The Turn-Taking system in British news interviews

Table of contents

1. Introduction

2. Linguistic Terms
2.1. Turn-Taking
2.2. Adjacency Pairs

3. Turn-taking in British news interviews

4. Conclusion

5. Bibliography

6. Appendix

1. Introduction

Being informed is what most people consider the most important thing in these days. Everybody wants to know everything about present events and their circumstances, future conditions regarding politics, etc. and also be able to discuss them with other people. This often requires having a lot of background knowledge. But as the facts are often too complex and not all information needed is accessible to each and every person, there has to be some way to easily inform all people who are interested in a comprehensible manner. What once was the herald, who was sent out by the king, may now be the interviewer in a TV-Show or on the radio. He gets the information directly from politicians who are involved or in charge, victims of a crime, eye-witnesses and so on. In so doing, he passes everything he learns to the audience who may not only consist of a third person participating in the discussion but of thousands or sometimes millions of people.

Consequently news interviews have become more and more important as they convey important issues to the public e.g. considering elections or a party manifesto. Since these interviews need to get across important facts it is imperative that in no case they are blurred by a disorderly held conversation. To guarantee this, the interview has to follow specific rules which preserve its informational character. News interviews often consist of an interviewer and one or more interviewee(s) who alternately talk in a specific way: one asks a question and the other person(s) give(s) a more or less appropriate answer. This may change on some occasions, though, depending on the number of participants, their knowledge about the topic or their degree of involvement in the current case.

Pragmatics, as a branch of linguistics, deals with language and communication and provides us with the possibility to analyze this orderly communication in a scientific context. From a linguistic point of view the constant change of the active speaker may be described by a turn-taking system and by the use of adjacency pairs.

Working with the terms 'turn-taking' and 'adjacency pairs' I want to examine to what extent the linguistic turn-taking system applies to British news interviews. Whether they do or do not follow this idea of a communicational ordering or even have their own turn-taking system shall be found out by looking at an excerpt from a British news interview.

2. Linguistic Terms

2.1. Turn-Taking

Conversation in general is characterized by turn-taking, which basically means the following: a person A is talking to a person B. When A is finished, it is B's turn to talk. Consequently, when B has finished speaking A may take the floor again.^[1] Though this description might give the impression of a fairly easy system, there is much more behind it. In view of the fact that I want to analyse the turn-taking in British news interviews, I want to give a more detailed explanation of the facts around the term 'turn-taking'.

Whenever people talk to each other there are gaps between their respective utterances. A quite interesting fact is that these gaps are more frequent than an overlap of talk, which, in other words, means that two people are talking at the same time. How are these (in most cases) orderly transitions achieved? Explaining this accurate timing and minimal amount of overlap is a key element of the turn-taking system. Levinson states that

[...] the mechanism that governs turn-taking [...] is a set of rules with ordered options which operates on a turn-by-turn basis [...] Such an allocational system will require minimal units [...] over which it will operate, such units being the units from which turns at talk are constructed.^[2]

and calls this a local management system. Considering this thesis it becomes clear that there has to be an existing system or an agreement about units which carry a certain meaning. Every sentence, clause, noun phrase and thus any possible utterance is such a unit, regardless of how long it takes to be uttered. As these units are constructed of syntactical components (e.g. sentences, phrases), they are called turn-constructional components or TCC. At the end of every single unit the active speaker may but must not necessarily change from one to the other. This transition point is called transition relevance place or TRP. The unit uttered prior to the TRP has to be characteristic in a way that it is possible to predict its end and therefore an upcoming TRP. This is the only reasonable explanation for the relatively short gaps between units uttered by e.g. different people participating in a conversation. According to Levinson “[t]he exact characterisation of such units still requires a considerable amount of linguistic work [...]“^[3] and yet the end of a unit seems to be easily projectable for everybody. Predicting a TRP is much easier, if the TCC holds a certain indication or even clear utterance that at its end someone else is invited to continue the conversation. This may be achieved by asking questions, making offers or requests and adding an address tag like a name or 'you' to indicate whom the speaker chooses as succeeding speaker. As soon as a TRP turns up the turn-taking component or TTC becomes applicable. According to Levinson there are specific rules operating on the turn-units:

The first rule applies right at the first TRP of any turn

(a) The Speaker selects another person and stops talking, leaving the floor to the other one.
(b) If the Speaker selects nobody as a successor, somebody self-selects.
The first self-selector may speak first.
(c) The Speaker may but need not continue to talk, if neither (a) or (b) apply.

The second rule says that after application of 1(c) at the next TRP 1 (a)-(c) may apply again. If the Speaker still follows 1(c), the next TRP (a)-(c) again may apply and so forth until a change of the active speaker has happened.^[4]

As these rules apply to all kinds of conversation it becomes clear, why gaps between turns are so small and why the active speaker changes that fast. Still overlaps may occur in conversation. In most cases they are precisely placed and either have the character of competing first starts or happen out of an erroneous prediction of a TRP. The former overlap may be longer than the latter because two or more people may want to gain control of the discussion simultaneously. These two kinds of overlap may be called inadvertent overlap and they stand in contrast to violative interruptions. However, if an overlap occurs, there is also a resolution system for resolving overlap. For example, if overlap occurs, one speaker stops talking immediately waiting for the other to fall silent. He then continues, while choosing the part of his utterance as starting point that was somewhat 'obscured' by the other person's utterance. In cases where the interrupting person does not show any sign of stopping soon, there is some sort of an allocational competitive system. This means that both speakers try to be the 'stronger ones' by increasing vowel length, slowing down their speech and speaking increasingly louder. A typical representation of that strategy is a political debate with politicians shouting at each others.

2.2. Adjacency Pairs

As already noted there are certain ways of selecting the next speaker by indicating or addressing the respective person. This technique is particularly well represented by the system of adjacency pairs or AP, as I will refer to them from now on. An AP consists of a first part or first and a second part or second which build up a paired utterance. Whenever somebody produces a first part and stops speaking someone else has to respond to that by uttering a second part that refers to the first. As Levinson states “the existence of such paired utterances is obvious, but a precise specification of the underlying expectations upon which the regularities are based is not [...] easy“^[5] and this paper would go beyond its scope if I tried to specify these for each and every kind of conversation.

Adjacency pairs are a fundamental unit of conversation and some people even consider them to be the fundamental unit. Simple examples of APs would be greetings, question-answer or offer-acceptance compounds. Still there are other occasions on which a second is not immediately produced after hearing the first. There may be insertion sequences which postpone the utterance of the second while preliminaries are sorted out. Thus a request and its agreement or rejection may be several utterances apart, and the insertion sequences sort out conditions for and the content of the second part. During the process of speaking the necessary second parts are kept in mind as their production is strongly expected. If a speaker fails to provide the correct second he gives a reason for this to show his orientation towards this expectation. So there is not really a tight rule for producing correct seconds to uttered first parts but rather a fulfilling of expectations by different means. Otherwise a missing second part would cause an incoherent discourse.^[6]

There are always several appropriate seconds to, for example, a question like answers, questions, requests, etc. However, those second parts may either be preferred or dispreferred seconds giving the possible seconds some sort of ranking. It is important to note that this term does not refer to one's personal desires but is closely related to the linguistic concept of markedness. Preferred seconds are unmarked whereas dispreferred seconds are marked by a higher amount of structural complexity. So the latter are usually delivered after a delay, or with some preface which marks them as dispreferred seconds or even by telling why the preferred second cannot be performed.^[7] Thus, according to Levinson

by ordering seconds as preferreds and dispreferreds, the organization allows the notion of an adjacency pair to continue to describe a strict set of expectations despite the existence of many alternative seconds to most kinds of first parts.[...]^[8]

3. Turn-taking in British news interviews

The turn-taking system in mundane i.e. everyday conversation leaves the management of the features like ordering, length and content of turn to the participants. However, as already said, news interviews have to have an informative and, as a consequence, an ordered character. Where in a normal conversation a TRP may be chosen by anyone as an opportunity to take the floor the TRPs in institutionally designed discourses are pre-allocated. Within an interview the interviewer IR and the interviewee IE are expected to either ask questions or give answers. This system of turn-type preallocation requires a set of rules that to some extent is quite similar to that concerning TRPs in mundane conversation.^[9]

Greatbatch lists up several such rules. I want to use those rules which apply to two-party news interviews to illustrate this turn-taking system.^[10] The respective rules can be found as headlines in bold print.

The data used for the examples is taken from an excerpt of a news interview that was broadcasted on BBC 1 on the 25th of November 1990.^[11] The interviewer is called A and the interviewee (John Major) will be referred to as B. IRs and IEs systematically confine themselves to producing turns that are at least minimally recognisable as questions and answers, respectively.

[...]

^[1] Cf. Stephen C. Levinson (1983). Pragmatics. Cambridge: Cambridge University Press, 296.

^[2] S. C. Levinson (1983:297)

^[3] S. C. Levinson (1983:297)

^[4] Cf. S. C. Levinson (1983:298)

^[5] S. C. Levinson (1983:303)

^[6] Cf. S. C. Levinson (1983:306)

^[7] Cf. S. C. Levinson (1983:307)

^[8] S. C. Levinson (1983:308)

^[9] see page 5 of this paper

^[10] Cf. David Greatbatch (1988). “A turn-taking system for British news interviews“. In: Language in Society 17, 401 – 430, 404.

^[11] The transcription of this news interview can be found in the appendix.