DEVELOPMET OF DISCOURSE THEORIES
By:
Zainul Mujahid
English Lecturer of STIBA MALANG
Abstarct: Discourse analysis (DA), or discourse
studies, is a general term for a number of approaches
to analyze written, spoken, signed language use or any
significant semiotic event. The objects of discourse
analysis—discourse, writing, talk, conversation,
communicative event, etc.—are variously defined in
terms of coherent sequences of sentences,
propositions, speech acts or turns-at-talk. Contrary to
much of traditional linguistics, discourse analysts not
only study language use 'beyond the sentence
boundary', but also prefer to analyze 'naturally
occurring' language use, and not invented examples.
This is known as corpus linguistics or text linguistics.
DISCOURSE ANALYSIS: ITS ORIGIN AND DEVELOPMENT
Discourse analysis is both an old and new discipline. Its
origins can be traced back to the study of language, public speech,
and literature more than 2000 years ago. One major historical source
is undoubtedly classical rhetoric, the art of good speaking. Whereas
the grammatica, the historical antecedent of linguistics, was
229
concerned with the normative rules of correct language use, its sister
discipline of rhetorica dealt with the percepts for the planning,
organization, specific operations, and performance of public speech
in political and legal settings. Its crucial concern, therefore, was
persuasive effectiveness. In this sense, classical rhetoric both
anticipates contemporary stylistics and structural analyses of
discourse and contains intuitive cognitive and social psychological
notions about memory organization and attitude change in
communicative contexts (Dijk, 1985:1). Some scholars consider the
Austrian emigre Leo Spitzer's Style Studies of 1928 the earliest
example of discourse analysis (DA); it was translated into French by
no less than Michel Foucault. But the term first came into general
use following the publication of a series of papers by Zellig Harris
beginning in 1952 and reporting on work from which he developed
transformational grammar in the late 1930s. Formal equivalence
relations among the sentences of a coherent discourse are made
explicit by using sentence transformations to put the text in a
canonical form. Words and sentences with equivalent information
then appear in the same column of an array. This work progressed
over the next four decades (see references) into a science of
sublanguage analysis (Kittredge & Lehrberger, 1982), culminating in
231
a demonstration of the informational structures in texts of a
sublanguage of science, that of immunology, (Harris et al. 1989) and
a fully articulated theory of linguistic informational content (Harris
1991). During this time, however, most linguists decided a
succession of elaborate theories of sentence-level syntax and
semantics (Wikipedia, 2010).
Discourse analysis, however, grew out of work in different
disciplines in most of the humanities and social sciences in the 1960s
and early 1970s, including linguistics, pragmatics, sociolinguistics,
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.
(McCarty, 1994:5). This phase is also claimed as the origins of
modern discourse analysis (Dijk, 1985:2). At a time when linguistics
was largely concern with the analysis of single sentences, Zellig
Harris published a paper with the title ‘Discourse Analysis’ (Harris,
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 far cry from the discourse analysis we
are used to nowadays. Also importance in the early years was the
emergence of semiotics and the French structuralist approach to the
study of narrative. In the 1960s, 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 [see Levinson,
1983; leech, 1983] (Wikipedia, 2010). The work or critical studies of
Harris, Hymes, Austin, Searle, and Grice, thus, show the significance
of discourse analysis as a new cross-discipline that makes the inquiry
to be not tied to one discipline only.
British discourse analysis was greatly influenced by M.A.K
Halliday’s functional approach to language (e.g. Haliiday, 1973),
which in turn has connections with the Prague School of linguistics.
Halliday’s frame work emphasizes 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 Birmingham, who developed a model for the
description of teacher-pupil talk, based on a hierarchy of discourse
units. Other similar work has dealt with doctor – patient orientations,
as well as monologues. Novel work in the British tradition has also
233
been done on intonation in discourse. The British work has
principally followed structural-linguistic criteria (Wikipedia, 2000).
Meanwhile, Michel Foucault became one of the key theorists of the
subject, especially of discourse, and wrote The Archaeology of
Knowledge.
American discourse analysis has been dominated by work
within
the
ethnomethodological
tradition
(the
founder
of
ethnomethodology is Harold Garfinkel), which emphasizes the
research method of close observation of groups of people
communicating in natural settings. It examines types of speech event
such as story telling, greeting rituals and verbal duels in different
cultural and social settings (e.g. Gumperz and Hymes, 1972). What
is called conversational analysis within the American tradition can
also be included under the general heading of discourse analysis. In
conversational analysis, the emphasis is not upon building structural
models but on the close observation of the behavior of participants in
talk and on patterns which recur over a wide range of natural data.
The work of Goffman (1976; 1979), and sacks, Schegloff and
Jefferson (1974) is important in the study of conversational norms,
turn taking, and other aspects of spoken interaction. Alongside the
conversation analysts, working within sociolinguistic tradition,
Labov’s investigations of oral story telling have also contributed to a
long history of 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 facepreserving phenomena in talk, overlapping with British work in
pragmatics (Wikipedia, 2010).
SPHERE OF INTERESTS OF DISCOURSE ANALYSTS
The range of inquiry of discourse analysis not only covers
linguistic issues, but is also concerned with other matters, such as:
enabling computers to comprehend and produce intelligible texts,
thus contributing to progress in the study of Artificial Intelligence.
Out of these investigations a very important concept of schemata
emerged. It might be defined as prior knowledge of typical situations
which enables people to understand the underlying meaning of
words in a given text. This mental framework is thought to be shared
by a language community and to be activated by key words or
context in order for people to understand the message. To implement
schemata to a computer, however, is yet impossible (Cook 1990:69).
235
Discourse
analysts
carefully
scrutinize
universal
circumstances of the occurrence of communicative products,
particularly within state institutions. Numerous attempts to minimize
misunderstandings between bureaucrats and citizens were made,
resulting in user-friendly design of documents. The world of politics
and features of its peculiar communicative products are also of
concern to discourse analysts. Having carefully investigated that area
of human activity scholars depicted it as characterized by frequent
occurrence of face saving acts and euphemisms. One other sphere of
life of particular interest to applied linguists is the judicature and its
language which is incomprehensible to most common citizens,
especially due to pages-long sentences, as well as peculiar
terminology. Moreover, educational institutions, classroom language
and the language that ought to be taught to enable learners to
successfully comprehend both oral and written texts, as well as
participate in real life conversations and produce native-like
communicative products is the domain of discourse analysis. Last
but not least, influence of gender on language production and
perception is also examined (Renkema 2004, Trappes-Lomax 2004).
Spoken Language Analysis
The examination of oral discourse is mainly the domain of
linguists gathered at the University of Birmingham, who at first
concentrated on the language used during teacher - learner
communication, afterwards altering their sphere of interest to more
general issues. However, patterns of producing speech characteristic
of communities, or members of various social classes within one
population were also of ethnomethodologists' interest. A result of
such inquiries was discovering how turn taking differs from culture
to culture as well as how standards of politeness vary. In addition,
manners of beginning discussions on new topics were described
(McCarthy 1991:24).
What is more, it was said that certain characteristics are
common to all societies, for instance, indicating the end of thought
or end of utterance. The words that are to point the beginning or the
closing stages of a phrase are called 'frames'. McCarthy (1991:13)
claims that it is thanks to them that people know when they can take
their turn to speak in a conversation. However, in spite of the fact
that frames can be noticed in every society, their use might differ,
which is why knowledge of patterns of their usage may be essential
237
for conducting a fluent and natural dialogue with a native speaker.
Moreover, these differences are not only characteristic of cultures,
but also of circumstances in which the conversation occurs, and are
also dependent on the rights (or 'rank') of the participants (McCarthy
1991:13).
Apart from that, it was pointed out that some utterances are
invariably interrelated, which can enable teachers of foreign
languages to prepare learners adequately to react as a native speaker
would. Among the phrases whose successors are easy to anticipate
there are for instance: greeting, where the response is also greeting;
apology with the response in the form of acceptance or informing and acknowledging as a response. Such pairs of statements are
known as adjacency pairs. While the function of the reply is
frequently determined by the former expression its very form is not,
as it depends on circumstances in which the conversation occurs.
Thus, in a dialogue between two friends refusal to provide help
might look like that: no way! I ain't gonna do that!, but when mother
asks her son to do something the refusing reply is more likely to take
different form: I'm afraid I can't do that right now, can you wait 5
minutes? Frequently used phrases, such as "I'm afraid", known as
softeners, are engaged when people want to sound more respectful.
Learners of a foreign language should be aware of such linguistic
devices if they want to be skillful speakers (McCarthy 1991:121).
Written Texts Analysis
Since the examination of written language is easier to
conduct than the scrutiny of oral texts, in that more data is available
in different genres, produced by people form different backgrounds
as well as with disparate purposes, it is more developed and of
interest not only to linguists but also language teachers and literary
scholars. Each of them, however, approaches this study in a different
way, reaching diverse conclusions, therefore only notions that are
mutual for them and especially those significant for language
methodology are accounted for here. What is worth mentioning is
the fact that in that type of analysis scholars do not evaluate the
content in terms of literary qualities, or grammatical appropriateness,
but how readers can infer the message that the author intended to
convey (Trappes-Lomax 2004:133).
Apart from differences between written and spoken language
described beforehand it is obviously possible to find various types
and classes of discourse depending on their purpose. Written texts
239
differ from one another not only in genre and function, but also in
their structure and form, which is of primary importance to language
teachers, as the knowledge of arrangement and variety of writing
influences readers' understanding, memory of messages included in
the discourse, as well as the speed of perception. Moreover, written
texts analysis provides teachers with systematic knowledge of the
ways of describing texts, thanks to which they can make their
students aware of characteristic features of discourse to which the
learners should pay particularly close attention, such as cohesion and
coherence. In addition, understanding these concepts should also
improve learners' writing skills as they would become aware of traits
essential for a good written text.
One of the major concerns of written discourse analysts is the
relation of neighboring sentences and, in particular, factors attesting
to the fact that a given text is more than only the sum of its
components. It is only with written language analysis that certain
features of communicative products started to be satisfactorily
described, despite the fact that they were present also in speech, like
for instance the use of 'that' to refer to a previous phrase, or clause
(McCarthy 1991:37). As mentioned before written language is more
integrated than the spoken one which is achieved by more frequent
use of some cohesive devices which apart from linking clauses or
sentences are also used to emphasize notions that are of particular
importance to the author and enable the reader to process the chosen
information at the same time omitting needless sections (Salkie
1995:XI).
TEXT AND DISCOURSE: WHAT IS TEXT LINGUISTICS?
WHAT IS DISCOURSE ANALYSIS - DEFINITION?
DISCOURSE FEATURES, AND DISCOURSE TYPES
Text Linguistics and Discourse Analysis
To define and describe the scope of the study of Text
Linguistics (TL) and Discourse Analysis (DA) and to establish the
differences between them both is not an easy task. Suffice it to say
that the term text and discourse are used in a variety of ways by
different linguists and researchers: there is a considerable number of
theoretical approaches to both TL and DA and many of them belong
to very different research traditions, even when they share similar
basic tenets.
241
In everyday popular use it might be said that the term text is
restricted to written language, while discourse is restricted to spoken
language. However, modern Linguistics has introduced a concept of
text that includes every type of utterance; therefore a text may be
magazine, article, a television interview, a conversation or a cooking
recipe, just to give a few examples.
Crystal (1997) defines TL as “the formal account of the
linguistic principles governing the structure of texts”. De
Beaugrande and Dresser (1981) present a broader view; they define
text as a communicative event that must satisfy the following criteria:
Cohesion, which has to do with the relationship between text
and syntax (grammatical relationship between parts of a
sentence essential for its interpretation). Phenomena such as
conjunction, ellipsis, anaphora, cataphora or recurrence are
basic for cohesion.
2) Coherence, which has to do with the meaning of the text.
Here we may refer to elements of knowledge or to cognitive
structures that do not have a linguistics realization but are
implied by the language used, and thus influence the
reception of the message by the interlocutor (the order of
statements relates one another by sense).
3) Intentionality, which relates to the attitude and purpose of
the speaker or writer (the message has to be conveyed
deliberately and consciously).
4) Acceptability, which concerns the preparation of the hearer
or reader to assess the relevance or usefulness of a given text
1)
(indicates that the communicative product needs to be
satisfactory in that the audience approves it).
5) Informativeness (-vity), which refers to the quantity and
quality of new or expected information (some new
information has to be included in the discourse).
6) Situationality, which points to the fact that the situation in
which the text is produced plays a crucial role in the
production and reception of the message (circumstances in
which the remark is made are important).
7) Intertextuality, which refers to two main facts: a) a text is
always related to some preceding or simultaneous discourse;
b) texts are always linked and grouped in particular text
varieties or genres [e.g. narrative, argumentative, descriptive,
etc.] (reference to the world outside the text or the
interpreters' schemata).
Nowadays, however, not all of the above mentioned criteria
are perceived as equally important in discourse studies, therefore
some of them are valid only in certain methods of the research
(Beaugrande 1981, cited in Renkema 2004:49).
In spite of the considerable overlap between TL and DA
(both of them are concerned with notion of cohesion, for instance)
the above criteria may help us make a distinction between them.
Tischer et al (2000) explain that the first two criteria
(cohesion and coherence) may be defined as text-internal, whereas
the remaining criteria are text-external. Those approaches oriented
243
towards ‘pure’ TL give more importance to text-internal criteria,
while the tradition in DA has always been to give more importance
to the external factors, for they are believed to play an essential role
in communication.
Some authors, such as Halliday, believe that text is
everything that is meaningful in a particular situation: “By text, then,
we understand a continuous process of semantic choice” (1978: 137).
In the “purely” text linguistics approaches, such as a cognitive
theories of text, texts are viewed as “more or less explicit epiphenomena of cognitive processes” (Tischer et al., 2000: 29), and the
context plays a subordinate role.
It could be said that the text-internal elements constitute the
text, while the text-external ones constitute the context. Schiffrin
points out that all approaches within Discourse Analysis view text
and context as the two kinds of information that contribute to the
communicative content of an utterance, and she defines these terms
as follows:
I will use the term “text” to differentiate linguistic material
(e.g. what is said, assuming a verbal channel) from the
environment in which “sayings” (or other linguistic
productions) occur (context). In term of utterances, then,
“text” is the linguistic content: the stable semantic meanings
of words, expressions, and sentences, but not the inferences
available to hearers depending upon the contexts in which
words, expressions, and sentences are used. … Context is
thus a word filled with people producing utterances: people
who have social, cultural, and personal identities, knowledge,
beliefs, goals and wants, and who interact with one another in
various socially and culturally defined situation (1994: 363).
Thus according to Shrhiffrin, DA involves the study of both
text and context. One might conclude, then, that TL only studies the
text, while DA is more complete because it studies both text and
context. However, as has been shown, there are definitions of text
(like de Beaugrande’s) that are very broad and include both
elements, and that is why it would be very risky to talk about clearcut differences between the two disciplines. De Beaugrade’s (2000)
definition of TL as “the study of real language in use” does not differ
from many of the definitions of DA presented by Schiffrin within its
functional approach, some of which are the following:
The study of discourse is the study of any aspect of language
use (Fasol, 1990: 65).
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 these forms are designed to serve in human
affairs (Brown & Yule, 1983: 1).
Discourse … refers to language in use, as a process which is
socially situated (Candlin, 1997: ix).
245
Thus, we see that the terms text and discourse are sometimes
used to mean the same and therefore one might conclude that TL and
DA are the same, too. It can be said, nevertheless, that the tendency
in TL has been to present a more formal and experimental approach,
while DA tends more towards a functional approach. Formalists are
apt to see language as mental phenomenon, while functionalists see
it as a predominantly social one. As has been shown, authors like
Schiffrin integrate both the formal and the functional approaches
within DA, and consequently, DA is viewed as an all-embracing
term which would include TL studies as one approach among others.
Slembrouck points out the ambiguity of the term discourse
analysis and provides another broad definition:
The term discourse analysis is very ambiguous. I will use it in
this book to refer mainly to the linguistic analysis of naturally
occurring connected speech 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 use in
social contexts, and in particular with interaction or dialogue
between speakers (2005: 1).
Another important characteristic of discourse studies is that
they are essentially multidisciplinary, and therefore it can be said
that they cross the Linguistics border into different and varied
domains, as van Dijk notes in the following passage:
… discourse analysis is for me essentially multidisciplinary,
and involves linguistics, poetics, semiotics, psychology,
sociology, anthropology, history, and communication
research. What I find crucial though is that precisely because
of its multi-faceted nature, this multidisciplinary research
should be integrated. We should device theories that are
complex and account both for textual, the cognitive, the
social, the political and historical dimension of discourse
(2002: 10).
Thus, when analyzing discourse, researchers are not only
concerned with “purely” linguistic facts; they pay equal or more
attention to language use in relation to social, political and cultural
aspects. For this reason, discourse is not only within the interests of
linguists; it is a field that is also studied by communication scientists,
literary critics, philosophers, sociologists, social psychologists,
political scientists, and many others. As Barbara Johnstone puts it:
… I see discourse analysis as a research method that can be
(and is being) used by scholars with a variety of academic
and non-academic affiliations, coming from a variety of
disciplines, to answer a variety of questions (2002: xi).
247
Furthermore,
Mills
(1997)
elaborates
that
the
term
‘discourse’ has become common currency in a variety of disciplines:
critical theory, sociology, linguistics, philosophy, social psychology
and many other fields, so much so that it is frequently left undefined,
as if its usage were simply common knowledge. It is used widely in
analyzing literary and non-literary texts and it is often employed to
signal certain theoretical sophistication in ways which are vague (1).
Since its introduction to modern science the term 'discourse' has
taken various, sometimes very broad, meanings. In order to specify
which of the numerous senses is analyzed in the following
dissertation it has to be defined. Originally the word 'discourse'
comes from Latin 'discursus' which denoted 'conversation, speech'.
Thus understood, however, discourse refers to too wide an area of
human life, therefore only discourse from the vantage point of
linguistics, and especially applied linguistics, is explained here
(Wiśniewski, 2006).
There is no agreement among linguists as to the use of the
term discourse in that some use it in reference to texts, while others
claim it denotes speech which is for instance illustrated by the
following definition: "Discourse: a continuous stretch of (especially
spoken) language larger than a sentence, often constituting a
coherent unit such as a sermon, argument, joke, or narrative"
(Crystal 1992:25). On the other hand Dakowska, being aware of
differences between kinds of discourses indicates the unity of
communicative intentions as a vital element of each of them.
Consequently she suggests using terms 'text' and 'discourse' almost
interchangeably betokening the former refers to the linguistic
product, while the latter implies the entire dynamics of the processes
(Dakowska 2001:81). According to Cook (1990:7) novels, as well as
short conversations or groans might be equally rightfully named
discourses.
It is interesting therefore to trace the ways in which we try to
make sense of the term. The most obvious way to track down its
range of meanings is through consulting a dictionary and some
experts’ theoretical concept:
Discourse 1. verbal communication; talk, conversation; 2. a
formal treatment of a subject in speech or writing; 3. a unit of
text used by linguists for the analysis of linguistic phenomena
that range over more than one sentence; 4. to discourse: the
ability to reason (archaic); 5. to discourse on/upon: to speak
or to write about formally; 6. to hold a discussion; 7. to give
forth (music) (archaic) [14th century, from Medieval Latin.
discursus: argument, from Latin, a running to and fro
discurrere] (Collin Concise English Dictionary, 1988).
249
Discourse: 1. a conversation, especially of a formal nature;
formal and orderly expression of ideas in speech or writing;
also such expression in the form of a sermon, treatise, etc.; a
piece or unit of connected speech or writing (Middle English:
discours, from Latin: act of running about) (Longman
Dictionary of English Language, 1984)
This sense of the general usage of discourse as having to do
with conversation and ‘holding forth’ on a subject, or giving a
speech, has been partly due to the etymology of the word.
Meanwhile, Geoffrey Leech and Michael Short argue that:
Discourse is linguistic communication seen as a transaction
between speaker and hearer, as an interpersonal activity
whose form is determined by its social purpose. Text is
linguistic communication (either spoken or written) seen
simply as a message coded in its auditory or visual medium
(cited in Hawthorn, 1992: 189)
And Hawthorn himself comments on this opposition between
text and discourse:
Michael Stubbs (1993) treats text and discourse as more or
less synonymous, but notes that in other usages a text may be
written, while a discourse is spoken, a text may be non
interactive whereas a discourse is interactive … a text may be
short or long whereas a discourse implies a certain length,
and a text must be possessed of surface cohesion whereas a
discourse must be possessed of a deeper coherence. Finally,
Stubbs notes that other theorists distinguish between abstract
theoretical construct and pragmatic realization, although,
confusingly, such theorists are not agreed upon which of
these is represented by the term text.
Roger Fowler contrasts discourse with ideology, as follows:
Discourse is speech or writing seen from the point of view of
the beliefs, values and categories which it embodies; these
beliefs etc. constitute a way of looking at the world, an
organization or representation of experience – ‘ideology’ in
the neutral non-pejorative sense. Different modes of
discourse encode different representations of experience; and
the source of these representations is the communicative
context within which the discourse is embedded. (cited in
Hawthorn, 1992:48).
Thus, when we try to define discourse, we may resort to
reference to dictionaries, to disciplinary context of utterance or to
terms which are used in contrast discourse, even though none of
these strategies produces a simple, clear meaning of the term, but
rather only serves to show us the fluidity of its meaning.
4.2
Features of Discourse.
Since it is not easy to unambiguously clarify what a discourse
is it seems reasonable to describe features which are mutual to all its
kinds. To do it thoroughly Saussurean concepts of langue and parole
251
are of use. Ferdinand de Saussure divided the broad meaning of
language into langue, which is understood as a system that enables
people to speak as they do, and parole - a particular set of produced
statements. Following this division discourse relates more to parole,
for it always occurs in time and is internally characterized by
successively developing expressions in which the meaning of the
latter is influenced by the former, while langue is abstract. To list
some additional traits: discourse is always produced by somebody
whose identity, as well as the identity of the interpreter, is significant
for the proper understanding of the message. On the other hand
langue is impersonal that is to say more universal, due to society.
Furthermore, discourse always happens in either physical, or
linguistic context and within a meaningful fixed time, whereas
langue does not refer to anything. Consequently, only discourse may
convey messages thanks to langue which is its framework
(Wiśniewski, 2006).
Types of Discourse
Not only is discourse difficult to define, but it is also not easy
to make a clear cut division of discourse as such. Therefore,
depending on the form linguists distinguish various kinds of
communicative products. A type of discourse might be characterized
as a class of either written or spoken text, which is frequently
casually specified, recognition of which aids its perception, and
consequently production of potential response (Cook 1990:156).
Discourse analysis, moreover,
is not only connected with the
description and analysis of spoken interaction. In addition to all our
verbal encounters we daily consume hundreds of written and printed
words: newspaper articles, letters, stories, recipes, instructions,
notices, comics, billboards, leaflets pushed through the door, and so
on. We usually expect them to be coherent, meaningful
communications in which the words and/or sentences are linked to
one another in a fashion that corresponds to conventional formulate,
just as we do with speech; therefore discourse analyst are equally
interested in the organization of spoken and written interaction. One
of such divisions, known as the Organon model, distinguishes three
types of discourse depending of the aspect of language emphasized
in the text. If the relation to the context is prevailing, it conveys
some knowledge. Thus, it is an informative type of discourse. When
the stress is on a symptom aspect the fulfilled function is expression,
as a result the discourse type is narrative. Last but not least in this
253
division is argumentative discourse which is characterized by the
accent on the signal aspect.
This
distinction
due
to
its
suitability
for
written
communicative products more than for spoken ones, faced
constructive criticism whose accurate observation portrayed that
there are more functions performed. Consequently there ought to be
more types of discourse, not to mention the fact that these often mix
and overlap. Thorough examination of the matter was conducted,
thus leading to the emergence of a new, more detailed classification
of kinds of spoken texts.
The analysis of oral communicative products was the domain
of Steger, who examined features of various situations and in his
categorization divided discourse into six types: presentation,
message, report, public debate, conversation and interview. The
criteria of this division include such factors as presence, or absence
of interaction, number of speakers and their relation to each other
(their rights, or as Steger names it 'rank'), flexibility of topic along
with selection and attitude of interlocutors towards the subject matter
(Wiśniewski, 2006).
However, it is worth mentioning that oral discourse might
alter its character, for instance in the case of presenting a lecture
when students start asking questions the type changes to interview,
or even a conversation. Using this classification it is possible to
anticipate the role of partakers as well as goals of particular acts of
communication.
The above mentioned typologies do not exhaust the possible
division of discourse types, yet, nowadays endeavor to create a
classification that would embrace all potential kinds is being made.
Also, a shift of interest in this field might be noticed, presently
resulting in focus on similarities and differences between written and
spoken communication (Renkema 2004:64).
Written and Spoken Discourse
Apart from obvious differences between speech and writing
like the fact that writing includes some medium which keeps record
of the conveyed message while speech involves only air, there are
certain dissimilarities that are less apparent. Speech develops in time
in that the speaker says with speed that is suitable for him, even if it
may not be appropriate for the listener and though a request for
255
repetition is possible, it is difficult to imagine a conversation in
which every sentence is to be rephrased. Moreover, talking might be
spontaneous which results in mistakes, repetition, sometimes less
coherent sentences where even grunts, stutters or pauses might be
meaningful. The speaker usually knows the listener, or listeners, or
he is at least aware of the fact that he is being listened to, which
enables him to adjust the register. As interlocutors are most often in
face-to-face encounters (unless using a phone) they take advantage
of extralinguistic signals as grimaces, gesticulation, expressions such
as 'here', 'now', or 'this' are used. Employment of nonsense
vocabulary, slang and contracted forms (we're, you've) is another
feature of oral discourse. Among other significant features of speech
there are rhythms, intonation, speed of uttering and, what is more
important, inability to conceal mistakes made while speaking
(Crystal 1995:291, Dakowska 2001:07).
In contrast, writing develops in space in that it needs a means
to carry the information. The author of the text does not often know
who is going to read the text, as a result he cannot adjust to readers'
specific expectations. The writer is frequently able to consider the
content of his work for almost unlimited period of time which makes
it more coherent, having complex syntax. What is more, the reader
might not instantly respond to the text, ask for clarification, hence
neat message organization, division to paragraphs, layout are of vital
importance to make comprehension easier. Additionally, owing to
the lack of context expressions such as 'now' or 'here' are omitted,
since they would be ambiguous as texts might be read at different
times and places. One other feature typical of writing, but never of
oral discourse, is the organization of tables, formulas, or charts
which can be portrayed only in written form (Crystal 1995:291).
Naturally, this division into two ways of producing discourse
is quite straightforward, yet, it is possible to combine the two like,
for example, in the case of a lesson, when a teacher explains
something writing on the blackboard, or when a speaker prepares
detailed notes to be read out during his speech. Moreover, some of
the foregoing features are not so explicit in the event of
sophisticated, formal speech or a friendly letter.
Discourse Expressed Formally and Informally
The difference in construction and reception of language was
the basis of its conventional distinction into speaking and writing.
Nevertheless, when the structure of discourse is taken into
257
consideration more essential division into formal and informal
communicative products gains importance. Formal discourse is more
strict in that it requires the use of passive voice, lack of contracted
forms together with impersonality, complex sentence structure and,
in the case of the English language, vocabulary derived from Latin.
That is why formal spoken language has many features very similar
to written texts, particularly absence of vernacular vocabulary and
slang, as well as the employment of rhetorical devices to make
literary-like impact on the listener (Wiśniewski, 2006).
Informal discourse, on the other hand, makes use of active
voice mainly, with personal pronouns and verbs which show feelings
such as 'I think', 'we believe'. In addition, contractions are frequent in
informal discourse, no matter if it is written or spoken. Consequently
it may be said that informal communicative products are casual and
loose, while formal ones are more solemn and governed by strict
rules as they are meant to be used in official and serious
circumstances.
The relation of the producer of the message and its receiver, the
amount of addressees and factors such as public or private occasion
are the most important features influencing selecting either formal or
informal language. Therefore, it is not unreasonable to assume that
the contemporary learner, who may easily travel and use his
linguistic skills outside class, will encounter mainly informal
discourse, which due to its flexibility and unpredictability might be
the most difficult to comprehend. Accordingly, it seems rational to
teach all varieties of language relying on authentic oral and written
texts (Cook 1990:50).
THE SIGNIFICANCE OF DISCOURSE ANALYSIS IN
LANGUAGE TEACHING AND LEARNING
To attain a good command of a foreign language learners
should either be exposed to it in genuine circumstances and with
natural frequency, or painstakingly study lexis and syntax assuming
that students have some contact with natural input. Classroom
discourse seems to be the best way of systematizing the linguistic
code that learners are to acquire. The greatest opportunity to store,
develop and use the knowledge about the target language is arisen by
exposure to authentic discourse in the target language provided by
the teacher (Dakowska 2001:86).
Language is not only the aim of education as it is in the case
of teaching English to Polish students, but also the means of
259
schooling by the use of mother tongue. Having realized that
discourse analysts attempted to describe the role and importance of
language in both contexts simultaneously paying much attention to
possible improvement to be made in these fields.
It has also been settled that what is essential to be successful
in language learning is interaction, in both written and spoken form.
In addition, students' failures in communication which result in
negotiation of meaning, requests for explanation or reorganization of
message contribute to language acquisition. One of the major
concerns of discourse analysts has been the manner in which
students ought to be involved in the learning process, how to control
turn-taking, provide feedback as well as how to teach different skills
most effectively on the grounds of discourse analysis' offerings
(Trappes-Lomax 2004:153).
Application of Discourse Analysis to Teaching Grammar
There are a number of questions posed by discourse analysts
with reference to grammar and grammar teaching. In particular, they
are interested in its significance for producing comprehensible
communicative products, realization of grammar items in different
languages, their frequency of occurrence in speech and writing
which is to enable teaching more natural usage of the target
language, as well as learners' native tongue (McCarthy 1991:47).
While it is possible to use a foreign language being unaware
or vaguely aware of its grammatical system, educated speakers
cannot allow themselves to make even honest mistakes, and the more
sophisticated the linguistic output is to be the more thorough
knowledge of grammar gains importance. Moreover, it is essential
not only for producing discourse, but also for their perception and
comprehension, as many texts take advantage of cohesive devices
which contribute to the unity of texts, but might disturb their
understanding by a speaker who is not aware of their occurrence.
Anaphoric reference, which is frequent in many oral and
written texts, deserves attention due to problems that it may cause to
learners at various levels. It is especially important at an early stage
of learning a foreign language when learners fail to follow overall
meaning turning much attention to decoding information in a given
clause or sentence. Discourse analysts have analyzed schematically
occurring items of texts and how learners from different
backgrounds acquire them and later on produce. Thus, it is said that
261
Japanese students fail to distinguish the difference between he and
she, while Spanish pupils have problems with using his and your.
Teachers, being aware of possible difficulties in teaching some
aspects of grammar, should pay particular attention to them during
the introduction of the new material to prevent making mistakes and
errors (McCarthy 1991:36).
The most prominent role in producing sophisticated
discourse, and therefore one that requires much attention on the part
of teachers and learners is that of words and phrases which signal
internal relation of sections of discourse, namely conjunctions.
McCarthy (1991) claims that there are more than forty conjunctive
words and phrases, which might be difficult to teach. Moreover,
when it comes to the spoken form of language, where and, but, so,
then are most frequent, they may take more than one meaning, which
is particularly true for and. Additionally, they not only contribute to
the cohesion of the text, but are also used when a participant of a
conversation takes his turn to speak to link his utterance to what has
been said before (McCarthy 1991:48).
The foregoing notions that words crucial for proper
understanding of discourse, apart from their lexical meaning, are also
significant for producing natural discourse in many situations
support the belief that they should be pondered on by both teachers
and students. Furthermore, it is advisable to provide learners with
contexts which would exemplify how native users of language take
advantage of anaphoric references, ellipses, articles and other
grammar related elements of language which, if not crucial, are at
least particularly useful for proficient communication (McCarthy
1991:62).
Application of Discourse Analysis to Teaching Vocabulary
What is probably most striking to learners of a foreign
language is the quantity of vocabulary used daily and the amount of
time that they will have to spend memorizing lexical items. Lexis
may frequently cause major problems to students, because unlike
grammar it is an open-ended system to which new items are
continuously added. That is why it requires close attention and,
frequently, explanation on the part of the teacher, as well as patience
on the part of the student.
Scholars have conducted in-depth research into techniques
employed by foreign language learners concerning vocabulary
263
memorization to make it easier for students to improve their
management of lexis. The conclusion was drawn that it is most
profitable to teach new terminology paying close attention to context
and co-text that new vocabulary appears in which is especially
helpful in teaching and learning aspects such as formality and
register. Discourse analysts describe co-text as the phrases that
surround a given word, whereas, context is understood as the place
in which the communicative product was formed (McCarthy
1991:64).
From studies conducted by discourse analysts emerged an
important idea of lexical chains present in all consistent texts. Such a
chain is thought to be a series of related words which, referring to the
same thing, contribute to the unity of a communicative product and
make its perception relatively easy. Additionally, they provide a
semantic context which is useful for understanding, or inferring the
meaning of words, notions and sentences. Links of a chain are not
usually limited to one sentence, as they may connect pairs of words
that are next to one another, as well as stretch to several sentences or
a whole text. The relation of words in a given sequence might be that
of reiteration or collocation, however, analyst are reluctant to denote
collocation as a fully reliable element of lexical cohesion as it refers
only to the likelihood of occurrence of some lexical items.
Nevertheless, it is undeniably helpful to know collocations as they
might assist in understanding of communicative products and
producing native-like discourse (McCarthy 1991:65).
Since lexical chains are present in every type of discourse it
is advisable to familiarize learners with the way they function in, not
merely because they are there, but to improve students' perception
and production of expressive discourse. Reiteration is simply a
repetition of a word later in the text, or the use of synonymy, but
what might require paying particularly close attention in classroom
situation is hyponymy. While synonymy is relatively easy to master
simply by learning new vocabulary dividing new words into groups
with
similar
meaning,
or
using
thesauri,
hyponymy
and
superordination are more abstract and it appears that they require
tutelage. Hyponym is a particular case of a more general word, in
other words a hyponym belongs to a subcategory of a superordinate
with narrower meaning, which is best illustrated by an example:
Brazil, with her two-crop economy, was even more severely hit by
the Depression thanother Latin American states and the country was
on the verge of complete collapse (Salkie 1995:15). In this sentence
the word Brazil is a hyponym of the word country - its superordinate.
265
Thus, it should not be difficult to observe the difference between
synonymy and hyponymy: while Poland, Germany and France are all
hyponyms of the word country, they are not synonymous. Discourse
analysts imply that authors of communicative products deliberately
vary discursive devices of this type in order to bring the most
important ideas to the fore, which in case of English with its wide
array of vocabulary is a very frequent phenomenon (McCarthy 1991,
Salkie 1995).
One other significant contribution made by discourse analysts
for the use of vocabulary is noticing the omnipresence and
miscellaneous manners of expressing modality. Contrary to popular
belief that it is conveyed mainly by use of modal verbs it has been
proved that in natural discourse it is even more frequently
communicated by words and phrases which may not be included in
the category of modal verbs, yet, carry modal meaning. Lexical
items of modality inform the participant of discourse not only about
the attitude of the author to the subject matter in question (phrases
such as I believe, think, assume), but they also give information
about commitment, assertion, tentativeness (McCarthy 1991:85).
Discourse analysts maintain that knowledge of vocabularyconnected discourse devices supports language learning in diverse
manners. Firstly, it ought to bring students to organize new items of
vocabulary into groups with common context of use to make them
realize how the meaning of a certain word might change with
circumstances of its use or co-text. Moreover, it should also improve
learners' abilities to choose the appropriate synonym, collocation or
hyponym (McCarthy 1991:71).
Application
of
Discourse
Analysis
to
Teaching
Text
Interpretation
Interpretation of a written text in discourse studies might be
defined as the act of grasping the meaning that the communicative
product is to convey. It is important to emphasize that clear
understanding of writing is reliant on not only what the author put in
it, but also on what a reader brings to this process. McCarthy (1991)
points out that reading is an exacting action which involves
recipient's knowledge of the world, experience, ability to infer
possible aims of discourse and evaluate the reception of the text.
267
Painstaking research into schemata theory made it apparent
that mere knowledge of the world is not always sufficient for
successful discourse processing. Consequently, scholars dealing with
text analysis redefined the concept of schemata dividing it into two:
content and formal schemata. Content, as it refers to shared
knowledge of the subject matter, and formal, because it denotes the
knowledge of the structure and organization of a text. In order to aid
students to develop necessary reading and comprehension skills
attention has to be paid to aspects concerning the whole system of a
text, as well as crucial grammar structures and lexical items. What is
more, processing written discourse ought to occur on global and
local scale at simultaneously, however, it has been demonstrated that
readers employ different strategies of reading depending on what
they focus on (McCarthy 1991:168).
Top-down and bottom-up text processing
Distinguishing noticeably different approaches to text
processing led to distinction of manners of attending to written
communicative products. Bottom-up processes are those which are
involved in assimilating input from the smallest chunks of discourse:
sounds in speech and letters in texts, afterwards moving to more and
more general features. This technique is frequently applied by lowerlevel learners who turn much attention to decoding particular words,
thus losing the more general idea, that is the meaning of a given
piece of writing. In the same way learning a new language begins:
first the alphabet, then words and short phrases, next simple
sentences, finally elaborate compound sentences. While it is
considered to be a good way of making learners understand the
language, a wider perspective is necessary to enable students to
successfully produce comprehensible discourse (Cook 1990,
McCarthy 1991).
Alternatively, top-down processing starts with general
features of a text, gradually moving to the narrower. This approach
considers all levels of communicative products as a total unit whose
elements work collectively, in other words, it is more holistic. Not
only does the information in a text enable readers to understand it,
but it also has to be confronted with recipient's former knowledge
and expectations which facilitate comprehension. It is important to
make students aware of these two ways of dealing with written
discourse and how they may be exploited depending on the task.
When learners are to get acquainted with the main idea of a
particular communicative product they should take advantage of top-
269
down approach, while when answering detailed true-false questions
they would benefit from bottom-up reading (Cook 1990, McCarthy
1991).
Types of text
Obviously, all texts have a certain feature in common,
namely they are indented to convey some meaning. This function,
however, might be fulfilled in a number of different ways: a road
sign 'stop', and a six hundred pages long novel are both texts which
might serve that purpose, yet, there are certain characteristics that
distinguish them. The above example presents the idea somewhat in
the extreme, although, enumerating several other common types of
texts might affirm that the notion of text is a very broad one and is
not limited to such varieties as those that can be found in language
course books (Cook 1990, Crystal 1995).
Differences between texts might be striking, while menu is
usually easy to read, legal documents or wills are not. All of them,
however, have certain features that others lack, which if explained by
a qualified teacher might serve as a signpost to interpretation.
Additionally, the kind of a given text might also provide information
about its author, as for example in the case of recipes, warrants or
manuals, and indirectly about possible vocabulary items and
grammar structures that can appear in it, which should facilitate
perception of the text. Having realized what kind of passage learners
are to read, on the basis of its title they should be able to predict the
text's content, or even make a list of vocabulary that might appear in
the communicative product. With teacher's tutelage such abilities are
quickly acquired which improves learners' skills of interpretation and
test results (Cook 1990, McCarthy 1991, Crystal 1995)
Patterns in text
Having accounted for various kinds of associations between
words, as well as clauses and sentences in discourse, the time has
come to examine patterns that are visible throughout written
communicative products. Patterning in texts contributes to their
coherence, as it is thanks to patterns that writing is structured in a
way that enables readers to easily confront the received message
with prior knowledge. Salkie (1995) indicates that the majority of
readers unconsciously makes use of tendencies of arranging texts to
approach information.
271
Among most frequently occurring patterns in written
discourses there are inter alia claim-counterclaim, problem-solution,
question-answer
or
general-specific
statement
arrangements.
Detailed examination of such patterning revealed that problemsolution sequence is frequently accompanied by two additional parts,
namely background (in other words introduction) and evaluation
(conclusion). While in some elaborate texts the background and the
problem might be presented in the same sentence, in other instances when reader is expected to be familiar with the background, it might
not be stated in the text itself. Although both cohesive devices and
problem-solution patterns often occur in written communicative
products only the former are designated as linguistic means, since
patterning, when encountered, has to be faced with assumptions,
knowledge and opinion of the reader (McCarthy 1991, Salkie 1995).
One other frequently occurring arrangement of texts is based
on general-specific pattern which is thought to have two variations.
In the first one a general statement is followed by a series of more
specific sentences referring to the same broad idea, ultimately
summarized by one more general remark. Alternatively, a general
statement at the beginning of a paragraph might be followed by a
specific statement after which several more sentences ensue, each of
which is more precise than its predecessor, finally going back to the
general idea (McCarthy 1991:158).
As McCarthy (1991: 161) points out, the structure of patterns
is fixed, yet the number of sentences or paragraphs in a particular
part of a given arrangement might vary. Furthermore, one written
text might contain several commonplace patterns occurring
consecutively, or one included in another. Therefore, problemsolution pattern present in a text might be filled with general-specific
model within one paragraph and claim-counterclaim in another. As
discourse analysts suggest making readers aware of patterning might
sanitize them to clues which enable proper understanding of written
communicative products.
DISCOURSE ANALYSIS: GRAMMATICAL COHESION &
LEXICAL COHESION
Grammatical Cohesion
Our feeling that a particular stretch of language in some way
hangs together, or has unity, (that it is, in other words, discourse)
cannot be accounted for in the same way as our feeling for the
acceptability of a sentence. In order to account for discourse, we
273
need to look at feature outside the language: at the situation, the
people involved, what they know and what they are doing. These
facts enable us to construct stretches of language as discourse; as
having a meaning and unity for us. The way we recognize correct
and incorrect sentences is different. We can do this through our
knowledge of grammar without reference to outside fact (Cook,
1989:14).
Again Cook (1989: 14) insists that we can describe the two
ways of approaching language as contextual, referring to facts
outside language, and formal, referring to facts inside language. A
way of understanding this difference may be to think of formal
features as in some way built up in our minds from the black marks
which form writing on the page, or from the speech sounds picked
up by our ears, while contextual features are somewhere outside this
physical realization of the language – in the world, or pre-existing in
the minds of the participants. Stretches of language treated only
formally are referred to as text.
Now although it is true that we need to consider contextual
factors to explain what it is that creates a feeling of unity in stretches
of language of more than one sentence, we can not say that there are
no formal links between sentences in discourse. Here, we try to
categorize these formal links and then examine how far they will go
in helping to explain why a succession of sentences is discourse, and
not just a disconnected jumble. Formal links between sentences and
between
clauses
(formal
links
within
text
that
links
sentences/clauses) are known as cohesive devices. The purpose of
cohesive devices, meanwhile, is to hold parts of the discourse
together in order to achieve overall unity.
Among the most common cohesive devices exposed by Cook
(1989: 15 – 21); McCharthy (1991: 35 – 46); Carter (1997) and
Halliday and Hasan (1976) are:
a. verb form
b. parallelism
c. referring expressions
d. substitution
e. ellipsis
f. conjunction
275
(Note: Cook, McCharty and Carter include lexical cohesion as a part
of grammatical cohesion but Halliday and Hasan exclude it since it is
independent)
a. Verb form
Cohesion is attained in a piece of discourse by the forms of
the verbs used. The form of the verb in one sentence limits the
choice of verbs form in the next. Depending on the time frame to
which the piece of discourse alludes, the verbs used in it need to
conform to the particular tense.
e.g.
Muaz: Right, who’s going to arrange the transport?
Well, someone’s gotta do it.
Rifau: I’m not going to.
Muaz: Come on, will you?
All the verbs are in the present. There is an apparent degree of
formal connection between them, a way in which the first tense
conditions all the others.
It would be quite strange if the conversation had been:
Muaz: Right, who’s going to arrange the transport?
Well, someone had gotta do it.
Rifau: I shan’t have been going to.
Muaz: Don’t! Come on, will you?
b. Parallelism
Another link in discourse is created by parallelism, a device which
effects/suggests a connection simply because the form of one
sentence or clause repeats the form of another. This is often used in
speeches, prayers, poetry and advertisements. It can have a powerful
emotional effect, and it is also very useful as a memory aid (aidememoire).
e.g.

Pregnant women get emotional very easily. They get
emotional if you raise your voice at them. They get emotional
if you do something romantic. They get emotional if you
notice they are eating lots. They get emotional if you tell
them they look beautiful. They get emotional if you are too
protective. They get emotional if you are not protective
enough.
277

‘Teach us, Good Lord, to give and not to count the cost, to
fight and not to heed the wounds, to toil and not to seek for
rest, to labor and to ask for no reward, save that of knowing
that we do they will’ (St Richard’s Prayer)
Parallelism, which achieves cohesion by an echo of form, is not
restricted to grammatical form.
e.g.
a.
Georgie Porgie pudding and pie
Kissed the girls and made them cry
When the boys came out to play
Georgie Porgie ran away
b.
Rushdhy: We would like to take this opportunity to wish you the best
of
luck
Zakittey: Go break a leg man!
c. Referring expressions (Reference)
These are words whose meaning can only be discovered by
referring to other words or to elements of the context which are clear
to both sender and receiver. The most obvious example of them is
third person pronouns (she/her/hers/herself; he/him/his/himself;
it/its/itself; they/them/their/their/themselves).
eg.
There was a banana on the table. So I ate it.
We would assume the speaker had eaten the banana, not the table
(even though the word table is nearer) because we know that people
are more likely to eat banana than table.
The opening lines of a famous English novel, Jude the
Obscure, by Thomas Hardy, show difference types of reference at
work:
The schoolmaster was leaving the village, and everybody
seemed sorry. The miller at Cresscombe lent him the small
white tilted cart and horse to carry his goods to the city of his
destination, about twenty miles off, such a vehicle proving of
quite sufficient size for the departing teacher’s effects.
The italicized items refer. For the text to be coherent, we
assume that him in ‘lent him the small white tilted cart’ is the
schoolmaster introduced earlier; likewise, his destination is the
schoolmaster’s. Referents for him and his can be confirmed by
looking back in the text; this is called anaphoric reference. Such a
also links back to the cart in the previous sentence. The novel opens
279
with the schoolmaster leaving the village. Which schoolmaster?
Which village?. It needs more information
Referring expressions can be anaphoric, cataphoric or exophoric
Anaphoric references (looking backward) are those that refer
back to elements mentioned earlier in the discourse (ana means
“upwards” and phor means “to carry.”). Exercises which involve
looking back in texts to find the referent of, for example, a pronoun,
have long been common in first and second language teaching and
testing. Usually items such as he/she or them can be decoded without
major difficulty; other items such as it and this may be more
troublesome because of their ability to refer to longer stretches of
text and diffuse propositions not necessarily paraphrasable by any
direct quotation from the text. Problems can also arise where lowerlevel learners are so engaged in decoding the individual utterance,
clause or sentence that they lose sight of the links back to earlier
ones.
e.g.
a. It rained day and night for two weeks. The basement
flooded and everything was under water. It spoilt all our
calculation.
Here it seems to mean ‘the events of two weeks’, or ‘the fact that it
rained and flooded’, that is, the situation as a whole rather than any
one specified entity in that situation.
b. The passengers presented themselves for checking before
boarding the plane.
Cataphoric references (looking forward) are those that refer
to
elements
that
are
mentioned
after
them
(cata
means
‘downwards’/’forward’). Cataphoric reference is the reverse of
anaphoric reference and relatively straightforward, but language
learners may lack awareness or confidence to put it into use in
constructing texts, and need to have the feature explicitly taught or
exercised. There is, too, the danger of its overuse or its use in
unnatural contexts. As always, it is a question of training the learner
to observe features of language above sentence level where these
might not necessarily be automatically transferred from L1,
especially since, in English, reference often involves the definite
281
article and demonstratives, which do not translate easily into many
other languages.
Consider these opening lines of a news article:
She claims Leo Tolstoy as a distant cousin. Her grandfather
was Alexei Tolstoy – the famous ‘Red Count’ who sided with
Lenin’s revolutionaries. Now, Tatyana Tolstaya has put pen
to paper, in her case to demonstrate that someone from the
family can write compactly. In her stories of ten to twelve
typewritten pages, ‘I somehow try to show the whole life of a
person from birth to death,’ she says.
(Newsweek, 21 Sep 1987: 12)
We do not establish who she is until the second sentence. Forwardlooking or cataphoric reference of this kind often involves pronouns
but it can involve other reference items too, such as the definite
article:
The trip would hardly have been noteworthy, except for the
man who made it. In mid-July a powerful American financier
flew to Mexico City for a series of talks with high- level
government officials, including President Miguel de la
Madrid and his finance minister, Gustavo Petricioli.
(Newsweek, 21 Sep 1987: 44)
Both examples of cataphoric reference were found in the same issue
of Newsweek, which underlines the most characteristic function of
cataphoric reference: to engage and hold the reader’s attention with
a ‘read on and find out’ message. In news stories and in literature,
examples of cataphoric reference are often found in the opening
sentences of the text.
e.g.
This is what she said – our teacher is a raving lunatic!
Exophoric references are those that refer to elements outside
the text (exo means ‘outside’). We have mentioned the possibility of
referring ‘outward’ from texts to identify the referents of reference
items when backward or anaphoric reference does not supply the
necessary information. Outward, or exophoric reference often directs
us to the immediate context, as when someone says ‘leave it on the
table please’ about a parcel you have for them. Sometimes the
referent is not in the immediate context but is assumed by the
speaker/writer to be part of a shared world, either in terms of
knowledge or experience. In English the determiners often act in this
way:
283
a. The government are to blame for unemployment.
b. She was using one of those strimmers to get rid of the
weeds.
It would be odd if someone replied to (a) with the question ‘Which
government?’. It is assumed by the speaker that the hearer will know
which one, usually ‘our government’ or ‘that of the country we are in
/ are talking about’. The same sort of exophoric reference is seen in
phrases such as the Queen, the Pope, the army, and in sentences such
as ‘We always take the car since we can just put the kids, the dog
and the luggage into it’. A learner whose L1 has no exact equivalent
to English the may need to have this central use of the article taught
explicitly. On the other hand, speakers of languages with extended
use of definite articles to cover general nouns in situations where
these would not be marked as definite in English sometimes produce
utterances which, to the English ear, seem to be making exophoric
reference, such as ‘Do you like this folk music?’ when no music is to
be heard.
Exophoric reference (especially in the press) is often to a
‘world of discourse’ connected with the discourse of the moment,
but not directly. An example of a text referring to such an assumed
shared world is extract, which talks of ‘the entire privatization
program’; readers are assumed to know that this refers to the British
government’s sell-off in 1989 of the entire public water service into
private hands:
Eighty per cent of Britain’s sewage works are breaking
pollution laws, according to a report to be published this
week.
The cost of fulfilling a government promise to clean them up
will run into billions, and put the entire privatization
program at risk.
(The Observer, 4 December 1988: 3)
Exophoric (exophora – adj) references will often be to a
world shared by sender and receiver of the linguistic message,
regardless of cultural background, but equally often, references will
be culture-bound and outside the experiences of the language learner
(e.g. British references of the City, the Chancellor, and so on). In
these cases the learner will need to consult some source of
encyclopaedic information or ask an informant. This aspect of
language learning is a gradual familiarization with the cultural
285
context of L2. Language teachers and materials writers will need to
monitor the degree of cultural exophoric references in texts chosen
for teaching to ensure that the referential burden is not too great.
More examples and explanations
In brief, we can say that If the pronoun is referring back to
something, this is called anaphoric reference; if the pronoun is
referring to something coming later, this is called cataphoric
reference. Meanwhile, where a reference items moves us outside a
text, so that we can only make full sense of the text by referring to its
context, this is called exophoric reference (The use of a pronoun or
other word or phrase to refer to someone or something outside the
text. Adjective: exophoric = extralinguistic); where we stay within
the text, not needing any support from outside, this is called
endophoric reference (intalinguistic) (Carter, 1997: 189 -198). Here
is an example of each:

Tom said that he was going home (anaphoric reference)

I couldn’t believe it – the house was a complete wreck
(cataphoric reference)

"That man over there says that women need to be helped into
carriages, and lifted over ditches, and to have the best place
everywhere. . . . (exophoric reference)

The monkey took the banana and ate it. ("It" is anaphoric
under the strict definition - it refers to the banana).

Pam went home because she felt sick. ("She" is anaphoric - it
refers to Pam).

What is this? ("This" can be considered exophoric - it refers
to some object or situation near the speaker).

The dog ate the bird and it died. ("It" is anaphoric and
ambiguous - did the dog or bird die?).

The Prime Minister of New Zealand visited us yesterday. The
visit was the first time she had come to New York since
1998. (anaphoric reference)
When taken in context, anaphor resolution can become
increasingly complex. Consider the three examples:
We gave the bananas to the monkeys because they were
hungry.
We gave the bananas to the monkeys because they were ripe.
We gave the bananas to the monkeys because they were here.
287
In the first sentence, "they" refers to "monkeys", whereas in the
second
sentence,
"they"
refers
to
"bananas".
A
semantic
understanding that monkeys get hungry, while bananas become ripe
is necessary when resolving this ambiguity. Since this type of
understanding is still poorly implemented in software, automated
anaphora resolution is currently an area of active research within the
realm of natural language processing. The third sentence isn't easily
resolved either way.

I saw Sally yesterday. She was lying on the beach ("she" is
an endophoric expression because it refers to something
already mentioned in the text, i.e. "Sally").
By contrast, "She was lying on the beach," if it appeared by itself,
has an exophoric expression; "she" refers to something that the
reader is not told about. That is to say, there is not enough
information in the text to independently determine to whom "she"
refers. It can refer to someone the speaker assumes his audience has
prior knowledge of or it can refer to a person he is showing to his
listeners. Without further information, in other words, there is no
way of knowing the exact meaning of an exophoric term.
Notice:
"In discourse in general, the third person pronouns may be
either endophoric, referring to a noun phrase within the
text, . . . or exophoric, referring to someone or something
manifest to the participants from the situation or from their
mutual knowledge ('Here he is,' for example, on seeing
someone who both sender and receiver are expecting) "In
songs, 'you' . . . is multi-exophoric, as it may refer to many
people in the actual and fictional situation. Take for example:
Well in my heart you are my darling,
At my gate you're welcome in,
At my gate I'll meet you darling,
If your love I could only win.
(Traditional)
This is the plea of one lover to another. . . . The receiver of
the song is apparently overhearing one half of a dialogue. 'I'
is the singer, and 'you' is her lover. Alternatively, and most
frequently, especially away from live performance, the
receiver projects herself into the persona of the addresser and
hears the song as though it is her own words to her own lover.
Alternatively, the listener may project herself into the
persona of the singer's lover and hear the singer addressing
her."
(Guy Cook, The Discourse of Advertising. Routledge, 1992)
A type of exophora, homophora relates to a generic phrase
that obtains a specific meaning through knowledge of its
context; a specific example of homophora can variably be a
"homophor" or a "homophoric reference". For example, the
289
meaning of the phrase "the Queen" may be determined by the
country in which it is spoken. Because there are many
Queens throughout the world, the location of the speaker
provides the extra information that allows an individual
Queen to be identified.
The precise origin of the term is not fully clear, but it is
probably intended to suggest a referring expression that
always has the same (Greek hómos) referent (within a given
cultural context, of course). It (or rather homophoric) seems
to have been first used in the influential book by M.A.K.
Halliday and R. Hasan, Cohesion in English (Longman, 1976,
pp. 71 and 73).
Repetition of words can create the same sort of chain as
pronouns, and there are sometimes good reasons for preferring it. In
Britain, mother tongue learners of English are discouraged from
using repetition on the ground that it is ‘bad style’, and encouraged
to use a device known as ‘elegant repetition’ where synonymous or
more general words or phrases are used. So instead of writing:

The pineapple … the pineapple … the pineapple … the
pineapple ….
They might write:

The pineapple … the luscious fruit … our meal … the
tropical luxury …
…

Dice the mangoes into large cubes. Arrange the diced mango
cubes on a plate, in any pattern you like. Pour whipped cream
over the diced mango cubes.
Referring expressions, repetitions and elegant repetitions all
establish cohesion by creating ‘chains’ of connected words/clauses
running through discourse. Such lexical chains need not necessarily
consist of words which mean the same, however. They may also be
created by virtue of some formal semantic connection (good, for
example, associates with its opposite bad; animal with any example
of an animal like horse; violin with orchestra of which it is a part),
or it may be because words are felt to belong to some more vaguely
defined lexical group (rock star; world tour; millionaire; yacht).
This last kind of connection, though it is sometimes treated as a kind
of cohesion, is really too dependent upon individual experience and
knowledge to be treated as a formal link.
291
e.g.
Four legs good. Two legs bad.
Here are seven kinds of repetitions described by Carter et al (1997:
172):
Varieties
Examples
Direct repetition (exactly the juice … juice
same word repeated)
Synonyms, or near-synonyms saving … investement
(use of words with similar
meanings).
Antonyms (opposites).
Loved … hated
Specific-general
reference
(words referring to the same
thing or person, but where one
has more detail than the other).
Ordered series (words that we
Put them into … spoon out
a fire proof dish … the dish
(going from general to specific):
apples … cooked apples
regulo 6 … 3; 10 minutes … 15
know as a set series – for minutes
example, the days of the week,
months of the year, or the
seasons).
Whole-part (where one term apples … pips
names a part of an item that the
other word describes in full).
e. Substitution
Another type of formal link (cohesive device) between
sentences is the substitution of words like do or so for a word or
group of words which have appeared in an earlier sentence. It would
be very long-winded if we had always to answer a question like Do
you like mangoes? with a sentence like Yes I like mangoes or Yes I
think I like mangoes. It is much quicker, and it means the same, if we
say Yes I do or Yes I think so. Unfortunately, much traditional
language teaching, in its zeal for practicing verb tenses and using
new vocabulary, has connected exclusively on longer forms (Answer
with a full sentence please!) and deprived students of briefer, more
authentic options.
e.g.
Shahid: Which shirt shall I wear today?
Ihusana: The maroon one.
Shahid: You also like that one then?
293
Ihusana: Yes, I do! I think It suits you better.
Shahid: I think so too.
f. Ellipsis
Ellipsis (substitution by zero) is the omission of elements
normally required by the grammar which the speaker/writer assumes
are obvious from the context and therefore need not be raised. This is
not to say that every utterance which is not fully explicit is elliptical;
most messages require some input from the context to make sense of
them. Ellipsis is distinguished by the structure having some missing
element. If two people have to stack and label a pile of items and one
says to the other ‘you label and I’ll stack’, the fact that label and
stack are usually transitive verbs requiring an object in the surface
structure is suspended because the context ‘supplies’ the object.
Another way of saying this is, of course, that structures are only fully
realized when they need to be, and that ellipsis is a speaker choice
made on a pragmatic assessment of the situation, not a compulsory
feature when two clauses are joined together.
e.g.
Zubaida: Have you been shopping girls? What did you buy on your
[shopping] spree?
Ziya: I bought jewelry and Zee [bought] clothes,
Zee: Clothes? You already have more than enough [clothes], and you
bought more [clothes]?
We shall concentrate here on the type of ellipsis where the
‘missing’ element is retrievable verbatim from the surrounding text,
rather in the way that anaphoric and cataphoric references are, as
opposed to exophoric references. For example:
a. The children will carry the small boxes, the adults [will
carry] the large ones.
where ‘will carry’ is supplied from the first clause to the second.
This type of main-verb ellipsis is anaphoric; in English we would not
expect:
b. The children the small boxes, the adults the large ones.
English does have the kind of cataphoric ellipsis suggested by our
rejected example (b), but usually only in front-placed subordinate
clauses:
295
c. If you could, I’d like you to be back here at five thirty.
English has broadly three types of ellipsis: nominal, verbal
and clausal. Nominal ellipsis often involves omission of a noun
headword:
d. Nelly liked the green tiles; myself I preferred the blue.
Ellipsis within the verbal group may cause greater problems.
Two very common types of verbal-group ellipsis are what Thomas
(1987) calls echoing and auxiliary contrasting. Echoing repeats an
element from the verbal group:
e. a. Will anyone be waiting?
b. Jim will, I should think.
Contrasting is when the auxiliary changes:
f. a. Has she remarried?
b. No, but she will one day, I’m sure.
Thomas also makes the point that in English, varying degrees of
ellipsis are possible within the same verbal group:
g. a. Should anyone have been told?
b. John should/should have/should have been.
These variants are not directly translatable to other languages and
will have to be learned.
With clausal ellipsis in English, individual clause elements
may be omitted; especially common are subject pronoun omissions
(‘doesn’t matter’, ‘hope so’, ‘sorry’, can’t help you’, etc.) Whole
stretches of clausal components may also be omitted:
h. He said he would take early retirement as soon as he could
and he has.
For this type of sentence, many languages will require at the very
least some kind of substitute for the main verb and an object pronoun
such as to produce a form roughly equivalent to ‘He said he would
take early retirement as soon as he could and he has done it.’
297
Ellipsis not only creates difficulties in learning what
structural omissions are permissible, but also does not seem to be
readily used even by proficient learners in situations where native
speakers naturally resort to it.
g. Conjunction
We include conjunction here in our discussion of
grammatical contributions to textuality even though it is somewhat
different from reference, ellipsis and substitution. A conjunction
does not set off a search backward or forward for its referent, but it
does presuppose a textual sequence, and signals a relationship
between segments of the discourse.
Discourse analysts ask the same sorts of questions about
conjunctions as they do about other grammatical items: what roles do
they play in creating discourse, do the categories and realizations
differ from language to language, how are they distributed in speech
and writing, what restrictions on their use are there which are not
reflected purely through sentence analysis, and what features of their
use are inadequately explicated in conventional grammars?
Here are some conjunctions described by Carter (1997: 217);
with a brief explanation for each group of what they are telling the
reader to do:
Type of
Meaning
Examples
Conjunction
additives/alternatives
add/give
an and, or, furthermore,
alternative
in addition, likewise,
in other words
Adversative
contradict,
but,
yet,
concede
however,
though,
on
the
for
this
contrary
Causal
one
idea/event so,
causes another
then,
reason, consequently,
it follows that, as a
result
Temporal
one event follows one day, then, finally,
another in time
up to now, the next
299
day
Continuatives
please continue to well, now, of course,
follow the text
anyway, surely, after
all
Additionally, Halliday (1985: 302-9) offers a scheme for the
classification of conjunctive relations and includes phrasal types as
well as single-word everyday items such as and, but, or, etc. Here is
a simplified list based on Halliday’s three category headings of
elaboration, extension and enhancement.
Type
Elaboration
Sub-type
Apposition
Examples
in other words
Extension
clarification
Addition
or rather
and/but
Enhancement
variation
spatio-temporal
alternatively
there/previously
causal-conditional
consequently/in
case
e.g.
that

First, peel and clean the potatoes. Then bake them. A
conventional oven is best. However, a microwave oven will
also do. Potatoes burn easily, so make sure you keep a close
watch on the timer!

He was insensitive to the group’s needs. Consequently there
was a lot of bad feeling. (single word conjunction)

He was insensitive to the group’s needs. As a consequence
there was a lot of bad feeling. (adverbial phrase as
conjunction)

As a consequence of his insensitivity to the group’s needs,
there was a lot of bad feeling. (adverbial phrase plus
nominalization)

The bad feeling was a consequence of his insensitivity to the
group’s need. (lexical item within the predicate of the clause)

She’s intelligent. And she’s very reliable. (additive)

I’ve lived here ten years and I’ve never heard of that pub.
(adversative: but could substitute)

He fell in the river and caught a chill. (causal)

I got up and made my breakfast. (temporal sequence)
301
6.1
Lexical Cohesion
Halliday and Hasan second categorization is lexical cohesion
(the semantic relationship between lexical items) which they classify
into two types: reiteration and collocation. Reiteration is classified
into four types: (a) the same word, (b) a synonym/near-synonym, (c)
a superordinate, and (d) a general word. For example, ‘a boy’ can be
replaced in the following sentences with ‘the boy’ (the same word),
‘the lad’ (a synonym/near-synonym), ‘the child’ (a superordinate),
and ‘the idiot’ (a general word) (1976: 279-280). Meanwhile, they
recognize collocation as an important part of creating cohesion in
connected text. Collocation refers to the semantic and structural
relation among words, which native speakers can use subconsciously
for comprehension or production of a text. They argue the case of
collocation as follows:
The cohesive effect … depends not so much on any
systematic relationship as on their tendency to share the same
lexical environment, to occur in COLLOCATION with one
another. In general, any two lexical items having similar
patterns of collocation – that is, tending to appear in similar
context – will generate a cohesive force if they occur in
adjacent sentences (Halliday & Hasan, 1976: 286).
A ‘cohesive force’ will produce a ‘cohesive tie,’ which is the
relationship between a cohesive item and the item it presupposed in a
text. In other words, collocational links between lexical items create
cohesion.
In response to Halliday and Hasan (1976), other researchers
have discussed lexical cohesion (Gutwinski 1976, Carell 1984, Hoe
1991, Martin 1992, and Cook 1994). However, cohesion can be
concluded as “the means by which texts are linguistically connected”
(Carter 1998: 80). It is significant to recognize that lexical cohesion
cannot exist without sentences. That is cohesive words should be
discussed not only as the meaning relations which hold between
items, but also as the explicit expression of those meaning relation
within a text. Ultimately, it is necessary to consider cohesion as “a
set of discourse semantic systems” (Martin 2001: 37).
In relation to lexical cohesion, repetition of words can create
the same sort of chain as pronouns, and there are sometimes good
reasons for preferring it. In Britain, mother tongue learners of
English are discouraged from using repetition on the ground that it is
303
‘bad style’, and encouraged to use a device known as ‘elegant
repetition’ where synonymous or more general words or phrases are
used. So instead of writing:

The pineapple … the pineapple … the pineapple … the
pineapple ….
They might write:

The pineapple … the luscious fruit … our meal … the
tropical luxury …
…

Dice the mangoes into large cubes. Arrange the diced mango
cubes on a plate, in any pattern you like. Pour whipped cream
over the diced mango cubes.
Referring expressions, repetitions and elegant repetitions all
establish cohesion by creating ‘chains’ of connected words/clauses
running through discourse. Such lexical chains need not necessarily
consist of words which mean the same, however. They may also be
created by virtue of some formal semantic connection (good, for
example, associates with its opposite bad; animal with any example
of an animal like horse; violin with orchestra of which it is a part),
or it may be because words are felt to belong to some more vaguely
defined lexical group (rock star; world tour; millionaire; yacht).
This last kind of connection, though it is sometimes treated as a kind
of cohesion, is really too dependent upon individual experience and
knowledge to be treated as a formal link.
e.g.
Four legs good. Two legs bad.
Here are seven kinds of repetitions described by Carter et al (1997:
172):
Varieties
Examples
Direct repetition (exactly the juice … juice
same word repeated)
Synonyms, or near-synonyms saving … investement
(use of words with similar
meanings).
Antonyms (opposites).
Loved … hated
Put them into … spoon out
305
Specific-general
reference
(words referring to the same
thing or person, but where one
has more detail than the other).
Ordered series (words that we
a fire proof dish … the dish
(going from general to specific):
apples … cooked apples
regulo 6 … 3; 10 minutes … 15
know as a set series – for minutes
example, the days of the week,
months of the year, or the
seasons).
Whole-part (where one term apples … pips
names a part of an item that the
other word describes in full).
Renkema (1993: 39), furthermore, elaborates reiteration
includes not only repetition repetition but also but also synonymy.
Reiteration can also occur through the use of a word that is
systematically linked to a previous one, for example, “young” and
“old”. In general, reiteration is divided into the following types:
E. repetition (often involving reference)
A conference will be held on national environmental policy. At
this conference the issue of salination will play an important role.
F. synonymy (often involving reference)
A conference will be held on national environmental policy. This
environmental symposium will be primarily a conference dealing
with water.
c. hyponymy (e.g., the relation of flower to tulip)
We were in town today shopping for furniture. We saw a lovely
table.
d. metonymy (part vs. whole)
At its six-month check up, the brakes had to be repaired. In
general, however, the car was in good condition.
e. antonymy (e.g., “white” vs. “black”)
The old movies just don’t do it anymore. The new ones are more
appealing.
In conjunction with collocation, meanwhile, Renkema (1993:
40) insists that it deals with the relationship between words on
the basis of the fact that these often occur in the same
surroundings. Some examples are: “sheep” and “woold”,
“cpngress” and “politician” or “college” and “study”.
e.g.
a. Red Cross helicopters were in the air continuously. The
blood bank will soon be desperately in need of donors.
307
b. The hedgehog scurried across the road. Its speed surprised
me.
In conclusion, a few words need to be said about the
difference between cohesion and coherence. Cohesion always deals
with the connections evident in the discourse. In many cases,
however, there are connections between successive sentences which
are not apparent in text elements. This is called coherence. Look at
the following example (without any conjunction) as an illustration of
this effect:
He is not going to school. He is sick.
This link between these two sentences relies on knowledge, namely,
that being sick can be the cause of absence from school. On the basis
of this knowledge, it is possible to make a connection between these
two sentences. Coherence, therefore, is the connection based on
knowledge that is in the mind of the reader of listener.
6.2.1 Lexical Cohesion and Text
Brown & Yule (1983) focus on the relationship between
cohesion and text, and indicate that lexical cohesion is not always
necessary for text to produce semantic relations between sentences,
as in the following example:
A: There’s the doorbell.
B: I’m in the bath.
(Brown & Yule 1983: 196)
These sentences have no lexical cohesion, but readers will
understand that the sequence of sentences continues a text. This
means that text can exist without lexical cohesion, though lexical
cohesion cannot exist without text. Brown & Yule (1983) explain
this case as follows:
[T]he reader may indeed use some of the formal expressions
of cohesive relationships present in the sentences, but he is
more likely to try to build a coherent picture of the series of
events being described and fit the events together, rather than
work with the verbal connections alone (Brown & Yule
1983: 197).
Moreover, an example of the inadequacy of cohesive ties between
sentences has to be considered. Brown & Yule’s quotation from
Enkvist (1978) is shown here:
309
I bought a Ford. A car in which President Wilson rode down
the Champs Elysees was black. Black English has been
widely discussed. The discussions between the presidents
ended last week. A week has seven days. Everyday I feed my
cat. Cats have four legs. The cat is on the mat. Mat has three
letters. (Enkvist 1978: 197)
Even though this text has lexical cohesion, it cannot be called a
coherent text. This means that a text including lexical cohesion
cannot always produce coherence. Here, the text fails to deliver any
message to the reader. As Bown & Yule (1983) advocate, cohesive
ties do not always lead reader to coherent interpretation of what they
have read. Melmely, it is significant to teach L2 learners how to
understand the coherence of a text when reading/writing. Cohesion is
never necessary nor sufficient to create coherence, though most
discourse includes cohesion. It is necessary to recognize that
“[c]ohsion is a manifestation of certain aspects of coherence, and a
pointer towards it, rather than its cause or necessary result” (Cook
1994: 34). That is, cohesive ties have to be considered as a
“manifestation of how we are making sense of the message in the
text” (Carter & McCharty 1988: 204). This means that it is necessary
to understand cohesive ties semantically, as well as grammatically.
Hence, it can be considered that a knowledge of lexical cohesion
might help L2 learners understand discourse.
CRITICAL DISCOURSE ANALYSIS: HISTORY, SCIENTIFIC
PEERS GROUP, and WHAT IS CDA?
History of Critical Linguistics and Critical Discourse Analysis
The 1970s saw the emergence of a form of discourse and text
analysis that recognized the role of language in structuring power
relations in society. At that time, much linguistic research elsewhere
was focused on formal aspects of language which constituted the
linguistic competence of speakers and which could theoretically be
isolated from specific instances of language use (Chomsky, 1957).
Where the relation between language and context was considered, as
in pragmatics (Levinson, 1983), with a focus on speakers’
pragmatic/sociolinguistic competence, sentences and components of
sentences were still regarded as the basic units. Much sociolinguistic
research at that time was aimed at describing and explaining
language variation, language change and the structures of
communicative interaction, with limited attention to of social
hierarchy and power (Labov, 1972; Hymes, 1972). In such a context,
attention to texts, their production and interpretation and their
relation to societal impulses and structures, signaled a very different
kind of interest.
311
Kress (1990: 84-97) gives an account of the theoretical
foundations and sources of critical linguistics. He indicates that the
term CL was ‘quite self-consciously adapted’ from its sociophilosophical counterpart, as a label by the group of scholars
working at the University of East Anglia in the 1970s. By the 1990s
the label CDA came to be used more consistently with this particular
approach to linguistics analysis. Kress shows how CDA by that time
was ‘emerging as a distinct theory of language, radically different
kind of linguistics’. He lists the criteria that characterize work in the
critical discourse analysis paradigm, illustrating how these
distinguish such work from other politically engaged discourse
analysis. Fairclough and Wodak (1997) took these criteria further
and established ten basic principles of CDA program. In the
contributions in this volume, we find an even more extensive
elaboration of these programmatic claims and proposals.
Many of the basic assumptions of CL/CDA that were salient
in the early stages, and were elaborated in later development of the
theory, are articulated in Kress’s work. These include assumptions
such as:

language is a social phenomenon;

not only individuals, but also institutions and social
groupings have specific meanings and values, that
are expressed in language in systematic ways;

texts are the relevant units of language in
communication;

readers/hearer are not passive recipients in their
relationship to texts;

there are similarities between the language of
science and the language of institutions, and so on
(Kress, 1989)
Kress
concentrates
on
the
‘political
economy’
of
representational media: that is, an attempt to understand how various
societies value different modes of representation, and how they use
these different modes of representation. A central aspect of this work
is the attempt to understand the formation of the individual human
being as a social individual in response to available ‘representational
resources’.
Fowler et al. (1979) shows how tools provided by standard
linguistic theories can be used to uncover linguistic structures of
power in texts. Not only in news discourses, but also in literary
313
criticism Fowler illustrates that systemic grammatical devices
function in establishing, manipulating and naturalizing social
hierarchies.
Fairclough (1989) sets out the social theories underpinning
CDA and, as in other early critical linguistic work, a variety of
textual examples are analyzed to illustrates the field, its aims and
methods of analysis. Later Fairclough (1992, 1995) and Chouliariki
and Fairclough (1999) explain and elaborate some advances in CDA,
showing not only how the analytical framework for investigating
language in relation to power and ideology developed, but also how
CDA is useful in disclosing the discursive nature of much
contemporary social and cultural change. Particularly the language
of mass media is scrutinized as a site of power, of struggle and also
as a site where language is apparently transparent. Media institutions
often support to be neutral in that they provide space for public
discourse, that they reflect states of affairs disinterestedly, and that
they give the perceptions and arguments of the news makers.
Fairclough shows the fallacy of such assumptions, and illustrates the
mediating and contrasting role of the media with a variety of
examples.
Van Dijk’s earlier work in text linguistics and discourse
analysis (1977, 19881) already shows the interst he takes in texts and
discourse as basic units and social practices. Like other critical
linguistic theorists, he traces the origins of linguistic interest in units
of language larger than sentences and in text- and contextdependency of meanings. Van Dijk and Kintsch (1983) considered
the relevance of discourse to the study of language processing. Their
development of a cognitive model of discourse understanding in
individuals, gradually developed into cognitive models for
explaining the construction of meaning on a social level. In the
Handbook of Discourse Analysis van Dijk (1985) collected the work
of a variety of scholars for whom language and how it functions in
discourse is variously the primary object of research, or a tool in the
investigation of other social phenomena. This is in a way a
documentation of the ‘state of the art’ of critical linguistics in the
mid-1980s, which then led to the new handbook (1997).
A Story about the Formation of a ‘Scientific Peer Group’
CDA as a network of scholars emerged in the early 1990s,
following a small symposium in Amsterdam, in January 1991. By
315
Chance and through the support of the University of Amsterdam,
Teun van Dijk, Norman Fairclough, Gunther Kress, Theo van
Leeuwen and Ruth Wodak spend two days together, and had the
wonderful opportunity to discuss theories and methods of discourse
analysis and specifically CDA. The meeting made it possible for
everyone to confront each other with the very distinct and different
formation, differences and sameness were exposed; differences
toward other theories and methodologies in discourse analysis (see
Titscher et al., 2000), and sameness in programmatic way which
could frame the differing theoretical approaches of the various
biographies and schools of the respective scholars.
Of course, the start of CDA network is also marked by the
launch of van Dijk’s journal Discourse and Society (1990) as well as
through several books, like Language and Power by Normal
Fairclough (1989), Language, Power, and Ideology by Ruth Wodak
(1989) or Teun van Dijk’s first book on racism, Prejudice in
Discourse (1984). But the Amsterdam meeting determined an
institutional beginning, an attempt both to start an exchange program
(ERASMUS for three years) and multiple joint projects and
collaborations between the different scholars and approaches as well
as special issue of Discourse and Society (1993), which brought the
above mentioned approaches together. Since then, much has
changed, the agenda as well as the scholars involved. New Journals
have been launched, multiple overviews have been written, and
nowadays CDA is an established paradigm in linguistics.
Since this first meeting (of course, CDA and CL has existed
before, but not as much an international, heterogeneous, closely knit
group of scholars) annual symposia take place and have
accompanied the emergence of this paradigm, which is bound
together more by a research agenda and program than by some
common theory or methodology. More scholars have taken part in
these conferences, and more researchers have started with research in
CDA, like for example Ron Scollon. Scholars from the Germanspeaking world seldom took part because the conferences were
always held in English. Nevertheless, Utz Maas as well as Siegfried
Jager and their approaches have been understood and acknowledged
(see Fairclough and Wodak, 1997; Titscher et al., 2000).This
explains the wide variety of different approaches in this book, both
theoretically and empirically, and the range of linguistics tools used
to analyzed discourse.
What is CDA?
317
The terms Critical Linguistics (CL) and Critical Discourse
Analysis (CDA) are often used interchangeably. In fact, in recent
times it seems that the term CDA is preferred and is used to denote a
theory formerly identified as CL. CDA regards ‘language as social
practice’ and further CDA conceptualizes languages as a form of
social practice, and attempts to make human beings aware of the
reciprocal influences of language and social structure of which they
are normally unaware (Fairclough, 1989; Wodak, 19989, and van
Dijk, 1993), and takes consideration of the context of language use
to be crucial (Wodak, 2000c; Benke, 2000). Moreover, CDA takes a
particular interest in the relation between language and power. The
term CDA is used nowadays 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. This research is
specifically considers institutional, political, gender and media
discourse (in the broadest sense) which testify to more or less overt
relations of struggle and conflict.
This shared perspective relates to the term ‘critical’ which in
the work of some ‘critical linguists’ could be traced to the influence
of the Frankfurt School Jurgen Habermas (Thomson, 1988: 71ff.;
Fay, 1977:203; Anthonissen, 2001). Nowadays, however, it is
conventionally used in a broader sense denoting, as Krings argues,
the practical linking of ‘social and political engagement’ with the
sociologically informed construction of society’ (Krings et al,
1973:808), while recognizing, in Fairclough’s words ‘that, in human
matters, interconnections and chains of cause and effect may be
distorted out of vision. Hence “critique” is essentially making visible
the interconnectedness of things’ (Fairclough, 1985: 747; Connerton,
1976: 11-39).
Thus, CL and CDA may be defined as fundamentally
concerned with analyzing opaque as well as transparent structural
relationships of dominance, discrimination, power and control as
manifested in language. In other words, CDA aims to investigate
critically social inequality as it is expressed, signaled, 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 legitimate relations of organized power. In so far as the
legitimations of power relations, … are not articulated, … language
is also ideological’ (Habermas, 1977:259 in Wodak and Michael,
2001:2).
319
In contrast to other paradigms in discourse analysis and text
linguistics, CL and CDA focus not only on texts, spoken or written,
as objects of inquiry. A fully ‘critical’ account of discourse would
thus require a theorization and description of both the social
processes and structures which give rise to the production of a text,
and of the social structures and processes within which individuals or
groups as social historical subjects, create meanings in their
interaction
with
texts
(Fairclough
and
Kress,
1993:
2ff).
Consequently, three concepts figure indispensably in all CDA: the
concept of power, the concept of history, and the concept of
ideology.
Unlike some of the research in pragmatics and traditional
sociolinguistics in which, according to critical linguists, context
variables are somewhat naively corrected with an autonomous
system of language, CL and CDA try to avoid positing a simple
deterministic relation between texts and the social. Taking into
account the insights that discourse is structured by dominance; that
every discourse is historically produced and interpreted, that is, it is
situated in time and space; and that dominance structures are
legitimated by ideologies of powerful groups, the complex approach
advocated by proponents of CL and CDA makes it possible to
analyze pressures from above and possibilities of resistance to
unequal power relationships that appear as societal conventions.
According to this view, dominant structures stabilize convention and
naturalize them, that is, the effects of power and ideology in the
production of meaning are obscured and acquire stable and natural
forms: they are taken as ‘given’. Resistance is then seen as the
breaking of conventions, of stable discursive practices, in acts of
‘creativity’ (Fairclough and Kress, 1993 : 4ff).
In CDA nowadays a huge continuity, of course, exists with
CL (see, for example, Fairclough and Wodak, 1977; Blommaert and
Bulcean, 2000) which developed in the 1970s and 1980s, primarily
at the University of East Anglia, with Roger Fowler, Tony Trew and
Gunther Kress. The continuity is visible mostly in claim that
discourses are ideological and that there is no arbitrariness of signs.
Functional systemic linguistics proved to be most important for the
text analysis undertaken by this school (see Halliday, 1978).
Other roots of CL and CDA lie in classical rhetoric, text
linguistics and sociolinguistics, as well as in applied linguistics and
pragmatics. The notions of ideology, power, hierarchy and gender,
and static sociological variables were all seen as relevant for an
interpretation or explanation of text. The subjects under investigation
321
differ for the various departments and scholars who apply CDA.
Gender issues, issues of racism, media discourses or dimensions of
identity research have become very prominent (see Wodak et al.,
1999; Blommaert and Verschueren, 1999; Martin Rojo and van Dijk,
1997; Pedro 1977 many editorials in Discourse and Society over the
years, specifically the debate between Emanuel Schegloff and
Michael Billig in issues 2-4, 1999/2-4, 2000). The methodologies
also differ greatly: small qualitative case studies can be found as well
as large data corpora, drawn from filed work and ethnographic
research.
The general principles of CDA, eventually, may be
summarized as follows (Wodak 1996: 17-20):

CDA is concerned with social problems. It is concerned with
language or language use per se, but with the linguistic
character of social and cultural process and structures.
Accordingly CDA is essentially interdisciplinary.

Power-relations have to do with discourse (Faucault, 1990;
Bourdieu, 1987), and CDA studies both power in discourse
and power over discourse.

Society and culture are dialectically related to discourse:
society and culture are shaped by discourse, and at the same
time constitute discourse. Every single instance of language
use reproduces or transforms society and culture, including
power relations.

Language use may be ideological. To determine this it is
necessary to analyze texts to investigate their interpretation,
reception and social effects.

Discourses are historical and can only be understood in
relation to their context. At the metatheoretical level this
corresponds to the approach of Wittgenstein (1984: 7),
according to which the meaning of an utterance rests in its
usage in a specific situation. Discourses are not only
embedded in a particular culture, ideology or history, but are
also connected intertextually to other discourses.

The connection between text and society is not direct, but is
manifest through some intermediary such as the socio-
323
cognitive one advanced in the socio-psychological model of
text comprehension (Wodak, 1986).

Discourse analysis is interpretative and explanatory. Critical
analysis implies a systematic methodology and a relationship
between the text and its social conditions, ideologies and
power relations. Interpretations are always dynamic and open
to new contexts and new information.

Discourse is a form of social behavior. CDA is understood as
a social scientific discipline which makes its interests explicit
and prefers to apply its discoveries to practical questions.
BIBLIOGRAPHY
Austin, J. L. 1962. How to Do Things with Words. Oxford: Oxford
University Press.
Brown, G & Yule, G. 1983. Discourse Analysis. Cambridge.
Cambridge University Press
Carter, Ronald et al. 1997.Working with Texts. London: Routledge
Chomsky, N. 1985. Syntactic Structure. S-Gravenhage: Mouton.
Collings Concise English Dictionary
Collins: Glasgow.
Cook,
1988. Patrick Hanks (ed),
Guy. 1989. Discourse. Oxford: Oxford University
Press.Crystal, D. 1992. Introducing linguistics. Harlow:
Penguin.
Crystal, D. 1995. The Cambridge encyclopedia of the English
Language. Cambridge: CUP.
Dakowska, M. 2001. Psycholingwistyczne podstawy dydaktyki
języków obcych. Warszawa: PWN.
Enkvist, N.E. 1978. ‘Coherence, Pseudo-coherence, and Noncoherence.’ Cohesion and Semantics. J.O. Ostman Abo,
Finland, Abo Akademi Foundation.
Fay, B. 1987. Critical Social Science. London: Polity Press.
Foucault, Michel. 1972. The Archeology of Knowledge and the
Discourse on Language, trans. A.M. Sheridan. London:
Tavistock.
Gauker, Ch. 2003. Words without meaning. Massachusetts: The MIT
Press.
Gee, J. P. 2001. An introduction to discourse analysis. London:
Routledge.
Gutwinski, W. 1976. Cohesion in Literary Texts. The Hague,
Mouton.
325
Halliday, M.A.K and R. Hasan. 1976. Cohesion in English. London.
Longman
Harris, Z. 1952. Discourse Analysis. Language, 28, 1-30.
Hawthorn, Jeremy. 1992. A Concise Glossary of Contemporary
Literary Theory. London: Edward Arnold.
Hymes, D. 1964. Towards Ethnographies of Communication. In J.J.
Gumperz and D. Hymes (Eds), The Ethnography of
Communication. American Antrophologist, 66, 1-34.
Renkema, Jan. 1993. Discourse Studies: An Introductory Textbook.
Amsterdam: John Benjamins Publishing Company.
Kress, G. 1990. ‘Critical Discourse Analysis’, Annual Review of
Applied Linguistics, 11: 87 – 97.
Longman Dictionary of the English Language. 1984. Heather Gay et
al. (ed). Harlow: Longman.
Martin, J.R. 2001. ‘Cohesion and Texture’. The Handbook of
Discourse Analysis. D. Schiffrin, D. Tannem, and H.E.
Hamilton. Oxford: Blackwell
McCarthy, Michael. 1994. Discourse Analysis for Language
Teachers. Cambridge: Cambridge University Press.
Mills, Sara. 1997. Discourse. New York: Routledge.
Foucault, Michel. 1972. The Archeology of Knowledge and the
Discourse on Language, trans. A.M. Sheridan. London:
Tavistock.
Renkema, J. 2004. Introduction to discourse studies. Amsterdam:
John Benjamins Publishing.
Rogers, R. (ed.). 2004 An introduction to critical discourse analysis
in education. Mahwah: Lawrence Erlbaum Associates, Inc.
Salkie, R. 1995. Text and Discourse analysis. London: Routledge.
Searle, J. R. 1969. Speech Acts. Cambridge: Cambridge University
Press.
Scollon, R. 2001. Mediated Discourse. The nexus of practice.
London: Routledge.
Thomson, J. B. 1988. Critical Hermeneutics. Cambridge: Cambridge
University Press.
Trappes-Lomax, H. 2004 "Discourse analysis". The handbook of
applied linguistics. 135-164.
Van Dijk, Teun A. 1993. Elite Discourse and Racism. Newbury
Park: Sage.
Van Dijk, Teun A. 1984. Prejudice in Discourse. Amsterdam:
Benjamins.
Van Dijk, Teun A. (ed). 1985. ’Handbook of Discourse Analysis’,
Vol. 2: Dimensions of Discourse. London: Academic Press.
327
Van Dijk, Teun A. Van. 1985. Introduction: Discourse Analysis as a
New Cross-Discipline. Journal RSE Analysis, Vol. I,
Academic Press, Inc.Carter, R. 1993. Introducing applied
linguistics. Harlow: Penguin.
Vetter, Eva, Ruth Wodak, Michael Meyer, and Stefan Titscher.
2000. Methods of Text and Discourse Analysis. London:
SAGE Publications.
Wodak, Ruth. 1986. Language Behavior in Therapy Groups. Los
Angeles: University of California Press.
Wodak, Ruth. 1989. Language, Power and Ideology. Amsterdam:
Benjamins.
Wodak, Ruth. 1995. ‘Critical Linguistics and Critical Discourse
Analysis’, in Jef Verschueren, Jan-Ola Ostman and Jan
Blommaert (eds), Handbook of Pragmatics. Manual.
Amsterdam: Benjamins, pp. 204 – 10.
Wodak, Ruth. 2000. ‘Does Sociolinguistics Need Social Theory?
New Perspective on Critical Discourse Analysis’, Keynote
speech at SS2000, Bristol, April 2000, p. 132 – 147
Wodak, Ruth. 2000. ‘Rise of Racism – an American or a European
Phenomenon?’ Discourse and Society, 11: 5 – 6.
Wodak, Ruth. 2001. ‘What CDA is about – a Summary of its
History, Important Concepts and its Development’, Method
of Critical Discourse Analysis by Ruth Wodak & Michael
(editors). London: Sage Publication.
Wodak, Ruth. 1996. Disorders of Discourse. London: Longman