Journal of Pragmatics 38 (2006) 196–215

www.elsevier.com/locate/pragma

Discourse representation in political interviews:

The construction of identities and relations
through voicing and ventriloquizing§
Gerda Lauerbach
Institut für England- und Amerikastudien, Fachbereich Neuere Philologien, Johann Wolfgang Goethe-
Universität, Grüneburgplatz 1, D-60629 Frankfurt am Main, Germany
Received 1 August 2002; received in revised form 1 October 2003; accepted 1 June 2005

Abstract

This paper studies the construction of identities and relations through the representation of
discourse. After introducing a distinction between voicing and ventriloquizing, interviews with
politicians of the British Conservative and Labour Parties by BBC and ITV television during their
election night specials of 1997 are analyzed. Interviewer practices differed with respect to the
politicians of the two major British parties: The losing Conservatives were presented with critical and
controversial voices from within their own party, so that an inner-party debate was constructed,
portraying the party as deeply divided. In the absence of dissent within the winning Labour Party,
Labour politicians were confronted with ventriloquisms about what they themselves might fear, think
and plan about sensitive issues of policy, so that doubt was cast on Labour’s united front, and future
inner-party debates and their issues were foreshadowed. Beyond the construction of debate and
dispute, the practices of voicing and ventriloquizing also had the effect of personalizing and
dramatizing political discourse and, in the interplay between interviewer and interviewee, implicitly
constructing identities and relations. The practice of ventriloquizing in particular, normally used with
babies and pets, was used in the data for women, particularly heavy losers, the socially deprived – and
Labour.
# 2005 Elsevier B.V. All rights reserved.

Keywords: Intertextuality; Represented discourse; Ventriloquizing; Constructing identity; Political interview

§
This is a revised version of papers given at the Georgetown Round Table for Languages and Linguistics 2001,
and at the International Conference on Language, the Media and International Communication, Oxford 2001.
E-mail address: Lauerbach@em.uni-frankfurt.de.

0378-2166/$ – see front matter # 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.pragma.2005.06.015
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 197

1. The constructive role of the media

In the reporting of politics, the constructive role of the media has often been remarked
on, as has the media’s changing function within democracy from mediator or watchdog to
almost a political institution itself. For instance, with respect to election campaigns,
Blumler and Gurevitch (1995:134) have noted that the editorial process of television often
yields a severely boiled down version of the campaign. One of the journalistic ‘‘packaging
devices’’ employed is the continual juxtaposition of statements made by the parties, about
which they say:
. . . in continually counterposing one party’s stand on an issue with that of its
rivals the journalists appear to be promoting inter-party communication. It is as if
the news personnel are trying to create a dialogue between the parties, where it
otherwise might not have existed, and to trigger the politicians into comments and
actions that might not have been forthcoming otherwise. Television journalists
thus help to orchestrate the campaign, even if they did not write the original
score.
What happens after the election, when the campaign has been fought, the polls have
closed and the major television stations start their election night specials? There is a
multitude of voices being orchestrated by one or more presenters in the studio, who
communicate with experts, news personnel and interviewees both in the studio and on
outside locations. Over the space of several hours, while the results of the election are
taking shape, the live election night broadcast weaves a complex web of intertextual
relations, made up of various media, formats and texts, in various semiotic modes.
Processes that might be observed over the course of days and weeks during an election
campaign may be condensed into the space of a few hours during election night.
This paper will focus on the analysis of political interviews that were conducted in
two such television election nights, namely those broadcast by the major national
television channels BBC and ITV after the British parliamentary elections of 1997. A
particular interviewer practice will be studied, that of embedding utterances of other
speakers in their questions, as well as the way in which this is dealt with by interviewees.
The practice of representing the discourse of others is central to the journalistic
‘‘packaging device’’ of juxtaposing positions mentioned by Blumler and Gurevitch. It is
prominent in election night interviews, but the question is whether it serves different
functions from those observed by Blumer and Gurevitch for election campaigns when
the election is over. What are the general forms and functions of the practice as employed
by British interviewers during election nights, and what are its particular effects in this
specific set of data? This question can be further differentiated: Does the form of the
practice differ in relation to the type of interviewee? Specifically, is there formal
variation depending on whether the interviewee is a politician of the winning Labour
Party or belongs to the losing Conservative Party? And do the functions of the device,
likewise, depend on the type of interviewee? Does the content of the interviewer
question also play a role? In short, who gets quoted to whom on which topics and with
which strategic aims on the part of the journalists? And how do the interviewees
respond? A further question, one which cannot be pursued within the confines of this
198 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

study but which is relevant for the interculturally comparative project that it is part of ,1
is how the findings of this paper compare to interviewer practices in US-American and
German television election nights.
The paper is organized as follows: In the next section, different forms of discourse
representation will be introduced and briefly discussed with respect to their functions in
media discourse. In section 3, which forms the main part of the paper, the data will be
analyzed within a theoretical framework of critical discourse analysis supplemented by a
Gricean perspective. The focus will be on the type of interviewee, on the type of voice
quoted and on the topics concerned. A section each on Conservative, Labour and a third
group of interviewees will be followed by section 4, in which the interviewers’ practices
and their functions will be compared in relation to the type of interviewee. The paper will
conclude with a short summary and questions for future research.

2. Voicing and ventriloquizing

What is involved in representing discourse on a linguistic level is reported speech in its

variants of direct, indirect and free indirect speech. These categories can be succinctly
demonstrated by following Fairclough (1988:126), and using his examples as given below.
Following traditional analyses, Fairclough first makes a distinction between direct
discourse or DD, as in example (a), and indirect discourse or ID, as in example (b):

(a) Mrs. Thatcher warned Cabinet colleagues: ‘I will not stand for any
backsliding’. (DD)
(b) Mrs. Thatcher warned Cabinet colleagues that she would not stand for any
backsliding. (DD)

Fairclough also has a category for cases where another voice is introduced without being
marked as represented discourse and which he calls unsignaled (UNSIG), as in example
(c). This corresponds to what is traditionally called free indirect discourse. He also
introduces a category for cases of ‘slipping’ between direct discourse and indirect
discourse or DD(S), as in example (d). In spoken language, this last category can be
difficult to detect, but may be signaled by special ‘scare-quote’ prosody:

(c) Mrs. Thatcher will not stand for any backsliding (as a newspaper headline).
(d) Mrs. Thatcher warned Cabinet colleagues that she would ‘not stand for any
backsliding’.

Representing the discourse of others functions as a device whereby speakers can distance
themselves from what is being expressed, positioning themselves in a Bakhtinian dialogic
universe of voices other than their own (White, 2000). In Goffman’s (1974, 1981) terms, a

1
This paper is part of the project ‘‘Television Discourse’’, supported by the German Science Council, and
directed by the author. The goal of the project is a comparative discourse analysis of television election night
coverage in the USA, Great Britain and Germany.
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 199

figure other than the speaker is being animated without the speaker being understood to be
either the author of the words or to be responsible for them. This type of representing
discourse will be called ‘voicing’ here. However, if a kind of mock-representation is done
by putting one’s own words into the mouths of others, e.g. speaking for pets or prelinguistic
children in baby talk, Goffman (1974:536) speaks of ‘‘say-foring’’ or ‘‘ventriloquizing’’:
Although baby talk is sometimes used between consenting adults as part of
affectionate styling (now perhaps mostly found on TV) and very often used to
children as a means of dispelling the fear they might have in dealing with adults,
another function is at issue here: its use as a mock-up in which a speaking adult acts
out a response that a nonspeaking child might make if he could (or would) talk. A
similar form of ventriloquism is used to animate pussy cats, teddy bears, and other
lovable objects.
The practice however is not restricted to ‘lovable objects’ and those who are unable to
speak for themselves. It is a particularly vivid way of enacting one’s own discourse through
another, and one that is not traditionally considered in linguistic and pragmatic analyses of
reported speech.
Obviously, the ability to do this sort of thing greatly increases the strategic potential of
communicators, and numerous functions have been described for the representation of
discourse in the literature.2 The functions of the practice depend to a large extent on the
type of activity or genre engaged in, as well as on the addressees. For news reporting it may
be done differently in the broadsheets than in the rainbow press (cf. Fairclough, 1988). For
news interviews, Clayman (1988) has shown how interviewers routinely employ the
practice. It allows them to incorporate a point of view or opinion into their questions,
inviting the interviewee to react to it, without endorsing the opinion themselves. Thus, they
can, at least superficially, maintain the neutral and impartial stance required by their
professional code.

3. Analysis

The data to be analyzed are from the night of the British General Elections of 1997,
transcribed from the video-recordings of 6–8 h of election night broadcasts by BBC and
ITV television. The political interviews analyzed from these data are conducted with
leading politicians of the two major British parties, the Conservative Party or Tories and the
Labour Party. 1997 was the election in which the Labour Party in the shape of Tony Blair’s
New Labour won its first landslide victory (the second was in 2001), ousting the
Conservatives after 18 years of Tory rule.
Who gets quoted in these interviews? This differs for Conservative and Labour
interviewees. For the Conservatives, the sources quoted are politicians of their own party
who have spoken out on topics that are controversial within the party. These critical voices

2
Such as mitigating or accounting for face-threating acts, warranting arguments, characterizing individuals,
translating into recipients’ register, maintaining social relations of dominance, etc., e.g. Leech and Short (1981),
Short (1988), Fairclough (1992), Caldas-Coulthard (1994).
200 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

may also be left vague or anonymous, or they may even be those of the interviewees
themselves. The aim is to trigger reactions that the interviewees would not otherwise show
in public and that can in turn be fed back into the night’s discourse. The effect is that a
debate is being constructed, not between the different parties as Blumler and Gurevitch
(1995) found for election campaigns, but within the Conservative party itself. Furthermore,
in the interplay between the provocative voices quoted by the interviewer and the defensive
strategies of the interviewees, complex negotiations are done regarding identities, roles and
relations between party members. These negotiations implicitly touch on loyalties,
conflicts, alliances and relative power. For Labour the situation is different: Against a
background of strong party discipline and an overwhelming victory taking shape, no
critical voices are to be heard from within the party. It is interesting to see how the
interviewers nevertheless manage to construct contrasting positions: they make them up by
ventriloquizing. The interesting question is what the effects of this practice are.
What are the controversial topics for whose negotiation the practice of representing
discourse is employed? Controversial topics can be formally identified in interviews
because they are marked by the interplay of interviewee evasion and interviewer insistence,
which leads to complex sequential structure on the one hand, and complex question and
response structures on the other (Clayman, 1988; Heritage and Roth, 1995; Jucker, 1986;
Lauerbach, in press-a, in press-b). According to these criteria, the sensitive topics in my
data for the Conservatives are: (1) to acknowledge defeat; (2) to analyze the causes for the
defeat, e.g. the campaign, divisions in the party (particularly over joining the European
Common Currency), policy, leadership; and (3) the future leadership of the Party. Sensitive
topics for Labour are (1) the acknowledgment of the scale of the victory and any danger
which such a substantial majority may pose to government (like unruly back-benchers); (2)
the discussion of future positions in the Cabinet; (3) the re-modeling of Old Labour into
New Labour; (4) the promises of the manifesto, and (5) the degree of radicality to be
expected from a Labour Government. These topics are recycled throughout the night in
2–4 min interviews. In a number of these interviews, the practice of discourse representation
is employed by interviewers.
In the following presentation and analysis of interview excerpts, interviews with
Conservative politicians in section 3.1 precede those with Labour politicians in section 3.2.
For reasons that will become clear during the analysis, a third category of participants will
need to be dealt with in section 3.3: women, losers and the socially deprived.3 The findings
will then be compared in section 4.

3.1. Voicing the Conservatives

Much of the intertextual web that is spun by television journalists throughout election
night relies on quoting what one interviewee said to another. But does any sort of
interviewee utterance qualify for this type of textual recycling? To answer the question,
excerpt (1) is enlightening: It not only shows that the appropriate quote must first be
elicited from an interviewee, but also that the interviewee must be someone who is known
to be outspoken and controversial and who can be relied on to provide a provocative

3
Reminiscent of, but not to be confused with, ‘women, fire and dangerous things’, cf. Lakoff (1987).
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 201

soundbite. The example is analytically particularly interesting because it represents a rare

instance where an interviewer can be shown to orient to the type of quote that is suitable for
being presented to other interviewees: he explicitly earmarks a comment – which in
forceful rhetoric expresses a negative evaluation regarding the interviewee’s own party –
for recycling later.

(1) ITV97.I/0.58 Jonathan Dimbleby interviews Conservative candidate

Edwina Currie
JD Does, given your experience, does this exit poll that we’ve got make sense
(3 lines omitted), that kind of majority is not a surprise to you?
EC I think it will be over a hundred, it depends on how it pans out in the
different uh constituencies (3 lines omitted) but there is no doubt that
the uh/that the/the vote was decided a long time ago, I’m not so sure that
the campaign made a scrap of difference, except in this sense (.) that it’s
clear that going Euro-sceptic did not help the Conservatives one jot, it did
not bring in extra votes and in fact I think it drove some more voters away
JD ! Interesting thought, we’ll pursue that in/in more detail with others later . . .

The controversial topics here are the possible causes for the defeat of the Conservatives
and the Party’s stand on joining the European Common Currency, which had caused a deep
split in the party.
Excerpt (2) exhibits the classical form of voicing: There is first a frame in which the quote
is ascribed to a source, done here in a neutral manner with the reporting verb ‘say’, then comes
the quote itself, and then a questioning element: Do you agree with that? The position of frame
and question element may be reversed, and one or both of them may be omitted. The
controversial comment here is ascribed to Secretary of Defense Michael Portillo and
concerns again the topic of disunity in the party as a cause for the election defeat.

(2) BBC97I/1.22 David Dimbleby interviews Deputy Prime Minister

Michael Heseltine at Conservative Head Quarters
DD ! Michael Portillo says that it’s disunity that was the problem for the
Tory Party. Would you agree with that?
MH Well I think it’s important to have a (.) mature period of reflection, to look
at all the questions, all the answers. And, what is critical is that the
Party finds a way to (.) regroup, unite, and begin to fight back. We are
a hugely successful political party, and what I do think is a great danger
is (.) now that we sort of get into a sort of (unenthusing) process uh in the
uh heat of (.) election results uh which/uh frankly uh/ the analysis is
better done within the private discussions that must go on within the
Conservative Party.

The responding turn then is the place where the interviewee can disagree with the
controversial voice, or align with it, or seek some compromise solution and do neither. The
dispreference markers at the beginning of the response in (2) (well, I think, unfilled pause
202 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

(cf. Levinson, 1983:332–345) are a formal signal that this is a dispreferred topic for the
interviewee and that a disagreeing response is coming up. The disagreement is however
implicit and has to be inferred from the contrastive stress on I and from the rest of the turn
which lays out how things should be done in contrast to how the quotee conducted himself.
In this respect, there is implicit personal criticism of a party colleague, but the issue
question (the divisions in the party) is not attended to. As in the other cases to be discussed,
both interviewer question and interviewee response exhibit a high density of evaluative
language, or descriptions of behavior from which positive or negative evaluations can be
inferred.4
In excerpt (3), Conservative candidate Edwina Currie’s critical remark on divisions in
the Party and on Europe, which was earmarked for recycling in (1), is quoted to a member
of the (outgoing) Cabinet:

(3) BBC97.I/1.18 Jeremy Paxman interviews Health Secretary Stephen

Dorrell (C)
JP ! Do you agree with Edwina Currie that it was the divisions in your party
and particularly the behaviour of the Eurosceptics that did for you?
SD ! I don’t accept by any means that the result of the election has yet been
decided or at least that it is yet known. Er (.) I do agree with what I heard
Michael Portillo saying earlier on in your programme, namely that a party
that presents a united face to the electorate (.) is a party that strengthens
its claim to their support.

The response is typical of one of the Conservative interviewees’strategies for dealing with
such quotes during the night, namely not to accept defeat for as long as possible. That has the
advantage that all follow-up questions on causes, responsibilities, blame, and party
leadership can only be asked in hypothetical mode. Accordingly, there is a vehement negative
reaction in first position here, on the grounds that the results are not yet known. This can be
done directly and without mitigation since it contradicts no-one directly. However, indirectly
again the quotee is criticized – for speaking precipitately, and in a way negative to the Party to
boot. The interviewee then addresses the issue (divisions). The way he does this shows that
interviewees as well may make use of the practice of voicing – here by quoting an alternative
voice that purportedly rebuts the position taken by the first voice quoted by the interviewer.
This allows the interviewee to align himself with high-status right-wing anti-Europe Michael
Portillo, rather than with low-status left-wing pro-Europe Edwina Currie. Again, this is done
implicitly and the particular status relations and policy positions have to be inferred, requiring
considerable background knowledge concerning the factions within the Tory party and the
status of quotees. Interestingly, as regards the practice of selective or biased quoting, the
interviewee here chooses the ‘united-face’ soundbite from the same interview from which the
interviewer had picked his ‘divisions’ quote in excerpt (2). In excerpt (3), the controversy

4
For the study of evaluation in language, appraisal theory has recently supplied an adequate descriptive system,
which underlies the analysis in this paper, but for lack of space cannot be explicated here (but see below and
Hunston and Thompson, 2000a,b; Martin, 2000; White, 2000; Lauerbach, in press-b).
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 203

over the divisions in the Party is being acted out and personalized in a variant of the ‘‘he said –
she said’’ script of everyday discourse.
Interviewers can also confront interviewees with voices that criticize and blame them
personally, as in the interview in excerpt (4) with Kenneth Clarke, whose move against
Europe was widely thought to have caused the split in the Conservative Party. There is
criticism voiced by backbenchers (who remain anonymous), and Nicolas Winterton, who
in terms of being outspoken seems to be comparable to Edwina Currie. In distinction to the
previous cases, these quotes are not a general criticism of the divisions but a direct attack on
the person who is held responsible for them – and who is the addressee:

(4) ITV97.I/2.32 Jonathan Dimbleby interviews Kenneth Clarke (C) in

Rushcliffe
JD ! What do you have to/do/we’ve already had some (.) back benchers
speaking out in very contrary ways about the cause of it and Nicolas
Winterton (.) who never keeps his light under a bushel, as it were, h/has/has
has blamed it all on you
KC Well I think there’re plenty of people who work for the party up and down
the country who hope uh that those back benchers in the next
parliament (.) uh on the Conservative side behave a little more sensibly
than the back benchers on the (.5) Conservative side did at times in the last
Parliament, and I/ my own advice I’d give to Nicolas is to (.) stop
bursting out into recriminations within two hours of the polls’ clothing/closing
is not the right preparation for opposition, uh my own view is I would have
liked to see mo:re about the economy (.) in this election, I’d have seen/liked
to’ve seen/heard more about the economy over the next five years, I would
have liked to’ve heard a lot less about Europe, and obviously I’d have liked
to’ve heard a lot less about sleaze, but (.) no party can completely control
the agenda, the modern media, the modern election has a life of it’s own,
now you/you/you can’t just suddenly leap instantly to decisions about why
the electorate (.) have decided as they have.

The response is, firstly, mitigated criticism of the first voice (the backbenchers).
Regarding the second voice, there is, secondly, a strong counter-reaction and again, as in
(3), this is personalized and directed at the messenger, not the message (what the
interviewee is blamed for). The message is addressed in third position of the turn, and
amounts to evading personal blame by implicitly criticizing the campaign. This criticism is
however mitigated by some media-critical reflections on the difficulty of controlling the
campaign agenda in mass media democracies, which to some extent exculpates the
campaign manager.5 Note also that it is the issues of the campaign that are addressed here,
not the person who was responsible for it. There is very intricate maneuvering here in
distributing responsibility and blame, and this contributes to the audience’s image of the
web of inner-party relations that is being constructed throughout the night.

5
See Fetzer (this volume) on further functions of explicit reference to the media in political interviews.
204 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

Edwina Currie can be relied on to speak out on other sensitive topics as well – she also
provided a suitable soundbite on the topic of the campaign, with which another high-
ranking Conservative politician is confronted:

(5) BBC97.I/1.04 Jon Sopel interviews Deputy Chairman Michael Dobbs (C)
at Conservative Head Quarters
JS ! Mr. Dobbs, Edwina Currie this evening has said that the Tory campaign was
a campaign (.) run by kids, it’s been a shambles, she said.
MD Well, perhaps I can thank Edwina for all the helpful advice that she was
giving to us during the campaign a/a/and perhaps get on to more (.) serious
subjects. Erm, I mean, if these exit polls are/are accurate e:r and we are
going to lose tonight uh I can still stand here and say I take tremendous
pri:de in having helped fight elections for eighteen years. You know,
when we (.) fought for the nineteen seventy-nine election we took over a
country which was (.) crucified by discontent, split apart, and if we are
handing the country over (.) to another government tonight it’s a country
which has a strong economy, which is/ there are not riots in the street=
JS But if you have done so well, why have you done so badly? . . ..

This is reacted to by sharp rebuttal which is (1) directed not at the criticism itself but at
Edwina Currie as its source, and (2) aggravated by using heavy irony: perhaps I can thank
Edwina for all the helpful advice she was giving to us during the campaign. This is a case of
classical Gricean irony which, unlike other cases discussed by Sperber and Wilson (1986) in
criticizing Grice’s account, does arise from an infringement of the maxim of quality and,
triggered by prosodic cues and contextual knowledge, leads to an inference of the opposite of
what is said: the advice Edwina Currie gave was unhelpful, damaging, etc. Interestingly, it is
not only propositional meaning that is reversed in the process, but also illocutionary force:
The explicit performative I can thank, expressing an interpersonally positive act, in its
ironical reading needs to be interpreted as a socially negative act like criticize or accuse. This
is followed by an agenda change to expound the merits of the Conservative Party, which in
turn draws a sarcastic interviewer challenge: But if you have done so well, why have you done
so badly?
In excerpt (6) both the voice of Nicolas Winterton blaming Kenneth Clarke for the split
in the Party, and that of Lord Tebbit blaming Brian Mawhinney, the Tory campaign
manager, for his management of the election campaign, are being played back to the
interviewee:

(6) ITV97.I/1.31 Jonathan Dimbleby interviews Foreign Secretary Malcolm

Rifkind (C) in Edinburgh Pentlands
JD ! Uhm on/ just uh one or two of your senior people are already speaking out,
Nicolas Winterton blaming it on Ken Clarke, N/Norman Tebbit, L/Lord
Tebbit saying I don’t know who was leading the campaign, I don’t think
Brian Ma/Mawhinney did either, what do you make out of all that?
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 205

MR Well I know Ken Clarke very well, he’s a man of great integrity, of great
honesty, he speaks his mind and I think that’s something which uh ought
to be admired not criticized or reviled.

The response is interesting because it engages only with the first voice and comes out
strongly in support of Clarke – again: of the person, not of the message, i.e. not of the
divisive policies on Europe that Clarke has been accused of. Following Grice’s (1975) first
maxim of quantity and the maxim of relation, an inference can be drawn that an
endorsement of this policy is significantly absent. What is also conditionally relevant but
missing is any interviewee reaction to the second part of the question – again inferences can
be drawn from this regarding the interviewee’s position on the campaign, or on Lord
Tebbit, or on Brian Mawhinney, or on any combination of the three.
Excerpt (7) is an example of the interviewee being confronted with a controversial
comment he himself made earlier that night. The interviewer quotes back to him something
he had said on the radio, incorporating the very words he used into an accusatory question:

(7) BBC97.I/1.22 David Dimbleby interviews Deputy Prime Minister

Michael Heseltine (C)
DD ! And why should the electorate have sleepwalked, in your words,
to disaster?
MH Well those are the sort of questions we’ll have to ask ourselves and, as I said,
I don’t myself go for this sort of (. . .) technique uh within hours of the
polls closing. I think these will require mature reflection [among] very
senior colleagues who want to play their part in it. And there’s absolutely
nothing to be gained from the debates when one person says one thing,
one says another, one asks one question, one gives a different answer,
that is a way which is very fascinating to the media but it is not
the best interest of the Conservative Party.
DD But of course it did sound as if you weren’t blaming your colleagues
but blaming the electorate when you said they were sleepwalking to
disaster if they voted Labour.

The question elicits a criticism of precipitate public judgments on election night and,
against the evidence, a flat denial that the interviewee is prone to them. Interestingly, party
hierarchy is invoked in respect of who should be part of post-mortem discussions. This
excludes the lower ranking party members and, outside the party, the media. Apart from the
explicit reference to party hierarchy, the excerpt is also interesting for the explicit criticism
of the media. A strong counter attack is launched on the media’s practice of orchestrating a
fictive debate by quoting interviewees to each other. The interviewer challenges this
response, but only on the grounds that the question has not been answered, while ignoring
the interviewee’s critical comment on journalistic practice.
But of course Heseltine is absolutely right in his analysis – the media are indeed
orchestrating a debate, and they are doing it by playing different voices against each other,
not between the parties, as Blumler and Gurevitch (1995) have noted for election
campaigns, but within the Conservative Party. The mechanism is quite simple: a source
206 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

which is known to be outspoken is interviewed on sensitive topics early in the night and his
or her provocative comments are then fed to other interviewees to elicit from them more
provocative comments to be recycled again, and so on (see also Clayman and Heritage,
2002:227 who call this interviewer practice ‘split-hunting’). This example also shows how
journalists not only rely on their own increasing corpus of quotes of the night but also
monitor other media outlets like radio and rival television channels for controversial
soundbites.
To summarize the forms and functions of discourse representation in the interviews
with Conservative politicians, a critical and controversial voice is quoted directly or
indirectly by the interviewer to the interviewee. The voice can be a party member’s but
may also be the interviewee’s own. If it is a party member’s, this member may be
relatively high or low in the party hierarchy, and may be friend or foe of the interviewee.
What the voice is quoted with is a negative opinion on the Party, specific party members,
or the interviewee himself. The negative evaluation can be explicit in the semantics (the
campaign was a shambles; blames it all on you) or implicit. If implicit, it may be
reconstructible without the situational (but not without the cultural) context: the
electorate sleepwalked to disaster is just not something a politician can say in a
democracy. The negative evaluation can also be dependent on the situational context, as
in excerpt (6): I don’t know who was leading the campaign and I don’t think Mawhinney
did either requires the specific knowledge that Mawhinney was the campaign manager in
order to be understood.
How the interviewees respond to this depends not so much on what is quoted but who
the sources are. The evidence for this is that in most cases in the data, the responses deal
with the individuals voiced and not with the issues that the quotes are about. Depending on
the relation between interviewee and source, complex negotiations of identity and relations
are conducted in the responses with direct, indirect and implicit means. The interplay of
disputing voices between interviewer question and interviewee response has the effect of
personalizing and dramatizing politics, and is designed to entertain the audience and keep
them watching into the early hours. But both participants in this, politics and the media,
profit from a symbiotic relationship (cf. Boorstin, 1987), in which publicity is exchanged
for information and entertainment. It may therefore be presumed that the politicians play
their part in this as well.

3.2. Labour

Turning now to interviews with Labour politicians, their situation was different from
that of the Conservatives not only in that they were winning a landslide victory while the
Tories were facing a crashing defeat. While Labour’s victory may have been also due to a
general fatigue with the Tories and a desire for change after 18 years of Conservative rule, it
was generally agreed that the root for Labour’s success lay in Tony Blair’s refashioning
what came to be known as Old Labour into what was called in the campaign New Labour or
The Third Way. This shifted the Labour Party from the left of the political spectrum firmly
into its middle. However, the move was not uncontroversial within the Party. It had required
firm party discipline to push it through, and campaign strategies such as sending left-wing
strongman John Prescott off on a bus to tour the North of England, where he could do little
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 207

harm. During election night however, Labour’s sweeping success silenced all dissenting
voices.
Where then to look for the means of creating a debate, where to find the controversial
topics and the perfect soundbites to quote? One way is to go back in history and dig up
something that was said by Labour in a comparable situation, as in (8).

(8) BBC97.I/1.41 David Dimbleby interviews Shadow Chancellor Gordon

Brown (L)
(topic: landslide victory, large majority)

DD ! But uh i/i/th/there’s not an element in that ‘‘we’re going to be more radical

than people thought’’, that ‘‘we are the masters now’’, the famous phrase
that was uttered the last time Labour had a landslide like this.
GB No no no not at all, not at all. We will honour our manifesto, we’ll fulfill it.
I think the feature of this election is that uh people thought the
Conservatives, and widely so, had betrayed the trust that they had got
from the British people in nineteen ninety-two, it was a record of broken
promises, we are determined that there is a bond of trust between
the new government and the British people, . . ..

Another way is to make quotes up. The legitimate means of doing this are provided
by the practice of ventriloquizing mentioned above, of openly putting words into
another’s mouth. Ventriloquizing does not recycle or sum up what was actually said but
what might have been said, might be thought, feared, intended, but which, it is
suggested, is not explicitly admitted, or is purposefully not talked about. In these
ventriloquisms, strong negative evaluation is directed at the Labour party, functioning as
criticism or blame. Obviously the preferred response to this is strong and unambiguous
rejection, contrasting the negative evaluation in the question with strong positive
evaluation in the response. Looking again at (8), it is actually not clear who exactly said
what is quoted. The words, with a first person plural subject referring to the Labour
Party, implicate party members as speakers. Yet what is said sounds too exaggerated and
impetuous to be plausible as something a political party would say about itself. Thus (8)
may be a case of recycling a previous ventriloquizing already, a second-degree
ventriloquism, as it were.
In (9) possible self-doubt and anxiety in the face of having to govern after 18 years in
opposition are voiced as something John Prescott might think:

(9) ITV97.I/1.05 Jonathan Dimbleby interviews John Prescott (L)

JD I uh sa/ uh I’ve never s/ well actually I’ve seen you smile quite a lot uh but I
(.)/try and keep a serious face for a second while I ask you isn’t it in fact/
you’ve been in opposition, you’ve loved opposition, you’ve gone around
s:tomping in opposition, suddenly isn’t it quite daunting to think (.)
! ‘‘we’ve actually got to run the show now for the first time (.) for a very
long time?’’
208 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

JP Of course it is, I mean people at this scale have given us such trust and
such (.) support, and uhm we’ve given our promises they’ve been carefully
thought out and we shall deliver them, make no mistake about that.

As in (8), the response is again strongly unequivocal, yet somewhat paradoxically in

agreement. This is however due to the fact that the interviewee has subtly refocused the
question: what is daunting is not the possible lack of experience after 18 years in opposition
but the scale of the victory. And in (8) and (9) alike, the rejection of possible criticism or
doubt is followed by strong commitment to the Party’s election promises and determination
to serve the people.
The next two examples focus on the shift from Old Labour to New Labour, which was
not uncontroversial in the Party itself. In (10) the interviewer again ventriloquizes the Old
Labour supporter Prescott with a critical comment he might well be expected to make but
emphatically has not made – again this elicits a strong unequivocal response, this time
disagreeing:

(10) ITV97.I/1.05 Jonathan Dimbleby interviews John Prescott (L)

JD You don’t sometimes think to yourself (.) or begin to think to yourself
when you look at this likely result, I stress it’s still an exit poll of course
basically uhm
! ‘‘did we need the quite so new Labour as we are in order to get it, couldn’t
we’ve been a little/a touch more old Labour and still got here?’’
JP No I think this is remarkable, results coming in around the country and uh
Tony Blair is going to be right (.) to uh have uh brought about those
changes as he did.

In (11) the ventriloquism is not an addressee or ingroup ventriloquism but an outgroup

one: the fictive voice is not that of a member of the Labour Party, but of the constituents of
the interviewee, and it is made to utter strong criticism regarding New Labour’s taxation
plans and the consequences for poor urban areas. The interviewer works this into a
conducive question, the expected answer to which would be a strong no:

(11) ITV97.I/2.24 Jonathan Dimbleby interviews Labour MP Harriet Harman

JD Did he turn the party on its head basically?
HH Well, he’s modernized the party, I mean people here in Camberwell and
Peckham never gave up (.) on that we could end the/ the divisions in society
and that we could have uh (.) opportunity for all, but (.) the party wasn’t
trusted-
JD But your/your people (.) Ha/ Harriet Harman in/ in your area which is a
very (.) uhm deprived area of London, one might expect that they would
be frustrated by New Labour and might say
! ‘‘Why can’t we have (.) more taxes in order to fund better services,
why can’t the rich be socked (.) so that we can have more resources here’’
and you’re not/ you don’t get that at all?
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 209

HH We want/ people here say they want Labour in government, and our pledge is
on unemployment, on hospitals and on schools and to cut VAT are exactly the
agenda for people here, but it was the modernization of the party that was
necessary to/to bring (.) those pledges into action tha/ that was what was
necessary and that’s what (.) the leadership we’ve had (.) has provided (.)
I hope.

In contrast to the previous examples, this unequivocal no is not forthcoming. Obviously,

the strongest statement the interviewee is able to make is people here say they want a
Labour government, from which the inference can be derived that she is unable to commit
herself to the stronger rejection. As in the previous excerpts, this is then followed by
endorsement of the party platform.
Summarizing the forms and functions of discourse representation with respect to
Labour politicians, it is striking that they are ventriloquized throughout. The voices that are
ventriloquized by the interviewer utter affective attitudes such as fear, anxiety, doubt, or
over-confidence about the landslide victory, and criticism about the transition from Old
Labour to New Labour. This is responded to by the interviewees with strong unequivocal
rebuttal, followed by a renewal of Labour’s election promises and commitments. No
maneuvering vis-à-vis the ventriloquized voice in indirect or otherwise mitigated
disagreement is necessary since the voice is purportedly one’s own. The exception is (11),
where it is the constituents of the interviewee that are voiced with criticism, which calls for
a more indirect rebuttal.
What are the effects of this practice? Firstly, in analogy with what is done with the
Conservatives, there is an attempt by the interviewers to construct an internal debate within
the Labour party. This fails, however, and is presumably expected to fail. What it does
succeed in is, secondly, to sow a grain of doubt regarding the exemplary unity of the Party,
to foreshadow possible debates in the future, and to name some of the potential issues. A
third effect of the practice is to portray Labour in a particular way, as discussed below.

3.3. Women, losers and the socially deprived

The practice of ventriloquizing was however not restricted to politicians of the Labour
Party. There were three Conservatives as well – two female, one male. One of the women
was again the (in)famous Edwina Currie, and the other was the equally (in)famous Lady
Thatcher:

(12) ITV97.I/0.58 Jonathan Dimbleby interviews Conservative candidate

Edwina Currie
JD Now uh uh uh what made you(.5)/you caused tremendous rage in the
party hierarchy, what made you write that article (.) and say publicly
! ‘‘we’re done for’’?
EC I’d like to put the records straight, I didn’t write an article at all . . .
(13) ITV97.I/ 2.39 Jonathan Dimbleby and political expert Michael
Brunson discuss the interview with Kenneth Clarke (C)
210 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

JD Uhm I was very interested, I expect you were, in what Ken Clarke had
to say, Mike.
MB Well, he was (.)/you/he was characteristically forthright, wasn’t he,
our Ken, never minces his words, he was talking there about ‘‘a period
of constructive opposition’’, but I also detected there is something else. Did
you notice the way he said that basically he wished that they’d stuck on the
economy. Now for example I am hearing that Lady Thatcher (.) was
frantically sending in messages during this campaign saying:
! ‘‘For goodness sake go on the economy’’, and I’ve also heard
from Labour Headquarters that that was their worst fear that if the Tories
had hammered them every day on the state of the economy, then they really
would have had (.) a hell of a fight on their hands, but of course because
of the state of the Conservative Party it split off in all these other issues.

The man was Prime Minister John Major, who – as the media kept stressing – had led his
party to the worst defeat since the Duke of Wellington in 1832. If the Conservative Grandes
interviewed before were on the losing side of this election, Major was the loser par
excellence, and he resigned as leader of the party the following day. The topic of the
sequence in (14) is whether Major should resign soon or not, and he is ventriloquized with
what he may be saying in resigning:

(14) ITV97.I/2.35 Jonathan Dimbleby interviews Defence Secretary Michael

Portillo (C)
JD But do you think he ought to stay/I mean from the Party’s point of view
would it be a good
! thing if John Major stays/stays on as leader or if he (.) swiftly says
‘‘okay over to everyone else?’’

The practice of ventriloquizing is a very personal and intimate device: it claims to be

able to get inside the character so treated, to speak for them and to publicize their innermost
intentions, plans, fears or critical stance, and all of this in a linguistically colloquial,
personal register. As my data show, the verbal mimicry involved also extends to features of
social and gender dialect (e.g. sock them for the poor constituents in (11), for goodness sake
for Lady Thatcher in (13), we’re done for for Edwina Currie in (12), we’ve got to run the
show now for John Prescott in (9)). By the form of the words chosen, it also paints the
character in a certain light – as seen by the speaker.
One can assume, based on Goffman’s analysis of ventriloquizing as directed to pets
and prelinguistic children, that the practice works downwards on the social scale, rather
like joking and complimenting. But this would, if uncomfortably, apply to only one set
of those who are ventroliquized in my data – women, losers, and the socially deprived:
Lady Thatcher and Edwina Currie (who are both women and political losers), the soon
to be ex-Prime Minister John Major, and the socially deprived in Harriet Harman’s
constituency.
What does this tell us about the high-ranking Labour politicians that are also dealt with
in this manner? They are in the process of winning a landslide victory and will soon be
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 211

members of the Government. Yet the identity that is being constructed for them is quite at
odds with this. Perhaps the question should be asked differently: What does this tell us
about how ITV sees Labour? For it is indeed a practice that is restricted to the interviews
done by ITV – there is no ventriloquizing done at all by the BBC in our data. Based on the
findings of this paper, a hypothesis can be formulated that the condescending style which
is implicated in the practice of ventriloquizing is but one facet of a low-status identity that
ITV constructs for Labour during its election night 1997. Further empirical evidence from
our election night data will need to be looked at in order to confirm or disconfirm this
hypothesis. Such evidence will come from other practices within political interviews with
members of both Labour and Conservatives (such as degrees of indirectness and
confrontativeness within interviewer questions and question sequences), but also from
other genres in which Labour is not talked with but talked about, such as expert
interviews, analysis and commentary. It will also be enlightening to analyze the political
interviews done during election night 2001, when Labour was reconfirmed with another
landslide after 4 years of Government, and see if discourse practices have changed with
the redistribution of power. The question this will address is whether the condescending
style has to do with power rather than with Labour party politics and ideology, which ITV
(and, since it is a partly commercial channel, its advertising customers) may disagree
with.

4. Comparison

The discourse representation practices of the interviewers in my data do not construct a

dialogue or debate between the parties, as found by Blumler and Gurevitch (1995) for election
campaigns. What they do create is dialogue within the parties by emphasizing the already
highly visible rifts within the Conservative Party and creating possible controversies within
Labour. These two strands of dialogue run side by side through the night’s discourse, no
attempt is made to engage the two major parties in debate with each other. The debates are
constructed between real voices within the Conservative Party, and between fictive ones
within Labour. The effects of this practice are to personalize and dramatize political discourse
in various respects:
For the Conservatives, in formally direct, indirect and free indirect speech, party
members are quoted or voiced with controversial statements. This has the general effect of
tying policy to person in various ways:
- It dramatizes the discourse by selecting quotes that are highly critical of specific
individuals and that are expressed in emotional and colloquial language.
- It plays proponents of different points of view against each other personally.

Beyond these general effects, there are also the context-specific ones tied to the
construction of the controversial topics of election night 1997: what caused the defeat, who
was responsible, and who is willing to name the culprits.
In the responses of the interviewees the practice
- results in interviewees disagreeing with or supporting the person whose view is voiced,
rather than engaging with the position expressed, and
212 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

- it leads to interviewees producing direct or complex indirect/implicit responses in which

they negotiate their identity and status in relation to the individuals quoted, depending on
the rank these individuals occupy in the party hierarchy and on their positioning in
conflicting party factions, relative to the interviewee.

For Labour things are different: In the absence of critical voices from within the Labour
Party itself, it is the practice of ventriloquizing that is employed by interviewers to
construct a controversial dialogue with a hidden inner voice in an attempt to cast doubt on
Labour’s united front, to foreshadow future debates within the party, and name their
possible issues and protagonists.
The effects of this practice are also to personalize and dramatize, but in a different
manner. As with voicing, there are general and particular effects of the practice of
ventriloquizing. The general effects are that
- it constructs the ventriloquees as rather low on the social scale,
- and that it does this from a perspective that the speaker assumes is shared by the audience.

The particular effects are specific to this election night:

- They regulate who is included in the group that is constructed as being low on the social
scale, i.e. women, heavy losers, the urban poor – and Labour.
- Through the semantics of what is ventriloquized, Labour is portrayed in a particular
manner, namely:
- not only as inexperienced in government, but also
- as ridden by self-doubt and anxiety on the one hand,
- and as rearing to go radical on the other.
- As betraying the values of Old Labour,
- and as being insensitive to the plight of the poor.
- In employing this practice the ITV interviewer implicates that these views are shared by
the majority of the television audience he is addressing.

Predictably, the interviewees respond to this with unequivocal disagreement as well as

by stressing their commitment to their election promises and their determination to serve
the electorate well, thus constructing themselves as competent, determined and
trustworthy.
So the practice of discourse representation can be employed to dramatize and personalize
political discourse on television, and the way this works is through constructing identities and
relations of the individuals and groups concerned. The cumulative results of this practice in
the data studied here are: the picture of a hopelessly split and infighting Tory Party on the one
hand, and of an insecure and not-to-be trusted Labour Party on the other.

5. Concluding remarks

This paper has shown that the practice of discourse representation as an interviewer
resource serves aims beyond constructing interviewer neutrality, as studied by Clayman
(1992). Also, in political interviews of election nights, it is not employed in order to create a
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 213

debate between the political parties, as noted by Blumler and Gurevitch (1995) for election
campaigns, but to highlight disputes within the parties. A further function of the practice
turned out to be the creation of disputes between individuals and the construction and
negotiation of identities and relations of the interviewees within their intra-party affiliations.
In its variant of ventriloquizing, the practice was shown to create a debate even within
individuals. These debates, however, are meant to stand metonymically for debates within
their political party as a whole. The general effect of the device is one of personalizing and
dramatizing political discourse, thereby offering the audience a preferred interpretation of the
political process that will influence their conceptualization of politics.
Because the practice can be precisely defined in formal terms and analyzed for its
functions in clearly delimitable genres, it is also suitable to serve as a category for
intercultural comparisons such as the ones undertaken in the comparative television
discourse project directed by the author (see note 1). The aim of this project is to compare
discursive practices in the election nights of the British parliamentary elections of 1997, the
German ones of 1998, and the American presidential elections of 2000. An interesting
question to be pursued as a follow-up of the findings presented in this paper is for instance
whether there are differences across these cultures in the way the media use discourse
representation to construct identities and interpersonal relations with respect to power and/
or party ideology. How do German television stations and their interviewers for instance
portray the winners and losers in a comparable situation to the British elections of 1997,
when in the German parliamentary elections of 1998 the Social Democrats ousted the
Conservatives after 16 years of Christian Democrat Chancellor Kohl? How do US-
American interviewers and journalists talk to and about their winning and losing politicians
during the memorable election night 2000 which, unlike any election night before, did not
yield a decision for President and instead led to a 5-week post-election dispute in the State
of Florida? Is discourse representation used in the same functions of creating debates and
constructing identities and relations across these cultures? And if so, is it used to create the
same types of identities and relations, or different ones?
Political scientists and media scholars have pointed out the symbiotic relationship
between politics and the media in mass media democracies (e.g. Boorstin, 1987). Even
media practicioners characterize the relation as a mutual dependency culture.6 If it is
reasonable to assume that such a relationship will be enacted and reflected in discourse
practice, it follows that the answers to questions such as the ones formulated above should
yield insights not only into culturally specific discourse practices, but also highlight
culturally specific relations between the media and politics as well as characterize specific
cultures of political communication.

Acknowledgements

I thank discussants at these conferences for their critical questions, as well as my

students at Frankfurt for the odd stimulating puzzlement, and three anonymous referees for
very helpful comments on an earlier version of this paper.

6
Personal communication from a BBC political editor.
214 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215

Appendix

Abbreviations and transcription conventions in the excerpts cited are as follows:

ITV97.I/0.58 Independent Television, 1997, tape I, counter 0.58

BBC etc. British Broadcasting Corporation
(C) Conservatives
(L) Labour
(.) brief pause, too short to be measured
(.5) pause of 0.5 s.
/ break-off
= next turn latch-on
(. . .) unintelligible
! points to phenomenon under discussion
vowel: lengthened vowel
bold type emphatic stress
italics ventriloquism

Overlap and interruptions were infrequent in these interviews. Where they did occur, they
were considered in the analysis (as e.g. contributing to the force of a responding rejection)
but removed from the transcript for better legibility (as for instance in excerpt 8). The same
procedure was followed with regard to most comments pertaining to prosody, non-verbal
behaviour, camera work, and editorial work like split screens, cutting, etc., since they did
not contribute significantly to the points being made here.

References

Bakhtin, Mikhail, 1981. The Dialogical Imagination. University of Texas Press, Austin.
Blumler, Jay G., Gurevitch, Michael, 1995. The Crisis of Public Communication. Routledge, London.
Boorstin, Daniel, 1987. Das Image – Der Amerikanische Traum. Rowohlt, Reinbek.
Caldas-Coulthard, Carmen Rosa, 1994. On reporting reporting: the representation of speech in factual and
factional narratives. In: Coulthard, Malcolm (Ed.), Advances in Written Text Analysis. Routledge, London/
New York, pp. 295–308.
Clayman, Stephen, 1988. Displaying neutrality in television news interviews. Social Problems 35, 474–
492.
Clayman, Stephen, 1992. Footing in the achievement of neutrality: the case of news interview discourse. In: Drew,
P., Heritage, J. (Eds.), Talk at Work. Cambridge University Press, Cambridge, UK, pp. 163–198.
Clayman, Stephen, Heritage, John, 2002. The News Interview: Journalists and Public Figures on the Air.
Cambridge University Press, Cambridge.
Fairclough, Norman, 1988. Discourse representation in media discourse. Sociolinguistics 17, 125–139.
Fairclough, Norman, 1992. Discourse and Social Change. Polity Press, Cambridge.
Goffman, Erving, 1974. Frame Analysis—An Essay in the Organization of Experience. Northeastern Press,
Boston.
Goffman, Erving, 1981. Footing. In: Goffman, Erving (Ed.), Forms of Talk. Philadelphia University Press,
Philadelphia, pp. 124–159.
Grice, Herbert Paul, 1975. The logic of conversation. In: Cole, P., Morgan, J. (Eds.), Speech Acts (=Syntax and
Semantics 3). Academic Press, New York, pp. 41–58.
 G. Lauerbach / Journal of Pragmatics 38 (2006) 196–215 215

Heritage, John, Roth, Andrew, 1995. Grammar and institution: questions and questioning in the broadcast news
interview. Research on Language and Social Interaction 28 (1), 1–60.
Hunston, Susan, Thompson, Geoffrey, 2000a. Evaluation: An introduction. In: Hunston, Susan, Thompson,
Geoffrey (Eds.), pp. 1–27.
Hunston, Susan, Thompson, Geoffrey (Eds.), 2000b. Evaluation in Text. Authorial Stance and the Construction of
Discourse. Oxford University Press, Oxford.
Jucker, Andreas, 1986. News Interviews: A Pragmalinguistic Analysis. Benjamins, Amsterdam.
Lakoff, George, 1987. Woman, Fire and Dangerous Things: What Categories Reveal about the Mind. University of
Chicago Press, Chicago.
Lauerbach, Gerda, in press-a. Opting out of the media-politics-contract: discourse strategies in confrontational
political interviews. In: Bondi, M. (Ed.), Selected Papers from the 8th IADA Conference. Bologna 2000.
Niemeyer, Tübingen.
Lauerbach, Gerda, in press-b. Political interviews as a hybrid genre. TEXT.
Leech, Geoffrey, Short, Michael, 1981. Style in Fiction. Longman, London.
Levinson, Stephen, 1983. Pragmatics. Cambridge University Press, Cambridge.
Martin, Jim, 2000. Beyond exchange: Appraisal Systems in English. In: Hunston, Susan, Thompson, Geoffrey
(Eds.), pp. 142–175.
Short, Michael, 1988. Speech presentation, the novel and the press. In: van Peer, W. (Ed.),The Taming of the Text.
Routledge, London, pp. 61–81.
Sperber, Dan, Wilson, Deidre, 1986. Relevance: Communication and Cognition. Blackwell, Oxford.
White, Peter, 2000. Dialogue and inter-subjectivity: reinterpreting the semantics of modality and hedging. In:
Coulthard, M., Cotterill, J., Rock, F. (Eds.), Dialogue Analysis VII: Working With Dialogue. Selected Papers
from the 7th International Association of Dialogue Analysis Conference. Niemeyer, Tübingen, pp. 67–80.

Gerda Lauerbach is Professor of English Linguistics and Applied Linguistics at the Goethe University,
Frankfurt/Main. She holds a PhD in Linguistics from Frankfurt University, where she also did her Habilitation
with a study A Grammar for Interaction. Her teaching and research focus on critical discourse analysis, pragmatics
and semantics. She has published on the semantics and pragmatics of conjunction, on discourse practices and
ideology, on genre, preference and inference, on argument and rhetoric, and on the interaction of the visual and verbal
modality in classroom and television discourse. The focus of her current work is media discourse, latest publications
are on political interviews and on press conferences. She is currently working on a project on the cross-cultural
analysis of television discourse, focussing on election nights on US, British and German television, and on the
discursive construction of the American (post-)Election 2000/the Florida Recount by CNN International.