1.1. Introduction
In this introductory unit we are going to look at a number of definitions of discourse and to try to define some key terms used in discourse analysis, with the aim of clarifying its scope in such a way that it can deal with a wide range of problems and phenomena, but in a more systematic and coherent way.
Discourse analysis refers, in a very basic sense, to talk. What most people do most of the time is talk, because to do anything requires talk and, often, texts, both in private and public spheres. However, until recently, little attention has been given to what people actually say and do in particular everyday circumstances. People talk about the world, about their work, about others and their relations with them. And in talking, they do things. That was Austin’s (1962) revolutionary insight into an aspect of language that had not been fully recognized: the pragmatic function of language, what language does to make social life possible. This turn to language, or to discourse, has had the effect of breaking down barriers between different social sciences concerned with the analysis of everyday social life.
1.2. Definitions of discourse
Discourse analysis is widely recongnised as one of the most vast and least defined areas of linguistics. One reason for this is that the understanding of discourse is based on scholarship from a number of academic disciplines that are quite different from one another, such as, the philosophy of language, pragmatics, sociolinguistics.
In a very general sense, two definitions are prevalent in the field, which underlie two different assumptions about the general nature of language and the goals of linguistics: the so-called ‘structuralist’ or ‘formalist’ approach, and the so-called ‘functional’ approach. The former defines discourse analysis as ‘language above the level of the sentence’ (Stubbs, 1983, and many others). Those who practise the formalist approach to language analyse discourse to find constituents that have certain relationships with one another and occur in a number of arrangements (the sort of linguistic analysis at the level of phonetics, morphology and syntax). We may call this type of approach to the analysis of language ‘sentence linguistics’, and it will not be included in the scope of our course for reasons to be explained further down.
The latter approach defines discourse analysis as ‘language in use, for communication’ (Cook, 1989). Its view is that analysis of discourse cannot be restricted to the description of linguistic forms independent of the purposes/functions which they are designed to serve.
Another definition is that ‘discourse is utterances’ (D.Schiffrin, 1994). Discourse is seen as ‘above’ the sentence (larger than other units of language). In other words, the utterance (not the sentence) is considered the smallest unit of which discourse is comprised, meaning that discourse arises as a collection of inherently contextualised units of language use. We are looking at the construction of meaning, i.e we are talking about ‘utterance meaning’ and ‘speaker meaning’ (and also about how the hearer interprets the meaning of an utterance).
Examples of this view of language are the Speech act approach (based on the philosophy of language), the Ethnography of communication, and pragmatics (the study of meaning in use). The question that such scholars ask is: What gives stretches of language unity and meaning?
Let’s now look at the differences between ‘sentence linguistics’ and ‘discourse analysis’ from the point of view of the types of data they use when analysing language (cf. Cook, 1989).
Differences between ‘sentence linguistics’ data (which applies rules of semantics and grammar) and discourse analysis data:
Sentence linguistics data Discourse analysis data
-isolated sentences - any stretch of language felt to be unified
- grammatically well-formed - achieving meaning
- without context - in context
- invented or idealised - observed From Cook, 1989:12
Here are, according to Cook (1989) some arguments concentrating on artificially constructed sentences in relation to language teaching and linguistics: • They are the best for the study of a foreign language because they isolate it from context • Actual language is ‘degenerate’ and deviates from the rules of grammar (Chomsky) • The treatment of language in terms of sentences has been successful in revealing how language works (by giving examples of grammatically correct, but somehow peculiar sentences such as: ‘Sincereity may frighten the boy’ – Chomsky, 1965:63).
Now, here are some arguments for studying language in use, in context, on which the present course relies: • It is more to producing and understanding meaningful language (to communicating) than knowing how to make or recongnise correct sentences • Being a communicator, having ‘communicative competence’ (Dell Hymes, 1967) involves more than just being able to construct correct sentences.
1.3. Key terms
In this section you will be given some key terms that are currently being used when dealing with pragmatics and discourse analysis. They will be detailed in further units of the course.
Language as action
This is a basic assumption about the nature of discourse, assumption which has come to be taken for granted by both social scientists and laypersons. A major source of the view that language in action is Austin’s (1962) theory of speech acts. Austin pointed out that utterances not only have a certain ‘meaning’ (i.e. they refer to states, persons, events, etc.), they also have force, that is, they also do things. In other words, language is action.
Functions of language
As already mentioned above, utterances may be seen to perform certain functions, or acts. Let’s look at the following example:
Eg.1
SPEAKER: Can you pass the salt? HERARER: /passes the salt/ (taken from Schiffrin, 1994:6)
The speaker’s utterance can be understood (functions) as both a question (about the hearer’s ability to pass the salt) and as a request (for the hearer to pass the salt). This is an example of how one and the same utterance may have different functions, since the two understandings are largely separable by context (the former associated, for example, with tests of physical ability, the latter with dinner table talk). The speech act theory is the approach to discourse that focuses upon knowledge of the underlying conditions and interpretations of acts through words. Contexts may help separate multiple functions of utterances from one another.
Context
There are very many definitions of context, but here are some concepts used in the literature:
- ‘co-text’, or the linguistic context.
Consider the following example:
Eg. 2
A: ‘Has the post been ?’
B: ‘I didn’t hear the letterbox’ (Source: Levinson, 1983)
We can understand B’s reply only in relation to A’s question. In order to understand this exchange as meaningful talk, we have to infer that there is some further knowledge, shared by the two participants. Thus, B’s reply becomes meaningful if A knows that whenever the postman arrives and introduces the mail into the letterbox, B can hear the letterbox. Thus, B’s answer may be interpreted by A as ‘No’.
- ‘physical context’ - the actual setting in which the interaction takes place. Meaningful discourse may also been interpreted as such, depending on the actual place or setting in which it takes place. For instance, we understand the utterance in example 1 above (Can you pass the salt?’) as a request and not as a questions if it takes place at a dinner table and not in a testing laboratory or a medical interview.
- ‘social context’: the social and personal relationships of the interactants with one another.
Consider the following example, in which the two interactants clearly state how the direction of talk is influenced by their social relationships as magistrate and defendant during a court hearing.
Eg 3:
Magistrate: I’m putting it to you again - are you going to make an offer – uh uh to discharge this debt?
Defendant: Would you in my position?
Magistrate: I - I’m not here to answer questions - you answer my question (Harris 1984:5)
- ‘cognitive context’ - the background knowledge and shared knowledge held by participants in the interaction .
Discourse vs. genre
In this course the term ‘discourse’ is being seen as a process through which a stretch of talk or a piece of writing becomes meaningful. However, scholars may also use the word ‘discourse’ (also having a plural, in this understanding) seen as a product. In this sense, discourse is defined as characteristics which arise from the content (what is being talked, written about). For example, we may refer to medical discourse, political discourse, etc.
‘Genre’, as distinct from ‘discourse/s’, is defined as characteristics which arise from the social occasion it is part of. For example, we may refer to ‘interviews’, ‘advertisements’ and ‘jokes’, as genres.
Text
Text is being referred to as ‘the verbal record of a communicative event’ (Brown and Yule, 1983). In other words, it is any use (stretch) of language that holds together. Texts can be written or spoken (the verbal record, the written transcription of the tape-recording).
Instead of a conclusion, we could say that pragmatics and discourse analysis concern a set of methods designed for the close analysis of talk and writing, but also a perspective on the nature of language and its relationship to some central issues that belong to various other disciplines. There are two basic assumptions about discourse: language as action, and the functions of the language.
TASKS Definitions of discourse analysis
TASK 1. (Source: Cook, 1989:6)
Some of the following are invented examples, for language teaching or grammatical analysis, and some pieces of language which were actually used to communicate. Is there any way of telling which is which?
1. John considers the analyst a lunatic.
2. Which of you people is the fish?
3. Please don’t throw me on the floor!
4. I wish someone had told me he is a vegetarian: I could have made an omelet.
5. Chicken and vegetable…hot…medium…hot…er rice…er two poppodamus and a…what’s a bhindi bhaji?
TASK 2 (Source: Cook, 1989:10)
Here is part of a verbal exchange between two people, which was recorded and then transcribed:
A: Right, (.hhh) who’s goin’ to lift the bottom? Well...come o’...someone’s got to take ‘old of it.
B: I ain’t goin’ to.
A: Don’t jus’...Come on will you?
1. What does the kind of language used tell you about the interactants?
2. What does Don’t jus’ (false start) suggests to you about the person?
3. What do you think the two people are doing? Is it important to know?
4. What does the transcript suggest to you concerning the social relationships between the two?
