Professional Documents
Culture Documents
Criticism of CDA
Criticism of CDA
London:
Bloomsbury Academic, 2013. 77–86. Bloomsbury Collections. Web. 1 Jul. 2021. <http://
dx.doi.org/10.5040/9781472545213.ch-006>.
Copyright © Stephanie Taylor 2013. You may share this work for non-commercial purposes only,
provided you give attribution to the copyright holder and the publisher.
6 Criticisms of discourse analysis
77
78 What is discourse analysis?
A second reason why some (but not all) analysts would agree with the
above statement is that words and language have a special status in society.
When people talk to each other their language use may incorporate ges-
tures (such as a nod of the head) and sounds which are not words (‘Mm’
or similar as an acknowledgement or agreement) but these are meaningful
within a communication system which is primarily verbal. There are special
cases in which an entire ‘conversation’ can be conducted without words,
but they are rare exceptions. Communication through language use is a
vital, rule-governed social practice in which people necessarily become
skilled in order to function as members of society. Talk and language are
therefore obvious forms of data for a social scientist who wants to study
society. In an analysis of such data, a researcher can draw on her or his
own expertise as a language user, as well as on previous findings about
the conventions of conversation. The importance of this argument can
be underlined by comparison with a research project in which the par-
ticipants are asked to draw a picture or take a photograph. They are being
asked to engage in an unfamiliar practice with few or no conventions. (Of
course this may not be quite the case if they are skilled artists or photogra-
phers, and the analyst shares those skills, but that would be an exceptional
situation and highly unusual in social research.) The consequence will be
that neither the participants nor the analyst will be able to ‘do’ things
with the media or interpret what is produced as either a communication
or a socially meaningful activity. Alternatively, the researcher may invite
them to talk about what they have produced, so that the analytic focus
becomes, again, language.
However, some discourse analysts would disagree strongly with the
statement that ‘Discourse is just about words’. Their argument, broadly, is
that there is no neat separation between the meanings in language and in
the social world more generally. To talk about people, things and the mate-
rial world is to bring them into language (and therefore ‘into culture’ in the
words of Edwards et al., 1995). Even if people and things are not specifically
named and described by a research participant, they inevitably function
within larger systems of meaning. Previous chapters have discussed exam-
ples of the connections between certain named or nameable identities
(a good learner, a healthy eater), the values and knowledge which makes
the identities available, the institutions through which the knowledge is
established and disseminated and so on. Some of this complex formation
may operate without being explicitly brought into language: indeed, one
Criticisms of discourse analysis 79
purpose of the research may be to identify and name what has previously
been invisible and unacknowledged. Words and language are inevitably
linked with the social world and its ongoing activities. In this view, to say
‘discourse analysis is just about words’ is to ignore the social meanings
which are attached to language and words yet not confined to them.
These issues arise partly because of the range of data and analytic
approaches encompassed within the category ‘discourse analysis’.
Researchers who are interested in the difficult experiences of people
who have been socially marginalized may feel that a focus on words and
language amounts to a trivialization of their participants’ difficulties. To
study details of wording, grammar and punctuation may seem inappropri-
ate, even offensive, and a turning away from vital, apparently non-verbal
aspects of their participants’ lives such as feelings, their poor health or
physical injuries, bad living conditions or incarceration. A further motiva-
tion here may be a wish to look at the ‘whole’. It may seem disrespectful to
investigate only small details of an account.
One counter-argument is that the apparently non-verbal aspects of
communication are generally connected to language, so language data
can provide appropriate evidence. Another counter-argument is that
all research involves the selective reduction of a complex social world to
data. For example, a census or survey reduces people and their situations
to categories and statistics, yet it can prompt help and improvement of
their welfare. Discourse analytic research has always had a strong connec-
tion with social issues. As the previous chapters have indicated, discourse
researchers have challenged prejudice, racism and other injustices. The
analysis of words and language, often at the level of fine detail, is their
starting point for doing this.
A final point to note here is the issue of emotions, more recently linked
to the study of affect. Some psychosocial researchers have suggested
that discourse analysis ‘leaves out’ emotion. The assumption here is that
the taking up or contesting of a subject position is a verbal or intellectual
action which is separate from feelings. This argument is possibly a conse-
quence of an artificial separation of language data from the continuum
of meanings, lived experience and so on with which it is implicated, as
outlined earlier. A similar separation appears in the common contrasting
of the rational with the emotional. A counter-argument would be that
the types of discursive resources discussed by many analysts are laden
with values and other affective associations so that the kind of identity
80 What is discourse analysis?
and a woman and decide that it is inevitably ‘about’ gender, Schegloff argued
that a close analysis on CA principles could show whether or not gender
was relevant, or salient, by revealing what the speakers themselves ‘orient
to’ or ‘make relevant’. This is also an argument against ‘reading off’ certain
behaviours as gendered. If a man speaks at the same time as a woman, is this
inevitably evidence of interruption and a performance of a gendered identity
of dominance? Might it not be an example of overlapping talk as a common
form of agreement in the talk of people who position themselves as equals?
Schegloff’s arguments are a useful challenge to the over-free interpreta-
tion of data. Given that any two speakers can be categorized in terms of
multiple differences (e.g., gender, age, sexuality, nationality, place of birth,
occupation, religion, body features such as height, education), how reason-
able is it to claim that their conversation is relevant evidence for a study
of any particular category? On the other hand, there seems no reason to
assume that people may not ‘orient to’ several different issues at once
simultaneously, so that talk may be ‘about’ gender as well as its ostensible
topic. Theorists of intersectionality (e.g., Phoenix and Pattynama, 2006)
argue that categorization is always complex; for example, the relevant iden-
tification will be as a raced, classed, gendered person of a certain age group.
Discourse research on prejudice and racism has shown how categories may
merge, so that ‘culture’ becomes synonymous with ‘race’. More subtly, the
conventional categories of research may be overtaken by new lived catego-
ries of identity and difference, such as the ‘postcodes’ currently invoked in
conflicts between some young people in London. A discourse analyst needs
to be alert to all of these possibilities, to be alive to the possible differences
between participants’ views and her own, and to pause repeatedly to inter-
rogate the assumptions and concerns she brings to the research.
These further issues can be followed up in the sources listed in the
recommended reading list at the end of the book. The more general point
to note is that yes, discourse analysis is interpretive as is other research,
qualitative or quantitative. Part of its theoretical foundation consists of
challenges to the notion of a simple, objective truth.
Summary
Chapter 6 has addressed some common criticisms of discourse analysis.