What is CA?
Conversation Analysis (CA) is a field of linguistic study that looks at how we 'do' things through conversation. CA claims that our subconscious awareness of how conversation works allows us to achieve our aims through everyday talk. For example, we generally share an understanding of requests, of how they are a means by which speakers seek to achieve particular goals. A lawyer who wants information about the scene of a crime can assume that if they make a request for information from a witness, then the witness will realise that they are expected to accept or decline this request by providing the desired information (or not). CA suggests that this sort of shared knowledge allows us to produce an 'orderliness' in conversation, and that this orderliness in turn serves as the basis for the smooth working of society.
What does CA do?
CA studies tend to concentrate on naturally occurring talk, which is then converted into detailed and comprehensive transcriptions. These transcriptions allow researchers to look at the structure of conversation in minute detail, with each change in voice quality or intonation and every overlapping piece of talk scrutinised for what it might tell the researcher about the progression of a particular conversation. The majority of research in CA ignores contextual information, in order that the resultant findings are as objective as possible, and made purely based on what occurs in the conversation being analysed.
The following passage is taken from a transcript of an infamous BBC Newsnight interview between presenter Jeremy Paxman and Conservative MP Michael Howard. The passage is a helpful introduction to the type of transcription used in CA, and will allow us to introduce some of the features of language that analysts commonly look at:
1 JP: Did you threaten [to overrule]
2 MH: [I- ] I was not entitled
3 to instruct Derek Lewis and I did not instruct
4 him .hh [an the]
5 JP: [Did you] threaten to overrule [(him)]
6 MH: [>The<
7 the truth of the matter i:s thet (.) Mister
8 Marriott was not suspende[d. I did not-]
9 JP: [Did you] threaten to
10 overrule him.
11 MH: I did not overrule Der[ek (Lewis).]
12 JP: [Did you threaten] to
13 overrule [him.]
14 MH: [I took] advice on what I could or
15 could not d[o::]
16 JP: [Did] you threat[en to overrule him=
17 MH: [and I acted=
18 JP: =Mister Howard.]
19 MH: =scrupulously] in accordance with that advice I
20 did not overrule D[erek]
21 JP: [Did] you threaten to
22 overrule him
(Taken from Clayman, S. and Heritage, J. (2002))
At first the Newsnight transcript might look intimidatingly complex. However, once you are aware of the standard transcript practices in CA, it becomes much easier to understand a transcript that you haven't seen before. Below is a short guide to some CA transcription conventions, each of which can be found in the Newsnight transcript. It might help to watch the original interview (from 4.14-4.40) in order to see how the transcription represents what was said.
Words contained in square brackets overlap with the corresponding words in square brackets in the next turn
I did not
Words that are underlined are uttered with emphasis
What to do:::
Colons after a word represent an extended sound
Where there is an equals sign at the end of a line, this means that the next line of transcript representing the same speaker continues directly from this preceding line
Words in open brackets are uttered with a sense of uncertainty
thet (.) Mister
A full-stop within a pair of brackets represents a short pause
Words within inward-pointing triangular brackets are rushed
What a CA transcript can tell us
The transcript from the Newsnight interview demonstrates the value of CA transcription. It can test our conceptions not just of a particular conversation, but also our assumptions about certain types of conversation. For example, many people will have certain ideas about how a news interview works – for instance, interviewers are in control, putting the questions to politicians and other figures, for the latter to answer – and typical features of how participants tend to talk in this particular type of conversation – for example, politicians avoiding giving direct answers to questions. From the transcript above, we can see that while the interviewer, Jeremy Paxman, is in control insomuch as he is asking the questions (or question!), the interviewee, Michael Howard, exerts control over what exactly is discussed by beginning turns at talk that overlap with the end of Paxman's question, as well as by providing answers that are evasive. We can also see how the interviewer seeks to assert his authority on the conversation. For example, at lines 12-13, Paxman emphasises the word 'threaten' ("Did you threaten to overrule h[im"), in an attempt to get Howard to respond to the precise meaning of the question.
For analysis of the Paxman-Howard interview, and discussion of the methodology of CA, see the following sociolinguistic study, available online at Dawson Era:
Clayman, Steven and John Heritage. 2002. The News Interview: Journalists and Public Figures on the Air. Cambridge: Cambridge University Press, 255-258.
For an accessible overview of CA work, see either of the following texts:
Hutchby, Ian and Robin Wooffitt. 2008. Conversation Analysis (second edition). Cambridge: Polity.
Liddicoat, Anthony J. 2007. An Introduction to Conversation Analysis. London: Continuum.