An Introduction to Applied Linguistics

goodness sake, I can’t get a word in!’). There are also ways of not taking the turn even 
when one has the chance to, for example by just saying Mmm. Vocalizations while 
another person is speaking, such as Mmm, uhuh, yeah, sure, right, are called ‘back-
channel’ responses (Yngve, 1970; McCarthy, 2001) and show that the listener is 
still following the speaker and wishes him or her to continue. Another important 
aspect of turn-taking is the way interlocutors predict one another’s turns and 
often complete the speaker’s utterance for them. Also, they often overlap with 
the speaker as they complete the speaker’s utterance even though the speaker 
is still talking. Neither back-channels nor completions or overlaps are normally 
perceived as interruptions or as rude. For conversation analysts, they represent co-
operative activity by participants to facilitate communication.
Patterns in Turn-taking: Adjacency Pairs
In conversation analysis, the most basic pattern is the ‘adjacency pair’, which is 

---

59
Discourse Analysis
a pair of turns that mutually affect one another. Examples of everyday adjacency 
pairs are greeting–greeting, compliment–thanks, apology–acceptance. Such pairs consist 
of two parts: a first pair-part and a second-pair-part:
First pair-part 
Second pair-part
A: Good morning 
➝ 
B: Hi, good morning
A: Congratulations on the new job 
➝ 
B: Oh, thanks
These adjacency pairs proceed smoothly and are well-formed in terms of the 
cultural contexts in which they typically occur in English: a greeting gets a 
greeting in return, and congratulations prompt a thank-you. These are examples 
of ‘preferred sequences’. But consider this:
A: Hi, how’s it going?
➝ B: Drop dead!
This would probably be perceived as a ‘dispreferred sequence’, a problem for the 
speakers. Sometimes it is necessary to produce a dispreferred second pair-part (for 
example, declining an invitation or offer). When this occurs, hard work is usually 
involved to make the sequence as little-damaging to the participants’ ‘face’ (sense 
of personal worth) as possible. Apart from ritual adjacency pairs (often connected 
with politeness, small talk, openings and closings, etc.), other common types 
include ‘solidary routines’ (for example, A: I have a terrible headache, B: Oh, I’m 
sorry, can I do anything?) and ‘converging pairs’ (for example, A: I just love that green 
sweater, B: Oh, so do I, isn’t it great!) (see Pomeranz, 1984).
Conversation analysts are also interested in conversational openings and 
closings (Schegloff and Sacks, 1973) and how interactants manage the topics they 
want to talk about (Gardner, 1987). (See Chapter 12, Speaking and Pronunciation, 
for more on turn-taking and topic management.)
A major contribution of conversation analysis has been to make everyday 
interaction a subject worthy of academic research. The strength of the observations 
of conversation comes, in part, from the fact that they are always based on actual 
recorded data of naturally occurring interactions, transcribed in meticulous detail 
(albeit usually giving the prosodic features of intonation and rhythm a very 
cursory treatment). Believing that intuition is an extremely unreliable guide for 
work in conversation, conversation analysis has always rejected experimental 
methods of collecting conversational data, such as simulating dialogues or setting 
up artificial interactive contexts, and has challenged discourse analysts to access 
the data offered by everyday life. This has implications for the language teaching 
classrooms: as much as possible, language learners should be given access to 
authentic spoken extracts, as so often the concocted examples provided in text 
books do not resemble real conversation at all.
Sociolinguistic Approaches: Ethnography and Variation 
Theory
Anthropological linguistics and sociolinguistics are concerned with studying 
not the isolated sentence but how language creates effective communication in 
the contexts of everyday life. The three sociolinguistic approaches to discourse 
analysis are ethnographic approaches, interactional sociolinguistics and variation 
theory. We will briefly outline two of these approaches, and refer readers to 
Schiffrin (1994), Eggins and Slade (1997), as well as to Chapter 9, Sociolinguistics.

---

60 An Introduction to Applied Linguistics
Ethnography
Ethnographic approaches to conversation have been led by Hymes (Hymes, 
1972a, b; Saville-Troike, 1989) and are concerned with ‘the situation and uses, 
the patterns and functions, of speaking as an activity in its own right’ (Hymes, 
1974: 3). Hymes developed a schema for analysing context that has the ‘speech 
event’ in which language occurs as its prime unit of analysis:
The speech event is to the analysis of verbal interaction what the sentence is to grammar 
. . . It represents an extension in the size of the basic analytical unit from the single 
utterance to stretches of utterances, as well as a shift in focus from . . . text to . . . 
 interaction. 
Hymes, 1972: 17
Speech events include interactions such as a conversation at a party or ordering 
a meal, etc. Any speech event comprises several components and these are 
listed in the grid in Table 4.2. With each letter acting as an abbreviation for a 
different component of communication, Hymes’s grid has become known as the 
‘SPEAKING grid’.
S
setting scene
temporal and physical circumstances subjective definition of 
an occasion
P