An Introduction to

Discourse Analysis
Theory and method

Fourth Edition

James Paul Gee

First published in the USA and Canada 1999
Second edition published 2005
Third edition published 2011
This edition published 2014
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

and by Routledge
711 Third Avenue, New York, NY 10017

Routledge is an imprint of the Taylor & Francis Group, an informa business

© 1999, 2005, 2011, 2014 James Paul Gee

The right of James Paul Gee to be identified as author of this work has
been asserted by him in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this book may be reprinted or

reproduced or utilised in any form or by any electronic, mechanical, or
other means, now known or hereafter invented, including photocopying
and recording, or in any information storage or retrieval system,
without permission in writing from the publishers.

Trademark notice: Product or corporate names may be trademarks or

registered trademarks, and are used only for identification and
explanation without intent to infringe.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication Data

A catalog record for this book has been requested

ISBN: 978-0-415-72125-7 (hbk)

ISBN: 978-0-415-72556-9 (pbk)
ISBN: 978-1-315-81967-9 (ebk)

Typeset in Berkeley
by Saxon Graphics Ltd, Derby
CHAPTER TWO
What is Discourse Analysis?

! Introduction 17
! What Does the Word “Discourse” Mean? 17
! Another Meaning for the Word “Discourse” 19
! What Speakers/Writers and Listeners/Readers Do 20
! Identities 22
! Webs of Association 26
! The Approach in this Book 29
! Readings 29
 What is Discourse Analysis? 17

Introduction
This chapter is an overview of what the word “discourse” means – or, at least,
what it will mean in this book – and an introduction to the perspective on
discourse analysis this book will develop. Some readers will be helped by an
initial “big picture” view from the top. Others may prefer to enter the water
more slowly. If readers find this chapter too arid, they should start the book
with the next chapter and return to this one when then feel ready for a “big
picture” view.
In the last chapter, we said that when we use language we are saying, doing,
and being. When I am a professor in a meeting with a student, I say and do
professor things as I enact my identity as a professor. When I game (play video
games), I say and do gamer things as I enact my identity as a gamer. When I
“bird” (go bird watching), I say and so birder things as I enact my identity as
a birder. To say, do, and be something I need other people and things in the
world in order say, do, and be. I need other professors (and students and
colleges), other gamers (and games), and other birders (and birds).

What Does the Word “Discourse” Mean?

The word “discourse” is used in different ways in different academic areas.
And different types of discourse analysis are used in a variety of different
disciplines. We humans can analyze nearly anything in terms of meaning. We
can treat anything as a sign or symbol and give it some meaning. For example,
babies, mountains, and stars can be, for us humans, not just things in the
world, but symbolic of meanings like innocence, majesty, and infinity.
However, to linguists “discourse” names a part of language that has an intimate
relation to syntax (“syntax” means the structure of language, the way words
and phrases combine together into sentences).
So what is syntax, and what has discourse got to do with it? Think about
painting. If you want to paint a picture, you have to compose different parts of
it to fit together in the ways you want. All the elements compose one thing, a
painting (though the eye takes in bit by bit, in pieces). Now think of an
animated film. Each frame in the film has to be composed (drawn) just like a
painting. The designer has to choose what elements to put into each frame and
how they should fit together. Perhaps a loving couple to the left, a large tree
to the right, and a waning moon at the top. Each frame is a picture that could
stand by itself. However, animated films are made up of many different frames
that flow (very fast) through time, one after the other. The art is in both the
drawing and the sequence of drawings. So the film makers have two choices
to make: first, what to put in each frame (picture) and second, how to sequence
frames one after the other to tell a story or achieve an effect.
Language, like film, flows in sequence through time. In language, we call a
frame (a single picture) a “sentence”. “Syntax” names the “rules” (conventions)
18 What is Discourse Analysis?

we follow when we compose the parts of a sentence. Like a painting or a film

frame, we choose what elements to put in the sentence and how they should
combine or fit together. So, in “Microsoft’s new operating system is loaded
with bugs” I combine elements (noun phrases, verb phrases, subjects and
objects, and so forth) in one way, and in “Microsoft loaded its new operating
system with bugs” I combine elements in a different way. Each sentence
follows the conventions of English grammar. Each says a different thing. The
first makes the bugs sound like a mistake, and the second makes the bugs
sound like something Microsoft did on purpose.
Like film, language moves through time. So having composed a sentence
(like a frame in film), we have to choose how to order our sentences one after
the other to tell a story or a joke, make an argument and an excuse, write a
report or a rant, and so on through a great many other possibilities.
Now we can see one meaning linguists have given the word “discourse”.
Discourse is the sequence of sentences. It is the ways in which sentences
connect and relate to each other across time in speech or writing. As we speak
or write we choose what words and phrases we will put into or “package into”
sentences. Discourse concerns how various sentences flowing one after the
other relate to each other to create meanings or to facilitate interpretation. If
we applied the word to film, it would mean a sequence of frames that compose
a part or all of a film.
For example consider the examples below:

1. The destruction of my home in the fire took only an hour.

2. My home was destroyed in the fire. It took only an hour.

Much the same information is expressed in (1) and (2). However, that
information is all combined into one sentence in (1) and is expressed in two
separate sentences in (2). The information that my house was destroyed
relates to the information that it took only an hour syntactically in (1). The
same two pieces of information are related by discourse (or “at the level of
discourse”) in (2).
Why say it one way or the other, why use (2) rather than (1)? This decision
depends not just on what the speaker or writer wants to say, but on who the
speaker or writer takes the listener or reader to be (e.g., friend, reporter, first
responder, neighbor, stranger, etc.), and what he or she wants the listener or
reader to feel, think, and possibly do (about the situation and about the
speaker/writer).
Some people think that the notion of a “sentence” is relevant only to writing.
This is not, in my view, true. Sentences work differently in speech and in
writing, as we will see in this book, but syntactic relations and intonation do
demarcate sentences in speech, albeit more flexibly and loosely than they do
in writing.
 What is Discourse Analysis? 19

What is a sentence at the most fundamental level? It is a way to background

and foreground information in the same “unit,” much like in a painting or
film frame. It is a way to treat some information as foregrounded (the main
point) and other information as subordinated to the main information, that is,
as background. Thus, consider the examples in (3) and (4):

3. Though money often determines the outcome of elections in the U.S., we

still call the U.S. a democracy.
4. Though we still call the U.S. a democracy, money often determines the
outcome of elections.

In (3), the speaker or writer is assuming that money often determines the
outcome of elections in the U.S.; this is information that is background and
that the speaker or writer asks the listener or reader to assume or take for
granted. The speaker or writer is claiming (asserting) that nonetheless we still
call the U.S. a democracy. This is information that is in the foreground and
that the speaker or writer asks the listener to treat as the main claim to be
discussed or disputed.
The situation is just the reverse in (4). Here the information that we still
call the U.S. a democracy is assumed and backgrounded and the information
that money often determines the outcome of elections is asserted and
foregrounded.
It is the little word “though” that helps determine which piece of information
in (3) and (4) is background and to be assumed and taken-for-granted, and
what information is foregrounded and asserted as the main claim. Linguists
call structures like “Though money often determines the outcome of elections,”
or “Though we still call the U.S. a democracy,” “subordinate clauses”.
Why do we combine sentences one after another? Why do we combine
frames in a movie one after the other? Filmmakers combine frames to create
bigger things than a single frame, to create things like scenes and episodes and
stories. We combine sentences to create bigger things than a single sentence,
to create things like conversations, reports, stories, jokes, arguments, or
meaningful parts of them (e.g., the set-up of a joke, the premises of an
argument, the finale of a story, the first stanza of a poem, and so forth).

Another Meaning for the Word “Discourse”

There is another – and partially related – meaning linguists have given to the
word “discourse”. This meaning is this: discourse is language-in-use (language
actually used in specific contexts). When we study language-in-use, we study
language not just as an abstract system (“grammar”), but in terms of actual
utterances or sentences in speech or writing in specific contexts of speaking
and hearing or writing and reading.
20 What is Discourse Analysis?

Of course, in most cases, whether in speech or writing, we use language to

produce more than one sentence and our sentences follow one after each
other. This is where the first meaning of the word “discourse” kicks in:
relationships across sentences. But, of course, we can utter or write just one
sentence, as in “Get out quick!” or “Can I have some coffee, please?” When
linguists study language-in-use – and use the term “discourse” for this – they
are concerned with the relationship between language and context, with the
ways in which contexts help determine the full extent of what we mean or can
be taken to have meant. For example, in a context where liquid coffee has
spilled, “Clean it up” means go get a mop, while in a context where a can or
tin of coffee grains or beans has broken open, it means go get a broom.
This sense of the word “discourse” – namely, language-in-use – is like the
study of how people actually interpret films as they watch them. Such a study
would have to investigate the clues and cues (how frames have been composed
and how they have been sequenced) filmmakers use to help shape how film
goers interpret their films and how film goers use these clues or cues to make
meaning from the film in the actual contexts of viewing and their lives. So,
too, with language. Discourse analysts study how speakers and writers use
clues or cues (namely, syntax and discourse) to shape the interpretations and
actions of listeners and readers.
Some linguists use the term “discourse analysis” for the study both of the
connections among and across sentences as they follow one after the other,
and for the study of language-in-use in specific contexts. Other linguists use
the term “discourse analysis” just for the first meaning (the connections
among and across sentences as they follow one after the other), and they use
the term “pragmatics” for the study of language in context, for how context
gives meanings to words and words give significance to context. I will use the
term “discourse analysis” for both, though readers should keep in mind that
in their further readings what they read will sometimes be labeled “pragmatics”
and not “discourse”.

What Speakers/Writers and Listeners/Readers Do

In the first chapter, I pointed out that, when we speak or write, we
simultaneously say something, do something, and are something. Different
approaches to discourse analysis tend to foreground saying (information),
doing (action), or being (identity). This book foregrounds identity. We speak
and listen, write and read, as particular kinds of people: for example, as
students, scholars, politicians, gamers, birders, women, feminists, anime fans,
and so on through a nearly infinite list. We also design (shape) our language
in terms of who we take ourselves to be talking or writing to. For example, am
I talking to my doctor as a medical professional or a friend, to my wife as a
academic colleague or an intimate? Such identities organize our social worlds.
 What is Discourse Analysis? 21

This book will argue that we interpret saying and doing in terms of identities
in this sense – as different kinds of people or roles in society. I cannot really
tell what you are trying to do or what you are really intending to say or imply
unless I know who you are and who you think I am or want me to be. It is
consequential whether one Native-American, for example, is talking to
another as a fellow Native-American, or as fellow military veterans, or as a
Native-American talking to someone he refuses to acknowledge as also a
Native-American.
When we speak or write we actively design our language to say and do what
we want to or hope to. We are like artists or musicians, composing with words
rather than paint or musical notes, trying to communicate and achieve effects.
There are two key jobs – two key types of work – we carry out as speakers or
writers:

1. “Recipient Design”: When we speak or write we design our language with

due consideration for whom we take our recipient or recipients to be. We
speak and write differently for friends than we do for strangers. Scientists
speak and write differently for fellow scientists than they do for non-
scientists. We talk differently to service workers than we do to our boss.
2. “Position Design”: We also often design our language in terms of how we
would like to our recipient(s) to be, think, feel, and behave. This means
that the speaker/writer seeks to invite or hail the listener/reader to assume
a particular identity, to be a particular type of recipient that the speaker/
writer wants. We try to “position” our listeners or readers in certain ways.
We do not just design what we say and write for whom we take our
listener or reader to be. We sometimes actively try to entice them to be
who or what we want or need them to be. We try to “position” others to
be and do what we want them to be and do. We might speak in a way to
try to get our doctor or teacher to be (“invite our doctor or teacher to be”)
less “official” and more “humane” and “personal”. We might write in a
way that tries to get readers to be more liberal or conservative or more
religious or less religious than they may actually be. We entice listeners
and readers, if only for a while, to take on a new or different identity that
may lead to new or different beliefs or actions. We seek to persuade,
motivate, change, and even manipulate others. This, too, is a core part of
social life and social change.

There are also two jobs – two core types of work – for listeners and readers:

3. “Situating Meaning”: This means the listener/reader gives words, phrases,

clauses, and sentences, and groups of these, specific situated meanings
based on a construal of what constitutes the relevant parts of the context
as the context exists in the world and as it is actively created, construed,
and constructed in and through language and interaction. If there is liquid
22 What is Discourse Analysis?

coffee on the floor, that context makes us take “Clean it up” to mean “Get
a mop”. If it is coffee grains or beans that are on the floor, that context
makes us take “Clean it up” to mean “Get a broom”. “Clean up” means
different things if we take the context to be about our own hygiene or
about a mess in our bedroom. Later in this book, we will see that language-
in-use has the interesting property of both reflecting the situations in
which we use language and helping to create these same situations or
what they mean or portend.
4. “Response Design”: As listeners and readers we need to prepare a response
to what we are listening to or reading. This means crafting a response
based on how speakers have done their jobs and how we as listeners or
readers have carried out our task of situation meaning (what we have
made of the context). This response may be held in mind, shown partially
on the body without language, or delivered in action or language. In
speech, listeners are always reacting and responding (e.g., by nods, eye
gaze, body posture, and vocalizations) while the speaker is speaking and,
thus, both speaking and listening are active design roles. A listener or
reader has to, in the end, take his or her “turn” and respond. This response
is based on how the utterance or text was designed by the speaker or
writer. The response might be to walk away, to get really engaged in the
conversation, or to argue with the speaker. The response might be to
throw the book away, blog about it, or admire the writer and even change
one’s views. The response might be to help or hinder the speaker or writer
in some way. If the response is language, it has to use Recipient Design
and perhaps Position Design, considering how the original speaker or
writer carried out these two types of designs. We can reject or accept the
original speaker’s or writer’s view of who we are (Recipient Design) and
the ways in which he or she seeks to position us (Position Design).

The main point here is that I want you to see speaking and writing and
listening and reading as active designing, as work we can also engage with
playfully and even artistically. These are all things we do like artists designing
art works to elicit a certain response and interpretation.

Identities
Identity is a complex concept (and complex set of practices in the world). We
can talk about identity at two levels, one quite broad and one much more
narrow and nuanced:

5. “Social Distance”: We can distinguish here among intimates, associates,

and strangers. Who counts as an intimate, associate, or stranger differs by
social groups and cultures and can change over time, of course. Associates
are people we know, but not well enough, or associate with, but not
 What is Discourse Analysis? 23

closely enough to count as intimates. They are in the middle between

intimates and strangers. People often tend to treat associates more politely
than intimates (whose regard they sometimes take for granted) and
strangers (whose regard they care less about than in the case of those who
they are more likely to interact with again in the future). We can also talk
about social distance in terms of a continuum with solidarity/bonding at
one pole and status/deference at the other.
6. “Socially significant kinds of people”: The identities we are talking about
here are identities that are enacted and recognized by different social
groups and social and cultural formations in society. They are situated in
the sense that they are enacted, recognized, and construed in specific
(partially conventional) ways in, and for, specific contexts as these
contexts exist, and as they are simultaneously construed and constructed.
Such identities are multiple – we all have many, can lose or reject some,
and can gain new ones. They can be partial, negotiated, contested,
attempted, improvised, innovated, imposed, freely chosen, hybrid,
mastered, in progress, or fossilized. They can be lies, truths, or a little or
a lot of both. Socially significant types of people may or may not have
commonly agreed upon labels, such as “lawyer,” “theoretical physicist,”
“African-American,” “radical feminist,” “Christian fundamentalist,”
“tough guy,” or “Latino street gang member”. Such labels are most often
rough and cover a variety of more specific types (e.g., “African-American
who is an insider to Black Street Culture” or “Radical feminist of the
postmodern English Department sort”). Identities are very often more
fluid than labels, and both identities and labels for them are negotiated,
contested, and change over time. Both are creatures of history and often
malleable conventions.

Each human language has a grammar. This grammar is the set of all linguistic
resources available to that language’s speakers and writers. These resources
include the language’s lexicon (words), phonology, morphology, syntax, and
semantics.
The grammar of any language is used, recruited, adapted, and transformed
differently by different social groups in order to carry out specific tasks,
practices, work, and to enact or recognize specific socially significant and
meaningful identities. These adaptations of grammar give rise to different
styles, varieties, or registers of language, to what I will call different “social
languages”. We can distinguish between two big classes of social languages.

7. “Non-vernacular (‘specialist’) social languages”: They are things like the

“way with words” used by lawyers, physicists, musicians, carpenters,
gang members, video gamers, and a great many more, i.e., socially
significant types of people. Each social language draws on the grammar of
the language – as well as on some of its own inventions – to shape language
24 What is Discourse Analysis?

to its own goals, values, needs, and work. For example, all speakers know
how to make and use relative clauses or nominalizations in English, but
these grammatical types are recruited differently and pattern with other
grammatical resources differently in different social languages.
8. “Vernacular social languages”: One type of social language – one variety or
style of language – is special. This is a person’s “vernacular social
language”. This is the variety of language a person uses when he or she
wants to enact being an “everyday person” speaking or writing not as a
specialist of any sort, but as an everyday person with “common sense”.
Different social groups have different vernacular social languages (often
different dialects). They have different ways of using language to enact
“everyday person”. And, of course, what constitutes “common sense” can
vary across social groups (and need not be consistent within them).

Since Recipient Design, Position Design, Situating Meaning, and Response

Design involve identities (either of the “social distance” sort or of the “socially
significant type of person” sort), it is crucial to realize that identities cannot be
enacted solely in language. Language is melded with other things in the act of
enacting and recognizing identities. For example, a minister cannot marry
someone just by saying the right words; he or she must also be a minister in
the right way and at the right time. A professor can say all the right things in
a lecture but will not be enacting the identity – or fully accepted in it – if he is
lecturing drunk in a thong. A street gang member cannot use serious
intimidating language while holding a Nerf Blaster.
To enact identities people have to talk the right talk, walk the right walk,
behave as if they believe and value the right things, and wear the right things
at the right time and right place. Identity is a performance. Like all performances
it will not work unless at least some people recognize what you are and what
you are doing in your performance.
Speaking and writing in discourse is always like lines in a play: the actors
and the audience both need to know who the actors are supposed to be, what
they are supposed to be doing, and what it all portends. Of course, avant-garde
plays can leave all this very vague for effect. But leaving it too vague in real life
can be dangerous. If you are not convincing as a cop in a rough bar you can
die. If you are ambivalent as a lover you can lose your mate. Most of life is not
avant-garde, which is not to say none of it is.
So when we communicate with each other we design not just with language
but with everything at our disposal. We design with clothes, gestures, bodies,
environments, props, tools, technologies, objects, the social display of beliefs
and values, and configurations of all these which we create or use as we find
them, for our purposes. So discourse is interactive identity-based communication
using language. We therefore need another term for interactive identity-based
communication using both language and everything else at human disposal. We
call this “Discourse” with a capital “D” or “big D Discourse”:
 What is Discourse Analysis? 25

9. “Big ‘D’ Discourse”: When two people are engaged in discourse (language in
interaction in context) they are communicating with each other via enacting
and recognizing socially significant identities. The identities are socially
significant because various and different social groups construct, construe,
use, negotiate, contest, and transform them in the world and in history. So
when two people interact, so too do two (or more) Discourses. It is as if
socially significant forms of life (identities), formed in history via social
work, talk to each other – continue a long-running conversation they have
been having, by using different human bodies and minds at different times.
When John and Jane talk, men and women (of certain types) and other
socially significant identities (e.g., doctor/nurse, gay/straight, white/black,
football fan/baseball fan, etc.) talk to each other as well, using John and
Jane. But John and Jane are not dupes of Discourses. The conversation
among Discourses is in their hands for the time and place and as they act it
out, they can change it, as can any good actor in any good play.

D/discourse theory is about seeing interactive communication through the

lens of socially meaningful identities. Speakers/writers use language, bodies,
and things (“context”) in the world to enact socially significant identities.
Listeners/readers pay attention to language, bodies, and things in the world
(“context”) to recognize such identities (successfully or not). It is often better
to say that speakers and writers are bidding to get a certain socially significant
identity (or set of them) recognized by listeners/readers. Bids can succeed, fail,
or partially succeed or fail. They can invite or bring about negotiation or
contestation from listeners/readers. Identities are lived and transformed (and
come and go, as well) in history via social interactions among responsive
bodies and minds.
Of course, some written language relies more on language alone than does
most oral language, where bodies, places, and things are more readily apparent.
But written language still has to communicate in specific contexts, contexts
composed of more than language. A textbook is usually read differently – and
recruits different identities – than does a policy document, a newspaper, or an
anime manga. And each can be used in different ways than they usually are
(e.g., the manga can be used as a textbook in a course on anime).
Readers may well ask where the notion of “culture” fits into D/discourse
analysis. “Culture” is a word with too many meanings and often brings with it
too strong assumptions about uniformity. Being a (particular type) of “tough
guy at a biker bar” is a Discourse just as much as being a (particular) type of
“anime otaku” or being a (particular type) of “special ed – ‘SPED’ – student”.
There is probably no point in calling all these cultures. In any case, D/discourse
theory is concerned with enacting and recognizing socially significant
identities. It is about recognition of “kinds of people” in performances in
context. “Culture” often covers much else (too much, really) and yet is often
concentrated on beliefs and values, not bodies, things, and the world, as well.
26 What is Discourse Analysis?

Discourses can be big or little (e.g., runners, gamers, lawyers, politicians,

evangelicals, gang members, or birders). There is no end to them. They come
and go in history. They are rooted in conventions that allow us for a time and
place to enact and recognize being certain socially significant socio-historical
types of people. We run out of labels for them because the human capacity to
recognize types and type of types – and to transform them in practice – is
nearly (and maybe really) infinite (just as is our capacity to recognize faces).
Recognition is both a psychological process and a social process and can be
studied as either or both.

Webs of Association
When we interpret oral or written language in specific contexts of use, based
on how language is designed and on context, we build webs of associations in
our minds. We build (and change) these webs as speech or writing unfolds
across time in context. We associate ideas or themes with both the words we
hear and the elements of the contexts we are in. These associations are often
based in our past experiences. Some people have called these webs of
associations “discourse records” or “mental models of discourse”. In this book
we will see that such webs are made by the work we do to situate meaning
(determine what things mean based on and in specific contexts of use) and by
what we will later call “social models”.
I will give a simple example to exemplify how the web of associations we
build as discourse unfolds helps us interpret language and situations and to
enact and recognize Discourses.
I, a 65 year-old college professor, enter a Mexican restaurant for a late lunch
in the college town attached to my university. As I enter the restaurant, I store
– in my discourse record – interpretations/construals of the context:

• Mexican
• Restaurant
• Lunch
• College town
• Slightly upscale
• Slanted towards young college-aged people and drinking
Etc.

These elements and their associations – and a great many other elements – are
placed in my discourse record. I sit at an empty bar, and a college-aged white
male comes to me and says “What can I get you, boss?” I now add, among
others, the following to the discourse record:

• Service encounter
• Young college-aged white male
 What is Discourse Analysis? 27

• Old man–young man

• Dressed in jeans
• Informal tone
• Jocular tone
• “Boss” = hierarchy, deference, status
• Wants to know what I will order
• He expects a drink order first, since I chose to sit at the bar
• He expects I may or may not also order food

All the associations so far would be typical for a person like me in a college
town like mine. All would be easy to discover via interviews or ethnographic
work. But I have other associations that are more closely aligned to other
Discourses in which I participate:

• Mexican restaurant in Arizona

• Anglo (not Mexican or Mexican-American) bartender/waiter
• No Mexican people in sight
• Perhaps this will not be “authentic Mexican” food
• Arizona has had lots of political turmoil over Hispanic immigration and
the treatment of Mexican immigrants and Spanish speakers.
• The term “boss” is sometimes used as an address by “lower status” people
to “higher status” ones, in particular sometimes it has been historically
used by African-Americans to whites as a sign of deference in an oppressive
racial culture. [For example, I have been asked by homeless people in
cities: “Can you spare some change, boss?”]

I am aware that these latter associations stem from my affiliations with other
Discourses and experiences that the bartender/waiter, much younger than me,
may or may not be aware of or care about, or may care about in a quite different
way than I do.
I order a beer and ask for a menu. The bartender/waiter and I have enacted
identities in a Discourse. It is a service encounter where hierarchy and
deference is used, but masked by a jocular and informal tone. I am aware that
this restaurant Discourse can be contrasted with others I am capable of
participating in or recognizing. For example, there is the “pretend friend”
service encounter where a young waiter gives me her or his first name and
asks mine and thereafter treats me as if we are on friendly terms. There is the
more traditional service encounter where the waiter is distantly courteous and
efficient, and there is the older-fashioned one where the waitress (always a
waitress) says something like, “What can I do for you, Hon”. There are more,
such as the sort where the waiter would say, “What can I get you, brother?” I
know what the current Discourse I am enacting means partly by placing it in
relation to these others.
28 What is Discourse Analysis?

These Discourses have and need have no formal name or label. They need
not be clearly discrete or unambiguous. We just have to get through the
encounter without anyone crying “foul” and “breaking” the Discourse
encounter (which can, of course, happen). For example, I am well aware that
the Discourse encounter I am participating in in the Mexican restaurant can
be “broken” if I take the bartender/waiter’s informal jocular deference too
seriously and seek to actually be a “boss” and order him around.
For me, the encounter is slightly uncomfortable because of the more
political associations I make, but uncomfortable in a way I am used to in
Arizona and the United States, since we are all used to having to leave unsaid
all sorts of aspects of our everyday lives that might, under more inspection,
“bother” us. For example, many Americans have had the experience of being
in a restaurant or hotel in Washington D.C., our nation’s capital, where every
service employee is African-American and every manager and “higher up” and
all the customers are white. We “take it for granted” because to mention it
would “break” the service encounter Discourses we are participating in.
Discourses involve people communicating via language and other stuff
(e.g., my clothes and the bartender/waiter’s clothes, my sitting at the bar and
ordering a drink in the middle of the day, my age, the pop culture decorations,
and so forth). Discourses also involve communication – using our minds,
bodies, and environments – among themselves across history. The service
encounter I had is part of a larger trend in U.S. and Western history to “mask”
(hide) power, hierarchy, and status without removing them. It is part of a
variety of youth-oriented collegiate Discourses that involve drinking and
sexual ambience. It is also part of how we deal with age in a youth-oriented
culture. Discourses are “talking” all the time, though if they cannot inhabit
minds and bodies, they die in history.
The example is meant to show how much has to go on to “pull off” a
Discourse even for the smallest encounter. Furthermore, what I have
delineated here is only a very small part of what is going on. However, we
should keep in mind that for D/discourse analysis we need not uncover
everything, only enough to make the point or argument we are trying to make
in some convincing way. D/discourse data is grist for arguments and
hypotheses about social life, not important just in and of itself.
D/discourse analysis is one tool for studying social life. It is best put together
with other tools (e.g., ethnography, history, sociology, psychology, etc.). It
uncovers the order in social life, an order that is changing and always prone to
being broken. It shows us producing and reproducing identities in history.
D/discourse analysis has a special feature. As analysts we are doing more
overtly and formally what all of us are doing (less overtly and formally) in our
lives in order to mean things to each other. D/discourses are about how we
know what we are to each other and what we are doing with each other in
encounters (in speech or writing). All of us, just as humans, are D/discourse
analysts.
 What is Discourse Analysis? 29

The Approach in this Book

The material in this chapter is meant to be a general overview of the issues any
approach to discourse analysis has to deal with. You can compare and contrast
different approaches to discourse analysis by asking how each such approach
deals with the issues we have discussed here.
In this book we will explicate Recipient Design in terms of how speakers
and writers enact specific socially meaningful identities. We will explicate
Position Design in terms of how speakers and writers use language to get
listeners and readers to view themselves and the world in certain ways and to
act in it in certain ways. We will explicate webs of associations in terms of
what we will call “figured worlds” or “cultural models,” as well as in terms of
how people situate meaning in specific contexts (how they assign specific
meanings to words or phrases based on the contexts in which they were used).
However, it should be pointed out that webs of association can be further
explicated by work on cognition and the brain that we will not cover in this
book (see Gee 2004 for more on this and references to the literature).
Response Design – how listeners and speakers plan and carry out responses
to what speakers say or writers write – is less directly discussed in most
approaches to discourse analysis, though it is a main topic of discussion in
literary criticism. However, discourse analysis tends to deal with Response
Design by considering how what a previous speaker has said, or what a writer
has written, shapes the language in terms of which listeners or readers respond
in speech or in their minds. It is, thus, at the core of how we analyze the flow
of interaction across time and the ways in which people in conversation (with
other people or with written texts) co-construct (build together) meaning as
time unfolds and within specific contexts.

Readings
Bourdieu, P. (1990). In other words: Essays towards a reflexive sociology. Stanford:
Stanford University Press. [The most accessible text by Bourdieu]
Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press.
Gee, J. P. (2004). Situated language and learning: A critique of traditional schooling.
London: Routledge.
Gee, J. P. (2012). Social linguistics and literacies: Ideology in Discourses. Fourth Edition.
New York: Routledge.
Hacking, I. (1986). Making up people, in T. C. Heller, M. Sosna, & D. E. Wellbery,
with A. I. Davidson, A. Swidler, & I. Watt Eds. Reconstructing individualism:
Autonomy, individuality, and the self in Western thought. Stanford, Calif.: Stanford
University Press, pp. 222–236. [Hacking’s work on identity as a whole – he has
several fascinating books – is worth reading, but start here]
Jones, R. (2012). Discourse analysis: A resource book for students. London: Routledge.
Rogers, R. (2011). An introduction to critical discourse analysis in education. Second
Edition. Mahwah, NJ. Lawrence Erlbaum.
Wodak, R. & Meyer, M. (2009). Methods of critical discourse analysis. Third Edition.
London: Sage.