2011: 1). A criti-
cal analysis might explore issues such as gender, ideology and identity and how these are
reflected in particular texts. This might commence with an analysis of the use of discourse
and move from there to an explanation and interpretation of the discourse. From here,
the analysis might proceed to deconstruct and challenge the texts, tracing ideologies and
assumptions underlying the use of discourse, and relating these to different views of the
world, experiences and beliefs (Clark 1995).
 Critical Discourse Analysis 187
9.1 Principles of critical discourse analysis
There is no single view of what critical discourse analysis actually is, so it is difficult to
present a complete, unified view on this. Fairclough and Wodak (1997), however, describe a
number of principles for critical discourse analysis which underlie many of the studies done
in this area. These include

social and political issues are constructed and reflected in discourse;

power relations are negotiated and performed through discourse;
discourse both reflects and reproduces social relations;
ideologies are produced and reflected in the use of discourse.

Each of these is discussed in the sections which follow.

Social and political issues are constructed and reflected in discourse

The first of Fairclough and Wodak’s principles is that critical discourse analysis addresses
social and political issues and examines ways in which these are constructed and reflected
in the use of certain discourse strategies and choices.
Recently, I received mail in my letter box about a proposal to build 125 apartments on
top of a shopping mall which is very near where I live. There was a letter from the local
council and a pamphlet from a local protest group, each of which expressed very different
views on the development. One was very factual (the letter from the council) which aimed
to remain ‘neutral’ on the topic in that it did not express a particular point of view on the
project. It just outlined the procedures for the development, and how the public would be
consulted about it, largely through written submissions and a public meeting that would
be held in the near future. The pamphlet from the protest group, by contrast, outlined the
problems the development would create for the neighbourhood such as overshadowing of
properties, increased demand for on-street parking and lack of privacy from windows in
the apartments that would overlook people’s back gardens. Both texts, then, referred to the
same event but chose very different ways of approaching it which were, in turn, reflected in
their discourse. One gave the impression it was neutral on the topic whereas the other had
a very particular take on the new development and what it would mean for people living in
the area.
A further example of this is Teo’s (2005) study of slogans for Singapore’s ‘Speak Mandarin’
campaign. In this campaign there is clearly a view that Singaporeans of Chinese decent
should speak Mandarin despite the fact that at the time of the launch of the campaign only
a small percentage of them actually spoke Mandarin as their first language. The aim of the
campaign was to connect Chinese Singaporeans with Chinese cultural traditions as well as
help counter ‘negative effects of westernisation’ (123). The campaign was also motivated by
188 Discourse Analysis

an economic policy which aimed at attracting foreign investment, especially from China.
These arguments were captured in slogans such as Mandarin: Window to Chinese Culture,
Speak Mandarin, It’s an Asset and Speak Mandarin: Your Children’s Future Depends on
Your Effort. Mandarin was also presented as cool and of contemporary relevance, as well
as a ‘stepping-stone to greater business opportunities’ with the Chairman of the Promote
Mandarin Council saying that Mandarin is ‘“cool” in more ways than one’, ‘Mandarin is
definitely “in”’ and Mandarin is ‘a store of linguistic and cultural treasure waiting to be
explored’ (134). The discourse of the campaign, thus, constructs the view of Mandarin as a
language that has both cultural, social and, in particular, economic value for the people of
Singapore.

Power relations are negotiated and performed through discourse

The next principle of critical discourse analysis is that power relations are both negotiated
and performed through discourse. One way in which this can be looked at is through an
analysis of who controls conversational interactions, who allows a person to speak and how
they do this.
In the case of the building of the apartments near my house the two different texts I
received gave very different impressions of how people were encouraged to speak on the
topic and negotiate different points of view on the development. The letter from the Council
said that people would be free to speak at the public meeting, but that they would be required
to register their intention to speak at the start of the meeting. The letter did not mention
that the full Council plus representatives of the developers who had put forward the pro-
posal would be at the meeting. The pamphlet encouraged residents to write to their local
councillors about the issue and gave the names and address of each of the members of the
local council. The area in which I live however is very multicultural. I am not certain the
elderly Vietnamese couple who live next door to me would have written to the councillors,
or would have felt their voices would have been heard had they gone to the meeting. They
would most likely not have felt they had the power to change things, nor were they in a
position to influence the outcomes of this discussion. The 90-year old woman who lives
by herself on the other side of my property may also not have felt comfortable going to the
meeting and standing up in front of everyone to have her say, as unhappy as she was about
the development.
Hutchby (1996) examined issues of power in his study of arguments in British radio talk
shows. As Hutchby and Wooffitt (2008) point out, the person who speaks first in an argu-
ment is often in a weaker position than the person who speaks next. The first person has to
set their opinion on the line whereas the second speaker merely has to challenge the oppo-
nent to expand on, or account for the claims. In a radio talk-back programme it is normally
the host that comes in the second position and has the power to challenge the caller’s claim,
 Critical Discourse Analysis 189
or to ask them to justify what they have just said. The following example shows how a talk-
back show host does this simply by saying Yes and So?:

Caller: I: have got three appeals letters here this week. (0.4) All a:skin’ for donations. (0.2). hh
Two: from tho:se that I: always contribute to anywa:y,
Host: Yes?
Caller: .hh But I expect to get a lot mo:re.
Host: So?
Caller: .h Now the point is there is a limi[t to . . .
Host: [What’s that got to do – what’s that got to do with
telethons though.
Caller: hh Because telethons. . . .
(Hutchby 1996: 489)

The host does this again in the next example where What’s that got to do with it? challenges
the caller and requires them to account for what they just said:

Caller: When you look at e:r the childcare facilities in this country, hh we’re very very low, (.)
i-on the league table in Europe of (.) you know of you try to get a child into a nursery it’s
very difficult in this country . . . hh An’ in fa:ct it’s getting wor::se.
Host: What’s that got to do with it.
Caller: .phh Well I think whu- what ‘at’s gotta d-do with it is. . . .
(Hutchby 1996: 490)

The caller can take the second speaking part in this kind of interaction only when the host
has moved, or been manoeuvred, into first position by giving an opinion of their own. If this
does not happen, it is hard for the caller to take control of the conversation and challenge the
control of the host. This kind of analysis, then, shows how power is brought into play, and
performed, through discourse (Hutchby 1996).

Discourse both reflects and reproduces social relations

A further principle of critical discourse analysis is that discourse not only reflects social
relations but is also part of, and reproduces, social relations. That is, social relations are both
established and maintained through the use of discourse.
The letter from my Council that I referred to earlier was written with authority and con-
tained a lot of technical detail, setting up a very clear power imbalance between the writer
and readers of the text. It was signed ‘Director – Planning and Development’ – and gave no
actual name for people to call to speak to. The pamphlet from the protest group, however,
was much more informal and gave an email address to write to for further information
and advice on what to do to change the situation. The social relations produced (and repro-
duced) through the two texts were, thus, quite different.
Page’s (2003) study of representations in the media of Cherie Blair, wife of the former
British Prime Minister Tony Blair, illustrates this further. Page shows how representations
190 Discourse Analysis

of Cherie Blair in the media as a lawyer, a wife and, especially, a working mother aim to
establish a certain relationship between her and the public and, in particular, other working
mothers. While Cherie Blair is largely presented by the media as a success story for manag-
ing her role as a working mother, as Page points out, working mothers are more typically
presented in negative terms in everyday discourse in a way that produces quite different
readings of the term and, in turn, different views of working women who have children.
Stokoe’s (2003) study of neighbourhood disputes shows, equally, how terms such as mother
and single women can be used to make moral assessments about women as well as per-
petuate ‘taken-for-granted “facts” about women’s appropriate behaviour’ (339) and social
relations with other people. The use of language in this way both reflects and reproduces
certain social views and relations. It, equally, reinforces social and gendered stereotypes and
inequalities (Page 2003).

Ideologies are produced and reflected in the use of discourse

Another key principle of critical discourse analysis is that ideologies are produced and
reflected in the use of discourse. This includes ways of representing and constructing soci-
ety such as relations of power, and relations based on gender, class and ethnicity.
Each of the pieces of mail I received on the development project near my house was quite
different. The letter from the Council presented the addition of 125 new apartments (plus
additional retail and commercial floor space and three eight-storey residential towers on top
of the already five-storey mall) at the end of my street as a neutral event that would have no
consequences for me or for where I am living. The pamphlet from the protest group made it
clear what the consequences of this would be, outlining them in detail and strongly voicing
its opposition to the project.
Mallinson and Brewster’s (2005) study of how stereotypes are formed in everyday spoken
discourse is a further illustration of the ways in which ideologies are produced and reflected
in the use of discourse. As Mallinson and Brewster point out, negative attitudes towards
non-standard social dialects of English are often transferred to negative views of the people
who speak these dialects. A job applicant who speaks a non-standard dialect, for example,
may not be hired when an employer sees this use of discourse as a way of predicting the
applicant’s future occupational performance; that is, the view that ‘good workers’ speak
standard English and ‘bad workers’ do not.
In their study of US restaurant workers’ views of their customers, Mallinson and Brewster
found that the (white) workers viewed all black customers as the same, in negative terms,
and using stereotypes to form their expectations about future interactions with black cus-
tomers, and the broader social group of African Americans. This was clear in the ‘discourse
of difference’ (Wodak 1997) that they used as they spoke about their black customers and
distanced themselves from them. The workers’ views of rural white Southerner customers
were similarly stereotyped, although they talked about this group in somewhat different
 Critical Discourse Analysis 191
ways, referring to where they lived, the ways they dressed and their food and drink pref-
erences as a way of justifying their claims about them. In both cases, the workers’ use of
discourse privileged their own race and social class, reflecting their ideological, stereotyped
views of both groups of customers.
Fairclough and Wodak also argue that all texts need to be considered in relation to the
texts that have preceded them and those that will follow them. They also need to be under-
stood by taking sociocultural knowledges of the texts and the matter at hand more broadly
into consideration. In the letter from my Council, there was no mention that this was the
third time the development company was attempting to have their proposal approved, and
that there had been two previous public meetings on the topic where the application had
been rejected. The pamphlet from the protest group, however, made this very clear. A criti-
cal analysis of these communications then, is a form of social action in that it attempts
to intervene and bring about change in both communicative and sociopolitical practices
(Fairclough and Wodak 1997).
Critical discourse studies, then, aim to make connections between social and cultural
practices and the values and assumptions that underlie the discourse. That is, it aims to
unpack what people say and do in their use of discourse in relation to their views of the
world, themselves and their relationships with each other. Critical discourse analysis takes
the view that the relationship between language and meaning is never arbitrary in that the
choice of a particular genre or rhetorical strategy brings with it particular presuppositions,
meanings, ideologies and intentions (Kress 1991). As Eggins (1994: 10) argues:

Whatever genre we are involved in, and whatever the register of the situation, our use of language
will also be influenced by our ideological positions: the values we hold (consciously or uncon-
sciously), the biases and perspectives we adopt.

Thus, if we wish to complain about a neighbour we may chose a genre such as a neighbour
mediation session, or we may decide to air our complaint in a television chat show, as some
of the speakers did in Stokoe’s (2003) study of neighbour complaints. We may also do this
by complaining to another neighbour about them. Our intention in speaking to the other
neighbour may be to build up a ‘neighbourhood case’ against the person we are unhappy
with. If the neighbour we are complaining about is a single mother we may draw on other
people’s prejudices against single mothers, and our own biases and moral judgements about
them as an added rationale for complaining about the neighbour. The woman being com-
plained about may pick up on this, as did one of Stokoe’s subjects, Macy, in a neighbour
mediation session where she says ‘if I had a big bloke living with me . . . none of this would
happen’ (Stokoe 2003: 329). Macy does not allow her single status to be used as a reason to
complain about her.
In a further extract Stokoe shows how speakers may draw on the fact that their neigh-
bour has boyfriends (more than one) as added ammunition against her; that is, the view
192 Discourse Analysis

that women should be monogamous (but not men) and if a woman breaches this rule, they
should be held morally accountable for their behaviour. The following example illustrates
this. In this example Terry (T) is the chat show host and Margaret (M) is a member of the
audience who is complaining about her neighbour:

T: I want to know (.) what happened to you

(0.5)
M: after living very happily (.) in my (.) one bedroom flat for thirteen years (.) it was a three
storey block of flats and I was on the top floor (.) and the young woman was put in the flat
below me (0.5) I them had (.) seven and a half ye:ars (.) of sheer hell
T: what sort of hell?
M: loud music (.) night and day (.) it just depended=
T: =well that wasn’t the worst was it?
M: = (0.5) it was boyfriends (.) and lovemaking that . . . (333)

The rhetorical strategy here, then, is to draw on a moralizing discourse about women (and
especially, single women who have sex) as a way of legitimating complaints about female
neighbours, as well as building a defence for making the complaint (Stokoe 2003).
Resende (2009) provides a similar example of how, through the use of discourse, groups
of people are framed in particular ways. She examines a report of a meeting on homeless
people which was sent, as a circular, to residents of a middle-class apartment building in
Brazil. The circular reported on a meeting that had been convened by a local restaurant
owner who was concerned about homeless people living in the area and the financial impact
it was having on his business. What she found was the genre of ‘apartment circulars’ which
normally addresses issues such as building maintenance had been appropriated to make a
case for the removal of people from the area. It used terms such as ‘government authorities’
and ‘public security representative’ to give it authority so that the views expressed in the
circular would be taken as given and not open for discussion. Views were expressed cat-
egorically and disguised the main issue, the problem of living on the streets. The issue was
reframed in terms of individual and community comfort and avoided the underlying social
cause of the problem.
Meadows (2009) employed ethnographically-sensitive critical discourse analysis to exam-
ine the relationship between nationalism and language learning in an English language
classroom on the Mexico/US border. The students in the classes he examined were all
Mexican management-level employees who held positions of economic and social privilege
in their particular community. The data he collected included participant observations,
interviews, questionnaires, classroom activities and emails. The data collected was then
examined from a critical discourse perspective with the aim of exploring how the relation-
ship between nationalism and language learning played out in the classroom. Meadows
found that the classroom provided a site for the reproduction of nationalist border prac-
tices as well as a place in which hierarchies of privilege were reinforced. This was revealed
 Critical Discourse Analysis 193

TEXTS

DISCOURSE PRACTICES

SOCIOCULTURAL PRACTICES

Figure 9.1 The relationship between texts, discourse practices and sociocultural practices in a critical
perspective (adapted from Fairclough 1992: 73)

through the ways in which the students discussed border categories using, for example,
polar categories such as americano and mexicano, and negative attitudes to categories which
blurred these boundaries such as Spanglish, Mexican-American and paisano. The teacher,
further, often used nationalist categories in the classroom which the students positively
responded to. His study shows how language classrooms are not only about developing lan-
guage proficiency. They are also often closely linked to students’ investment in their worlds
outside the classroom.
A key focus, then, of critical discourse studies is the uniting of texts with the discourse
and sociocultural practices that the text reflects, reinforces and produces (Fairclough 1995).
The chart in Figure 9.1 summarizes this. Discourse, in this view, simultaneously involves
each of these dimensions.

9.2 Doing critical discourse analysis

Critical discourse analysis ‘includes not only a description and interpretation of discourse
in context, but also offers an explanation of why and how discourses work’ (Rogers 2004: 2).
Researchers working within this perspective

are concerned with a critical theory of the social world, the relationship of language and discourse
in the construction and representation of the social world, and a methodology that allows them to
describe, interpret and explain such relationships. (Rogers 2011: 3)
194 Discourse Analysis

A critical analysis, then, might commence by deciding what discourse type, or genre, the
text represents and to what extent and in what way the text conforms to it (or not). It may
also consider to what extent the producer of the text has gone beyond the normal bounda-
ries for the genre to create a particular effect.
The analysis may consider the framing of the text; that is, how the content of the text is
presented, and the sort of angle or perspective the writer or speaker is taking. Closely related
to framing is the notion of foregrounding ; that is, what concepts and issues are emphasized,
as well as what concepts and issues are played down or backgrounded in the text. Equally
important to the analysis are the background knowledge, assumptions, attitudes and points
of view that the text presupposes (Huckin 1997).
At the sentence level, the analyst might consider what has been topicalized in each of the
sentences in the text; that is, what has been put at the front of each sentence to indicate what
it is ‘about’. The analysis may also consider who is doing what to whom; that is, agent-patient
relations in the discourse, and who has the most authority and power in the discourse. It
may also consider what agents have been left out of sentences such as when the passive voice
is used, and why this has been done (ibid.).
At the word and phrase level, connotations of particular words and phrases might be
considered as well as the text’s degree of formality or informality, degree of technicality
and what this means for other participants in the text. The choice of words which express
degrees of certainty and attitude may also be considered and whether the intended audience
of the text might be expected to share the views expressed in the text, or not (ibid.).
The procedure an analyst follows in this kind of analysis depends on the research situa-
tion, the research question and the texts that are being studied. What is essential, however,
is that there is some attention to the critical, discourse and analysis in whatever focus is
taken up in the analysis (Rogers 2011).
Critical discourse analysis, then, takes us beyond the level of description to a deeper
understanding of texts and provides, as far as might be possible, some kind of explanation of
why a text is as it is and what it is aiming to do. It looks at the relationship between discourse
and society and aims to describe, interpret and explain this relationship (ibid.). As van Dijk
(1998) has argued, it is through discourse that many ideologies are formulated, reinforced
and reproduced. Critical discourse analysis aims to provide a way of exploring this and, in
turn, challenging some of the hidden and ‘out of sight’ social, cultural and political ideolo-
gies and values that underlie texts.

9.3 Critical discourse analysis and genre

One way in which a question might be approached from a critical perspective is by con-
sidering the genres that have been chosen for achieving a particular discourse goal.
Flowerdew (2004) did this in his study of the Hong Kong government’s promotion
 Critical Discourse Analysis 195
campaign of Hong Kong as a ‘world-class city’. He discusses the various genres that were
involved in constructing this view of Hong Kong. These included committee meetings,
policy speeches, commission reports, an inception report, public fora, exhibitions, focus
group discussions, presentations, a website, consultation documents, information leaf-
lets, consultation digests and videos. He discusses how each of these genres played a role
in the construction of this particular view of Hong Kong. He then carries out an analy-
sis of three different genres which made this claim: a public consultation document, the
Hong Kong annual yearbook and a video that was produced to promote Hong Kong as
‘Asia’s World City’.
Flowerdew shows how the Hong Kong bureaucracy developed and constructed this par-
ticular view, from the generation of an initial idea through to the public presentation of
this view. He also shows that while the official aim of the consultation process was to gain
feedback on the proposal, it was as much designed to win over the public to this view. The
public consultation document, Flowerdew shows, used a language of ‘telling’ rather than a
language of ‘asking’ (or indeed consulting). The tone of the text was prescriptive in its use of
the modal verb will, for example, as in every Hong Kong resident will, HK 2030 will involve
and This will ensure. The voice of authority, thus, was dominant in the use of the genre and
discouraged dissent from the view that it promoted.
The yearbook that Flowerdew examined extolled the virtues of Hong Kong and was
overtly promotional in nature. Other voices were brought in to give authority to this view
such as ‘perceptions of Hong Kong internationally’ and ‘our review on international per-
spectives of Hong Kong’. Who these views actually belonged to was not stated. This text,
interestingly, was produced before the actual public consultation process commenced and
suggests the government had already decided on the outcome of its consultation, before it
had actually commenced.
The voice on the video that was examined, as with the yearbook, was overwhelmingly
promotional. The difference between this and the yearbook was that the video used the
discourse of advertising and public relations to make its point, rather than the discourse
of bureaucracy. The video used short sharp pieces of text such as The pace quickens and
Horizons expand. These statements were accompanied by a series of images of technology,
architecture and nightlife that presented Hong Kong as a vibrant and fast-paced, modern
city. The mix of traditional Chinese and Western music on the soundtrack of the video gave
both an Asian and an international feel to the video. The video was also produced before
the public consultation actually took place. As Flowerdew points out, this is consistent with
branding theory which emphasizes the importance of gaining support and the belief of the
public in the promotion of a brand, or product. It is not, however, the sequence of genres
that might be expected to conform with a public consultation process. Flowerdew, then,
shows how the voices of three very different genres come together to impose, rather than
negotiate, a certain point of view on the readers and viewers of the texts, which formed part
of the campaign.
196 Discourse Analysis

9.4 Critical discourse analysis and framing

A further way of doing a critical analysis is to examine the way in which the content of a text
is used; that is, the way in which the content of the text is presented to its audience, and the
sort of perspective, angle and slant the writer or speaker is taking. Related to this is what is
foregrounded and what is backgrounded in the text; that is, what the author has chosen to
emphasize, de-emphasize or, indeed, leave out of the text (Huckin 1997).
Huckin (1997) looks at a newspaper report on a demonstration at a nuclear test site
in the United States in just this way. Figure 9.2 is the opening section of the text he
examined.
The demonstration described in this report is framed as a confrontation between the
group of protesters and law-officials. The report does not discuss the issue that motivated
the protest. The protesters and how many were arrested is presented to the readers. The
protesters are presented negatively, as trespassers, rather than as people with a concern
for the environmental future of their country. A ‘police versus protesters’ frame is fore-
grounded, and also presented, rather than the social, public health or environmental issues
they are protesting about. There is much that is backgrounded, or omitted from the text.
Information on nuclear testing planned for the site is left out, nor is anything mentioned
of the health issues faced by people living near the site. The role of the government is also
omitted from the text. The story, thus, presupposes that the most interesting and important
aspect of the story is the number of protesters that were arrested, not the issues they were
protesting about (ibid.).
Huckin goes on to examine topicalization in the text. As he shows, the topic of the sen-
tences support his claim that the text is ‘about’ protesters versus officials, not the issues

Nevada Officials Arrest 700 At Test-Site Gulf Protest

MERCURY, NV (AP) – More than 700 people were arrested Saturday during an anti-nuclear, anti-Persian Gulf buildup protest at
the Nevada Test Site, an official said.
Thousands turned out for the demonstration. Those arrested on misdemeanour trespass charges were taken to holding pens,
then transported by bus to Beatty, 54 miles north of the remote nuclear proving ground.
An Energy Department spokesman estimated the crowd at 2,200 to 2,500 people. A sponsor of the protest, American Peace
Test, said the crowd was 3,000 to 4,000 strong.
The turnout was one of the largest since anti-nuclear demonstrations began at the test site nearly a decade ago, but it failed
to match a turnout of 5,000 demonstrators in 1987, when 2,000 people were arrested on trespass charges.
The DOE spokesman, Darwin Morgan, said more than 700 people were arrested and would be released in their own recogni-
zance. ‘Some of the demonstrators were a bit more aggressive, kicking at the guards when they were brought out of the pens’
Morgan said.…

Figure 9.2 A newspaper report on an anti-nuclear demonstration (Huckin 1997: 85) Used with permission of
The Associated Press © 2011.
 Critical Discourse Analysis 197
that prompted the demonstration. In the following examples the topic of each sentence is
in italics:

More than 700 people were arrested Saturday during an anti-nuclear, anti-Persian Gulf buildup
protest at the Nevada Test Site.

Thousands turned out for the demonstration.

A sponsor of the protest, American Peace Test, said the crowd was 3,000 to 4,000 strong.

Throughout the text, it is the officials that largely have the agency; that is, it is they who
initiate the action. They do the arresting and decide if the protestors will be released. The
protesters only have agency when they are engaged in antisocial behaviour, such as kicking
the guards.
The text is mostly written in the semi-formal register of ‘objective’ news reporting. Events
are presented as factual, ‘without the slightest trace of uncertainty’. This has the effect of
making the issues that underlie the protest ‘completely closed to discussion and negotiation’
(89–90). As Huckin shows, the tactics used by the writer put a particular slant on the text
and encourage the reading of the text in a particular way. Analyses of this kind, thus, aim to
bring hidden meanings to the surface by unpacking the assumptions, priorities and values
that underlie texts.

9.5 Critical discourse analysis and

larger data sets
Much of the work in critical discourse analysis often draws its discussion from the analy-
sis of only a few texts which have sometimes been criticized for being overly selective and
lacking in objectivity. One way in which the scale of texts used for a critical analysis can be
expanded is through the use of texts that are available on the world wide web. Using mate-
rial from the world wide web is not without its problems, however. It is not always possible
to identify the source of texts on the web. It is also not always possible to determine which
texts have more authority on a topic than others on the web. It is also difficult to see some-
times who in fact is writing on the web. Texts on the web, further, also often rely on more
than just words to get their message across. The multimodal nature, thus, needs to be taken
account of in any analysis of material from the web. Texts on the web are also more subject
to change than many other pieces of writing. Each of these issues needs to be considered
when using data from the world wide web for a critical (or indeed any kind of) discourse
study (Mautner 2005a).
The world wide web has, however, been used productively to carry out critical discourse
studies which draw on the strengths of the web’s capacity to collect a lot of relevant data.
Mautner’s (2005b) study of ‘the entrepreneurial university’ is an example of this. Mautner
198 Discourse Analysis

did a search of the web for the term ‘entrepreneurial university’ to see who was using this
term, what genres it typically occurs in and how it is typically used. She used a search
engine to do this as well as carried out a trawl through the websites of 30 of the top UK uni-
versities to find further uses of the term. She also used a reference corpus to see what words
‘entrepreneurial’ typically collocated with, outside of her particular area of interest.
Mautner observes that the use of the term entrepreneurial university brings together
the discourses of business and economics with the discourse of the university. It is not
just the newer, seemingly more commercially driven universities, however, that are doing
this. The following example from the Oxford University website illustrates this:

Oxford is one of Europe’s most innovative and entrepreneurial universities. Drawing on an 800-
year tradition of discovery and invention, modern Oxford leads the way in creating jobs, skills and
innovation for the 21st century. (109)

The term ‘entrepreneurial university’ was not, however, used positively in all the texts that
Mautner examined. On occasions writers purposely distanced themselves from the term by
putting scare quotes around these words. Even those who were advocates of the entrepreneur-
ial university also showed they were aware of the potentially contentious nature of the term
by adding qualifying statements to their use of the term, such as ‘we still care about education
and society’ and ‘it isn’t about commercialisation’ (111). Studies such as this, then, show the
enormous potential of using the world wide web for the critical study of the use of discourse.

9.6 Criticisms of critical discourse analysis

Critical discourse analysis has not been without its critics, however. One argument against
critical discourse analysis has been that it is very similar to earlier stylistic analyses that
took place in the area of literary criticism. Widdowson (1998, 2004) for example, argues that
a critical analysis should include discussions with the producers and consumers of texts,
and not just rest on the analyst’s view of what a text might mean alone. Others have sug-
gested that critical discourse analysis does not always consider the role of the reader in the
consumption and interpretation of a text, sometimes mistaking themselves for a member
of the audience the text is aimed at (van Noppen 2004). Critical discourse analysis has also
been criticized for not always providing sufficiently detailed, and systematic, analyses of the
texts that it examines (Schegloff 1997).
There have been calls for critical discourse analysts to be more critical and demanding
of their tools of analysis, as well as aim for more thoroughness and strength of evidence
for the claims that they make (Toolan 1997). Others, however, have come to the defence of
critical discourse analysis arguing that its agenda is important and of considerable social
significance but that there are important details and arguments that still need to be care-
fully worked out (Stubbs 1997).
 Critical Discourse Analysis 199
Writers such as Cameron (2001) discuss textual interpretation in critical discourse anal-
ysis saying it is an exaggeration to say that any reading of a text is a possible or valid one.
She does, however, agree with the view that a weakness in critical discourse analysis is its
reliance on just the analyst’s interpretation of the texts. She suggests drawing more on recip-
ients’ interpretations in the analysis and interpretation of the discourse as a way of counter-
ing this. As Cameron (140) suggests, a critical discourse analysis

is enriched, and the risk of making overly subjective or sweeping claims reduced, by going beyond
the single text to examine other related texts and to explore the actual interpretations recipients
make of them.

As she points out, all discourse and all communication is interactive, and this needs to be
accounted for in the analysis.
Benwell (2005) aimed to deal with this in her study of the ways in which men respond
to the discourse of men’s lifestyle magazines. Drawing on a textual culture approach, two
groups of readers were interviewed about their reading habits, practices and dispositions
with reference to issues such as gender, sexism, humour and irony in articles and images
in the magazines. Conversation analysis and membership categorization analysis (Schegloff
2007) were used for the analysis of the data. One of the interviewees said, laughing, Lucky
this is anonymous! when admitting that he had responded to the influence of an advertise-
ment in a magazine and had gone and bought a skin care product, more commonly asso-
ciated with women. In his hesitation in revealing this, he also showed his alignment with
the constructions of masculinity promoted by the magazine as well as his affiliation with
the view that ‘“real” men do not use grooming products’ (164). The combination, thus, of
Benwell’s reading of the texts, with readers’ views in relation to the texts tells us more about
the texts themselves, as well as about how many men may read them.
A further way in which critical discourse studies could be enhanced is through a more
detailed linguistic analysis of its texts than sometimes occurs. Systemic functional linguis-
tics has been proposed as a tool for one way in which this could be done (Fairclough 2003,
Martin 2000). Corpus approaches have also been proposed as a way of increasing the quan-
titative dimension of critical discourse analyses (Mautner 2005a). Others have proposed
expanding critical discourse studies by drawing on work such as schema theory and work
in the area of language and cognition (McKenna 2004).
Threadgold (2003) proposes a greater bringing together of work in the area of cultural
studies with work in the area of critical discourse analysis, suggesting the issue of performa-
tivity (see Chapters 1 and 2 of this book) be given greater prominence in this work to give a
better explanation and understanding of what people ‘do’ in their use of spoken and written
discourse. Trautner (2005) did this in her examination of how exotic dancers do gender and
social class in their presentations of themselves to their clients. She found that gender and
social class are a central feature of the interactions that take place in dance clubs. They are
200 Discourse Analysis

reflected, she found, ‘in very concrete ways: in the appearance of dancers and other staff,
dancing and performance styles, and the interactions that take place between dancers and
customers’ (786). The notion of performativity, thus, provides an important way for think-
ing about language, identity, class, social memberships and, in turn, the critical analysis of
discourse.

9.7 Summary
This chapter has discussed key issues and principles in critical discourse analysis. It has
given examples of studies that have been carried out from this perspective, all of which
have aimed to uncover out-of-sight norms and values which underlie texts which are key to
understanding the roles that texts play in particular social, cultural and political contexts.
Suggestions have then been made for ways of doing critical discourse analysis. Criticisms of
critical discourse analysis have also been discussed and ways of responding to these criti-
cisms have been suggested.

9.8 Discussion questions

(1) To what extent do you think texts reflect hidden and ‘often out-of-sight’ values? Choose a text
which you think illustrates this and explain in what way you think this is done through the use
of the discourse.
(2) To what extent do you think that the way a text is ‘framed’ encourages a certain reading of it.
Choose a text which you think illustrates this. Discuss framing, foregrounding, backgrounding
and the presuppositions that underlie the way the text is presented to its audience.
(3) Choose a text which you feel encourages a certain reading from its use of illustrations, pictures,
layout and design, etc. How do you feel each of these resources aim to ‘position’ the reader in a
particular way?

9.9 Data analysis projects

(1) Choose a text which you feel would be useful to examine from a critical perspective. Analyse it
from the point of view of genre and framing. Link your analysis to a discussion of how you feel the
text aims to ‘position’ its readers. Read Huckin (1997) on critical discourse analysis to help with
this.
(2) Choose a text which you feel uses multimodal discourse such as layout, design and images to
communicate its message to its audience. Analyse your text highlighting the ways in which it
does this. Look at Van Leeuwen’s (2005) Introducing Social Semiotics for suggestions on how to
do this.
 Critical Discourse Analysis 201
9.10 Exercises
Exercise 1: Textual silences
Huckin (2002), in his article ‘Textual silence and the discourse of homelessness’ examines
newspaper reporting on homelessness in the United States. He defines textual silence as
‘the omission of some piece of information that is pertinent to the topic at hand’ (348).
One of the silences he discusses is manipulative silence, a strategy of deliberately concealing
relevant information from readers to the advantage of the writer. The writer, thus, decides
‘what to say and what not to say about the topic’ (356). In his study of 163 newspaper articles
and editorials on this topic he found the most common themes were causes of homelessness,
effects of homelessness, public responses to homelessness and demographics such as number
and types of homelessness. That is, these were the topics that were foregrounded (Huckin
1997) in the texts. Topics such as treatment of the causes of homelessness, for example, were
omitted, or generally ‘textually silent’.
Look at the following extracts from one of the editorials Huckin analyses in his article.
Which of the themes Huckin lists are represented on this text? What are some of the issues
that are not mentioned, but could have been featured in this text?

Sunday, 3 January 1999 – Final edition Editorial – page B6 (Editorial)

Off-Ramp Etiquette

YOU see them at the highway on- and off-ramps. These freeway destitutes, with their imploring eyes
and scribbled signs of woe, make passers-by uncomfortable. ‘Homeless single mom’, waits at one exit.
‘Need gas money’ and ‘Will work for food’ huddle in the rain and wind at other locations. Where to
stop the car, especially if you don’t plan to give?
Due to a combination of necessity, pluck and success, an increasing number of panhandlers place them-
selves prominently in your daily path: freeway exits, the grocery store, the post office. Cruel as it sounds,
don’t roll down the window, salve your conscience and offer a five-spot. Give the time or money to
recognized organizations catering to the homeless. . . .
Saying no to the guy on the off-ramp holding a hard-luck sign and leaning on a pair of crutches may
seem cold, mean-spirited. But good citizens can still give generously to a recognized shelter or food
bank. Give the panhandler a coupon to a fast-food restaurant. Buy the Real Change newspaper. Some
proceeds go to the sellers. That at least shows some initiative and dignity. Volunteer time at a food
bank or shelter.
Give gifts that feed, clothe and attempt to break the cycle of chronic homelessness. In the long run,
such acts of kindness are far more helpful and meaningful.

© 1999 The Seattle Times, All Rights Reserved. (2936473)

202 Discourse Analysis

Exercise 2: Migration and identity

Krzyzanowski and Wodak (2008), in a chapter titled ‘Multiple identities, migration and
belonging’ discuss ways in which belonging in relation to migration is constructed through
discourse. The data they analyse is based on focus group discussions held in eight European
countries (Austria, Cyprus, France, Germany, Italy, Sweden, Poland and the United
Kingdom). Look at the following extracts from their data. In what way is belonging (and not
belonging) constructed in their discourse? What is the point of difference between where
the speaker is living and their home country in each of these extracts?

(1) And since 7 years (.) I’ve been standing astride, one leg here and another there. We want to be
at home because we have nothing to live on, no work and here we are fine (1.0) but it is not our
home, I mean, not fully ours. Not because we are not fine here but because our roots are there (.)
in the Ukraine. We have friends there, a family. . . . (female speaker living in Poland)
(2) I know my religion keeps me apart from the English people because nearly every English person
was a protestant. (female speaker living in the United Kingdom)
(3) Having another culture is useful [. . .] it is important to know where one is from [. . .] many don’t
know. I like to talk about it, because if you don’t know your origins and your roots, it has no
sense. (female speaker living in Italy)
(4) Because you see other people with other ideas different from you (female speaker living in
Italy)

Exercise 3: Gender, identity and online chat rooms

In their paper ‘Constructing sexuality and identity in an online teen chat room’
Subrahmanyam, Greenfield and Tynes (2004) discuss ways in which gender roles and
expectations can become blurred, and indeed resisted, in online chat rooms. The following
data is an example of what they term ‘cyber-pickup’, where one of the participants identifies
a potential partner with whom they will ‘pair off’. They will then go off with that person
into a private instant message space.
In what ways do these extracts show resistance of traditional gendered expectations?
(The numbers refer to lines from the data set. FoxyR and Breethebrat are teenage girls.
DEREKH101 is a teenage boy).

210. FoxyR: wasssssssssup yallllllllllll

212. FoxyR: anybody here like 50 cent press 1234
214. FoxyR: 1234
221. FoxyR: wassssssssssup
222. FoxyR: wanna chat
262. FoxyR: any body wanna chat with a hot chick
264. FoxyR: press 1234
265. Breethebrat: if there r any m/13/Tx in here if so im me
266. DEREKH01: 1234
304. FoxyR: any body wa nna chat
 Critical Discourse Analysis 203
9.11 Directions for further reading
Bloor, M. and Bloor, T. (2007), The Practice of Critical Discourse Analysis. London: Hodder Arnold.
Bloor and Bloor’s book contains many useful exercises and activities. Sample texts are drawn from history, adver-
tising, literature, newspapers and television. Specialized terminology is explained and there is a glossary and
grammar appendix which outlines systemic functional terms used in the text.
Flowerdew, J. (2008), ‘Critical discourse analysis and strategies for resistance’, in V. K. Bhatia, J. Flowerdew and
R. Jones (eds), Advances in Discourse Analysis. London: Routledge, pp. 195–210.
This chapter gives details of linguistic features that critical discourse analysis might examine as part of its overall
project. Future directions of critical discourse analysis are also discussed.
Rogers, R. (ed.) (2011), An Introduction to Critical Discourse Analysis in Education (2nd edn). London: Routledge.
This second edition of Rogers’ book contains a wide range of chapters that are useful for understanding what
critical discourse analysis is and how to do it. The book explains key concepts and issues in critical discourse
analysis and provides many examples of critically oriented discourse studies.
van Leeuwen, T. (2008), Discourse and Practice: New Tools for Critical Discourse Analysis. Oxford: Oxford University
Press.
This book brings together van Leeuwen’s work in the area of critical discourse analysis, drawing on the work of
Foucault, Bernstein and Halliday. There are examples and sample analyses in each of the chapters.
Wodak, R. (2011), ‘Critical discourse analysis’, in K. Hyland and B. Paltridge (eds), Continuum Companion to Discourse
Analysis. London: Continuum, pp. 38–53.
Wodak’s chapter reviews current approaches and developments in critical discourse analysis. She also provides a
sample study as an example of how to do critical discourse analysis.

For an extended list of references and further readings see the companion website to this book.