Week 7:

Discourse and Conversation

Abdelmalek El Kadoussi
 Quote
Though a man succeeds, he should not (as is frequently
the case) engross the whole talk to himself; for that
destroys the very essence of conversation, which
is talking together .
(William Cowper, "On Conversation," 1756)
Course Outline
Conversation analysis (CA): Definitions, aims and objects
of analysis
Transcription of conversations: JTS symbols
Components of a conversational structure
Conversational openings
Conversational closing
Adjacency pairs
Turn taking
Preference organization
Repair
Feedback
Discourse markers
Issues and criticism of CA
Glossary
 Introduction

Contrary to what we generally assume, everyday conversation

is fairly well structured and organized.

Through habits of socialization and interaction, we

unconsciously develop and internalize principles, patterns
and norms of talk in conversation.

We recognize their existence every time they are broken.

 What is Conversation Analysis?
Conversation analysis is an approach to the analysis
of spoken discourse that looks at how people
develop, organize and manage their everyday
conversational interactions.

Conversation analysis works with recordings of

spoken data and carries out careful analyses of this
data.
 The Aim of Conversation Analysis

The objective of CA is “to uncover the often tacit

reasoning procedures and sociolinguistic
competencies underlying the production and
interpretation of talk in organized sequences of
interaction."
(Hutchby and Wooffitt 1998)
 What does CA Examine?
Conversation analysis examines different aspects of spoken discourse
such as:
1. How conversations begin (conversational openings)
2. How conversations end (conversational closings),
3. Sequences of related utterances (adjacency pairs),
4. Preferences for particular combinations of utterances (preference
organization),
5. The way speakers take turns (turn taking),
6. Speakers’ response to each other (Feedback),
7. How speakers fix conversational issues (repair),
8. Signposts of conversation coherence (discourse markers)
9. How listeners attend to speakers’ speech (response tokens).
 Transcription Codes and Conventions

Conversation analysts work on data (recorded text/conversation).

They start with data transcription (coding) while the

text/conversation is being produced or while listening to its
recording.

Analysts trace the recurrence of features like pitch, pause, stress,

intonation, pace ,…, assign them specific codes/signs (like: [, (.), <
>, (( )), ::, =, ....) and see how speakers react to them.

The aim is to understand how speakers manage their

conversational interactions.
 The Jefferson Transcription System (JTS)
(2004)
JTS is a conversational analysis code used by discourse and
conversation analysts annotating data or looking at speech
patterns.

The transcripts provide a detailed version of the complex

nature of everyday conversation.

There are many symbols used in the JTS.

 Commonly Used JTS Symbols

(.) A full stop inside brackets denotes a small pause with no significant length.
(0.2) A number inside brackets denotes a pause with timed length
[ Square brackets denote a point where overlapping speech occurs.
> < Arrows surrounding talk like these show high pace of the speech
< > Arrows in this direction show that the pace of the speech has slowed down
( ) pace between brackets denotes that the words were too unclear to transcribe
(( )) Double brackets denote some contextual information where no symbol of
representation was available.
Word or part of it is underlined denotes a raise in volume or emphasis
↑ An upward arrow denotes high pitch
↓ A downward arrow denotes low pitch
→ An arrow like this denotes a particular sentence of interest to the analyst
CAPITALS denote loud voice
Hum’h’ bracketed ‘h’ means that there was laughter within the talk
= The equal sign represents latched speech, no breaks b/b stretches of a talk
:: Colons represent prolonged speech, a stretched sound
 Example of Data Transcription and
Analysis
On YouTube, see the video : “Sex and the City”
(Carrie’s wedding ring).
Next slide contains the text of the conversation with the
transcription code next to it.
Focus on how the conversation is annotated/coded
Read the analysis of the conversation in your book (p.
93)
Then conclude by deducing conversational patterns like
cooperation, turn-taking, …
 Example of JTS Code

Text from “Sex and the City” (Video: Key

Carrie’s wedding ring)
↑ Shift into especially high
Ch: You’re getting engag↑ed pitch
Ca: I threw up I saw the ring and I NOW especially loud sounds
threw up (0.5) that’s not normal. relative to the surrounding talk
Sa: that’s my reaction to marriage. :: prolongation of the
Mi: what do you think you might do if immediately prior sound
he asks. (.) a brief interval (about a
Ca: I don’t know. tenth of a second) within or between
Ch: just say ye:::s:: utterances
Ca: well (.) it hasn’t been long enough (0.5) the time elapsed (by tenths
(0.5) has it? of seconds) between the end of the
Ch: Trey and I got engaged after only a utterance or sound and the start of
month= the next utterance or sound
Sa: =how long before you separated. know stress
Ch: we’re together NOW and that’s = latched utterances – no
what matters. (.) when it’s right you just break or gap between stretches of
know talk
Sa: Carrie doesn’t know. ? rising intonation
Ca: Carrie threw up= . falling intonation
Sa: =so it might not be right. , unfinished intonnational
contour
 Conclusion of Data Analysis
• Analysis shows how Carrie and her friends manage their
conversation in a cooperative manner.

• They let each other continue with what they want to say, rather
than compete for a place in the conversation.

• It also shows the strategies they use when they want to take a
turn in the conversation, such as not letting too much time to lapse
before speaking, in case another speaker should take the turn.
 Components of the Conversation
Structure
Conversational interactions have the following aspects:
1. Openings
2. Closings
3. Turn taking
4. Adjacency pairs
5. Preference organization
6. Feedback
7. Conversational ‘repair’
8. Discourse markers
9. Response tokens
Opening Conversations

Schegloff (1986) describes the ‘canonical opening’ for American private telephone
conversations as follows:

((ring)) (summons/ answer sequence)

R: Hello
____________________________________________________
C: Hi Ida? (identification/recognition sequence)
R: Yeah
____________________________________________________
C: Hi, this is Carla= (greeting sequence)
R: =Hi Carla.
____________________________________________________
C: How are you. (how-are-you sequence)
R: Okay:.
C: Good.=
R: =How about you.
_____________________________________________________
C: Fine. Don wants to know .. (reason for call sequence)

Most tel. conversations open with an identification/recognition sequence

 Closing Conversations

According to Button (1987), telephone closings among Americans usually go over

four turns of talk:
The pre-closing is often made up of two turn units consisting of items such as
‘OK’ and ‘all right’ with falling intonation.
The closing is made up of two further units, such as ‘bye bye’ and ‘goodbye.’

The closing may also be preceded by a number of pre-sequences, such as the

making of an arrangement, referring back to something previously said in the
conversation, the initiation of a new topic …

But in general, closings are complex interactional units which depend on

whether the speaker wants to close the conversation or not. (Button 1987,
Thornborrow 2001).
 Turn Taking
Turn-taking is the manner in which orderly conversation
normally takes place.

It's the notion that people in a conversation take turns in

speaking since normally one person speaks at a time
When studied by sociologists, however, the analysis goes
deeper, into topics such as how people know when it's their
turn to speak, how much overlap (participants speaking at the
same time) there is between speakers, when it's OK to have
overlap, and how to consider regional or gender differences.
Then the speaker may nominate another speaker, or another
speaker may take up the turn without being nominated.
 Turn Taking
Turn-yielding cues:
a. The completion of a syntactic unit,
b. The use of falling intonation, then pausing.
c. Through eye contact,
d. Body position
e. Body movement
f. Voice pitch.
Turn-requesting cues:
a. Utterances like: excuse me, can I say something? …
b. Body language,…
 Turn Taking
Turn-maintaining cues:
a. Not pausing too long at the end of an utterance and starting straight away
with saying something else.
b. Pausing during an utterance rather than at the end of it
c. Increasing the volume of what we are saying by extending a syllable or a
vowel
d. Speaking over someone else’s attempt to take our turn.

Turn taking varies according to the context, the number of participants,

the relationship b/n the speakers, knowledge of topic, ….

For more understanding of how data transcription informs about 2 turn

taking patterns, refer to your book, p. 98.
 Adjacency Pairs
An adjacency pair is a two-part exchange in which the second
utterance is functionally dependent on the first like in conventional
greetings, invitations, and requests. It is also known as the concept
of ’nextness’. Each utterance is spoken by a different person.
What is in the first part of the pair determines what needs to be in the
second part.
Adjacency pairs include such exchanges as question/answer;
complaint/denial; offer/accept; request/grant;
compliment/rejection; challenge/rejection, and instruct/receipt.
Adjacency pairs typically have three characteristics:
1. They consist of two utterances;
2. The utterances are adjacent, that is the first immediately follows the
second;
3. Different speakers produce each utterance
In each of the pairs the first speaker stops and allows the
second speaker to produce the expected second part to the
pair of utterances
Preference Organization
Preference is what the first speaker would prefer to have as a
response.
Some second pair parts may be preferred (expected answer) and
others may be dis-preferred (unexpected answer or no answer).
Example: A: would you like a drink? (invitation)
B: sure! (preferred response: acceptance)
When an unexpected answer happens, the dis-preferred second pair
part is often preceded by a ‘delay’, a ‘preface’ and/or an ‘account’.
Example:
A: Are you going out with anyone at the moment? (Question)
B: Uhhh . . . (Delay)
Well, kind of . . . (Preface)
There is someone I met a while back . . . (Account)
Actually, I’m getting married at the end of the year (Unexpected
answer)
 Feedback
Feedback refers to the ways in which listeners
show they are attending to what is being said.

This can be done, for example, by the use of

‘response tokens’ such as ‘mmm’ and ‘yeah’,
‘OK’, ‘I see’, ‘really?’ by paraphrasing what
the other person has just said.

It can also be done through nonverbal features

like body position or the use of eye contact.
 Repair
Repair refers to the way speakers correct things they or someone else has
said, and check what they have understood in a conversation. There are two
types:
1. Self repair ( the speaker corrects her/himself)
2. Other repair (the listener corrects the speaker).
Examples:
A: She called me at midday –oh, at half pat twelve. (self-repair)
B: I see.
A: She was at home…
B: Oh, she can’t have been at home, she was at work. (other repair)
A: You’re right, it was office time.
 Discourse Markers
Discourse markers (Schiffrin 1987, 2001, Fraser 1990, 1999)
are items in spoken discourse which act as signposts of
discourse coherence. They include:
Interjections such as oh, mmm, amm, errr,…
Conjunctions such as but, and, yet, …
Adverbs such as now, so, well, definitely, absolutely,
Lexical phrases such as y’know, kind of, somehow, (Schiffrin
2001).

They can be at the beginning, middle or end of an utterance

and can serve both as anaphoric (pointing back) and
cataphoric (pointing forward) reference in the discourse
(Mendoza-Denton 1999).
 10 Reasons we use Discourse
Markers
1. To describe sth: “It’s kind of,… how shall i put it,... like weird
2. To show that you’re thinking. “Hang on a moment, I have basically… very little
money”
3. To make a statement less harsh. “Well, you have, um, you have a little something
in your teeth.”
4. To make a statement stronger. “Actually, I think the models are absolutely
stunning!”
5. To make a statement weaker. “Well, if you ask me, She isn’t that cute, is she?
6. To express a different opinion. “Actually/to be frank with you/to be honest/in
fact, I would not buy it”.
7. To stall for time. “Uhh. Umm. Well, you see, how can I explain this,… my dog ate
it.”
8. To look for words. “Now, that’s an interesting question, now let me think, well,
you see, he’s …. err ... it’s on the tip of my tongue, … malignant. Yes absolutely,
that’s the word.”
9. To ask for feedback. “You know what, she killed her husband, can you believe it?”
10. To sound natural and fluent. Like native speakers ….
 Gender and CA
Conversation analysis has, in recent years, made a major
contribution to discussions of language and gender.
With the move from the view of language as a reflection of
social reality to a view of the role of language in the
construction of social reality and social identity.
It assumes that gender is constructed, as a joint activity,
in interaction.
Example:
A: Did you notice how she moved towards me as if she was
not a woman?
B: And her words? Her words were so unbelievable for a
lady.
 Criticism to CA
It dissociates the text from its context
It disregards issues of power, inequality and
social injustice
It is dry and reductionist
It is not socially engaged (without a cause)
 CA Glossary
The assumption that participants in a conversation normally attempt to
be informative, truthful, relevant, and clear: cooperative principle
The manner in which an orderly conversation normally takes place: turn
taking
A type of turn-taking in which the second utterance (for example, "Yes,
please") depends on the first ("Would you like some
coffee?"): adjacency pair
A noise, gesture, word, or expression used by a listener to indicate that
he or she is paying attention to a speaker: back-channel signal
A face-to-face interaction in which one speaker talks at the same time as
another speaker to show an interest in the conversation: cooperative
overlap
Speech that repeats, in whole or in part, what has just been said by
another speaker: echo utterance
A speech act that expresses concern for others and minimizes threats to
self-esteem: politeness strategies
 CA Glossary
A particle (such as oh, well, you know, and I mean) that's used in conversation to
make speech more coherent but that generally adds little meaning: discourse
marker
A filler word (such as um) or a cue phrase (let's see) used to mark a hesitation in
speech: editing term
The process by which a speaker recognizes a speech error and repeats what has
been said with some sort of correction: repair
The interactive process by which speakers and listeners work together to ensure that
messages are understood as intended: conversational grounding
Meaning that's implied by a speaker but not explicitly expressed: conversational
implicature
The small talk that often passes for conversation at social gatherings: phatic
communication
A style of public discourse that simulates intimacy by adopting features of informal,
conversational language: conversationalization