Discourse Analysis
Discourse Analysis
Discourse Analysis
Vuk Vukotic,
PhD student,
Institute of Lithuanian Language
Defining discourse today is not an easy task, not even after it has played an extremely
important role in the development of social sciences in the past five-six decades or so. To
illustrate this, let us take a geographical difference in understanding of the term discourse:
Giving a complete analysis of the position of DA in philosophy and the field of social
sciences is beyond the scope of this task, therefore I limit the aim of this paper a historical
overview of the development of and discourse analysis (DA) in the field of linguistics. I begin
from the relationship between discourse studies (DS) and structural linguistics of Ferdinand de
Saussure, continuing to the place it received in the development of many linguistic fields in the
past 70 years, as well as contemporary theoretical and methodological trends.
Relationship to structuralism
The beginnings of DA thought in the field of linguistic science can be traced back to
criticism directed towards Ferdinand de Saussure’s structuralist linguistic tradition. This was also
an argument for functionalist against formalist linguistics: Through study of discourse, we can
find usage-based explanations of languages’ inner logic. In this section, I will present the
development of discourse as a term used in linguistics and social sciences, as well as the main
features and theoretical basis of DS.
This brings us the next aspect introduced to DA, because van Dijk was referring to the
growing field of pragmatics. Discourse received different treatment in the Speech Acts Theory
(Austin, 1960, Searle 1969), which laid ground for understanding of discourse as linguistic
action, guided by wider conventions, social and cultural norms. Interest for language beyond the
borders of a sentence and text had become significant, but by now already with the focus on
utterance as a non-arbitrary. This is the logic we we find in Bakhtin (1986), who sees words and
sentences as elements of language and he calls elements of communication - utterances. For him,
utterance is defined not by the linguistic elements, but by the expressive values. A more concrete
development of the supra-sentence level of linguistic analysis was proposed by Beaugrande in
his "Introduction to Text Linguistics" (1981). He listed specific criteria for a text to be considered
discourse (Cohersion, Coherence, Intentionality, Acceptability, Informativeness, Situationality,
Intertextuality).
What also is often neglected is that this new understanding of discourse added another
aspect of language, which in the dominant structuralist tradition was not even considered a part
of the language system, namely phonetics. Intonation, for example, ss Couper-Kuhlen (2001)
points out, received first proper treatment in discourse analysis from an unexpected field, namely
language teaching. She further makes an argument that the book Discourse intonation and
language teaching (Brazil, Coulthard & Johns, 1980) was of key significance for the realization
of the importance of intonation in discourse.
Going beyond text – sociolinguistics and context
Another level beyond text, the dynamics of conversation was examined by Goffman in
his early works from non-linguistic such as drama, games and ethology (Psathas, 1995).
Arguably the first work to draft issues of everyday talk was Goffman’s work on self-
presenatation (1959), and the first systematic was later systematicized Lakoff’s work (Gordon &
Lakoff, 1971), Forms of talk (Goffman, 1981), and gained the established name Conversation
analysis (CA) in work of Silverman (1998). CA introduced important concepts such as turn-
taking, and introduced real-time analysis of discourse, using recorded and video material –
grounding this field of linguistics very rigorously into the empirical data.
Another aspect of language that was not analysed in Saussurean and Chomskyan
formalist traditions was context of language use, which, for the sake of comparing with structural
linguistics, can here be understood as another supra-linguistic aspect that influences language
and its use. In structuralism, it was denounced as arbitrary, situational, based on individual
choice, a relationship Saussure calls in absentia (1916: 171), meaning is has a psychological,
unobservable reality. In one way, we could say that parole was missing from analysis, but we
also meet serious criticism of the need for such a distinction, as we will see in the following
paragraphs. The first and strongest criticism, from the side of functionalist linguistics, was that
parole (language in context) is the only observable language, while langue isn’t (Weinreich,
Labov, & Herzog, 1968).
The sociolniguistc tradition had great impact on understanding the nature of discourse.
Context was introduced through Labov’s ground-breaking work, all which included discourse
analysis as the main research tool. Firstly, through analysing discourse markers of black New
York speakers, Labov (1968) found different Englishes in speech of New York. He thus
introduced contextual factors such as age, gender, profession to his analysis, starting thus a
tradition of variational sociolinguistics. Revisiting the ideas of the father of linguistics, what
these socilinguistc developments have shown is that the relation between langue and parole, is
not arbitrary or dependant solely on individual choices, but on a great number of variables,
which can be studied systematically. After Labov’s pioneering work, discourse and context
gained much more attention. Around the same time, Gumperz and Hymes had started developing
and innovative approach to studying communicative acts, in which context was systematically
included (arguably for the first time), and they edited a monograph entitled Directions in
sociolinguistics: the ethnography of communication (1972). This gave us a new approach,
heavily influenced by ethnography, called interactive sociolinuistics (Scheffrin, 1994). The
approach provided a method of researching culturally and socially specific meanings in
communication, discourse was now analysed as indexing of contextual information (age,
ethnicity, aggressiveness, collaborative-ness etc.). The goal of such research is to bring detailed
explanations of very specific contexts (such as the workplace) and has proven useful in analysing
cross-cultural communication and is often used in that sphere (Holmes, 2014: 178)
So far, we have seen elements that make up discourse in relation to text: it has expressive
functions, pragmatic functions, contextual dependency and contextual meanings. This led to
some the main definitions of discourse as “text in contetxt” (Van Dijk, 1990: 164), “language-in-
use” (Brown & Yule, 1983: xiii). To conclude this section, discourse was in the centre of focus
before discourse analysis became established as a field in linguistics, and linguistic theoretical
framework.
Linguistic study of discourse started gaining firmer theoretical grounds in the ninth
decade of the 20th century. Being such a young field of study and having multidisciplinary roots,
the development of DA in linguistics today non-linear and various, hence the goal of this section
is to present the ongoing debates in the development of DA.
One of the interesting things about discourse is that it has, just like communication
science, been developed in the age of the revolution of research methods in social sciences, and
is in essence multidisciplinary. This poses a question of whether discourse study into a “separate
discipline” is necessary at all. DA was developed from multiple disciplines, and no wonder it
cross-cuts so many disciplines within linguistics, humanities and social sciences. The newest
edition of a comprehensive reader on discourse (van Dijk i ed., 2011), features 18 chapters, each
on a different aspect of DS, of which I only have not mentioned discourse and semantics,
discourse and cognition, discourse racism, discrimination gender and power (studied in CDA),
discourse and culture and discourse and identity.
CDA builds upon the tradition of criticism of ideology, that same philosophical grounds
in which discourse analysis began in social research and social philosophy. It, therefore, takes its
main social-theoretical roots in works of Michelle Foucault, Jürgen Habermas, Pierre Bourdieu
(van Dijk, 2001: 364) and Antonio Gramschi (van Dijk, 2001: 355). It starts from the idea that
ideologies are expressed through discourse, not necessarily all the time, but this is the case most
often. Power analysis and social critique are central to CDA; “regular” or “common sense”
thinking about and doing of things is in the service of the powerful, and against the
discriminated, according to scholar considered to be the father of CDA, Norman Fairclough
(1989: 77). The focus of CDA research is for example, public discussion, political debates, hate
speech and discriminatory speech – the broader societal ideologies and power relations are
contained within language.
As we have seen so far, many schools of thought exist within DA, some even operate on
a very different philosophical basis (CDA, which is based on critique of ideology, which views
ideology as a “mistake” in reasoning, and a constructionist tradition, which views it as inherent
to human nature).
Blommaert, like many leading DA theoreticians, think that ideology and social relations
in the global era must not be ignored, and draws on a neo-Maxist (Wallerstein’s) explanation of
the globalized world and economy to illustrate that context is global. This is one of the main
theoretical divisions in DA today, and one point of criticism – it seems that CDA authors use
Foucault’s words, but analyse power relations held by actors, not taking into consideration
Foucault’s understanding of discourse as “subject-less” (Wodak, 2006: 602). Here I also wish to
mention the issue of the multidisciplinary nature of discourse. Lakoff (2004) calls for a theory of
discourse as interdisciplinary: She illustrates her point, using an analysis of the discourse of
apology (a particularly difficult type of discourse with many sub-types) to show that what is in
play is everything from phonology to neurology and sociology. Whether this is the solution to the
diverse nature and traditions in DA, we cannot know, but we know that discourse is gaining
much attention in the last years (major publications – from textbooks and readers to monographs
– on discourse are still published greatly), we can surely say that interesting theoretical debates
are yet to come.
References (chronologically):
Brazil, D., Coulthard, M., & Johns, C. (1980). Discourse intonation and language
teaching (pp. 1-82). London: Longman.
Van Dijk, T. A. (1980). Text and context: Explorations in the semantics and
pragmatics of discourse.
Norman, F. (1989). Language and power. London and New York: Longman.