Discourse Analysis and Stylistics

Introduction

Discourse analysis (DA), or discourse studies, is a general term for a number

of approaches to analyzing written, spoken or signed language use.

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 boundary1, but also prefer to analyze ’naturally
occurring’ language use, and not invented examples.

Discourse analysis has been taken up in a variety of social science disciplines,

including linguistics, anthropology, sociology, cognitive psychology, social
psychology, international relations and communication studies, each of which
is subject to its own assumptions, dimensions of analysis, and methodologies.

The term discourse analysis first entered general use as the title of a paper
published by Zellig Harris in 1952, although that paper did not yet offer a
systematic analysis of linguistic structures ”beyond the sentence level”. As a
new cross-discipline, DA began to develop in the late 1960s and
1970s in most of the humanities and social sciences, more or less at the
same time, and in relation with, other new (inter- or sub-) disciplines, such as
semiotics, psycholinguistics, sociolinguistics, and pragmatics.

Whereas earlier studies of discourse, for instance in text linguistics, often

focused on the abstract structures of (written) texts, many contemporary
approaches, especially those influenced by the social sciences, favor a more
dynamic study of (spoken, oral) talk-in-interaction.

Topics of interest to discourse analysts include:

1. The various levels or dimensions of discourse, such as sounds (intonation,

etc.), gestures, syntax, the lexicon, style, rhetoric, meanings, speech acts,
moves, strategies, turns and ot aspects of interaction

2. Genres of discourse (various types of discourse in politics, the media,

education, sci business, etc.)

3. The relations between discourse and the emergence of sentence syntax

4. The relations between text (discourse) and context

5. The relations between discourse and power

6. The relations between discourse and interaction

7. The relations between discourse and cognition and memory Critical
Discourse Analysis

Critical Discourse Analysis (CDA) is an interdisciplinary approach to the study

of disi which views ”language as a form of social practice” (Fairclough) and
focuses on the ways social political domination is reproduced by text and talk.
Critical discourse analysis is founded on idea that there is unequal access to
linguistic and social resources, resources that are conta

122
DISCOURSE ANALYSIS AND STYLISTICS

123

institutionally. The patterns of access to discourse and communicative events

is one essential

element for CDA.

Background

CDA developed within several disciplines in the humanities and social

sciences, such as ’critical linguistics’ (exemplified by the seminal book
Language and Control^ Roger Fowler).

Norman Fairclough’s Language and Power (1989; 2001) and Critical Discourse
Analysis (1995) articulate a three-dimensional framework for studying
discourse, ”where the aim is to map three separate forms of analysis onto
one another: analysis of (spoken or written) language texts, analysis of
discourse practice (processes of text production, distribution and
consumption) and analysis of discursive events as instances of sociocultural
practice”.

In addition to linguistic theory, the approach draws from social theory and
contributions from Karl Marx, Antonio Gramsci, Louis Althusser, and Pierre
Bourdieu in order to examine ideologies and power relations involved in
discourse. Fairclough notes ”that language connects with the social through
being the primary domain of ideology, and through being both a site of, and a
stake in, struggles for power” (1989). Van Dijk (1998) articulates ideology as
the basis of the social representations of groups, and iriore generally
advocates a sociocognitive interface between social structures and discourse
structures. Ruth Wodak emphasizes the importance of a historical dimension
in critical discourse studies, as she also has shown in her work on racism and
antisemitism. There is a new way for discourse to evolve, the so called
neocourse

Methodology

The methodology of CDA (also called radical styHstics) is applied in the

research into institutional discourse, esp. the language of politics (e.g.,
nukespeak of the 1980s) and administration (e.g., bureaucratese as a specific
style of language characterized by circumlocutions, euphemisms, buzzwords,
abstractions), journalism (journalese as a style of language full of cliches,
occasional neologisms, archness, sensationalizing adjectives, unusual or
faulty syntax, etc.), massmedia and advertising. CDA sees discourses
primarily as linguistic constructs and concentrates accordingly on their
linguistic properties (hence CDA is referred to as critical linguistics). The
following linguistic means have been given considerable attention: a)
modality and grammatical means for conveying attitude (modal auxiliaries
and modal adverbs, evaluative adjectives, verbs of knowledge, prediction and
evaluation), b) the model of transitivity, i.e., how the meaning is represented
in the clause in terms of the participants (actor, goal), process (action, event)
and circumstances, c) the pragmatic model of meaning, d) the category of
gender (disclosing sexist bias in language, e) the speech acts theory, f)
lexical means (metaphors, euphemisms, weasel words, words of abuse and
endearment, etc.) were successfully applied on the study of narrative fiction
and news stories. Highly persuasive strategies in advertising come under
scrutiny while noting the trend towards increased conversionalization’
(employing strategies of informal conversation) and evocation of casualness
and intimacy.

R. Fowler claims that news is always rendered from a particular angle; as a

social construct it reflects certain ’news values’ (Galtung and Ruge):
frequency, threshold, unambiguity, meaningfulness, consonance,
unexpectedness, continuity, composition, reference to elite nations, isference
to elite people, reference to persons, and reference to something negative.

N. hcirclough identifies three pervasive features of the contemporary ’orders’

of discourse: a) conversatiob-lization (colonization of the public domain by
the practices of the private domain/ b) democratization (reduction of overt
markers of power assymetry between people of unequal institutional prwer)
and c) marketization or commodification (the reconstruction of social life on a
market basis) In his Media Discourse, Fairclough provides a survey of
approaches forming an
124

A HANDBOOK OF STYLE AND STYLISTICS

analytical framework for the study of media discourse: linguistics,

sociolinguistics, CA, CDA, semiotics, critical linguistics, van Dijk’s ’social-
cognitive’ model and the cultural-generic approach.

Although CDA is sometimes mistaken to represent a ’method’ of discourse

analysis, it is generally agreed upon that any explicit method in discourse
studies, the humanities and social sciences may be used in CDA research, as
long as it is able to adequately and relevantly produce insights into the way
discourse reproduces (or resists) social and political inequality, power abuse
or domination. That is, CDA does not limit its analysis to specific structures of
text or talk, but systematically relate these to structures of the sociopolitical
context.

Critical discourse analysis (CDA) attempts to disclose the ways societal

structures and processes are encoded in the discourse practices.
Understanding discourse as a socio-semiotic process (M.A.K.Halliday),
exponents of CDA assume that public discourse not only reflects patterns of
(assymetrical) societal organization (i.e., hierarchy, power, hegemony), but
also helps construct, re-inforce and reproduce them. By raising the public
awareness of the encoded ideologies (systems of values and sets of beliefs)
which pervade the strategies of influence, manipulation, persuasion
employed by those in power (e.g., structures of politics, popular media and
consumer industry) these practices can not only be disclosed (’demystified’),
but also altered.

Discourse Analysis and Conversation Analysis

Discourse Analysis (DA) and Conversation Analysis (CA) have evolved as two
broad avenues of research which aspire after the description of conversation
(i.e., how it is organized and through what procedures its coherence is
established); they attempt to do so, however, from largely different
perspectives: while linguistically based DA tries *o extend descriptive
apparatus of traditional linguistics to analyse spontaneous talk, the
sociologically based CA studies conversation as a part of everyday social
practice (for the critique of the two approaches).

The methodology of DA (the term was first used by Zelig Harris in the early
1950s) is chiefly associated with the Birmingham model of DA which,
applying the principles of Halliday’s systemic grammar, developed a rank-
and-scale model of the description of classroom interaction.

In their model they relate three organizational levels of school interaction,

viz. pedagogic (nonlinguistic), discourse and grammatical levels, and suggest
the areas of overlap of their structural units: Levels: Pedagogic Discourse
Grammatical Units: course period lesson topic transaction exchange move
sentence act clause group word variation, where the term ”style” overlaps
morpheme

On the discourse level, the bottom unit of art represents the smallest types of
classroom activity (e.g., elicitation realized by question, reply by statement,
evaluation, etc.), move stands for a participant’s uninterrupted turn,
exchange as a basic constructional unit of interaction consists of related pairs
of moves, e.g., question and answer, transaction is made up of sequences of
exchanges united by a single task (e.g., teacher’s explanation) and lesson is
the highest unit of classroom discourse. The applicability of the model, which
casts light on the pedagogical process, has proved fruitful in analyses of
various other standardized types of intearction such as doctor-patient
consultations, shopkeeper-customer encounters, courtroom interaction (in
these, the term interaction is used instead of the term lesson). Its suitability
to approach talk in less constrained (informal) settings characterized by fewer
restrictive rules (Campare: the pre-defined distribution and sequencing of
turns between teacher and pupil, prosecutor and defendant), i.e., ordinary
conversation, has been challenged. The concepts of initiation (I) and response
(R) were suggested to signal the structural relatedness of successive
utterances; the simplest structure for an exchange being IR (e.g., question-
answer, bid-nomination pairs). The specificity of classroom interaction,
however, lies in a typical recurrence of three-part exchanges of the type IRF
in which teacher
DISCOURSE ANALYSIS AND STYLISTICS

125

initiates as well as closes the exchange by providing feedback (F) or

evaluation (e.g., a teacher comments on the quality of pupil’s answer: T(I):
What’s the capital of Britain ?P(R): London T(F): Correct, London). IRF
exchanges are not so common outside the rule-bound or information-oriented
settings - in casual conversation participants normally do not evaluate each
other’s responses. For example, in the exchange A(I): Is it raining outside?
B(R): Yes, it is. *A(F) Correct, it’s raining the feedback is semantically
anomalous.

The model of DA provides a useful insight into the nature of talk in organized
settings and points out some of its subtle internal workings, such as who is in
control of discourse, how roles of speaker and listener are passed over, how
new topics are introduced, etc. All these factors provide a wide space for
stylistically motivated options, e.g., individual teaching styles are marked by
idiosyncracifcj in structuring exchanges, marking the boudaries of topics,
providing metacommunication statements (which help pupils see the
structure, understand the purpose, etc.), providing or withdrawing the
feedback, balancing the dialogical vs. monological format, etc.
Conversational Analysis and Stylistics

Conversational analysis originated in the ethnomethodological tradition of

American sociology which studies procedures (’methods’) and rituals
underlying ordinary social activities and interactions which are used by the
members of a society (Gr. ethnos) themselves in order to produce and make
sense of their own social interaction. As an empirical and inductive discipline
CA focuses specifically on the verbal interaction which is a pervasive social
practice playing a prominent role in the (re)production of social reality. In
conrast with DA, CA avoids theoretization (it concentrates on how speakers
themselves interpret each other’s turns) and, by analysing detailed
transcripts of authentic ordinary speech events (i.e., conversations in the
broadest sense) attempts to identify systematic recurrence of patterns (turns,
pairs of turns, exchanges) which give conversational interactions elaborate
architectural design. Turn-taking

One of the essential observations of CA is that, when conversing, participants

obviously switch their roles of speaker and hearer, i.e., they take turns in the
process named turn-taking. How the system of turn-taking (TT) operates
however, is far from obvious: interlocutors use a variety of turntaking
strategies to indicate that they are ready to ’yield the floor’. Tum-yelding
clues include various verbal, nonverbal and paralinguistic signals (e.g.,
completion of a basic turn-constructional unit (clause), (re)establishment of
eye-contact, lowering the distance, termination of gestures, lowering the
pitch, pause) and occur (preferrably) at the transition-relevance places (TRP,
i.e., borders of syntactic, intonational, functional and thematic units). Turn-
beginning dues are realized, for example, by the beginning of syntactic units,
overloudness, gesturing, audible inhalation, etc. TT enhances the basic
requirement of efficient communication, viz. that only one party talks at a
time; this, however, does not completely prevent a possible overlap of
two/more turns (simultaneous talk) form occurring. Latching takes place when
there are no gaps in transitions between participants’ rums - these are tightly
timed. A difference has to be made between overlap, a ’byproduct’ of TT
occurring at or near TRP, and interruption of turn by a partner (whereby s/he
loses the floor); while overlaps are common and ’neutral’ in cooperative
behaviour since they may signal interest and involvement (e.g., in heated
debates), interruptions maybe considered as undesired violations of the
’etiquette of speaking’ (unless motivated by urgent situations). Adjacency
Pair and Preference Organization

The next phenomenon studied under the rubric of CA is adjacency pair (AP) -
a paired sequence of turns in which the second turn is conditionally relevant
on. the first, e.g., questionanswer (Q-A), greeting-greeting, request-
accept/tum down, etc. What is more, the occurrence of the second pair part is
expected and its ’official’ absence is marked and given some (conventional)
126

A HANDBOOK OF STYLE AND STYLISTICS

interpretation (e.g., ignorance, bad manners). Some APs provide two

alternative options as to their second pair part, e.g., invitation-
accept/decline; these options are, however, not equal as one of them is
’preferred’ (accept) while the other one is ’dispreferred’ (decline). The notion
of ’ (dis)preferredness’ is based on the observation that preferred alternatives
are structurally simpler (hence unmarked) while dispreferred ones are
structurally more complex (hence marked). For example, a decline of an
invitation (e.g., Maybe we could go to a movie) would commonly consist of _
the following steps: thanking for the invitation (Thanks for the invitation),
refusing it (I am afraid), giving reason why one cannot comply (I ’m busy
tonight) and suggesting an alternative (How about tomorrow night?). Also, a
decline would typically be accompanied with hesitation (weA) and pauses. In
contrast, an accept would be much shorter (Yes, let’s). The procedure
enabling speakers to differentiate between preferred and dispreferred
alternatives, i.e., preference organization, is a standard part of the inventory
of CA.

Local and Overall Organisation

Besides the features of local organisation of talk (i.e., those which operate
across two neigbouring turns, viz. AP and TT), CA also studies the overall
organisation of speech events, viz. their openings, main bodies and closings.
Of special significance is the summons-answer AP which forms a standard
preface to many types of interactions: summons as an attention-getting
device (e.g., Anybody there?, telephone ringing, or knock on the door)
requires an adjacent answer (e.g., Yeah?) after which a summoner is obliged
to fill in the first topic slot, i.e., provide the topic (i.e., announcement of the
reason for the summons). A standard part of the opening section is also a
reciprocal exchange of greetings; participants also have to solve the problem
of mutual identification and recognition.

It should be noted that one single turn usually performs several overlapping
functions, e.g., greetings normally offer sufficient clues for identification (e.g.,
voice signature in telephoning.

Since the provision of the ’topic’ is the summoner’s obligation, the TT system
is temporarily suspended and resumed only after the summoner has signalled
that s/he has done so. The main body of the interaction is then structured
around the first (i.e., ’privileged’) topic; in the chain-like series of turns
participants collaboratively develop the coherent ’content’ of the event. New
topics can be introduced via topical associations, often producing marked
topic ’jumps’.

Closings are technically as well as socially delicate stages of interaction -

both parties need to demonstrate their consent to close the undertaking and
they do so in such a way that noone’s face is threatended. Their readiness to
terminate the contact is announced via a pre-closings sequence or AP (e.g.,
A: Okay? B: Okay) some time before the actual closing (i.e., exchange of
greetings A:Bye, B:Bye) takes place.

Conversation analysts identify several other types of sequences. Participants

can check whether conditions for the realization of some APs are valid by the
initiation of presequences. For example, before an act of invitation takes
place (Maybe we could go to a movie), a participant may open a question-
answer AP (Do you have any plans for Saturday?) to check partner’s
availability. In case s/he is unavailable (I’m afraid I’m busy on Saturday), the
act of invitation is not realized. This procedure enables the participants to
protect each other’s face by withdrawing dispreferred options.

University Questions

1. How are Discourse Analysis and conversational Analysis related to

Stylistics?

2. Discuss the major principles in CDA and CA.

T
17