DISCOURSE, DISCOURSE ANALYSIS AND C.D.

2.1 Discourse in linguistics

In linguistics, discourse refers usually to the study of speech patterns and the
usage of language etc. To understand the speech patterns one need to be clear about
the term ‘discourse’ and ‘text’. Discourse, put simply, is structured collections of
meaningful texts (Parker, 1992). A text is a part of the process of discourse. It is the
product of any communication by writer/speaker. A text consists of cues for
interpretation processes and traces of production processes. As Fairclough (1989)
says this process includes in addition to the text the process of production, of which
the text is a product, and the process of interpretation, for which the text is a resource.
As a resource for the interpreter, the text consists of lexico-grammatical realisations
of three kinds of meaning relating to three basic language functions (the ideational,
interpersonal and textual functions of systemic linguistics). These lexico-grammatical
cues to ideational, interpersonal and textual meanings are interpreted with the help of
other resources beyond the text.

In using the term text, we refer not just to the written transcriptions but to “any
kind of symbolic expressions requiring a physical medium and permitting of
permanent storage” (Taylor & Van Every, 1993: 109). For a text to be generated, it
must be spoken, written, or depicted in some way. Only when such an activity
happens a text takes a shape, Taylor, Cooren, Giroux, & Robichaud, (1996: 7) say that
when such an activity happens text takes on material form and becomes accessible to
others. Therefore, talk is also a kind of text Fairclough(1995); van Dijk (1997a), and,
in fact, the texts that make up discourses may take a variety of forms, including
written documents, verbal reports, artwork, spoken words, pictures, symbols,
buildings, and other artifacts (e.g., Fairclough, 1995; Grant, Keenoy, & Oswick,
1998; Taylor et al., 1996; Wood & Kroger, 2000). Discourses cannot be studied
directly they can only be explored by examining the texts that constitute them
(Fairclough, 1992; Parker, 1992).

22
 A ‘text’ means the observable product of interaction i.e. a cultural object and
‘discourse’ means the process of interaction itself i.e. a cultural activity. The
distinction between ‘text’ and ‘discourse’ is an analytical one i.e. between the
observable materiality of a completed product and the ongoing process of human
activity Widdowson (1979); Brown and Yule (1983); Halliday (1985). Text is the
fabric in which discourse is manifested, whether spoken or written, whether produced
by one or more participants. Text refers to the observable product of interaction
(whether language production or interpretation). As already mentioned a text may be
either written or spoken. In the actual production and interpretation of a stretch of
language (a simple example being a conversation) the interactants have access to
historically prior texts. These are products of previous interaction, which make up the
interactional history and thus produce discourse. In reporting previously uttered
speech, for instance, a fragment of an earlier text is embedded in the current text.
Text, then, is a frozen observable substance, a concrete cultural object. This does not
mean that the text-product actually exists as marks on paper or impulses on magnetic
tape. It may only exist in the possibly mistaken memories of people; indeed, with the
texts of previous conversations this is usually the case. This interrelation of texts
produces discourse.

Parker’s defines discourse as “a system of statements which constructs an

object” (1992: 5). Discourse “‘rules in’ certain ways of talking about a topic, defining
an acceptable and intelligible way to talk, write or conduct oneself” and also “‘rules
out’, limits and restricts other ways of talking, of conducting ourselves in relation to
the topic or constructing knowledge about it” (Hall, 2001: 72). In other words,
discourses “do not just describe things; they do things” (Potter & Wetherell, 1987: 6)
through the way they make sense of the world for its inhabitants, giving it meanings
that generate particular experiences and practices (Fairclough, 1992; van Dijk,
1997b).

Discourse is produced and interpreted by specific people in specific

institutional and broader societal contexts. Language users may not be seen as unitary
subjects who produce intended meanings in discourse but also as decentered subjects
who are constituted by their actions. In presenting the decentred subject there is no

23
clear division between a person and her actions and she is not their inventor or sole
creator. In producing discourse, she is constrained by what has gone before. In
production and interpretation people draw upon a wide range of what Fairclough
(1989:24-25) refers to as “members resources”, or MR. These MR include “their
knowledge of language, representations of the natural and social worlds they inhabit,
values, beliefs, assumptions, and so on.” In this view, MR are both cognitive and
social, “The MR which people draw upon to produce and interpret texts are cognitive
in the sense that they are in people’s heads, but they are social in the sense that they
have social origins, they are socially generated, and their nature is dependent on the
social relations and struggles out of which they were generated - as well as being
socially transmitted and, in our society, unequally distributed.” Production and
interpretation, then, are cognitive processes related to social practices. Discourse has
been a very controversial term and different critiques have given various definitions
of discourse. It will be difficult to discuss all of them but I would briefly describe
Fairclough and Foucault as they are related to my study.

2.2 Foucauldian discourse

Production and interpretation are key aspects in this study also. This study
follows C.D.A. which has its root in Foucault and many others schools of thought.
Discourse as understood in C.D.A. is quite similar to Foucauldian discourse thus a
brief review of Foucauldian discourse follows to make discourse in C.D.A. clear.

Discourses in Foucault’s work are structures of possibility and constraint,

which impinge on, and indeed bring about social practices. A number of linguists use
this sense of discourse from social theory. Fairclough employs the term discourse in a
Foucauldian sense, to mean systematically organised sets of conventions forming
practices, in accounting for how language use is socially conditioned. A similar,
apparently unacknowledged conception underlies the work of Pecheux. A third
linguist, Gunther Kress, refers to the work of Foucault in characterising discourse as
sociocultural practice. Kress (1985) draws upon Foucault in describing the defining
and delimiting quality of discourse and mentions that discourses are systematically
organised sets of statements, which give expression to the meaning, and values of an

24
institution. Beyond that, they define, describe and delimit what it is possible to say
and not possible to say (and by extension what it is possible to do or not to do) with
respect to the area of concern of that institution, whether marginally or centrally. A
discourse provides a set of possible statements about a given area, topic, object,
process that is to be talked about. In that, it provides descriptions, rules, permissions
and prohibitions of social and individual actions. Discourses for Foucault are
historically constituted social constructions in the organisation and distribution of
knowledge. Knowledge does not arise out of things and reflect their essential truth; it
is not the essence of things in the world. Foucault argues that dominant members of
institutions maintain control through discourses by creating order, i.e. by being the
ones who make boundaries and categories.

Foucault produced historical analysis of discourse and power. He investigated

the exercise of social power in/through discourses, through the definition of objects
and social subjects themselves. His approach is anti-humanist. As post-structuralist
feminist, Weedon (1987:107), says “It is in the work of Michel Foucault that the
poststructuralist principles of the plurality and constant deferral of meaning and the
precarious, discursive structure of subjectivity have been integrated into a theory of
language and social power which pays detailed attention to the institutional effects of
discourse and its role in the constitution and government of individual subjects”.

In his work, Foucault attends to discourses of the social sciences, which he

argues have contributed substantially to what people are. That is, practices and
relations between people are brought into being as a result of the socially constructed
bodies of knowledge, which are the social sciences. He writes about how the
discourses of the social sciences have impinged physically on people, constructing
them as patients, legal subjects and sexual subjects and so on. In his archaeologies he
writes that the domains of knowledge, which form social subjects in taking human
beings as their subject, are not timeless but historical constructions. In The History of
Sexuality, for instance, his focus is the discursive constitution of sexual subjects in
the judicial system, in medical texts etc.. In The Birth of the Clinic, he examines
medical writing, exploring shifts in what it is to ‘do medicine’, what illness is, both
clinically and socially.

25
 In contrast with the analysis of discourse in linguistics, Foucault does not
analyse concrete text samples (Fairclough (1988)). However, in the Archaeology of
Knowledge (1972) he makes brief but interesting observations about the notion of a
concrete whole text. “The materiality of the book”, he says, is only one kind of unity,
and not the most significant; for example, a missal and an anthology of poems are
both books but the unity each derives from discourse is what constitutes them as
missal and anthology. The unity a single actual text has is weaker than the “discursive
unity of which it is the support”. The “discursive unity” is not homogeneous; to
illustrate this point he contrasts the relation between Balzac’s novels with the relation
between Joyce’s Ulysses and the Odyssey. There is more to a text than the concrete
book; it only exists in relation with other texts “The frontiers of a book are never
clear-cut it is caught up in a system of references to other books, other texts, other
sentences: it is a node within a network. ...It indicates itself, constructs itself, only on
the basis of a complex field of discourse.”

Foucault proposes that discourses consist of sets of statements whose unity is

based on objects, style, concepts and themes; but with an important proviso that their
unity is not fixed but constructed through dispersion and discontinuity. In this way,
Foucault avoids presenting a discourse as though it were a single continuous book,
naturally unfolding through time. For instance, in saying that a set of statements has
unity in terms of the objects/referents it describes we cannot assume a “well defined
field of objects” outlined in discourses. In discourses on madness, for example,
madness is not a single object; it is constructed in interplay of conditions enabling its
appearance as an object and subject of constant displacement. Objects are shaped in
social practices. Objects and statements cannot exist by themselves but rather exist in
an associated social field. A statement can only exist in connection with other
statements that it repeats, opposes, comments on etc. The associated field is made up
of all the formulations to which the statement refers (implicitly or not), either by
repeating them, modifying them, or adapting them, or by opposing them, or by
commenting on them. There can be no statement that in one way or another does not
reactualize others.

26
 Foucault then, proposes the examination of discourses as systematically
organised sets of statements by locating discontinuity and dispersion. He refers to a
particular system of dispersion between a group of statements and to a specific
regularity between object, types of statement, concepts and thematic choices as a
“discursive formation”.

2.3 Discourse formations

Following Foucault, discourses in his account are both historically constituted

and consisting of practices enacted by subjects. The substance of concrete discourse is
not spontaneously created, but pre-determined by “discourse formations”, which are
described as follows in Haroche C. (Henry and Pecheux (1971:102): “discourse
formations determine what can and must be said (articulated in the form of an
argument, a sermon, a pamphlet, a paper, a political statement etc.) from a given
position at a given conjuncture”.

This concept of “discourse formation” relates to Foucault’s “discursive

practice”. The discourse formation is the site of the constitution of meaning and also
the site of the constitution of the subject. He explains this with the mechanism of
interpellation; “individuals are “interpellated” as speaking-subjects (as subjects of
their discourse) by the discursive formations which represent “in language” the
ideological formations that correspond to them”.

2.4 Discourse as Social Practice (Fairclough)

A recent linguistic combination of the two senses of discourse as action and

convention is found in Norman Fairclough’s presentation of critical language study in
Language and Power (1989). He remarks on the “felicitous ambiguity” of the term
discourse, and the more general practice, to refer both to an actual enactment and to a
social convention governing actions: “the individual instance always implies social
conventions - any discourse or practice implies conventional types of discourse or
practice” Fairclough (1989:28).

He also plays on this ambiguity to underline the way social practices,

including discourse, are both enabling and constraining, providing the social

27
conventions within which it is possible to act. “The ambiguity also suggests social
preconditions for action on the part of individual persons: the individual is able to act
only in so far as there are social conventions to act within. Part of what is implied in
the notion of social practice is that people are enabled through being constrained: they
are able to act on condition that they act within the constraints of types of practice or
of discourse” Fairclough (1989:28).

The importance placed on the category subject in Fairclough’s model is

evident in the detailed consideration of interpersonal meaning and his emphasis on the
subject positioning of language-users. Social positions are set up in discourse-types
and subjects are constrained to act within them. These constraints are what make
action possible; subjects are enabled by being constrained. Fairclough points out
another felicitous ambiguity in the term subject. The subject is both an active agent
and passively shaped, Fairclough (1989:39) “In one sense of subject, one is referring
to someone who is under the jurisdiction of a political authority, and hence passive
and shaped: but the subject of a sentence, for instance, is usually the active one, the
doer, the one causally implicated in action.”

Fairclough splits the interpersonal dimension of meaning into two closely

related but distinct considerations, the social identities of interactants and their social
relationships. These enter into the process of discourse as in features with expressive
and relational value respectively. These two aspects of interpersonal meaning are
closely related but it is sometimes useful to distinguish between them. Subject
positions and relations between them are set up in discourse-types. A single
individual is placed in a wide range of subject positions. He/She is not an autonomous
entity who exists independently of these positions and social relations, but constituted
in the act of working within various discourse-types. From the beginning of her entry
into social life she is positioned within varied institutional and societal structures,
which bestow upon her specific social roles. As Fairclough (1989:108) says “The
social process of producing social subjects can be conceived of in terms of the
positioning of people progressively over a period of years indeed a lifetime in a range
of subject positions. The social subject is thus constituted as a particular configuration

28
of subject positions. A consequence is that the subject is far less coherent and unitary
than one tends to assume.”

Fairclough aims to account for how people are constituted in social struggle
through discourse, i.e. he is ultimately interested in social subjects; discourse is the focus
of attention because that is what subjects are constituted in. Like Foucault, he attends to
how the practices of the social sciences have shaped, and continue to shape, the
institutional discourses forming subjects. His contribution is in constructing a model of
discourse as social practice allowing detailed linguistic analysis of the interaction of
individuals as realisations of these subject-shaping practices.

A social order according to Fairclough is “a structuring of a particular social

space into various domains associated with various types of practice.” From a
discoursal perspective, a social order is an order of discourse. It is here we can see the
Foucauldian sense of discourses as conventional, as systematically organised sets of
conventions forming practices. Just as a social order determines types of practice,
(conventions) which in turn determine actual practices (actions); orders of discourse
determine the types of discourse that establish actual discourses.

Discourses, then, are structured in orders of discourse. This emphasises the

shaping of actual discourse production and interpretation by higher level structures
beyond the immediate situation of utterance and amounts to an expansion of the
notion of context to include the social formation. This three level structuring relating
orders of discourse, types of discourse and the actual activity of discourse also
enables Fairclough to account for the presence of more than one discourse-type in
discourse. He emphasises that discourses are not simply mechanical implementations
of discourse-types but the creative extension-through-combination of existing
resources. Indeed, actual discourses, which draw on a single discourse-type, are
“limiting cases rather than the norm” Fairclough (1989:31).

2.5 Resources for production and interpretation

Discourse is determined by the social conditions in which it is produced and

interpreted. Institutional and societal structures always impinge upon discourse,

29
bestowing specific social identities and power relations upon interactants and giving
them different access to language, to representations of knowledge/beliefs, etc. (MR).

In discourse interpretation, features of text and context serve as cues which

activate specific MR. Interpretation is achieved in the dialectical interplay of cues and
MR. It is a complex of different processes in which MR serve as interpretive
procedures for both the language-user and the analyst. Fairclough identifies six major
elements of MR functioning as interpretive procedures which relate to features of text
and context. He distinguishes between MR for interpreting the situational context, for
which subjects need knowledge of social orders, and MR for interpreting the
intertextual context, for which knowledge of interactional history is needed.

The situational context provides external cues which have to be interpreted on

the basis of MR: (Fairclough 1989:114) “Participants arrive at interpretations of
situational context partly on the basis of external cues - features of the physical
situation, properties of participants, what has previously been said: but also partly on
the basis of aspects of their MR in terms of which they interpret these cues -
specifically, representations of societal and institutional social orders which allow
them to ascribe the situations they are actually in to particular situation types.”

Participants interpretations of what the situation is determine the discourse-

types drawn’ upon. These in turn determine the kinds of procedure drawn upon for
interpretation of the text. Fairclough (1989:145) distinguishes between situational and
intertextual context. In addition to “what has previously been said” in the same stretch
of discourse, interactants also need to refer explicitly or implicitly to previous
discourses: “we also need to refer to intertextual context: participants in any discourse
operate on the basis of assumptions about the previous (series of) discourses to which
the current one is connected to, and their assumptions determine what can be taken as
given in the sense of part of common experience, what can be alluded to, disagreed
with, and so on.”

The four remaining elements of MR functioning as interpretive procedures

relate to the text. Conventions of phonology, grammar and lexis are resources which
provide procedures for interpreting the surface of utterance. Other kinds of resource

30
are semantics, pragmatics and cohesion, which provide procedures for interpreting the
meaning of utterance and its local coherence. Other resources are schemata, which
provide procedures for interpreting a text’s structure and ‘point’: its global coherence.

These six domains of interpretation are interdependent. At any given point,

among the resources for interpretation are previous interpretations; for example,
interpretation of ‘higher level’, global elements of text are dependent on interpretation
of local elements “ for instance, to interpret the global coherence and ‘point’ of a text,
you draw upon interpretations of the local coherence of parts of it; and to arrive at
these, you draw upon interpretations of utterance meanings; and to arrive at these, you
draw upon interpretations of the surface forms of utterances.”

The interdependency of interpretation in different domains is not one

directional: it is ‘top-down’ as well as ‘bottom-up’ (as in the example above):
(Fairclough 1989:145) For instance, interpreters make guesses early in the process of
interpreting a text about its textual structure and ‘point’, and these guesses are likely
to influence the meanings that are attached to individual utterances, and the local
coherence relations set up between them”. This interdependency leads to what is
called the intertextual context.

2.6 Intertextual context

Interpretation of the intertextual context involves the historical series to which

a text belongs. Here the interpreter draws upon MR relating to interactional history.
This means that common presuppositions are created for readers so that come to
common interpretations. Fairclough’s (1989:152) mentions that “the interpretation of
intertextual context is a matter of deciding which series a text belongs to, and
therefore what can be taken as common ground for participants, or presupposed. As in
the case of situational context, discourse participants may arrive at roughly the same
interpretation or different ones, and the interpretation of the more powerful
participant may be imposed upon others. So having power may mean being able to
determine presuppositions.”

Producers of mass media texts are placed in a position of power in having to

construct an ‘ideal reader’. Through presupposition, they are able to present specific

31
intertextual experiences as common ground, thus postulating an ‘audience with shared
moments in interactional histories which are taken as given rather than asserted. As
well as presupposing elements of the intertextual context, producers can contest them.
By negating assertions in a text, a producer can assume that these assertions “are to be
found in antecedent texts which are within readers’ experience.”(Fairclough
1989:155).

Intertextual context adds a historical dimension to “discourse as social

practice by placing a single text, a product of discourse, in a historical series with
other texts: “The concept of intertextual context requires us to view discourses and
texts from a historical perspective, in contrast with the more usual position in
language studies which would regard a text as analysable without reference to other
texts, in abstraction from its historical context.”

2.7 Tendencies in discourse change

Fairclough points to certain tendencies in discourse in contemporary society,

which I think, are of particular importance for women. There is a high degree of
integration among social institutions in modern capitalist society and these
institutions are responsible in legitimising certain kinds of social identity. This is
significant because particular kinds of ‘social identity and relationship are likely to
appear commonsensical and natural when they span across institutional orders of
discourse; i.e. when people are placed in them in all kinds of diverse institutions.
Some general tendencies are an increasing sophistication of discourse technologies,
the use of synthetic personalisation in addressing mass audiences and an increasing
tendency for the formation of subjects as consumers. I intend to draw upon these
observations in my sample analysis.

In examining how discourse practices shape subjects, Fairclough points to an

increasing tendency for subject formation through strategic, manipulative discourses.
This tendency is marked by a sophistication of subject-shaping practices, informed by
discourses of the social sciences: the development of discourse technologies.
Discourse technologies are discourse-types which span across institutional orders of
discourse, ‘colonising’ new areas (key examples being interviewing and counselling).

32
These scrutinising discourses construct the object of their scrutiny, bringing into
being the social subjects defined by the expertise of the human sciences, such as
psychology. A characteristic of these forms of discourse is to present the interests of
the dominant bloc as the interests of the population as a whole, so that existing social
conditions are legitimised. One such manipulative kind of discourse, which is
spreading, is advertising. As Fairclough says advertising firmly embeds the mass of
the population within the capitalist commodity system by assigning them the
legitimate and even desirable role of consumers.

Social subjects are positioned as consumers in an increasing variety of social

situations as commodification expands into new areas of social life. Fairclough
mentions that the capitalist economic domain has been progressively enlarged to take
in aspects of life which were previously seen as quite separate from production. The
commodity has expanded from being a tangible ‘good’ to include all sorts of
intangibles like educational courses, holidays, health insurance, and funerals are now
bought and sold in the open market in ‘packages’, rather like soap powders.

An increasingly common feature of types of discourse used to address mass

audiences is synthetic personalisation that is a compensatory tendency to give the
impression of treating each of the people addressed in mass as an individual. This
synthetic personalisation is extremely common in the mass media, especially in
magazine advertisements and articles, leaflets, front page headlines, etc. It involves
the construction of an ideal subject as if it were an actual individual and also the
construction of a persona or ideal subject for the producers.

This construction of a social subject and study of social practices or discourse to

unveil the hidden ideologies or motives behind any discourse is called discourse analysis.
What follows is a brief description of discourse analysis.

2.8 Discourse Analysis

The term discourse analysis has come to be used with a wide range of
meanings which cover a wide range of activities. It is used to describe activities at the
interaction of disciplines as diverse as sociolinguistics, psycholinguistics,

33
philosophical linguistics and computational linguistics. So there are many versions of
discourse analysis (van Dijk 1997).

One major division is between approaches which include detailed analysis of texts
and approaches which do not. Fairclough (1992b) used the term ‘textually oriented
discourse analysis’ to distinguish the former from the latter. Discourse analysis in social
sciences is often strongly influenced by the work of Foucault (Foucault 1972, Fairclough
1992b). Social scientists working in this tradition generally pay less attention to the
linguistic features of texts. Fairclough’s approach (2003) to discourse analysis (a version
of critical discourse analysis) is based upon the assumption that language is an irreducible
part of social life, dialectically interconnected with other elements of social life, so that
social analysis and research always has to take account of language. This means that one
productive way of doing social research is through a focus on language, using some form
of discourse analysis. His approach to discourse analysis has been to transcend the
division between work inspired by social theory which tends not to analyse texts and
work which focuses upon the language of texts but tends not to engage with social
theoretical issues. So, text analysis is an essential part of discourse analysis, but discourse
analysis is not merely the linguistic analysis of texts. Fairclough (2003: 2) sees discourse
analysis as “oscillating between a focus on specific texts and a focus on the order of
discourse, the relatively durable structuring of language which is itself one element of the
relatively durable structuring and networking of social practices”. Hoowever there are
different views of discourse analysis by different linguists. The focus of discourse
analysis, as Jaworski and Coupland (1999: 7) argue, will usually be “the study of
particular texts” (e.g. conversations, interviews, speeches, etc. or various written
documents), although discourses are sometimes held to be abstract value system which
will never surface directly as texts.

van Dijk (1985b: 2) argues that “what we can do with discourse analysis is
more than providing adequate descriptions of text and context. That is, we expect
more from discourse analysis as the study of real language use, by real speakers in
real situations, than we expect from the study of abstract syntax or formal semantics.
Together with psycho- and sociolinguistics, discourse analysis has definitely brought
linguistics to the realm of the social sciences”.

34
 Taking a primarily linguistic approach to the analysis of discourse, Brown and
Yule (1983) examine how humans use language to communicate and, in particular,
how addressers construct linguistic messages for addressees and how addressees work
on linguistic messages in order to interpret them. They (1983: 1) suggest, “the
analysis of discourse is, necessarily, the analysis of language in use. As such, it
cannot be restricted to the description of linguistic forms independent of the purposes
or functions which those forms are designed to serve in human affairs”.

Stubbs (1983: 1) uses the term discourse analysis to refer mainly to the
linguistic analysis of naturally occurring connected spoken or written discourse:
“Roughly speaking, it refers to attempts to study the organization of language above
the sentence or above the clause and therefore, to study larger linguistic units, such as
conversational exchanges or written texts. It follows that discourse analysis is also
concerned with language in use in social contexts and in particular with interaction or
dialogue between speakers”. Hatch (1992: 1) defines discourse analysis as “the study
of the language of communication spoken or written”.

For Gee (1999: 92) a discourse analysis essentially involves asking questions
about how language, at a given time and place, is used to construe the aspects of the
situation network as realized at that time and place and how the aspects of the
situation network simultaneously give meaning to that language. A discourse analysis
involves, then, asking questions about the six building tasks. The tasks through which
one uses language to construct and/or construe the situation network, at a given time
and place, in a certain way, are:

1. Semiotic building, that is, using cues or clues to assemble situated meanings
about what semiotic (communicative) systems, systems of knowledge and
ways of knowing, are here and now relevant and activated.
2. Word building, that is, using cues or clues to assemble situated meanings
about what is here and now (taken as) ‘reality’, what is here and now (taken
as) present and absent, concrete and abstract, ‘real’ and ‘unreal’, probable,
possible and impossible.
3. Activity building, that is, using cues or clues to assemble situated meanings about
what activity or activities are going on, composed of what specific actions.

35
4. Socioculturally-situated identity and relationship building that is, using cues or
clues to assemble situated meanings about what identities and relationships are
relevant to the interaction, with their concomitant attitudes, values, ways of
feelings, ways of knowing and believing, as well as ways of acting and
interacting.
5. Political building, that is, using the cues or clues to construct the nature and
relevance of various ‘social goods’ such as status and power and anything else
taken as a ‘social good’ here and now (e.g. beauty, humor, verbalness,
specialist knowledge, etc.).
6. Connection building that is using the cues or clues to make assumptions about
how the past and future of an interaction, verbally and non-verbally, are
connected to the present moment and to each other after all, interactions
always have some degree of continuous coherence Gee (1999: 85-6).

van Dijk (1985b:1) argues that discourse analysis is essentially a contribution

to the study of language in use: “Besides - or even instead - of an explication of the
abstract structures of texts or conversations, we witness a concerted interest for the
cognitive and especially the social processes, strategies and contextualization of
discourse taken as a mode of interaction in highly complex socio-cultural situations”.

These different views show that discourse analysis has now emerged as a
diverse area of study, with a variety of approaches in each of a number of disciplines
and scholars working in different disciplines that tend to concentrate on different
aspects of discourse.

2.9 A Historical Overview

A brief historical overview to the study of discourse analysis shows that it

grew out of work in different disciplines in the 1960s and early 1970s, including
linguistics semiotics, psychology, anthropology and sociology. Discourse analysts
study language in use: written texts of all kinds and spoken data, from conversation to
highly institutionalized forms of talk.

At a time when linguistics was largely concerned with the analysis of single
sentences, Zellig Harris published a paper with the title “Discourse analysis” (Harris

36
1952). Harris was interested in the distribution of linguistic elements in extended
texts and the links between the text and its social situation, though his paper is a far
cry from the discourse analysis which is used nowadays. Also important in the early
years was the emergence of semiotics and the French structuralist approach to the
study of narrative. In the 1960’s Dell Hymes provided a sociological perspective with
the study of speech in its social setting (e.g. Hymes 1964). The linguistic philosophers
such as Austin (1962), Searle (1969) and Grice (1975) were also influential in the
study of language as social action, reflected in speech-act theory and the formulation
of conversational maxims, alongside the emergence of pragmatics, which is the study
of meaning in context ( Levinson 1983, Leech 1983).

British discourse analysis has been greatly influenced by M.A.K. Halliday’s

functional approach to language (e.g. Halliday 1973), which in turn has connections
with the Prague School of linguists. Halliday’s framework emphasises the social
functions of language and the thematic and informational structure of speech and
writing. Also important in Britain were Sinclair and Coulthard (1975) at the
University of Brimingham, who developed a model for the description of teacher-
pupil talk, based on a hierarchy of discourse units. There exists that similar work
deals with doctor-patient interaction, service encounters, interviews, debates and
business negotiations, as well as monologues. The British work has principally
followed structural-linguistic criteria, on the basis of the isolation of units and sets of
rules defining well-formed sequences of discourse.

American discourse analysis has been dominated by work within the

ethnomethodological tradition, which emphasises the research method of close
observation of groups of people communicating in natural settings. It examines the
types of speech events such as storytelling, greeting rituals and verbal duels in
different cultural and social settings (e.g. Gumperz and Hymes 1972). What is often
called conversation analysis within the American tradition can also be included under
the general heading of discourse analysis.

Alongside the conversation analysts, working within the sociolinguistic tradition

Labov’s investigations of oral storytelling have also contributed to a long history of

37
interest in narrative discourse. The American work has produced a large number of
descriptions of discourse types, as well as insights into the social constraints of politeness
and face-preserving phenomena in talk, overlapping with British work in pragmatics
(McCarthy 1991: 5-6).

Michel Pecheux at the Laboratoire de Psychologie Sociale, University of Pans MI

(1969, 1975, has developed a heavily theorized account of discourse in France Pecheux et
al. 1979, Robin 1973) as a tool for ideological struggle. He states as his objective the
need to provide “the basis for a scientific analysis of discursive processes by articulating
through historical materialism the study of ideological superstructures, psychoanalytical
theory and linguistic research” (Pecheux 1975: 234). As part of this design and drawing
on Althusser’s work on the theory of ideology, he has reformulated the Saussurian
dichotomy ‘langue-parole’ as ‘langue/ processus discursifs’.

This shift foreshadowed in the work of Volosinov and Bakhtin (Bennett 1979: 75-
82) and their critique of Saussure (see also Guespin in Gardin, Baggioni and Guespin
1980), takes into account the distinct systems of linguistic value that exist in a single
language community, in langue (Haroche, Henry and Pecheux 1971). In other words, it
focuses on the different meanings that words and expressions (signifiers) can have
according to the ideological position of the users and the determining effects of the socio-
historical conditions (or ‘ideological formations’) in which the utterances are produced
that are themselves constitutive of meaning. Discursive processes are thus seen as part of
an ideological class relation Pecheux( 1975: 82), Pecheux et al. (1979: 23-24), Seidel
(1985: 46-7).

Also relevant to the development of discourse analysis as a whole is the work of

text grammarians, working mostly with written language. Text grammarians perceive
texts as language elements strung together in relationships with one another that can be
defined. Linguists such as van Dijk (1972), De Beaugrande (1980), Halliday and Hasan
(1976) have made a significant contribution in this area. The Prague School of linguists,
with their interest in the structuring of information in discourse, has also been influential.
Its most important contribution has been to show the links between grammar and
discourse.

38
 Discourse analysis has grown into a wide-ranging and heterogeneous discipline,
which finds its unity in the description of language above the sentence and an interest in
the contexts and cultural influences, which affect language in use. It is also now,
increasingly, forming a backdrop to research in applied linguistics and second language
learning and teaching in particular (McCarthy 1991: 6-7).

2.10 Approaches to Discourse Analysis

Discourse analysis has become increasingly popular in recent years, across a

range of disciplines. This must be seen against the background of what is often referred to
as the linguistic turn in philosophy and the social sciences, which has shaped much
twentieth century thought. Diverse developments can be included under this heading: the
emergence of symbolic logic in the work of Frege, Russell and Whitchcad, of
structuralism in linguistics and elsewhere, of philosophical hermeneutics and of the so-
called linguistic philosophy of Wittgenstein, Austin, Ryle and others.

Of course, discourse analysis is itself a term that covers a multitude of different

approaches. Coulthard (1985) identifies four versions: anthropological work on the
ethnography of speaking; speech act theory; ethnomethodological conversation analysis;
and the systemic linguistics-based approach associated with John Sinclair and himself.
This by no means exhausts the field: Levinson (1983) and Brown and Yule (1983) adopt
more pragmatics-based approaches; conversational analysis has been extended beyond its
original concerns by researchers working in social studies of science and other areas of
sociology and social psychology (see for example Potter and Wethereil 1987) and there
are quite distinct forms of discourse analysis associated with French structuralism and
post-structuralism (Macdonell 1986).

These different kinds of discourse analysis vary in several important ways: in

their focus, in what sorts of claim they make and in the kinds of technique they deploy.
At one extreme there are approaches focusing on language above the level of the
sentence, which rule ‘non-linguistic’ action out of account and rely on some established
form of linguistic analysis as a model. At the other end of the spectrum,
ethnomethodologists, structuralists and others see language as constituting social reality,
although in different ways. For them, the study of discourse is a way of studying society
and the analytic techniques they use reflect this.

39
 More recently, a new kind of discourse analysis has been announced: critical
discourse analysis. Few advocators of C.D.A. are Teun van Dijk, Norman Fairclough,
Ruthwodak, (van Dijk 1983a,b, 1985b, 1998c, Fairclough 1989, 1992b, 1995a, 2003.
Kress 1990, Luke 1995/6, Caldas-Coulthard and Coulthard 1996). The distinctiveness of
C.D.A. lies not so much in the analytic techniques it employs but lies in its attempt to
locate discourse within a particular conception of society and its ‘critical’ attitude
towards society Hemmersley (1997: 237).

Eggins and Slade (1997: 24) classified the different approaches to discourse
according to their disciplinary origins:

Sociology Conversational analysis

Ethnography

Sociolinguistics Interactional sociolinguistics

Variation theory

Speech act theory

Philosophy
Pragmatics

Birmangham School
Structural-functional
Systemic functional linguistics
Linguistics
Systemic functional linguistics
Social semiotic
Critical discourse analysis

Artificial intelligence

Approaches to discourse analysis according to disciplinary origins (adopted from

McCarthy et al. 2002: 60).

40
 Although each of the approaches listed above has made a significant
contribution to the understanding of discourse, however the present study will review
only those that are currently playing a major role in various contexts of applied
linguistics.

2.10.1 Speech Act Theory

Speech act theory focuses on the fact that by saying something one is also
doing something. Discourse analysts working in this tradition have elaborated
complex typologies of different sorts of speech act and have tried to explain different
aspects of communication, such as psychiatric interviews, by trying to identify the
intended meanings of a speaker’s utterance and the responses of hearers, Howarth
(2002: 6-7).

John Austin and John Searle, developed speech act theory from the basic belief
that language is used to perform actions and thus, its fundamental insights focus on
how meaning and action are related to language. Although speech act theory was not
first developed as a means of analyzing discourse, some of its basic insights have
been used by many scholars to help solve problems basic to discourse analysis
Schiffrin (1994: 49).An elaboration of speech act theory was offered by Labov and
Fanshel (1977) in their examination of psychiatric interview. Although their prime
concern was with the identification of speech acts and specifying the rules governing
their successful realization, they broadened the view that an utterance may only
perform one type of speech act at a time.

Labov and Fanshel (1977) explain communication in terms of hearers

accurately identifying the intended meaning of the speaker’s utterance and responding
to it accordingly. However, given the multifunctionality of utterances, one cannot be
sure that a hearer always picks up the right interpretation of an utterance, i.e., the one
that was intended by the speaker. In general, the problem of internationality and
variability in people’s discourse rules precluded developing a coherent framework for
explaining communication, beyond producing an inventory of such rules and speech
act types, Jaworski and Coupland (1999: 16).

41
 Much of the speech act theory has been concerned with taxonomising speech
acts and defining felicity conditions for different types of speech acts. For example
Searle (1969, 1979) suggested the following typology of speech acts based on
different types of conditions which need to be fulfilled for an act to obtain:
representative (e.g., asserting), directives (e.g., requesting), commissives (e.g.,
promising), expressives (e.g., thanking) and declarations (e.g., appointing)’. This
taxonomy was one of many and it soon became clear in speech act theory that a full
and detailed classification would be unwieldy given the multitude of illocutionary
verbs in English. Stipulating the felicity conditions for all of them appeared to be
not only a complex procedure but also an essentialising one - relying too heavily on
factors assumed to be essential in each case, when reality shows us that they are
variably determined by the precise social context, Jaworski and Coupland
(1999: 16).

Speech act theory is basically concerned with what people do with language and
with the functions of language. Typically, however, the functions focused upon are those
akin to communicative intentions (the illocutionary force of an utterance) that can be
performed through a conventional procedure and labeled. Even within this relatively
well-defined set of acts, the act performed by a single utterance may not be easy to
discover: some utterances bear little surface resemblance to their underlying illocutionary
force.

Despite the emphasis on language function, speech act theory deals less with
actual utterances than with utterance-types and less with the ways speakers and hearers
actually build upon inferences in talk, than with the sort of knowledge that they can be
presumed to bring to talk. Language can do things - can perform acts - because people
share constitutive rules that create the acts and that allow them to label utterances as
particular kinds of acts. These rules are part of linguistic competence, even though they
draw upon knowledge about the world, including an array of “social facts” (e.g.
knowledge about social obligations, institutions, identities), as well as knowledge about
the grammar of language Schiffrin (1994: 60).

42
2.10.2 Pragmatics

Pragmatics deals primarily with the detailed study of meaning in language.

Closely related to semantics, which is primarily concerned with the study of word and
sentence meaning, pragmatics concerns itself with the meaning of utterances in
specific contexts of use. It deals with three concepts: meaning, context and
communication, Levinson (1983).

Morris (1938: 30) defines pragmatics as “the science of the relation of signs to
their interpreters”. Pragmatics is defined by him as a branch of semiotics, the study of
signs (see Givon (1989: 9-25), for discussion of its earlier roots). Morris (1938: 81)
views semiosis (the process in which something functions as a sign) as having four
parts. A sign vehicle is a sign: a designatum is that to which the sign refers: an
interpretant is the effect in virtue of which the sign vehicle is a sign; an interpreter is
the organism upon whom the sign has an effect. To put it the other way something is a
sign of a designatum for an interpreter to the degree that the interpreter takes account
of the designatum in virtue of the presence of the sign. Morris identifies three ways of
studying signs i.e. ‘syntax’ is the study of formal relations of signs to one another,
‘semantics’ is the study of how signs are related to the object to which they are
applicable, pragmatics is the study of the relation signs to interpreter. Thus,
pragmatics is the study of how interpreters engage in the taking-account-of designate
(the construction of interpretants) of sign-vehicles Schiffrin (1994: 191).

Grice’s ideas (1957) about the relationship between logic and conversation lead to
Gricean pragmatics. Gricean pragmatics provides a set of principles that constrains
speakers’ sequential choices in a text and allows hearers to recognize speakers’ intensions
by making it easy to relate what speakers say (in an utterance) to its text and contexts. It
provides a way to analyze the inference of speaker’s meaning: how hearers infer
intentions underlying a speaker’s utterance.

The approach that Gricean pragmatics offers to discourse analysis is based on a

set of general principles about rationally based communicative conduct that tells speakers
and hearers how to organize and use information offered in a text, along with background
knowledge of the world (including knowledge of the immediate social context), to

43
convey (and understand) more than what is said - in brief, to communicate. The operation
of these principles leads to a particular view of discourse structure in which sequential
dependencies - constrains imposed by one part of a discourse on what occurs next - arise
because of the impact of general communicative principles on the linguistic realization of
speaker meaning at different points in time. Thus, what Gricean pragmatics offers to
discourse analysis is a view of how participant assumptions about what comprises a co-
operative context for communication (a context that includes knowledge, text and
situation) contribute to meaning and how these assumptions help to create sequential
patterns in talk (see Schiffrin (1994: 191-227).

Thomas (1995) distinguishes three types of meaning:

• abstract meaning (the meaning of words and sentences in isolation, e.g., the
various meanings of the word grass);

• contextual or utterance meaning (e.g., when two intimate persons hold their
faces very near each other and one says ‘I hate you’ while smiling, the
utterance really means I love you’);

• utterance force (i.e., how the speaker intends his/her utterance to be

understood: e.g., when X says to Y are you hungry?, X may intend the
question as a request for Y to make X a sandwich) (Jaworski and Coupland
1999: 15).

Thomas (1995: 22) focuses on utterance meaning and force, which are central
to pragmatics, which she defines as “the study of meaning in interaction” with the
special emphasis on the interrelationship between the speaker, hearer, utterance and
context.

Spencer-Oatey and Zegarac (2002: 74-5) indicate that pragmatics is concerned

not with language as a system or product per se, but rather with the interrelationship
between language form (communicated) messages and language users. It explores
questions such as:

• How do people communicate more than what the words or phrases of their
utterances might mean and how do people make these interpretations?

44
• Why do people choose to-say and/or interpret something in one way rather
than another?

• How do people’s perceptions of contextual factors (for example, who the

interlocutors are, what their relationship is and what circumstances they are
communicating in) influence the process of producing and interpreting
language?

Pragmatics thus, questions the validity of the ‘code-model’ of communication

that was developed within the discipline of semiotics. According to this view,
communication is successful to the extent that the sender and the receiver pair signals
and messages in the same way, so that the message broadcast in the form of a given
signal is identical to the one received when that signal is decoded.

Modern approaches to pragmatics recognize that human communication

largely exploits a code (a natural language), but they also try to do justice for the fact
that human communicative behavior relies heavily on people’s capacity to engage in
reasoning about each other’s intentions, exploiting not only the evidence presented by
the signals in the language code, but also evidence from other sources, including
perception and general world knowledge.

2.10.3 Interactional Sociolinguistics

International sociolinguistics as an approach to discourse is inextricably linked

to the names of the sociologist Erving Goffman (e.g., 1959, 1967, 1981) and the
anthropological linguist John Gumperz (e.g., 1982a, 1982b). International
socioloinguistics gives an approach to discourse that focuses upon situated meaning.
What Gumperz contributes to this approach is a set of concepts and tools that provide
a framework within which to analyse the use of language during interpersonal
communication; Gumperz views language as “a socially and culturally constructed
symbol system that both reflects and creates macro-level social meaning and micro-
level interpersonal meanings. Speakers use language to provide continual indices of
who they are and what they want to communicate” Schiffrin (1994: 133).

45
 The work of Erving Goffman also focuses upon situated knowledge, the self
and social context in a way that complements Gumperz’s focus on situated inference:
Goffman provides a sociological framework for describing and understanding the
form and meaning of the social and interpersonal contexts that provide
presuppositions for the interpretation of meaning.As Goffman’s work shows, for
example, all interactive activity is socially organized at multiple levels: all utterances
are situated within contexts such as ‘occasions’, ‘situations’, or ‘encounters’ that not
only provide structure and meaning to what is said, but may themselves be organized
by what is said (e.g. Goffman 1963). What Gumperz stresses is the interpretive
importance of contexts, including, of course, the occasion in which an utterance is
produced Schiffrin (1994: 133-4).

Much of Gumperz’s research has concentrated on ‘intercultural interaction’ and

especially, on the mechanisms of ‘miscommunication’. For example, he demonstrates
how seemingly irrelevant signaling details, such as falling rather than rising intonation on
a single word, can trigger complex patterns of interpretation and misinterpretation
between members of different cultural groups (see also Roberts et al. 1992). These
patterns of (mis) interpretation, which he labels ‘conversational inferencing’ depend not
only on the ‘actual’ contents of talk, but to a great extent on the processes of perception
and evaluation of a number of the signaling mechanisms, based on details of intonation,
tempo of speech, rhythm, pausing, phonetic, lexical and syntactic choices, non-verbal
signals and so on. Gumperz calls such features contextualisation cues and he shows that
they relate to what is said the contextual knowledge that contributes to the
presuppositions necessary to the accurate inferencing of what is meant (including, but not
limited to, the illocutionary force) Schiffrin (1994: 99-100).

In sum, Schiffrin (1994: 134) states that, “interactional socioloinguistics views

discourse as a social interaction in which the emergent construction and negotiation of
meaning is facilitated by the use of language”. Although the interactional approach is
basically a functional approach to language, its focus on function is balanced in
important ways. The work of Goffman forces structural attention to the contexts in
which language is used: situations, occasions, encounters, participation frameworks
and so on, have forms and meanings that are partially created and/or sustained by

46
language. Similarly, language is patterned in ways that reflect those contexts of use.
Thus, language and context co-constitute one another: language contextualizes and is
contextualized, such that language does not just function in context, language also
forms and provides context. One particular context is social interaction. Language,
culture and society are grounded in interaction: they stand in a reflexive relationship
with the self, the other and the self-other relationship and it is out of these mutually,
constitutive relationships that discourse is created.

2.10.4 Ethnography of Communication

The ethnography of communication is an approach to discourse that is based

on anthropology and linguistics. It focus upon a wide range of communicative
behaviors and built into its theory and methodology is an intentional openness to the
discovery of the variety of forms and functions available for communication and so
the way such forms and functions are part of different ways of life. According to
Hymes (1974a: 20) the ethnography of communication is not an approach that can
“simply take separate results from linguistics, psychology, sociology, ethnology, as
given and seek to correlate them”. Rather, it is an approach that seeks to open new
analytical possibilities. It seeks to do so by analyzing patterns of communication as
part of cultural knowledge and behavior: this entails a recognition of both the
diversity of communicative possibilities and practices (i.e. cultural relativity) and the
fact that such practices are an integrated part of what people know and do as members
of a particular culture (i.e. a holistic view of human beliefs and actions) Schiffrin
(1994: 137).

In anthropological tradition of ethnography of communication, as Gumperz

(1982a: 154) states, “socio-cultural knowledge is seen as revealed in the performance
of speech events defined as sequences of acts bounded in real time and space and
characterized by culturally specific values and norms that constrain both the form and
the content of what is said”. The key figure responsible for the development of the
ethnography of communication is Dell Hymes. Hymes’s definition of the ethnography
of communication consists of four elements:

• whether and to what degree something is grammatical (linguistic competence);

47
• whether and to what degree something is appropriate (social appropriateness);
• whether and to what degree something is feasible (psycholinguistic
limitations);
• whether and to what degree something is done (observing actual language
use).

(Jaworski and Coupland 1999: 25-6)

This far broader conceptualization of language and indeed of the purpose of

language study imposes a radically different methodology from Chomsky’s
linguistics, which is based on introspection and intuition. The object of inquiry for
Hymes is no longer the structure of isolated sentences, but ‘rules of speaking’ within
a community. Consequently, the sentence is replaced as a basic unit of analysis with a
three fold classification of speech communication (Hymes, 1972) :

• speech situations, such as ceremonies, evenings out, sports events, bus trips
and so on; they are not purely communicative (i.e., not only governed by rules
of speaking) but provide a wider context for speaking.
• speech events are activities which are “par excellence” communicative and
governed by rules of speaking, e.g., conversations, lectures, political debates,
ritual insults and so on.
• speech acts are the smallest units of the set, e.g. orders, jokes, greetings,
compliments, etc.: a speech act may involve more than one move from only
one person, e.g., greetings usually involve a sequence of two ‘moves’.

Hymes argues that ethnographers can analyze communicative patterns using

the traditional method of anthropological research i.e. participant observation. By
participating in a wide range of activities common to the life of a particular group of
people, one attempts to replace one’s own way of thinking, believing and acting with
a framework in which what is done by the members of another group starts to seem
‘expected’ and ‘natural’. The challenge faced by an anthropologist is thus, in some
ways, similar so that faced by any neophyte’ an anthropologist has to learn what
native members already know about how to “make sense out of experience” Schiffrin
(1994: 140).

48
 Linguists ignored the study of communicative patterns and systems of
language use for reasons quite different from those of anthropologists. Chomsky’s
(1957, 1965) reformulation of the goals of linguistic theory excluded the analysis of
performance, focusing theoretical interest instead on competence, i.e. tacit knowledge
of the abstract rules of language. Rather than concentrating linguistic theory on
competence, Hymes proposed that scholarship focus on communicative competence.
Knowledge of abstract linguistic rules is included in communicative competence. But
also included is the ability to use language in concrete situations of everyday life: the
ability to engage in conversation, to shop in a store, to interview (and be interviewed)
for a job, to pray, joke, argue, tease, warn and even to know when to be silent.
Furthermore, the study of language in use i.e. the study of how people are
communicatively competent and this contributes “in an empirical and comparative
way (to) many notions that underlie linguistic theory proper” Hymes (1974a: 20), also
Hymes (1981), simply because “it is not easy to separate areas of language that are
insulated from cultural and social processes, from those that are vulnerable to such
processes” Ochs (1988: 3).

2.10.5 Variation Analysis

Variationist approach stems largely from studies of variation and change in

language i.e. fundamental assumptions of such studies arc that linguistic variation
(i.e. heterogeneity) is patterned both socially and linguistically and that such patterns
can be discovered only through systematic investigation of a speech community.
Thus, variationists try to discover patterns in the distribution of alternative ways of
saying the same thing, i.e. the social and linguistic factors that are responsible for
variation in ways of speaking (see McCarthy et al. 2002).

Both the initial methodology and the theory underlying such studies are those
of William Labov (who has also developed a speech act approach to discourse).
Although traditional variationist studies have been limited to semantically equivalent
variants (what Labov (1972a) calls “alternative ways of saying the same thing”), such
studies have also been extended to texts. It is in the search for text structure, the

49
analysis of text-level variants and of how text constrains other forms that a
variationist approach to discourse has developed.

Although the linguistics in which this approach is grounded is ‘socially

realistic’ (a term used by Hymes (1974c: 196), Labov resists the term sociolinguistics
“since it implies that there can be a successful linguistic theory or practice which is
not social” (1972c: xiii). Furthermore, although social factors (e.g. social
stratification) are considered in actual analyses and in more general formulations of
patterns and explanations of distributions, the influence of sociology (its assumptions,
concepts and theory) was not heavily incorporated into early studies of language
variation and change. Thus, the influence of linguistics pervades the variationist
approach to discourse.

Labov (1972d) and Labov and Fanshel (1977, Chap. 3), propose rules that
connect actions and meanings to words. This is a major task of discourse analysis and
should be pursued in formal terms: “Linguists should be able to contribute their skill
and practice in formalization to this study. ... Formalization is a fruitful procedure
even when it is wrong: it sharpens our questions and promotes the search for answers”
(Labov 1972d: 298). Variationist formalizations of discourse rules include social
information with linguistic primitives because discourse is an area “of linguistic
analysis in which even the first steps towards. Rules cannot be taken unless the social
context of the speech event is considered” (Labov 1972e: 252).

Labov (1972b, Labov and Waletsky 1967) provided a systematic framework for
the analysis of oral narrative - a framework that illustrates quite well the variationist
approach to discourse units. This framework, as indicated by Schiffrin (1994: 283),
defines a narrative as a particular bounded unit in discourse and it defines parts to
narrative as smaller units whose identities are based on their linguistic (syntactic,
semantic) properties and on their role in the narrative.

Labov argued that a fully formed narrative (as summarized by Mesthrie et al.
2000: 193) may include the following:

1. Abstract, which summarizes the events to come or offers a preliminary

assessment of the significance of those events;

50
2. Orientation, which identifies the setting, characters and other background and
contextual details relevant to narrative;
3. Complicating action, a series of narrative clauses - the basic details of the
storyline;
4. Evaluation(s), which indicate the point of the story, or the reason(s) why the
speaker thinks the story is worth (retelling. Such material may occur at the
end, but may also be included at any point within the narrative;
5. Result or resolution, which resolves the story;
6. Coda, which signals the end of the narrative and may bridge the gap between
the narrative and the present time.

Variationists require data that allow the discovery of the highly regular rules
of language and the social distribution of variants governed by those rules. This type
of data-a variety of language termed the vernacular - emerges only during certain
social situations with certain interactional conditions. One such condition is when a
speaker tells a narrative of personal experience. Thus, the same discourse unit that is
useful for variationists because of its regular textual structure and because it enables
the definition of environments in which to locate specific linguistic variants, is also
useful as a source of vernacular speech in which patterns of linguistic variation and
change maybe discovered (Schiffrin 1994:290).

A variationist approach to discourse is thus, a linguistic approach that

considers social context under certain methodological and analytical circumstances.

2.10.6 Conversation Analysis

Conversation analysis (CA) offers an approach to discourse that has been

extensively articulated by sociologists, beginning with Harold Garfinkel who
developed the approach known as ethnomethodology and then applied specifically to
conversation, most notably by Harvey Sacks, Emanuel Schegloff and Gail Jefferson
(Gumperz 1982a).

Ethnomethodology means studying the link between what social actors ‘do’ in
interaction and what they ‘know’ about interaction. Social structure is a form of order
and that order is partly achieved through talk, which is itself structured and orderly.

51
Social actors have common-sense knowledge about what it is they are doing
interactionally in performing specific activities and in jointly achieving
communicative coherence. Making this knowledge about ordinary, everyday affairs
explicit and finding an understanding of how society is organized and how it
functions, is ethnomethodology’s main concern (Garfinkel 1967, Turner 1974,
Heritage 1984).

Conversation analysis differs from other branches of sociology because rather

than analyzing social order it seeks to discover the methods by which members of a
society produce a sense of social order. Conversation is a source of much of our sense
of social order, e.g. it produces many of the typifications underlying our notions of
social role (Ciccourel 1972). Conversation also exhibits its own order and manifests
its own sense of structure.

Conversation analysis is like interactional sociolinguistics in its concern with the

problem of social order and how language both creates and is created by social context. It
is also similar to the ethnography of communication in its concern with human
knowledge and its belief that no detail of conversation (or interaction) can be neglected a
prior: as unimportant. All three approaches also focus on detailed analysis of particular
sequences of utterances that have actually occurred. But conversation analysis is also
quite different from any of the approaches discussed thus far: conversation analysis
provides its own assumptions, its own methodology (including its own terminology) and
its own way of theorizing (Schiffrin 1994: 232).

C.A views language as a form of social action and aims, in particular, to discover
and describe how the organization of social interaction makes manifest and reinforces the
structures of social organization and social institutions (Zimmerman 1991, Drew and
Heritage 1992, Schegloff 1999. Hutchby and Wooffit; 1998). Hutchby and Wooffitt
(1998: 14), point out that talk in interaction is now commonly preferred to the
designation ‘conversation’, define C.A as follows: “Conversation analysts were the first
to provide systematic evidence for the cooperative nature of conversational processes and
to give interactional substance to the claim that -to use Halliday’s expression - words
have both relational and ideational significance”. (Gumperz 1982a: 160). The emphasis

52
in C.A in contrast to earlier ethnomethodological concerns has shifted away from the
patterns of ‘knowing’ towards discovering the ‘structures of talk’ which produce and
reproduce patterns of social action.

One central C.A concept is ‘preference’, the idea that, at specific points in
conversation, certain types of utterances will be more favored than others (e.g. the
socially preferred response to an invitation is acceptance, not rejection). Other
conversational features which conversation analysis has focused on, as Jaworski and
Coupland (1999: 20) indicate, include:

• openings and closings of conversations (Schegloff and Sacks 1999).

• adjacency pairs (i.e. paired utterances of the type summons-answer, greeting-
greeting, compliment-compliment response, etc.);
• topic management and topic shift;
• conversational repairs;
• showing agreement and disagreement;
• introducing bad news and processes of troubles-telling;
• (probably most centrally) mechanisms of turn-taking.

Conversation analysts drawing largely on Garnfinkel’s (1967) sociological

method of ethnomethodology, which is “the study of the way in which individuals
experience their everyday activities, endeavors to deduce from observation that
speakers are doing and how they are doing it” (Trask 1999: 57).

In sum, conversation analysis approaches to discourse consider the way

participants in talk construct systematic solutions to recurrent organizational problems
of conversation. The existence of those problems and the need to find such solutions
arises out of the ethnomethodological search for members’ knowledge of their own
ordinary affairs, knowledge that reveals and produces a sense of order and normalcy
in everyday conduct. Since, the sense of order that emerges is publicly displayed
through ongoing activity, one can examine the details of that activity for evidence of
its underlying order and structure searching not just for evidence that some aspect of
conversation ‘can’ be viewed in a certain way, but that it is viewed that way by
participants themselves, Levinson (1983: 318-19).

53
2.11 Critical Discourse Analysis

A brief introduction about C.D.A. is must before I explain in detail the model
of C.D.A. that I am following in this study. The roots of C.D.A. lie in classical
rhetoric, text linguistics and sociolinguistics as well as applied linguistics and
pragmatics. The notions of power, ideology, hierarchy and gender together with
sociological variables are all seen as relevant for an interpretation and explanation of
text. Gender issues, issues of racism, media discourses, political discourses,
organizational discourses or dimensions of identity research have become very
prominent now. C.D.A. takes a particular interest in relation between language and
power. The term C.D.A. is now used to refer more specifically to the critical
linguistic approach of scholars who find the larger discursive unit of text to be the
basic unit of communication.

Critical Discourse Analysis (C.D.A.) stems from a critical theory of language

which sees the use of language as a form of social practice. All social practices are
tied to specific historical contexts and are the means by which existing social
relations are reproduced or contested and different interests are served. It is the
questions pertaining to interests: How is the text positioned or what is its positioning?
Whose interests are served by this positioning? Whose interests are negated? What
are the consequences of this positioning? All these relate discourse to relations of
power. C.D.A. seeks to understand how discourse is implicated in relations of power.
Critical discourse studies stem from three overlapping intellectual traditions, each
emphasizing the linguistic turn in the social sciences. These traditions are discourse
studies (e.g., Benveniste, 1958/1971; Derrida, 1974; Foucault, 1969/1972; Pecheux,
1975), feminist post-structuralism (e.g., Butler, 1990; Davies, 1993), and critical
linguistics (e.g., Fowler, Hodge, Kress, & Trew, 1979; Halliday & Hasan, 1989;
Hodge & Kress, 1979/1993; Pecheux, 1975; Pennycook, 2001; Willig, 1999). Critical
Discourse Analysis focuses on how language as a cultural tool mediates relationships
of power and privilege in social interactions, institutions, and bodies of knowledge
(see, for example, Bourdieu, 1977; Davies & Harre, 1990; Foucault, 1969/1972; Gee,
1999; Luke, 1995/1996).

54
 Gee (2004) makes the distinction between the capitalized term “Critical
Discourse Analysis” (which the abbreviation C.D.A. represents) and “critical
discourse analysis” in lowercase letters, a distinction that is quite relevant to this
review. He argues that C.D.A. refers to the brand of analysis that has been informed
by Fairclough, Hodge, Kress, Wodak, van Dijk, van Leeuwen, and followers.
Lowercase “critical discourse analysis” includes a “wider array of approaches” Gee’s
own form of analysis (1992, 1994, 1996, 1999), that of Gumperz (1982), Hymes
(1972), Michaels (1981), and Scollon, & Scollon (1981), and the work of other
discourse analysts in the United States and elsewhere. These scholars are conducting
critically oriented forms of discourse analysis but do not specifically call their work
C.D.A.. Gee (2004:33) points out that critical approaches to discourse analysis “treat
social practices in terms of their implications for things like status, solidarity,
distribution of social goods, and power”. As language is a social practice and because
not all social practices are created and treated equally, all analyses of language are
inherently critical.

C.D.A. however is not a single theory or methodology. Rather researchers of

C.D.A. rely on different theories. There are various definitions of discourse, in
relation to ‘text’, ideology and power and all this makes C.D.A. dynamic. It is an
interdisciplinary approach to language and social study that explores the social
interaction which is manifested in linguistic forms. C.D.A. views discourse as a form
of social practice, which is in turn effected by and effects the discursive practices.
Discursive practices are shaped and affected by social institutions and social
structures. Discursive practices as part of discourse affect ideologies that are further
responsible for power relations.

C.D.A. has its roots and tenets in various traditional theories. It is a type of
discourse analytical research that primarily studies the way social power; abuse,
dominance and inequality are enacted, reproduced and resisted by text and talk in the
social and political context, van Dijk (2003). It is one of the approaches within D.A
that blends a textual linguistic analysis and a detail social analysis. Thus, C.D.A.
focuses on social problems and societal issues. It explains the social and discursive

55
structures. C.D.A. follows a unique approach to social issues since; it endeavors to
make visible explicit power relations, which normally are backgrounded in the social
relations and does this by emphasizing specially on the context of the text. C.D.A.
refers to extra linguistic factors as culture, society and ideology. The notion of
context thus, holds a lot of significance. Context encompasses all social psychological
and cultural dimensions. The notion of context gives rise to the assumption of
relationship between language and society. This relationship between society and
language is viewed as dialectical which is a substantive point that makes C.D.A. so
very distinct and fruitful. Critical linguistics supports a similar viewpoint.

Critical linguistics was the approach developed by a group at the University of

East Anglia in the 1970s (Fowler et al. 1979, Kress and Hodge 1979). They tried to
marry a method of linguistic text analysis with a social theory of the functioning of
language in political and ideological processes, drawing upon the functionalist
linguistic theory associated with Michael Halliday (1978, 1985) and known as
systemic functional linguistics, Fairclough (1992b: 25-6).

The differences between critical discourse analysis (C.D.A.) and other

sociolinguistic approaches may be most clearly established with regard to the general
principles of C.D.A.. First of all the nature of the problems with which C.D.A. is
concerned are different in principle from all those methods which do not determine
their interest in advance. It is a fact that C.D.A. follows a different and a critical
approach to problems, since it endeavours to make explicit power relationships which
are frequently hidden and thereby helps deriving results that are of practical
relevance.

One important characteristic that arises from the assumption of critical discourse
analysis is that all discourses are historical and can therefore, only be understood with
reference to their context. In accordance with this, critical discourse analysis refers to
such extra linguistic factors as culture, society and ideology. In any case, the notion of
context is crucial for critical discourse analysis, since this explicitly includes social,
psychological political and ideological components and thereby postulates an

56
interdisciplinary procedure. Beyond this, critical discourse analysis, using the concepts of
intertextuality and interdiscursivity, analyses relationships with other texts whereas this is
not pursued in other methods. Wodak (2001) mention that from the basic understanding
of the notion of discourse it may be concluded that critical discourse analysis is open to
the broadest range of factors that exert an influence on texts.

From the notion of context a further difference emerges concerning the

assumption about the relationship between language and society. C.D.A. does not take
this relationship to be simply deterministic but invokes an idea of mediation. There is
a difference between various approaches to discourse. Norman Fairclough (1992b)
defines the relationship in accordance with Halliday’s multifunctional linguistic
theory and the concept of orders of discourse according to Foucault, while Ruth
Wodak (2001a,b) and Paul Chilton (2004), like Teun van Dijk, introduces a
sociocognitive level. This kind of mediation between language and society is absent
from many other linguistic approaches, such as for example, conversation analysis.

A further distinguishing feature of critical discourse analysis is the specific

incorporation of linguistic categories into its analyses. As Chouliaraki and Fairclough
(1999: 93) state “to capture some contradictory character of discourse on late
modernity, critical discourse analysis should be open in its analysis to different
theoretical discourses which construct the problem in focus in different ways”.

Another characteristic of critical discourse analysis is its interdisciplinary

claim and its description of the object of investigation from widely differing
perspectives, as well as its continuous feedback between analysis and data collection.
Compared with other linguistic methods of text analysis, critical discourse analysis
seems to be closest to sociological and socio-psychological perspectives, although
these interfaces are not well defined everywhere (Meyer 2001: 15-6).

In sum, as Scollon (200la: 140) argues “C.D.A. is a programme of social

analysis that critically analyses discourse that is to say language in use as a means of
addressing problems of social change. The programme of C.D.A. is founded in the

57
idea that the analysis of discourse opens a window on social problems because social
problems are largely constituted in discourse”.

In the next section I discuss some foundational principles that are relevant in
any discussion of Critical Discourse Analysis. The discussion is structured around the
key constructs: “critical,” “discourse,” and “analysis” and what is Critical in C.D.A.?

The Frankfurt school, the group of scholars connected to the Institute of Social
Research at the University of Frankfurt, focused their attention on the changing nature
of capitalism and its relation to Marxist theories of economic determinism. Adorno,
Marcuse, and Horkheimer, the scholars most commonly connected with the Frankfurt
School initiated a conversation with the German tradition of philosophical and social
thought of Marx, Kant, Hegel, and Weber. While rejecting the strict economic
determinism (the view that economic factors determine all other aspects of human
existence) associated with Marxism, they continued the view that injustice and
oppression shape the social world. The Frankfurt school and scholars from across
disciplines engaged with critical theory and attempted to locate the multiple ways in
which power and domination are achieved, Kinchloe & McLaren (2003).

Thus, the Frankfurt school and other neo-Marxist scholars of society and
language (e.g., the Bakhtin Circle) opened the debate about whether language belongs
to the economic base or the cultural superstructure, and whether it is determined by
material conditions or, in fact, determines these conditions Ives (2004). It is important
to remember that at the same time that the Frankfurt school was rising in academic
popularity, the works of W. E. B. DuBois (1903/1990) and Carter Woodson
(1933/1990) also mounted serious challenges to the dominant Euro- American
scholarly paradigm. However, Horkheimer, Adorno, and Marcuse are commonly
associated with critical theory, whereas DuBois and Woodson remain invisible in the
scholarly canon in critical theory (Ladson-Billings, 2003). This is important because
critical theory, a set of theories that attempt to locate and confront issues of power,
privilege, and hegemony, has also been critiqued for reproducing power knowledge
relations and constructing its own regime of truth.

58
 Or, as Yancy (1998) puts it, critical theory is often “the words of white men
engaged in conversations with themselves” . Evidence of this can be seen in the
striking absence of issues of race in much of critical theory. Critical theory is not a
unified set of perspectives. Rather, it includes critical race theory, post-
structuralism, post-modernism, neo-colonial studies, queer theory, and so on.
Critical theories are generally concerned with issues of power and justice and the
ways that the economy, race, class, gender, religion, education, and sexual
orientation construct, reproduce, or transform social systems. Although there are
many different “moments” when research might be considered critical, the various
approaches to critical research share some assumptions. Critical theorists, for
example, believe that thought is mediated by historically constituted power
relations. Facts are never neutral and are always embedded in contexts. Some
groups in society are privileged over others, and this privilege leads to differential
access to services, goods, and outcomes. Another shared assumption is that one of
the most powerful forms of oppression is internalized hegemony, which includes
both coercion and consent Gramsci (1973); Ives (2004). Critical researchers are
intent on discovering the specifics of domination through power. However, power
takes many forms and they are ideological, physical, linguistic, material,
psychological and cultural. Critical theorists generally agree that language is central
in the formation of subjectivities and subjugation.

Post-structuralism, the intellectual movement with which Michel Foucault is

often associated was a rejection of the structuralist movement of the earlier 20th
century and is intimately related to critical theory. Structuralism assumed that
relationships existed between structures in systems and that examining those
relationships could help us to understand the entirety of a system. The theory of
structuralism permeated across disciplines and could be seen in studies of the
economy (Marx), language (Saussure), psychology (Freud), and anthropology
specifically, culture and kinship relations (Levi-Strauss). Foucault, once himself a
structuralist, broke from structuralism and argued that we cannot know something
based on a system of binaries and static relationships. Post-structuralism pointed out

59
the inevitable slipperiness of social constructs and the language that constructed and
represented such constructs, Peters & Burbules (2004). Foucault’s (1969/1972)
concept of discourse and power has been important in the development of C.D.A., as
discussed in the next section.

Critical discourse analysis regards language as social practice, Fairclough and

Wodak (1997) and takes consideration of the context of language use to be crucial
Wodak (2000); Benke (2000). Fairclough and Wodak (1997: 271-80) have put forward
an eight-point programme to define critical discourse analysis as follows:

1. Critical discourse analysis addresses social problems.

2. Power relations are discursive.
3. Discourse constitutes society and culture.
4. Discourse does ideological work.
5. Discourse is historical.
6. The link between text and society is mediated.
7. Discourse analysis is interpretative and explanatory.
8. Discourse is a form of social action.

In Wodak’s (2001: 2) term, critical discourse analysis may be defined as

fundamentally concerned with analysing opaque as well as transparent structural
relationships of dominance, discrimination, power and control as manifested in
language. In other words, C.D.A. aims to investigate critically social inequality as it
is expressed, signalled, constituted, legitimized and so on by language use (or in
discourse). Most critical discourse analysts would thus endorse Habermas’s claim
that “language is also a medium of domination and social force. It serves to
legitimize relations to organized power. In so far as the legitimations of power
relations, ... are not articulated, ... language is also ideological” Habermas (1977:
259).

Among all the approaches, models and methods of critical discourse analysis,
Fairclough’s (1989) C.D.A. model is apt and most suitable for the present project.

60
2.11.1 Fairclough’s Approach to C.D.A.

Fairclough (1989) sets out the social theories underpinning C.D.A. and, as in
other early critical linguistic work, a variety of textual examples are analyzed to
illustrate the field, its aims and methods of analysis. Later Fairclough (1992b, 1995a,
1998, 2003) and Chouliariki and Fairclough (1999) explain and elaborate some
advances in C.D.A., showing not only how the analytical framework for investigating
language in relation to power and ideology developed, but also how C.D.A. is useful
in disclosing the discursive nature of much contemporary social and cultural change.
Particularly the language of the mass media is scrutinized as a site of power, of
struggle and also as a site where language is apparently transparent.

Faiclough considers discourse as a form of social practice. This implies firstly

that language is a part of society and not somehow external to it. Secondly, that
language is a social process. And thirdly, that language is a socially conditioned
process, conditioned that is by other (non-linguistic) part of society. Since, this
approach to discourse can fulfill the objectives of this research project, in the
following sections; Fairclough’s models of discourse analysis will be explained in
detail.

Three - Dimensional Model

Corresponding to his three-layered model of discourse (text, discursive

practice, social practice), Fairclough (2001c:21) distinguishes three dimensions, or
stages, of critical discourse analysis:

• Description: the stage which is concerned with formal properties of the text.
• Interpretation: This is concerned with the relationship between text and
interaction, with seeing the text as the product of a process of production and
as a resource in the process of interpretation.
• Explanation: which is concerned with the relationship between interaction
and social context. It is related to the social determination of the processes of
production and interpretation and their social effects.

61
 These three stages will be discussed in detail as parts of a procedure for doing
critical discourse analysis.

DESCRIPTION

In the case of description, analysis is generally thought of as a matter of

identifying and labelling formal features of a text (features of vocabulary, grammar,
punctuation, turn-taking, types of speech act and the directness or indirectness of their
expression) in terms of the categories of a descriptive framework. Fairclough
(2001c:92-3) lists ten main questions (and some sub-questions) that can be asked of a
text to find the set of textual features which tend to be most significant for critical
analysis.

A. Vocabulary
1. What experiential values do words have?
• What classification schemes are drawn upon?
• Are there words which are ideologically contested?
• Is there ‘rewording’ or ‘over wording’?
• What ideologically significant meaning relations (synonyms,
hyponyms, and antonyms) are there between words?
2. What relational values do words have?
• Are there euphemistic expressions?
• Are there markedly formal or informal words?
3. What expressive values do words have?
4. What metaphors are used?
B. Grammar
1. What experiential values do grammatical features have?
• What types of ‘Process’ And ‘Participant’ predominate?
• Is the agency unclear?
• Are processes what they actually seem to be?
• Are nominalizations used?
• Are sentences active or passive?
• Are sentences positive or negative?

62
 2. What relational values do grammatical features have?
• What modes (declarative, grammatical question, imperative) are
used?
• Are there important features of relational modality?
• Are the pronouns we and you used and if so, how?
3. What expressive values do grammatical features have?
• Are there important features of expressive modality?
4. How are (simple) sentences linked together?
• What logical connectors are used?
• Are complex sentences characterized by ‘coordination’ or
subordination?
• What means are used for referring inside and outside the text?
C. Textual structures
1 What interactional conventions are used?
• Are there ways in which one participant controls the turns of others?
2. What larger-scale structures does the text have?
The significance and interest of each of these questions are explained by
Fairclough (2001c:94-l 16) in details.

INTERPRETATION

The relationship between text and social structure is an indirect, mediated one.
It is mediated first of all by the discourse which the text is a part of, because the
values of textual features only become real, socially operative, if they are embedded
in social interaction, where texts are produced and interpreted against a background of
commonsense assumptions (part of members resources) which give textual features
their values. These discourse processes and their dependence on background
assumptions are the concern of the second stage of procedure, interpretation. The
stage of interpretation is concerned with participants’ processes of text production as
well as text interpretation. From the point of view of the interpreter of a text, formal
features of the text are cues which activate elements of interpreters’ members
resources and that interpretations are generated through the dialectical interplay of

63
cues and members resources. In their role of helping to generate interpretations,
Fairclough refers to members resources as interpretative procedures.

The following diagram gives a summary view of the process of interpretation:

Interpretation (from Fairclough 2001c:l19)

In the right-hand column of the diagram, under the heading ‘Interpreting’, six
major domains of interpretation have been listed. The two in the upper section of the
diagram relate to the interpretation of context, while those in the lower section relate
to four levels of interpretation of text. In the left-hand column (Interpretative
procedures) are listed major elements of members resources (MR) which function as
interpretative procedures. Each element of MR is specifically associated with the
level of interpretation which occurs on the same line of the diagram. The central
column identifies the range of ‘Resources’ which are drawn upon for each of the
domains of interpretation on the right. Notice that in each case these resources include

64
more than the interpretative procedure on the left i.e. there are either three or four
inputs to each box.

The upper section of the diagram relates to the interpretation of context.

Participants arrive at interpretation of ‘situational context’ partly on the basis of
external cues, but also partly on the basis of aspects of their members resources in
terms of which they interpret these cues. How participants interpret the situation
determines which discourse types are drawn upon and this in turn affects the nature of
the interpretative procedures which are drawn upon in textual interpretation. But one
also needs to refer to ‘intertextual context’ participants in any discourse operate on
the basis of assumptions about which previous (series of) discourses the current one is
connected to and their assumptions determine what can be taken as given in the sense
of part of common experience, what can be disagreed with, alluded to and so on.

The boxes in the central column in the figure represents the ‘contents’ of each
box as a combination of the various ‘inputs’ (identified by the arrows) which feed into
it. Notice, firstly that linking each box with the domain of interpretation identified to
its right is a double-headed arrow. What this means is that, at a given point in the
interpretation of a text, previous interpretations constitute one part of the ‘resources’
for interpretation. This applies for each of the domains of interpretation.

Notice, secondly, that the boxes in the central column are also linked vertically
with double-headed arrows. What this means is that each domain of interpretation
draws upon interpretations in the other domains as part of its ‘resources’. This
interdependence is in part obvious for the four levels of text interpretation for
instance, to interpret the global coherence and ‘point’ of a text, one draws upon
interpretations of the local coherence of parts of it; and to arrive at these, one draws
upon interpretations of utterance meanings; and to arrive at these, one draws upon
interpretations of the surface forms of utterances. But there is also interdependence in
the opposite direction. For instance, interpreters make guesses early in the process of
interpreting a text about its textual structure and ‘point’ and these guesses are likely
to influence the meanings that are attached to individual utterances and the local
coherence relations set up between them. One may capture this by saying that

65
interpretations have the important property of being ‘top-down’ (higher-level
interpretations shape lower-level) as well as ‘bottom-up’.

There is a similar situation with the relationship between interpretations of

context and interpretations of text, interpreters quickly decide what the context is and
this decision can affect the interpretation of text; but the interpretation of context is
partly based upon and can change in the course of, the interpretation of text.

Interpretation can be seen as a complex process with various different aspects. It

is partly a matter of understanding what words or sentences or longer stretches of text
mean, understanding what speakers or writers mean (the latter involving problematic
attributions of intentions). But it is also partly a matter of judgment and evaluation: for
instance, judging whether someone is saying something seriously or not, judging whether
the claims that are explicitly or implicitly made are true; judging whether people are
speaking or writing in ways which accord with the social institutional set up etc. relations
within which the event takes place, or perhaps in ways which mystify those relations.
Furthermore, there is an explanatory element to interpretation one often tries to
understand why people are speaking or writing as they do and even identify less
immediate social causes. Having said this, it is clear that some texts receive a great deal
of more interpretative work than others, some texts are very transparent, other more or
less opaque to particular interpreters; interpretation is sometimes unproblematic and
effectively automatic, but sometimes highly reflexive, involving a great deal of conscious
thought about what is meant, or why something has been said or written as it has been
(Fairclough 2003:11).

For interpreting the situational context, Fairclough (1989: 147-8) suggests

four; questions which relate to four main dimensions of situation:

1. What is going on? This can be subdivided into 'activity', 'topic' and 'purpose'
(one could certainly make finer discriminations, but these will suffice for our
purposes). The first, activity, is the most general; it allows us to identify a
situation in terms of one of a set of activity types, or distinctive categories of
activity, which are recognized as distinct within a particular social order in a
particular institution. The activity types are likely to constrain the set of

66
 possible; topics, though this does not mean topics can be mechanically
predicted given the activity type. Similarly, activity types are also associated
with particular institutionally recognized purposes.

2. Who is involved? The question of 'who's involved' and 'in what relations' is
obviously closely connected, though analytically separable. In the case of the
former, one is trying to specify which 'subject positions' are set up; the set of
subject positions differs according to the type of situation. It is important to
note that subject positions are multi-dimensional. Firstly, one dimension
derives from the activity type. Secondly, the institution ascribes social
identities to the subjects who function within it. And thirdly, different
situations have different speaking and listening positions associated with them:
speaker, addressee, hearer, over hearer, spokesperson and so forth.

3. In what relations? When it comes to the question of relations, one looks at

subject positions more dynamically, in terms of what relationships of power,
social distance and so forth are set up and enacted in the situation.

4. What is the role of language? Language is being used in an instrumental way

as a part of a wider institutional and bureaucratic objective. The role of
language in this sense not only determines its genre but also its channel,
whether spoken or written.

Answering these questions helps interpreters arrive at interpretations of the

situational context and the way in which this determines decisions about which
discourse type is the appropriate one to draw upon.

What has been said about interpretation can be summarized in the form of
three questions which can be asked about a particular discourse:

1. Context: what interpretation(s) are participants giving to the situational and

intertextual contexts?
2. Discourse type(s): what discourse type(s) are being drawn upon (hence what
rules, systems or principles of phonology, grammar, sentence cohesion,
vocabulary, semantics and pragmatics; and what schemata, frames and scripts
have been drawn upon)?

67
3. Difference and change: are answers to questions 1 and 2 different for different
participants? And do they change during the course of the interaction?

The stage of interpretation corrects delusions of autonomy on the part of

subjects in discourse. It makes explicit, what for participants is generally implicit: the
dependence of discourse practice on the unexplicated common-sense assumptions of
members resources and discourse type. What it does not do on its own, however, is
explicate the relations of power and domination and the ideologies which are built
into these assumptions and which make ordinary discourse practice a site of social
struggle. For this, analyst needs the stage of explanation.

EXPLANATION

The objective of the stage of explanation is to portray discourse as part of a

social process, as a social practice, showing how it is determined by social structures
and what reproductive effects discourses can cumulatively have on those structures,
sustaining them or changing them. These social determinations and effects are
mediated by members resources, that is social structures shape members resources,
which in turn shape discourses and discourses sustain or change members’ resources,
which in turn sustain or change structures.

The stage of explanation involves a specific perspective on members resources

as they are seen specifically as ideologies. That is, the assumptions about culture,
social relationships and social identities which are incorporated with members
resources, are seen as determined by particular power relations in the society or
institution and in terms of their contribution to struggles to sustain or change these
power relations they are seen ideologically. The social structures which are in focus
are relations of power and the social processes and practices which are in focus are
processes and practices of social struggle. Explanation is a matter of seeing discourse
as part of processes of social struggle, within a matrix of relations of power.

Depending on whether the emphasis is upon process or structure, upon processes

of struggle or upon relations of power, explanation has two dimensions, discourses can be
seen as parts of social struggles and contextualize them in terms of these broader (non-
discoursal) struggles and the effects of these struggles on structures. This puts the

68
emphasis on the social effects of discourse, on creativity and on the future. On the other
hand, the power relationships which determine discourses; can be shown as relationships
that are themselves the outcome of struggles and are established (and, ideally,
naturalized) by those with power. This puts the emphasis on the social determination of
discourse. Both social effects of discourse and social determinants of discourse should be
investigated at three levels of social organization: the societal level, the institutional level
and the situational level. This is represented in figure below.

Societal Societal

Institutional MR Discourse MR Institutional

Situational Situational
Determinants Effects

Fig. Explanation (from Fairclough2001c:136)

The stage of explanation can be summarized in the form of three questions

which can be asked of a particular discourse under investigation:

1. Social determinants: what power relations at situational, institutional and

societal levels help shape this discourse?
2. Ideologies: what elements of members’ resources which are drawn upon have
an ideological character?
3. Effects: how is this discourse positioned in relation to struggles at the
situational, institutional and societal level? Are these struggles overt or covert?
Is the discourse normative with respect to members resources or creative?
Does it contribute to sustaining existing power relations, or transforming
them?

Following the conception of discourse which involves an interest in properties

of texts, the production, distribution and consumption of texts, social practice in
various institutions, the relationship of social practice to power relations and
hegemonic projects at the societal level, Fairclough (1992b) explains that discourse
analysis ought ideally to be an interdisciplinary undertaking. These facets of
discourse impinge upon the concerns of various social sciences and humanities,

69
including linguistics, psychology and social psychology, sociology, history and
political science.

2.11.2 The Model of the Present Study

Considering the three dimensions of discourse analysis (i.e. (i) analysis of

discourse practices, (ii) analysis of texts and (iii) analysis of the social practice of
which the discourse is a part), Fairclough (1992b: 225-238) suggests a guideline for
doing discourse analysis as follows:

(i) Discourse Practice

Each of the three dimensions of discourse practice is represented below.

Interdiscursivity and manifest intertextuality focus upon text production, intertextual
chains upon text distribution and coherence upon text consumption (see Fairclough
1992b, Chap. 3-4).

• Interdiscursivity: The objective is to specify what discourse types are drawn

upon in the discourse sample under analysis and how. If it is not clear whether
something is a genre, activity type, style, or discourse, The general term of
discourse type can be used. Main way of justifying an interpretation is through
text analysis, by showing that one’s interpretation is compatible with the
features of the text and more compatible than others (see Fairclough 1992b:
124-30).

– Is there an obvious way of characterizing the sample overall (in terms

of genre if so, what does it imply in terms of how the sample is
produced, distributed and consumed?)
– Does the sample draw upon more than one genre?
– What activity type(s), style(s), discourse(s) are drawn upon’? (can you
specify styles according to tenor, mode and rhetorical mode?)
– Is the discourse sample relatively conventional in its interdiscursive
properties, or relatively innovative?

• Intertextual Chains: The objective here is to specify the distribution of a

(type of) discourse sample by describing the intertextual chains it enters into,

70
 that is the series of text types it is transformed into or out of (see Fairclough
1992b: 130-2).

– What sorts of transformation does this (type of) discourse sample

undergo?
– Are the intertextual chains and transformations relatively stable, or are
they shifting, or contested?
– Are there signs that the text producer anticipates more than one sort of
audience?

• Coherence: The aim here is to look into the interpretative implications of the
intertextual and interdiscursive properties of the discourse sample. This could
involve the analyst in ‘reader research’, that is, research into how texts are
actually interpreted (see Fairclough 1992b: 83-4).

– How heterogeneous and how ambivalent is the text for particular

interpreters and consequently how much inferential work is needed?
(This leads directly to intertextual dimensions of the construction of
subjects in discourse: see ‘Social practice’ below.) Does this sample
receive resistant readings and from what sort of readers?

• Conditions of Discourse Practice: The aim is to specify the social practices

of text production and consumption associated with the type of discourse the
sample represents (which may be related to its genre: see the first question
under ‘Interdiscursivity’ above) (Fairclough 1992b: 78-80).

Is the text produced (consumed) individually or collectively? (Are there

distinguishable stages of production Are animator, author and principal the
same or different people?) What sort of non-discursive effects does this sample
have?

• Manifest Intertextuality: Manifest intertextuality is a grey area between

discourse practice and text it raises questions about what goes into producing a
text, but it is also concerned with features which are ‘manifest’ on the surface
of the text. The objective is to specify what other texts are drawn upon in the

71
 constitution of the text being analysed and how. Genres differ in the modes of
manifest intertextuality with which they are associated and one aim here is to
explore such differences (see Fairclough 1992b: 117-23, 128).

• Discourse Representation:

– Is it direct or indirect?
– What is represented aspects of context and style, or just ideational
meaning?
– Is the represented discourse clearly demarcated? Is it translated into the
voice of the representing discourse?
– How is it contextualized in the representing discourse?

• Presupposition:

– How are presuppositions cued in the text?

– Are they linked to the prior texts of others, or the prior texts of the text
producer?
– Are they sincere or manipulative?
– Are they polemical (such as negative sentences)? Are there instances of
metadiscourse or irony?

(ii) Text

• Interactional Control: The objective here is to describe larger-scale

organizational properties of interactions, upon which the orderly functioning
and control of interactions depends (see Fairclough 1992b: 152-8). An
important issue is who controls interactions at this level to what extent is
control negotiated as a joint accomplishment of participants and to what extent
is it asymmetrically exercised by one participant?

– What turn-taking rules are in operation? Are the rights and obligations of
participants (with respect to overlap or silence, for example) symmetrical
or asymmetrical?
– What exchange structure is in operation?

72
 – How topics are introduced, developed, established and is topic control
symmetrical or asymmetrical?
– How are agendas set and by whom? How are they policed and by whom?
Does one participant evaluate the utterances of others?
– To what extent do participants formulate the interaction? What functions
do formulations have and which participant(s) formulate(s)?

• Cohesion: The objective is to show how clauses and sentences are connected
together in the text. This information is relevant to the description of the
‘rhetorical mode’ of the text (Fairclough 1992b: 127) its structuring as a mode
of argumentation, narrative, etc. (Fairclough 1992b: 174-7).

– What functional relations are there between the clauses and sentences of
the text?
– Are there explicit surface cohesive markers of functional relations?
Which ‘types of marker (reference, ellipsis, conjunction, lexical) are
most used?

• Politeness: The objective is to determine which politeness strategies are most

used in the sample, whether there are differences between participants and
what these features suggest about social relations between participants
(Fairclough 1992b: 162-6).

– Which politeness strategies (negative politeness, positive politeness, off

record) are used, by whom and for what purposes?

• Ethos: The objective is to pull together the diverse features that go towards
constructive selves, or social identities, in the sample. Ethos involves not just
discourse, but the whole body. Any of the analytical categories listed here may
be relevant to ethos (Fairclough 1992b: 166-7).

• Grammar: Three dimensions of the grammar of the clause are differentiated

here: ‘transitivity’, ‘theme’ and ‘modality’. These correspond respectively to
the ‘ideational’, ‘textual’ and ‘interpersonal’ functions of language (Fairclough
1992b:64).

73
• Transitivity: The objective is to see whether particular process types and
participants are favoured in the text, what choices are made in voice (active or
passive) and how significant is the nominalization of processes. A major
concern is agency, the expression of causality and the attribution of
responsibility (Fairclough 1992b: 177-85).

– What process types (action, event, relational, mental) are most used and
what factors may account for this?
– Is grammatical metaphor a significant feature?
– Are passive clauses or nominalizations frequent and if so what functions
do they appear to serve?

• Theme: The objective is to see if there is a discernible pattern in the text’s

thematic structure to the choices of themes for clauses (Fairclough 1992b:
183-5).

– What is the thematic structure of the text and what assumptions (for
example, about the structuring of knowledge or practice) underlie it?
– Are marked themes frequent and if so what motivations for them are
there?

• Modality: The objective is to determine patterns in the text in the degree of

affinity expressed with propositions through modality. A major concern is to
assess the relative import of modality features for (a) social relations in the
discourse and (b) controlling representations of reality (Fairclough 1992b:
158-62).

– What sort of modalities are most frequent?

– Are modalities predominantly subjective or objective?
– What modality features (modal verbs, modal adverbs, etc.) are most
used?

• Word Meaning: The emphasis is upon ‘key words’ which are of general or
more local cultural significance; upon words whose meanings are variable and

74
 changing; and upon the meaning potential of a word - a particular structuring
of its meanings - as a mode of hegemony and a focus of struggle (Fairclough
1992b: 185-90).

• Wording: The objective is to contrast the ways meanings are worded with the
ways they are worded in other (types of) text and to identify the interpretative
perspective that underlies this wording (Fairclough 1992b: 190-4).

– Does the text contain new lexical items and if so what theoretical,
cultural or ideological significance do they have?
– What intertextual relations are drawn upon for the wording in the text?
– Does the text contain evidence of overwording or rewording (in
opposition to other wordings) of certain domains of meaning?

• Metaphor: The objective is to characterize the metaphors used in the

discourse sample, in contrast to metaphors used for similar meanings
elsewhere and determine what factors (cultural, ideological, etc.) determine the
choice of metaphor. The effect of metaphors upon thinking and practice should
also be considered (Fairclough 1992b: 194-8).

(iii) Social Practice

The analysis of social practice is more difficult to reduce to a checklist, so the

following heads should be seen only as very rough guidelines. The general objective
here is to specify: the nature of the social practice of which the discourse practice is a
part, which is the basis for explaining why the discourse practice is as it is; and the
effects of the discourse practice upon the social practice.

• Social Matrix of Discourse: The aim is to specify the social and hegemonic
relations and structures which constitute the matrix of this particular instance
of social and discursive practice; how this instance stands in relation to these
structures and relations (is it conventional and normative, creative and
innovative, oriented to restructuring them, oppositional, etc.?); and what
effects it contributes to, in terms of reproducing or transforming them.

75
 • Orders of Discourse: The objective here is to specify the relationship of the
instance of social and discursive practice to the orders of discourse it draws
upon and the effects of reproducing or transforming orders of discourse to
which it contributes. Attention should be paid to the large-scale tendencies
affecting orders of discourse Fairclough (1992b, Chap. 7).

• Ideological and Political Effects of Discourse: It is useful to focus upon the

following particular ideological and hegemonic effects (Fairclough 1992b: 86-
96).

1. Systems of knowledge and belief;

2. Social relations;
3. Social identities (selves).

TEXTUAL ANALYSIS

The textual analysis consists of investigating the following aspects in a text.

1. Social Events:

– What social event and what chain of social events, is the text a part of?
– What social practice or network of social practices can the events be
referred to, be seen as framed within?
– Is the text part of a chain or network of texts?

2. Genre:

– Is the text situated within a genre chain?

– Is the text characterized by a mix of genres?
– What genres does the text draw upon and what are their characteristics?

3. Intertextuality:

– Of relevant other texts/ voices, which are included, which are significantly
excluded?
– Where other voices are included? Are they attributed and if so, specifically
or non-specifically?

76
 – Are attributed voices directly reported (quoted), or indirectly reported?
– How are other voices textured in relation to the authorial voice and in
relation to each other?

4. Assumptions:

– What existential, prepositional, or value assumptions are made?

– Is there a case for seeing any assumptions as ideological?

5. Semantic/ Grammatical Relations between Sentences and Clauses :

– What are the predominant semantic relations between sentences and

clauses (causal-reason, consequence, purpose; conditional; temporal;
additive; elaborative; contrastive/ concessive)?
– Are there higher-level semantic relations over larger stretches of the text
(e.g. problem-solution)?
– Are grammatical relations between clauses predominantly paratactic,
hypotactic or embedded?
– Are particularly significant relations of equivalence and difference set up
in the text?

6. Exchanges, Speech Functions and Grammatical Mood:

– What are the predominant types of exchange (activity exchange, or

knowledge exchange) and speech functions (statement, question, demand,
offer)?
– What types of statement are there (statements of fact, predictions, hypo-
theticals, evaluations)?
– Are there metaphorical relations between exchanges, speech functions, or
types of statement (e.g. demands which appear as statements, evaluations
which appear as factual statements)?
– What is the predominant grammatical mood (declarative, interrogative,
imperative)?

77
7. Discourses:

– What discourses are drawn upon in the text and how are they textured
together?
– Is there a significant mixing of discourses?

– What are the features that characterize the discourses which are drawn
upon (semantic relations between words, collocations, metaphors,
assumptions, grammatical features-see immediately below)?

8. Representation of Social Events:

– What elements of represented social events are included or excluded and

which included elements are most salient? How abstractly or concretely are
social events represented?
– How are processes represented? What are the predominant process types
(material, mental, verbal, relational, and existential)?
– Are there instances of grammatical metaphor in the representation of
processes? How social actors are represented (active/passive, personal
impersonal, named/ classified, specific/ generic)?
– How are time, space and the relation between space-times represented?

9. Styles:

– What styles are drawn upon in the text and how are they textured together?
– Is there a significant mixing of styles?
– What are the features that characterize the styles that are drawn upon (body
language, pronunciation and other phonological features, vocabulary,
metaphor, modality or evaluation, see immediately below for the latter
two)?

10. Modality:

– What do authors commit themselves to in terms of truth (epistemic

modalities)? Or in terms of obligation and necessity (deontic modalities)?

78
 – To what extent are modalities categorical (assertion, denial etc.), to what
extent are they modalized (with explicit markers of modality)?
– What levels of commitment are there (high, median, low) where modalities
are modalized?
– What are the markers of modernization (modal verbs, modal adverbs, etc.)?

11. Evaluation:

– To what values (in terms of what is desirable or undesirable) do authors

commit themselves?
– How are values realized as evaluative statements, statements with deontic
modalities, statements with affective mental processes, or assumed values?

During discourse analysis there is a constant alternation of focus from the

particularity of the discourse sample, to the type(s) of discourse which it draws upon
and the configurations of discourse types to which it is oriented. .Analysis should be
directed at both: it should show features, patterns and structures which are typical of
certain types of discourse, restructuring tendencies in orders of discourse and ways of
using these conventional resources which are specific to this sample.

There are always alternative possible analyses for discourse samples and the
question arises of how analysts can justify the analyses they propose (how they can
‘validate’ them). There is no simple answer and all one can do is decide, given
alternative analyses, which seems to be preferable on the balance of evidence
available.

It was the goal of the preceding sections to give a brief outline of the core
procedures applied in the different approaches to C.D.A.. Finally, it should be pointed
out that, although there is no consistent C.D.A. methodology, some features are
common to most C.D.A. approaches: firstly they are problem oriented and not
focused on specific linguistic items. Yet linguistic expertise is obligatory for the
selection of the items relevant to specific research objectives. Secondly theory as well
as methodology is eclectic: both are integrated as far as it is helpful to understand the
social problems under investigation.

79
 Considering all above mentioned models for doing critical discourse analysis,
this research will follow more closely the models of Fairclough (1989) for analyzing
the corpus data from women magazines. However, C.D.A. cannot be restricted to any
one particular model but rather it’s a whole paradigm of interrelated aspects. Before I
begin with the analysis of the collected data, a detailed review of the studies
conducted using C.D.A. is very important.

80