Printer Friendly

Un estudio de elicitaciones y estilo del entrevistador en preguntas formuladas en entrevistas politicas televisivas en ingles.

The Caterpillar and Alice looked at each other for some time in silence: at last the Caterpillar took the hookah out of its mouth, and addressed her in a languid, sleepy voice.

'Who are you?' said the Caterpillar.

This was not an encouraging opening for a conversation. Alice replied, rather shyly, hardly know, Sir, just at presentuat least I know who I was when I got up this morning, but I think I must have been changed several times since then.' 'What do you mean by that?' said the Caterpillar, sternly. 'Explain yourself!' 'I can't explain myself, I'm afraid, Sir, ' said Alice, 'because I'm not myself, you see.'

Lewis Carroll, Alice's Adventures in Wonderland

1. Introduction

Since its infancy in the early 1950s, televised political interviews as a genre developed from a rather deferential style of interviewing -with both interviewers (IRs) and interviewees (IEs), political figures, strictly following a pre-arranged agenda of questions- to an adversarial style which put politicians in more challenging situations and difficult questions. It is no wonder then that being one of the most important forms of public speaking, broadcast interviews quickly attracted considerable research attention in Sociology, Linguistics and Communication Studies.

Different approaches have been taken to study broadcast talk, including news interviews, talk shows and phone-in programmes. Focusing on lexis, grammar, interviewer neutrality, speech acts, and discourse structure, researchers have tried to explain conventionalised speaking practices and how participants pursue their goals and tasks during an interview.

Conversation Analysis (CA) emerged in the 1960s as an approach to the study of social life and intelligible spoken interaction. In the field of Linguistics, CA has focused on identifying and analysing recurring patterns at micro- and macro-levels of different types of discourse, including medical consultations, small claims courts and broadcast interviews. In 1994, Tsui (University of Hong Kong) proposed one of the most comprehensive and detailed models of English conversation. Her work integrates and expands the contributions and analyses presented by outstanding previous researchers. In her taxonomy of discourse acts, Tsui distinguishes three primary Head Acts (and their corresponding sub-classes of acts) based on their structural location and prospected response, namely 'initiation' (I), 'response'(R) and 'follow-up'(F). Recently, in 2002, Huddleston and Pullum (University of Queensland and University of California, Santa Cruz, respectively) explored and discussed questions from a semantic and pragmatic perspective, giving an insightful account of clause types and illocutionary force. In the late 1990s and early 2000, Judy Elliot and Peter Bull (university of York) developed a categorisation of face threats in questions posed during political interviews. In their model, 19 different types of face threat were identified, grouped into three sub-classes of face which politicians need to defend: their own personal face, the face of the party and the face in relation to significant others.

These three descriptive frameworks, two linguistic and one sociological, will be employed in analysing an extensive (28-minute) televised political interview: journalist Jeremy Paxton interviewing Labour Party candidate Tony Blair in the 2005 UK General Election.

The aim of this study is threefold:

(a) to distinguish the most commonly occurring sub-classes of elicitations and their corresponding grammatical structure

(b) to identify the most frequently occurring categories of face threat generated in the elicitations

(c) to analyse the interviewer's neutrality and style of eliciting

The major focus of this paper will be to explore aspects of interviewer style when eliciting information from three different perspectives.

2. Theoretical Framework

2.1. General background

2. 1.1. Neutralism in interviews

In democracies, news organisations are often expected to aim for objectivity: Reporters try to cover all sides of an issue without bias, as compared to commentators or analysts, who provide opinion or personal point of view. Clayman and Heritage (2002: 321) explicate that a more appropriate standard for neutrality and objectivity should be fairness and accuracy. Under this standard, taking sides on an issue would be permitted as long as the side taken was accurate and the other side was given a fair chance to respond. Many professionals believe that true objectivity in journalism is not possible and reporters must seek balance in their stories (giving all sides their respective points of view), which fosters fairness.

Various techniques have been established in order to assess objectivity and impartiality in broadcast journalism, being the main focus the degree of balance in treatment of rival politicians, opposing camps on a controversial issue, and other factors such as topical content, timing, lexical choice and primacy in order of presentation. Elliot and Bull (1996) developed a typology for the analysis of face threats in questions posed during political interviews. Their Face Model provides a means of analysing the face structure of the questions and provides a greater understanding of the relationship between questions and responses, giving an explanation for why a politician will feel constrained to avoid giving certain answers. It has a number of practical applications and can be used in the comparative analysis of the performance of both interviewers and politicians. For instance, it can provide a measure of the toughness of an interview, using the presence or absence of a 'no necessary threat' (NNT). Following is the list of face threats category they propose:

(i) Personal Political Face

1. Negative impression of personal competence

2. Negative impression of public persona

3. Losing credibility

4. Contradicting personal past statements/policies

5. Personal difficulties in future

6. Failure to present a positive image of self

7. Failure to present a positive image of party

(ii) Party Policy

8. Negative impression of party

9. Contradiction between party's holidays, statements, etc.

10. Party difficulties in future

(iii) Face In Relation to Significant Others

11. Not supporting significant body of opinion in electorate

12. Not supporting sub-group of own party

13. Not supporting a colleague

14. Supporting a negatively valued other

15. Not supporting positively valued people/institutions

16. Not supporting a friendly country

17. Negative impression of the 'state of the nation'

The task of the interviewer being to elicit information and opinion from the interviewee, interviewers must attend to legal and, in some instances, moral constraints, and they should maintain impartiality and balance in their coverage of current affairs and should not comment in favour or against matters of public policy. In general terms, apart from organisations like the UK government agency OFCOM (Office of Communications), there are charters, licences and acts which set the terms of reference for television and radio organisations and the penalties for infringement of these obligations. Besides, broadcasters are well aware of the fact that the audience tends to favour neutralism and departures from neutralistic interviewing may result in a drop in rating.

Greatbatch (1998: 167) describes two major requirements for interviewers: "(1) refrain from the direct assertion of opinions on their own or their employers' behalf, and (2) refrain from overt affiliation with, or disaffiliation from, those expressed by interviewees". Naturally, IRs produce utterances that contain some assumptions, and might therefore be considered as biased by the audience. Sometimes, their conduct will not be viewed as neutral, for example if their style of questioning is aggressive or hostile. Greatbatch (1998: 168) also provides a list of characteristics of neutral IRs. These include:

--IRs should produce utterances that are least minimally recognizable as 'questions'.

--Distance themselves from evaluative statements by attributing them to third parties (who may or may not be named).

--Avoid responses which are characteristically produced by 'questioners' in private conversation, as well as other forms of broadcast talk, but which could be taken as indicators of agreement or disagreement with what an IE has said, such as acknowledgement tokens ('mm hm', 'huh huh', 'yes' etc), news receipt objects ('oh', really', 'did' you'' etc) and assessments.

Interestingly, Greatbatch (1998: 169) points out that IEs commonly collaborate in the maintenance of the neutralistic stance of the IRs, regardless of their own opinion about the IRs' motivation, and treat difficult questions as part of the modern style of interviewing. And in general when IEs perceive questions as biased, they make use of various strategies to respond without directly challenging the neutral position of the IRs. Furthermore, IEs defend their neutralism by, for example, enumerating facts, citing opinions of third parties, or directly defending their neutrality.

Important topics for future research include analysing the limits of neutralism and a comparative study of interviewing in different countries. Preliminary work in this area, for example, has begun to identify important differences between interviewing in the UK and USA. Heritage et al. (cited by Greatbatch 1998: 183), for example, have found differences in content and the level of formality in news interviews.

2.1.2. Politicians and control in the interview

Some important explorations have been made concerning the methods through which politicians avoid difficult questions and deal with challenges to face. Two important means are mentioned by Elliot and Bull (1996): agenda shifting and equivocation, as follows:

(a) Agenda shifting: Although interviews are ritualised and where it is the convention for IRs to pose questions and the IEs to limit themselves to responding to them, IEs can create opportunities to change topic and direct the course of the interaction to a more advantageous direction. Examples for evading the challenging aspects of a question include making requests for permission to speak and reformulating a question. IRs note these strategies and take remedial actions to lead the interaction to its original course.

(b) Equivocation: In a study of eight political interviews from the 1987 British General Election, Bull and Mayer (1993) confirmed the popular perception that politicians frequently fail to answer some questions in political interviews. Margaret Thatcher, for example, replied only to 37% and Neil Kinnock to only 39% of IR questions. In the same paper, both authors distinguished 30 different forms of equivocation, the most common one being 'attacking the question'.

2. 2. Conversation Studies

2.2.1. Conversation Analysis

Conversation Analysis (CA) emerged in the 1960s in sociology and anthropology, it was mainly used as a model and method to study how conversation is socially organised and managed by participants. The pioneering work of Sacks, Schegloff and Jefferson produced an important corpus of ethnomethodological studies that focused on the social organisation which underlies intelligible naturally occurring spoken interaction. CA used inductive research procedures to identify patterns in verbal and non-verbal interaction, but it did not work with a priori hypotheses. One of their fundamental contributions was the descriptive units of conversational interaction labelled as 'turn', '(adjacency) pair' and 'sequence'.

Although CA started as the study of ordinary conversation, it soon included the study of a wide range of social interactions, including medical consultations, psychiatric intake interviews, calls for emergency, business meetings and broadcast interviews. As for CA research on the news interviews, studies have focused on various aspects (Greatbatch 1998), such as allocation of opportunities to speak, the design of IR utterances and IE responses, topic introduction and change, and disagreement between IEs (Emmerson 2006).

2.2.2. Linguistic approaches to conversation

A different approach to the study of conversation has developed in the field of linguistics: an analysis of the relation between linguistic features and the contexts of the situation. Another feature is that linguists usually work with a priori theories and assumptions and use conversational data to confirm a theory or model.

Using a corpus taken from classroom data, Sinclair and Coulthard (1975) proposed a descriptive framework for analysing spoken discourse. They suggested the following descriptive units: 'acts', 'move', 'exchange', 'transactions' and 'lesson'. These units were ordered hierarchically, acts combining to form moves, moves combining to form exchanges and so on. For Sinclair and Coulthard (1975: 34), an act, as a unit, is characterised according to its function in the discourse: "... the discourse value of an item depends on what linguistic items have preceded it, what are expected to follow, and what do follow." That is why, for example, we can identify an utterance as a question, its function being to solicit a response. The two authors also point out that the structural location of an utterance in the discourse determines what type of act it will perform. The same utterance can be used as a reply to an elicitation or as an informative, for example. The structure of conversation

Following Sinclair and Coulthard (1975), the basic organisational unit of classroom discourse is a three-part exchange: initiation, response and follow-up. However, this organisation model does not work well when it comes to study non-classroom discourse. Coulthard and Brazil (1981) and Tsui (1994: 25) suggested that non-classroom discourse is mostly organised in terms of two-part exchanges with an optional third part. Tsui (1994: 25) gives the following example to show that the three parts in the exchange form a bounded conversational unit:
I   C: Can I just use your lighter?   (1st pair part)
    I've run out of matches.
R   B: Oh aye ahh                     (2nd pair part)
F   C: Ta                             (3rd pair part?)

It is interesting that in face-to-face interaction, the follow-up move is usually realised by non-verbal gestures such as a nod, a smile, an eyebrow raising and the like, hence the importance of paying attention to these when analysing televised interviews. Obviously, in other types of interaction, as in telephone conversations, the follow-up move must be verbalised.

Finally Tsui (1994: 43) concludes that "... conversation is a potentially three-part exchange with an optional fourth or fifth part", and that the structure of conversation can be seen as exchanges which have three elements of structure, an initiation, a response, and a follow-up, which is optionally recursive. A taxonomy of discourse acts

Tsui presents a taxonomy of acts based on the criteria of structural location and prospected response. She distinguishes three primary classes of acts which are head acts of the three moves of an exchange, as we can see in the table below (Tsui 1994: 61):
Elements of     I              R               F1

Move            Initiating     Responding      Follow-up (1)
Head act:       Initiating     Responding      Follow-up (1)
primary class   (Initiation)   (Response)
Head act:       Elicitation    Positive        Endorsement
subclass        Requestive     Negative        Concession
                Directive      Temporization   Acknowledgement

Elements of     F2

Move            Follow-up (2)
Head act:       Follow-up (2)
primary class
Head act:       Turn-passing

The class occurring at the head of the 'initiating move' can be labelled as 'initiating acts'; that occurring at the head of the 'responding move' can be identified as 'responding acts'; and that occurring at the head of the 'follow-up' can be identified as 'follow-up acts'.

(a) Sub-classes of initiating acts with their corresponding discourse functions include:

--Elicitations, which elicit an obligatory verbal/non-verbal response.

--Requestives, which solicit non-verbal actions and the addressee is given the option of carrying out the solicited action. These have often been referred to, in the speech act theory, as request, invite, ask for permission, and offer.

--Directives, which solicit non-verbal actions and the addressee is NOT given the option of carrying out the solicited action. They subsume those utterances which have been referred to as order, command, and instruct.

--Informatives, which have the function of providing information, reporting events or states of affairs, recounting personal experience and expressing beliefs, evaluative judgements, feelings, and thoughts.

(b) Sub-classes of responding acts

--Preferred and dispreferred seconds

CA proposes the notion of 'preference organisation', which considers that not all responding utterances are of equal status. Thus we might distinguish between 'preferred' seconds, which are given without delay and contain less linguistic material, and 'dispreferred' seconds, which take more time and contain more linguistic material. The example below contains three responses:

A: How about coming to the cinema with me tonight?

B1: I would love to.

B2: hehh Well that's awfully sweet of you but I don't think I can ... I've got to finish some reports ... and ... uhm ...

B3: um I don't know.

In B1, the acceptance is simple and given without delay. By contrast, the refusal in B2 is delayed using expressions like hehh, hh, uhm, well, mitigating refusals like I don't think I can and explanation of the refusal. Based on these linguistic features, we may say that there are two types of responding acts depending on the type of response: 'positive' and 'negative'.


Another type of response is example B3. It is neither a positive nor a negative responding act. The speaker is postponing the decision-making. This type of response is labelled as 'temporization', which is a dispreferred response and also contains fillers, particles, and the like.

(c) Subclasses of follow-up acts

--Endorsement and Concession

Compare the two examples below:

i) A: How about coming to the cinema with me tonight?

B: I would love to.

A: Great

ii) A: How about coming to the cinema with me tonight?

B: hehh Well that's awfully sweet of you but I don't think I can ... I've got to finish some reports ... and ... uhm ...

A: That's too bad.... Never mind ... maybe next weekend ...

In (i), B's response is followed by A's acceptance of the positive outcome of the interaction. By contrast, in (ii), B's negative response produces an utterance that accepts the negative outcome. These two follow-up moves are realised by different follow-up acts which cannot occur interchangeably and hence are labelled differently as endorsement and concession respectively.

Endorsements are typically realised using items like good, great, smashing, thank you, ta, etc., and they can also be used to make a comment on the information provided, as in:

A: Have you finished the report yet?

B: Ehh no I haven't.

A: Oh that's unbelievable.

--Acknowledgements are produced to indicate that the response has been heard, understood and accepted, and that the interaction has been successful. Examples include a closed set of items like okay, right, oh, I see, or a repetition of the response in low key. Acknowledgements should not be confused with responses to informatives.


A second follow-up move or follow-up moves subsequent to a first follow-up move are considered as turn-passing signals. Examples are items like yeah, okay, or alright and commonly produced in polite formulaic exchanges, such as well-wishing, greeting, and so on.


Sinclair and Coulthard (1975) introduced this term to describe utterances in the classroom. They describe them as follows:

An elicitation is an act the function of which is to request a linguistic response--linguistic, although the response may be a non-verbal surrogate such as a nod or raised hand. (p. 28)

--Subclasses of elicitations

According to Tsui (1994), elicitations, irrespective of their syntactic form, can be divided into six subclasses.

(a) Elicit: inform

These are elicitations which invite the addressee to supply a piece of information, e.g. yes-no questions, wh-questions and declaratives.

(b) Elicit: confirm

These elicitations invite the addressee to confirm the speaker's assumption(s). This subclass commonly includes tag interrogatives, declaratives and positive/ negative polar interrogatives.

c) Elicit: agree

They are elicitations which invite the addressee to agree with the speaker's assumption that the expressed proposition is self-evidently true. It is usually realised by tag interrogatives and negative polar interrogatives, both spoken with a falling tone.

(d) Elicit: commit

In addition to a verbal response, elicit: commit also elicits commitment of some kind, as in:

[right arrow] J: Can I talk to you?

S: Sure. Come in. Let's close the door. Have a seat.

(e) Elicit: repeat

These two subclasses refer to discourse itself. Elicit: repeat solicits a repetition and is realised by wh-interrogatives such as Who/When/Where/What did you say?, Say that again? or words such as Sorry?, Pardon?. Note that in What did you say? the word what is prominent and is usually realised with a rising tone.

(f) Elicit: clarify

Elicit: clarify requires a clarification of a preceding utterance or utterances. Common examples are produced with wh-questions such as What do you mean?, Which room? or Where?

--Responses and challenges to elicitations

Tsui (1994: 162) proposes that the illocutionary intent of A's elicitation is to get B to provide a piece of information. It presupposes that:

a. the speaker does not have the information and wants to (sincerely);

b. the speaker has the need and the right to ask for the information;

c. the speaker has reason to believe that the addressee has the information;

d. the speaker has reason to believe that the addressee is willing to supply the information.

Apart from these presuppositions, there are those which pertain to all illocutionary acts:

e. the addressee can hear what the speaker says;

f. the addressee can understand the meaning conveyed.

Sometimes, it occurs that a response does not fulfil one or more of these presuppositions. In that case we can say that it challenges its pragmatic presuppositions. In this model, moves which challenge the presuppositions of the preceding utterance are labelled 'challenging moves'. These commonly occur after an initiating move or after a responding move. The head act of a challenging move is performed by an initiating act. In sum, a challenging move is a kind of initiating move which challenges the presupposition of the preceding initiating move or responding move. Disagreements are often delayed and prefaced by fillers and false starts in order to reduce the facethreatening effect of a subsequent disagreement.

The following is a summary of elicitations and their corresponding responses and challenges, as defined by Tsui (1994: 165-193).

(a) Elicit: inform

A positive responding act provides the information that an elicit: inform seeks. By contrast, if the speaker expresses inability or reluctance to give the information he is performing a 'challenge' because it challenges the presupposition that the addressee has the information.

(b) Elicit: confirm

Its illocutionary intent is to get the addressee to confirm that the speaker's assumption is correct. It presupposes that:

a. the speaker believes that the expressed proposition is true, but certain things in the context have led him to doubt his beliefs;

b. the addressee is able to and will confirm that the speaker's assumption is true.

(c) Elicit: agree

Its illocutionary intent is to get the addressee to agree with the speaker's assumption that the expressed proposition is self-evidently true. It presupposes that:

a. the speaker believes that the expressed proposition is self-evidently true;

b. the addressee will agree with the speaker.

(d) Elicit: commit

Its illocutionary intent is to get the addressee to produce a verbal response which will commit him or her to the production of a further exchange(s) or a future action. It presupposes that:

a. the speaker sincerely wants the addressee to commit him/herself;

b. the addressee may be able and willing to commit himself.

(e) Elicit: repeat and elicit: clarify

Positive responses to them are a 'repetition' and a 'clarification' respectively.

2.3. Characterisation of Questions

We mentioned previously that elicitations solicit an obligatory verbal response or its non-verbal surrogate. This sub-class has often been referred to as 'questions' because of its interrogative form, and sometimes because it expects an answer or some verbal performance from the addressee.

There have been different approaches to the study of questions. Quirk et al. (1985) define questions as a semantic class which is used to obtain information on a specific point. They propose that there are three major classes of question according to the answer they expect: yes-no questions (subclasses being real yes-no questions, tag questions and declaratives), wh-questions and alternative questions. Tsui (1994) makes a critical analysis of these categories and considers the importance of intonation to mark the speaker's expectations, as suggested by Brazil (1985). Huddleston and Pullum (2002) provide an extensive and insightful discussion of questions based on semantic and pragmatic grounds. This chapter is a summary of their main ideas, especially those which are related to the focus of this paper. Some reference will be made to other contributions.
Huddleston and Pullum examine distinctions on four dimensions
(2002: 867), as shown below:

i     POLAR                    Alternative            Variable
      Is it breathing?         Is it alive or         Why isn't it
                               dead?                  moving?
ii    INFORMATION              Direction
      What time is it?         Shall I put some
                               music on?
iii   NEUTRAL                  Biased (declarative,
                               interrogative, etc.)
      Have you seen it?        Haven't you read
                               it yet?
iv    ORDINARY (NON-ECHO)      Echo
      What's he going to do?   He s going to what?

Polar and alternative make up the class commonly called 'closed questions', and both are expressed by 'closed interrogatives'. Variable questions are open questions, and are expressed by 'open interrogatives'.

As for dimensions ii-iii, the category in the left column is considered as the default, hence at a later stage we shall examine direction questions (where the answers have the force of directives, not statements), biased questions (where the speaker is biased in favour of one answer over another) and echo questions (which seek repetition or clarification of what has just been said).

2.3.1. Polar Questions

--Answers to polar questions

Typically a polar question has as answers a pair of polar opposites, positive and negative. The choice between YES and NO is determined by the answer.

--The form of polar questions

The prototypical form is a closed interrogative clause with rising intonation, as in Is it breathing? (R) Other possibilities include:

--Declarative (usually with rising intonation), as in Your aim that evening, then, was to go to the discotheque? (R)

--Coordination of declaratives, as in So you went to the party but your brother stayed at home? (R)

--Clause fragment, as in Another cup of tea? (R)

Sometimes polar questions use non-assertive forms like any, ever, etc. and are biased towards a positive answer. A negative answer would be contrary to that expectation, as we shall discuss later.

2.3.2. Alternative Questions

--Answers to alternative questions

Alternative questions have a set of two or more alternatives given in the question itself. For example, the answers to Is it right or wrong? are It's right and It's wrong.

--The form of alternative questions

Alternative questions usually have closed interrogative syntax. The essential feature is the coordinator or, which relates the alternatives. This or-coordination is normally marked by a rise on the first coordinate and a fall on the final one, as shown in the different forms of alternative questions below:
--Is it a boy (R) or a girl (F)?           (closed interrogative)
--Is it genuine (R) or is it a hoax (F)?   (coordination)
--You're staying here (R), or coming       (declarative)
  with us (F)?
--Tea or coffee?                           (clause fragment)

Another type of alternatives is polar-alternative questions, which consist of a positive and its negative counterpart. Compare the following examples: Are you ready to order or are you not ready?, Are you ready or not? (polar-alternative) and Are you ready? (polar).

The polar version is simpler and much more frequent: it can be regarded as the default version. Polar-alternative questions are generally used to convey emotive meaning, as impatience or petulance. In general, the less elliptical the form, the greater the emotive meaning is likely to be.

2.3.3. Variable Questions

--Answers to variable questions

Also known as 'open questions' or 'Wh-questions' (Quirk et al. 1985), variable questions have a prepositional content consisting of an open proposition. In general the set of answers is open-ended, although there are cases when the number of possible answers may be limited, as in Which of the two proposals suits you better?

--The form of variable questions

These questions have the form of an open interrogative clause. Some variable questions invite the addressee to repeat and/or clarify what was said before and take the discourse backwards as in What did you say?. These and other examples suggest that there are various sub-classes of variable questions.

In some special situations, for example during the cross-examination of a medical witness, there may be no inversion of the interrogative phrase, as in:
Fronted: Inversion                   In Situ: No Inversion

Where are those      [right arrow]   And those senses are
senses located?                      located where?

Finally, there are exclamatory questions, which are considered a minor type of question by Quirk et al. (1985). They usually have a final falling tone and can take the form of a negative polar question, as in Isn t it expensive! or a positive polar question, such as Is it expensive?

As illocutionary acts, questions and statements are different in that the former has a feature of doubt, and that its felicity condition is that the speaker should not know the answer to the question.

Since the category 'question' is vague, argues Tsui (1994: 80), in the present model those utterances which elicit a verbal response are called elicititations.

2.3.4. Direction Questions

The great majority of questions are 'information questions': when used as inquiries they seek to elicit information. The characteristic illocutionary force of their answers is that of a statement. There is also a kind of questions whose answers characteristically have the force of directives. They seek not information but direction, and are accordingly called 'direction questions'. These are distinguished from information questions by the illocutionary force of the answers.

Compare the following examples:

Information Question Direction Question
A: Did he open       [left and right arrow]   A: Shall I open the
the window?                                   window? (polar)
B: Yes, he opened                             B: Open the window.
the window
No, he didn't.                                Open it now.

A: Did he do it      [left and right arrow]   A: Shall I do it now or
then or later?                                later? (alternative)
B: He did it then.                            B: Do it now.

A: When did he       [left and right arrow]   A: When shall we come
come back?                                    back? (variable)
B: He came back                               B: Come back at six.
at six.

2.3.5. Biased Questions

For Huddleston and Pullum, the basic characteristic of biased questions is that "the speaker is predisposed to accept one particular answer as the right one" (2002: 879), be it positive, negative, etc.

The authors state that there are different degrees of bias and distinguish three types of bias:

(a) Epistemic bias, which refers to the fact that the speaker thinks, expects and knows that one answer is the right one. This term is applied in the Sfield of modality.

(b) Deontic bias, when the speaker judges that one answer ought to be the right one. Again, this is another concept taken from the field of modality.

(c) Desiderative bias, in which the speaker wants an answer to be the right one.

Identifying biased questions involves consideration of the context and assumption of the speaker's intentions. The grammatical structure may also be important, as in the case of declaratives and negative interrogatives, which are clearly strongly biased. Besides, there are certain items, like some and any, which also convey positive or negative bias. We shall take these items in turn.

--Declarative questions

This type of questions, which fall under the category of polar questions for Quirk et al. (1985), are items that are identical lexico-grammatically to declaratives but function as questions because they are usually spoken with rising intonation, the aim being to invite the hearer's verification. For Huddleston and Pullum (2002: 882) "positive declarative questions have an epistemic bias towards a positive answer, negative ones towards a negative answer." There are some lexical markers like no doubt, surely, of course, etc. that indicate confidence in the truth of the proposition expressed, as in They no doubt misunderstood her intentions? and There isn t any chance of her changing her mind, I take it?

--Negative interrogative questions

Strongly biased, they allow a range of interpretations depending on the context. Their epistemic bias can be towards either the negative or the positive answer. Consider the following example (2002: 883):

Didn't I tell you Kim would be coming?

Bias towards negative answer: I didn't tell you.

Context: I (probably) have omitted to tell you that Kim would be coming.

Bias towards positive answer: I did tell you.

Context: I remember quite well having said that Kim would be coming. My prediction was not accepted at the time but has now been shown by Kim's presence to have been correct and I am asking you to admit I was right.

Tsui (1994: 71) adds that the use of a declarative form suggests that the speaker has some assumptions. Following Brazil (1985) she emphasises the role of intonation (rising, falling, high key, etc), to ask for confirmation, or to ask a polar question or an information question.

--Positively-and negatively-oriented polarity-sensitive items (some vs any, etc.)

There are a number of lexical items that mark bias in questions. Compare:

Positive interrogative questions:
Negatively-oriented Item        Positively-oriented Item

Is anything wrong?              Is something wrong?
Has anybody told Ed about it?   Has somebody told Ed about it?
Have they gone yet?             Have they gone already?
                                And in various indirect speech acts:
                                Could you please do something about
                                that noise?
Would you like some coffee?

Positive declarative questions:

As they are strongly biased, they do not take negatively-oriented items.

You have ever been to Paris?

Negative interrogatives:

--Negatively-oriented items, like any, anybody, etc., usually give these a negative bias.

Haven't they seen anybody about it yet?

--Positively-oriented items can have either bias, though the positive case is more common.

Haven't you forgotten something?

2.3.6. Echo Questions

Echo questions are used to question whether one has correctly heard what the previous speaker said ('stimulus'). This doubt may arise because it was not perceptually clear or because its content is particularly surprising or remarkable to the hearer.

Most cases of echo questions include mainly polar or variable questions, and also alternative questions. Consider the following examples (2002: 886) and the role of intonation to mark differences:
Stimulus                  Echo Question

(i) A: She's leaving      B: She's leaving on Saturday? (R) (polar)
on Saturday
(ii) A: He gave it to     B: He gave it to Anne (R) or Anna? (F)
Anne.                                                 (alternative)
(iii) A: He's proposing   B: He's proposing to what? (R) (variable)
to resign.

The main difference between echo and ordinary questions is that the propositional content of echo questions is not the same as that which is actually expressed in the utterance, and therefore they are considered as indirect speech acts.

--Types of echoes

We can distinguish between two types of echoes: (i) repetition echo questions, as their answers include a repetition of the stimulus, and (ii) clarification echo questions, which seek clarification of some element in the stimulus. Consider (2002: 890):
Stimulus                         Echo Question

(i) A: I've finally solved the   B: You've finally solved what?
problem of the missing cents.    (repetition)
(ii) A: I've finally solved it.  B: You've finally solved what?

2.3.7. Tags and Parentheticals

It is useful to make a distinction between the following types of questions:
He's rather aggressive, isn't he?   (interrogative tag)
He's rather aggressive, don't you   (interrogative parenthetical)

Note that an interrogative clause is added as a supplement to another clause, changing the illocutionary force of the utterance. This other clause or 'anchor' can belong to any of the five major clause types.

(a) Tags

Tags are short interrogative clauses which may be negative or positive and are generally followed by a positive tag or a negative tag respectively. These are referred to as 'reversed polarity tags'. It is also possible to have tags with the same polarity as the anchor, which we shall call 'constant polarity tags'. The intonation of a tag is related to the illocutionary force of the utterance that contains it. There are two main patterns both with falling intonation on the anchor and tags being either rising or, more frequently, falling. Compare:
Positive Anchor               Negative Anchor

He was here, wasn't he (R)?   He wasn't here, was he (R)? (rising tag)
He was here, wasn't he (F)?   He wasn't here, was he (F)? (falling tag)

Each of these sentences has different assumptions and expectations.

--The rising tag

A positive anchor containing a tag with a rising tone is positively biased. It expresses doubt and invites the addressee to confirm the speaker's assumption. A negative anchor, on the other hand, has no bias towards an answer with the same polarity, as in It isn't raining again, is it?

--The falling tag

A tag spoken with falling tone invites the addressee to agree with the speaker. There are examples when the tag is an implicit invitation to provide an explanation, as in (i), an invitation for the hearer to admit something s/he didn't previously accept, as in (ii), an invitation to agree to some minor uncontroversial proposition, a rhetorical question, as in (iv), etc. Compare:

(i) A: Good gracious, you're up early this morning, aren't you?

B: Yes, I've got a train to catch.

(ii) A: I was all right all along, wasn't I?

(iii) A: It's a lovely day again, isn't it?

(iv) A: What a mess I've made of things, haven't I?

(b) Parentheticals

Some expressions like I think, don't you think?, wouldn t you say? can be appended parenthetically to an anchor clause. Parentheticals are used for different purposes:

--Declarative anchor + declarative parenthetical: It is quite safe, I think.

Many parentheticals are used to weaken or strengthen the speaker's commitment to the truth of the anchor preposition: I believe, I guess, it seems, etc or I'm sure, I have no doubt, etc.

--Declarative anchor + interrogative parenthetical: It is quite safe, don tyou think?

These have an effect much like that of a tag and seek confirmation that the anchor proposition is true. Other examples are: wouldn t you say, don t you reckon, am I right?, don tyou think so?

--Interrogative anchor + interrogative parenthetical: Is it safe, would you say?

The parenthetical serves to clarify the way the question expressed in the anchor clause is to be interpreted and answered. It is similar to asking in your opinion.

3. Methodology

3.1. Corpus

The corpus for this study is a televised interview from Newsnight BBC 2 (20 April 2005) in which journalist Jeremy Paxman questions Prime Minister Tony Blair. The interview takes 28 minutes and 18 seconds.

3.2. Procedure and analysis of the data

Following Tsui's model (1994), Huddleston and Pullum's account of questions (2002), and Elliot and Bull's categorisation of face-threatening acts (1996), the corpus was analysed as follows:

1. Transcription of the interview

2. Identification of turns

3. Identification of main initiating moves and responding moves

4. Identification of grammatical structure of main discourse acts

5. Identification of intonation contours

6. Identification of primary acts: initiation and responses

7. Identification of subclass of elicitations

8. Identification of questions carrying potential face threat

9. Quantitative and qualitative analysis

10. Conclusions

4. Presentation And Analysis Of Results

4.1. Structure of the interview

The interview under study corresponds to the first segment of the programme Newsnight BBC 2 featuring journalist Jeremy Dixson Paxman and Prime Minister Tony Blair. This programme was broadcast live from Leeds on Wednesday 20 April 2005. Newsnight BBC 2 ( is a British daily news analysis, current affairs and politics programme broadcast between 22:30 and 23:20 (Leeds, UK) on weekends. Journalist, news presenter and author, Paxman is one of the presenters in the programme.

Integrating the model presented by Tsui and Clayman and Heritage we can identify a general structure consisting of different phases. Consider Figure 1.


As shown above, we can distinguish three phases: opening, body, and closing.

Opening: facing the camera, Jeremy Paxton starts the interview with a monologue addressed to the audience. The purpose of this pre-questioning phase is both to state the topic of the interview (Headline) so as to engage the audience and then to introduce the interviewee, Prime Minister Tony Blair (Lead-in).

Body: characterised by various subclasses of elicitations and responses.

In this interview the Prime Minister is questioned on five major issues:

--The validity of his decision to go to war in Iraq although Intelligence did not have extensive and detailed reports on the existence of chemical or biological weapons.

--His proposal not to raise taxes after the election. In the previous campaign the Labour Party committed to not raising taxes but based on different grounds they did so.

--His proposal to set up forty large-scale casinos so as to increase the budget for social welfare.

--Failed asylum seekers and "illegals".

--Blair's credibility as a political figure.

Closing: the interviewer prepares the end of the interview with a prefatory question, which, in this case, is different to the previous ones in terms of depth and content. After abruptly saying the Prime Minister's name and expressing thanks (informative: expressive), Paxton addresses the audience to announce the topic of the next programme (informative: report) and finishes with a farewell (informative: expressive (goodwill).

4.2. Analysis of the elicitations by the interviewer

Paxman proves to have an abrasive and forthright style of interviewing, which is characterised by repetitive interruptions on the interviewee, inquisitive face threatening questions and an abrupt ending. This inquisitive style results in Blair being clearly defensive in many instances of the interview.

Out of the 188 turns, 95 were performed by the interviewer and 93 by the interviewee.

After a brief opening, Paxman starts the series of questions with a direct

(1) P: Prime Minister, is there 'anything you'd like to apologise for?

and then proceeds to focus on a number of controversial topics. Table 1 shows the frequency of classes and subclasses of acts identified in the three phases of the interview. Table 2 describes the occurrence of elicitations.

The quantification of the initiations (Table 1) shows that 77 (82.1%) of the subclasses of acts correspond to elicitations and within these (Table 2), the frequency of the subclasses elicit: inform (33.43.6%) and elicit: agree (30.38.5%) is remarkable. Similar results have been obtained in other studies (Romo 2006) and they coincide with Clayman and Heritage's definition of the interview as having one main objective: to obtain information for the audience.

Table 3 shows the frequency of main acts of elicitations performed by the interviewer and their respective grammatical realisations. Furthermore, this table provides a detailed account of these results considering the intonation they exhibit.

It is of interest to note the frequency of unbiased and biased questions. Table 4 describes different degrees of bias in questions.

Table 5 shows the grammatical realisation of the subclasses of elicitations together with the the different types of clause types (questions) found in the corpus.

A correlation of the information presented in Tables 2-5 generates a number of observations drawn from the different types of elicitations and question types, as follows:

(a) Elicit: inform

The interviewer mainly uses polar questions and variable questions to elicit information.

--The number of polar questions with a closed interrogative form is considerably high (24), especially those which have a neutral positive structure (20), in contrast to negatively biased polar questions (4). Compare, for example:

(1) P: Prime Minister, is there anything you'd like to apologise for?

(24) P: Did you see the Foreign Office legal advice which said, 'that military action against Iraq would be illegal without a further uN resolution?

(185) P: Isn't there a point where you think, God it's going to be great to be shot of all of this?

Earlier we discussed that in a biased question the speaker is predisposed to accept one particular answer as the right one. A neutral question does not have such bias towards one answer rather than another.

--Variable questions, which have the form of an open interrogative, is the second type of clause commonly used for elicit: inform acts. There are 11 examples, which is half the number of polar questions. Following common use, 9 out of the 11 questions have falling intonation and 2 are rising.

--There is also an example of a polar declarative question. This type of clause, which has rising intonation, is strongly biased and has an epistemic bias towards a positive answer, as in:

(64) P: And the billions of pounds that involved, you're prepared to make available? (R)

(b) Elicit: agree

Declaratives (17) and tags (12), both with falling intonation, are used to invite the interviewee to agree with the interviewer.

--There are affirmative and negative declaratives in this corpus:

(36) P: So, the short answer to the question is you don't accept any responsibility.

(86) P: So there// there could be any old report coming along after this election, which will necessitate you raising taxes again.

(115) P: Prime Minister, you have really no idea of how many failed asylum seekers there are illegally in this country.

(167) P: So that means, by the time there's a new American President coming on his first visit to Britain in 2009, you will still be in Downing Street.

--There are 11 examples of tags with reversed polarity used to elicit agreement with the speaker. Of these, 7 have the form positive anchor + negative tag and 5 have the form negative anchor + affirmative tag. Compare:

--Positive anchor + negative tag

(9) P: Okay, but you know, don't you, that just two weeks before you made that statement, the Joint Intelligence Committee said that 'intelligence remains limited'. (Embedded tag)

(80) P: And you did, didn't you?

(88) P: You are going to have to raise taxes after the election, aren't you?

(111) P: But you have no idea, haven't you.

--Negative anchor + positive tag

(5) P: All right, let's look at Iraq. When you told parliament that the intelligence was 'extensive, detailed and authoritative', that wasn't true was it.

(11) P: Well therefore it's not extensive, detailed and authoritative, is it?

(46) P: It didn't meet at all, did it

--Negative anchor + negative tag

(90) P: Well you can't give us a commitment, won't you?

According to Huddleston and Pullum (2002: 895), this type of question (a) seeks acknowledgement/agreement that the anchor is obviously true and (b) invites or forces the hearer to provide an explanation.

--There is an example of a negative polar with falling intonation. It shows a deontic bias towards a contextual positive answer: any reasonable person would.

(79) P: Urm, wouldn't then any reasonable person suppose that you therefore proposed to increase National Insurance contributions.

(c) Elicit: confirm

Almost one third (17.9%) of the elicitations aim at inviting the addressee to confirm the interviewer's assumptions, through negative polar questions, polar declaratives, and tags.

--We discussed above that negative polar questions are biased towards a negative answer. Notice the direction of the examples below:

(30) P: The Attorney General is a political appointment. Prime Minister, shouldn't you have seen the Foreign Office legal advice? (R)

(126) P: Do you- Do people not come to you and say, we think Prime Minister there may be a hundred thousand (R) or two hundred thousand (R) or fifty thousand (R), or five hundred thousand? (R)

--There are two kinds of polar declarative questions in the interview: negative and affirmative.

(26) P: Did you see... You didn't see that Foreign Office advice saying that an invasion would be illegal without a second UN resolution? (R)

(92) P: And there is no question on your watch, of a local income tax ever being introduced?

(134) P: And just one final time. You have no figure that you can give us for the number of refused asylum seekers who are in this country? (R)

(173) P: So if there is any deal between the two of you, it's a deal just in Gordon Brown's mind?

--There are two tags used to elicit confirmation, both with reverse polarity, negative anchor (hence negatively-oriented) and rising intonation:

(58) P: You haven't given George Bush any undertakings about anywhere else in the world have you?

(74) P: You're not expecting us to fall for that the second time, are you?

--One example of parenthetical was found. It is used to strengthen Paxman's commitment to the truth of the anchor preposition:

Negative declarative anchor + declarative parenthetical:

(180) P: But you came in -you came in- a young Prime Minister, talking about a young country. Now they talk about how you've got a fake tan. You haven't got a fake tan I take it.

3. Analysis of face threats in elicitations by the interviewer

One of the aims of this study was to identify those face threat categories which occurred most frequently in the questions posed by the interviewer. The frequency with which each face threat code occurred in the questions and the number of 'no necessary threat' (NNT) is displayed in Table 6.

Out of a total number of 78 questions, 61 (78.2%) carried some kind of threat and 17 (21.8%) were categorised as NNT.

NNT questions typically include a higher percentage of variable questions and some examples of polar questions

(44) P: How many times did the Ministerial Defence and Overseas Policy Committee meet in the run up to war?

(113) P: Well what is your idea Prime Minister?

(152) P: Let's look at economic migrants. Is there an upper limit to the number of economic migrants who should be allowed into this country?

(156) P: What does our economy need?

It can be seen that negative impression of personal competence/judgement/decision/ policy (26 examples) was the most frequently occurring face threat, followed by the threat and losing credibility (15) and of personal difficulties in future (8) It is of interest to note that the least frequent face threats were those associated with negative impression of public persona, not supporting positively valued people/institutions and failure to present a positive image of self.

Categories of Face Threat

In this study we identified a higher frequency of questions concerned with personal political face and face in relation to significant others. Interestingly, there were no questions concerned with the party policy. Table 7 gives a full account of this type of questions posed by Paxman in his role as interviewer. Following is a list of the main categories of face threat found in the corpus.

(1) Creating or confirming a negative statement or impression about personal competence/judgement/decision/policy

Paxman had a strong focus on calling into question Blair's personal competence from the beginning of the interview:

(1) P: Prime Minister, is there anything you'd like to apologise for?

(32) P: Do you accept any responsibility at all for the death of Dr David Kelly? (1)

It seems possible that Tony Blair was asked a high proportion of this type of question because of three factors. First, his vulnerability. In his previous government, Blair decided to follow the US into war in Iraq (2003), despite both the lack of well-substantiated reports of chemical weapons and not having sought enough advice. Second, Blair represents a previous government with policies and decisions that increased unemployment, took the country to war, and had a negative effect on the economy. Thirdly, The Labour Party never accepted responsibility for these wrong decisions.

Blair is unable to answer Paxman's initial question directly because answering affirmatively would demonstrate that there had been a personal misjudgement. In the subsequent questions the interviewer brings discussion of this topic until he considers he has forced Blair to provide the 'right answer': his political incompetence in the Iraq affair.

(2) Losing credibility

Closely related to the threat of creating a negative impression of personal competence and second in frequency of occurrence, Paxman's tough style is clear when he questions Blair repeatedly in his attempt to demonstrate that he had consciously misled the public over the war when he declared that the intelligence report was 'extensive, detailed and authoritative': There are many instances when Paxman tries to make Blair agree with him:

(5) P: All right, let's look at Iraq. When you told parliament that the intelligence was 'extensive, detailed and authoritative', that wasn't true was it (F)

(50) P: The problem is Prime Minister, that the next time the Joint Intelligence Committee come to you and say, we have extensive, authoritative intelligence of a threat or possible threat to this country, we urge preemptive action, you won't be able to sell it to the public, will you (F)

(74) P: Before the last election, you made exactly the same promise you're making this time. No increase in the basic rate of tax, no increase in the higher rate of tax. No commitment at all on National Insurance. You spoke to us very kindly before that election, and I suggested to you that any reasonable person would therefore conclude that after the election, you would raise the basic rate of National Insurance. You said that we shouldn't make such an assumption, and then you did it. You're not expecting us to fall for that the second time, are you? (R)

(3) Personal difficulties in future

There are 8 examples of questions (13%) referring to personal difficulties in the future occurring as potential threats.

(60) P: While we're on defence Prime Minister, the British independent nuclear deterrent is going to need replacing, probably a decision that has to be taken in the next government that takes office after May 5th. Will you replace it?

(88) P: You are going to have to raise taxes after the election, aren't you?

(161) P: Prime Minister, if you are returned to Downing Street on May 6th, can you at least give us a guarantee that within say twelve months of your handing in your cards as Prime Minister, there would be a General Election?

(4) Contradicting personal past statements/policies

Fourth in rank, 4 questions (6.5%) focused on contradicting Blair's defence and to question the validity of some of his assertions.

(11) P: Well therefore it's not extensive, detailed and authoritative, is it?

(36) P: So, the short answer to the question is you don't accept any responsibility.

This suggests that Paxman considers Blair is stretching the truth to some point.

(5) Negative impression of public persona

Only two examples were found in the interview. Negative impression of public persona is the result of Blair and the Labour Party's wrong policy in a number of issues. In the 2003 UK General Election Tony Blair's image was of a charismatic honest and well-intentioned politician. In the 2005 election campaign, the same image was exploited. However, opposition questioned the authenticity of this 'young and new' image.

(3) P: But do you accept that there is a trust issue, and that the reason opposition parties can talk about wiping the smirk off your face, is because you can't any longer say, look at me, I'm a pretty straight kind of guy ?

(181) P: But you came in -you came in- a young Prime Minister, talking about a young country. Now they talk about how you've got a fake tan. You haven't got a fake tan I take it.

(6) Failure to present a positive image of self

Although infrequent (3 examples), these questions are closely related to personal competence and losing credibility. Below is one of the most critical questions posed by Paxman during the interview: an elicit: inform variable question, where Blair risks suffering some face threat whatever his response:

(52) P: Why should they believe you again?

(7) Creating or confirming a negative impression about the party or its policies, actions, principles, etc.

Paxman prefers to focus on questioning Blair's competence rather than on the Labour Party he is representing. There is one example when Paxman refers to both Blair and the party he represents:

(42) P: All right. You keep referring to these enquiries. The enquiry that you set up under Lord Butler concluded that what went wrong there, was partly the consequence of your style of government and- and- especially of the Labour Party style of government, so called 'sofa government'. Will your next government, if you have one, be any different? (R)

(8) Not supporting a colleague/peer/subordinate/governmental office

By supporting the Joint Intelligence Committee report Blair would lack credibility, accept self-contradiction, and indirectly admit incompetence when he decided to take military action against Iraq.

(16) P: So was the JIC, the Joint Intelligence Committee Report wrong?

And when Paxton asks about the number of foreign workers that should be allowed to come into the country, Blair faces the threat of refuting what David Blunkett had said:

(157) P: Your previous Home Secretary, David Blunkett told us that he saw no obvious upper limit.

5. Conclusions

The main focus of this study was the interviewer's neutrality and style in a political interview. This involved an analysis of questions from three different but complementary perspectives: identification of (a) sub-classes of elicitations and their grammatical structure together with (b) questions carrying some type of face threat.

Huddleston and Pullum provide a semantic and pragmatic discussion of questions and generate an account of clause types and illocutionary force. Tsui's model was used to distinguish various types of elicitations. An integration of both models demonstrated that sub-class elicit: inform was higher in frequency (46%) and was performed mainly through polar questions and variable questions. Lower in frequency were (i) elicit: agree (36%), especially through declaratives and tags, and (ii) elicit: confirm (18%), especially through declaratives.

Consideration of the number of unbiased and biased questions demonstrated a high percentage of biased questions. Paxman focused his interview on asking a high percentage of questions which carried face threat (78.2%) as opposed to questions with no necessary threat (21.8%). In the category of questions with a potential threat, rather than questioning Blair on party policy (1%) or face in relation to others (3.3%), Paxman focused on personal political face (95.7%). In this sub-category, questions carrying face threat in relation to personal competence (43%), losing credibility (25%) and personal difficulties in the future (13%) were frequent.

Apparently, Paxman focused on face-threatening questions in relation to personal competence and credibility because of Blair's vulnerability as a politician. In the 2001 General Election campaign Blair appeared as a charismatic honest figure and promised to give a "younger look" to the system. During his government, however, a number of unpopular actions were taken, like the invasion of Iraq (2003), which resulted in a terrorist attack in London, unemployment, and a style of government described as "sofa government" by the opposition. In the 2005 election, as part of the campaign, Blair, representing the Labour Party, made similar promises.

It is interesting that Paxman frequently adopted a specific technique which tended to cause problems for the interviewee. In contrast to a lower proportion of unbiased no necessary threat questions to elicit: inform, he uses a high number of biased questions (especially declaratives and tags) to elicit: agree and elicit: confirm, sometimes preceded by a highly critical statement which could not easily be rebutted by the politician since it carried some obvious truth. Any attempt to deny or evade the truth of the statement would therefore lead to losing his credibility. Generally, to facilitate the expression of harsh criticism and contribute to the maintenance of an impression of interviewer neutrality, Paxman prefaces the question with a 'footing', citing references.

Another characteristic in his style is that questions are repeated insistently if the interviewee evades the expected answer or tries to deflect attention from the actual question.

Ideally, an interviewer should perform a high proportion of unbiased elicit: inform acts, principally by variable and polar affirmative questions in contrast to a low number of biased elicit: agree and elicit: confirm acts. In this interview, there is a high frequency of face-threatening elicitations and biased clauses. In this interview there are a number of characteristics exhibited by Paxman which are associated with his style of questioning:

--a high frequency of elicit: agree acts with statements that cannot be easily rebutted by the politician since they reflect an obvious truth

--highly biased eliciting clauses

--critical comments

--emphasis on making a negative assessment concerning Blair's actions

--insistence on an expected answer

--questions which might threaten especially the interviewer's competence and credibility.

These strategies are part of his tough style of interviewing and also account for the media portraying him as an aggressive and incisive journalist.

The three models presented here are complementary in the analysis of elicitations during televised political interviews. They also have strong potential applications in terms of providing an insightful description of interviewer style. In a future study these models could be applied in an analysis of interviewer-interviewee interaction so as to identify connections between elicitations and responses.

Recibido: enero 2009 Aceptado: marzo 2009


Brazil, D. C. 1985. The communicative value of intonation. Discourse Analysis Monograph no. 8. University of Birmingham: ELR.

Bull, P. E. and K. Mayer. 1993. How not to answer questions in political interviews. Political Psychology 14: 651-666.

Clayman, S. and J. Heritage. 2002. The news interview. Cambridge: Cambridge University Press.

Coulthard, R. M. and D. C. Brazil. 1981. Exchange Structure. In R. M. Coulthard and M. M. Montgomery (Eds.), Studies in discourse analysis. London: Routledge. Pp. 82-106.

Elliot, J. and P. Bull. 1996. A question of threat: face threats in questions posed during televised political interviews. Journal of Community and Applied Social Psychology Vol. 6: 49-72.

Emmerson, S. 2006. Interviewers' challenging questions in British debate interviews. Journal of Pragmatics 39: 570-591.

Greatbatch, D. 1998. Conversation analysis: neutralism in British news interviews. In A. Bell, and P. Garret (Eds.), Approaches to media discourse. Oxford: Blackwell Publishers Ltd. Pp. 163-185.

Huddleston, R. and G. Pullum. 2002. The Cambridge grammar of the English language. Cambridge: Cambridge university Press.

Mindich, D. T. Z. 1998. Just the facts: how "objectivity" came to define American journalism. New York: New York University Press.

Quirk, R., S. Greenbaum, G. Leech, and J. Svartvik. 1985. A comprehensive grammar of the English language. London: Longman.

Romo, A. 2006. Analisis discursivo de la entrevista televisiva. Tesis para optar al grado de Magister en Linguistica con mencion en Lengua Inglesa. Universidad de Chile.

Sinclair, J. McH. and R.M. Coulthard. 1975. Towards an analysis of discourse. The English used by teachers and pupils. London: Oxford University Press.

Tsui, A. B. M. 1994. English conversation. Oxford: Oxford University Press.

(1) David Christopher Kelly was an employee of the UK Ministry of Defense, an expert in biological warfare, and a former United Nations weapons inspector in Iraq. Being under investigation, he committed suicide (July 17, 2003) days after appearing before a Parliamentary committee.

Para correspondencia, dirigirse a la direccion postal: General Jofre 462, Santiago, Chile, o al correo electronico <>

Carlos Aspee Palacios

Universidad de Santiago de Chile
TABLE 1: Primary acts and subclasses of acts performed
by the interviewer

SUBCLASSES OF ACTS              No.   %

ELICIT: inform                  33    36.2%
ELICIT:agree                    30    32.0%
ELICIT: confirm                 14    14.6%
INFORMATIVE: report              7     7.5%
INFORMATIVE: assessment          3     3.2%
REQUEST: proposal                3     3.2%
INFORM: expressive (goodwill)    2     2.1%
REQUEST: action                  1     1.2%

TOTAL                           93     100%

TABLE 2: Frequency of elicitations

SUBCLASSES OF     No.    %

ELICIT: inform    33    46%
ELICIT: agree     30    36%
ELICIT: confirm   14    18%

TOTAL             77   100%

TABLE 3: Grammatical realisations of the subclasses of elicitations

                     CLAUSE TYPES: QUESTIONS

                             (Rising       intonation

ELICIT: inform        21          1
ELICIT: agree          1                       17
ELICIT: confirm        2         11
TOTAL                 24         12            17

CLASSES AND              TAG       VARI-

                     R        F

ELICIT: inform                     11
ELICIT: agree                 12
ELICIT: confirm      2
TOTAL                    14        11

(R = Rising intonation, F = Falling intonation)

TABLE 4: Questions and degrees of bias

Degree of bias    Type of biased      No.        %

Unbiased          Variable            11    11   14%
Slightly biased   Polar affirmative   20    36   46%
                  Polar negative       4
                  Polar declarative   12
Highly biased     Declaratives        18    31   40%
                  Tags                13
TOTAL                                       78   100%

TABLE 5: Grammatical realisation of the subclases of elicitations
(R = Rising intonation, F = Falling intonation)

OF ACTS           POLAR              POLAR
                                     DECLARATI    DECLARATI
                                     VE           VE

                  Aff.     Neg       Aff.   Neg   Aff.   Neg

                  R    F   R     F   R      R     F      F

ELICIT: inform    13   7   1         1            1
ELICIT: agree                    1                8      9
ELICIT: confirm            2         4      7

TOTAL             13   7   3     1   5      7     9      9

OF ACTS           TAG                       VARIABLE   Total

                  (+) anchor   (-) anchor
                  +            +
                  (-) tag      (+) tag

                  R   F        R     F      R   F      No.   %

ELICIT: inform                              2   9      34    43.6%
ELICIT: agree         7              5                 30    38.5%
ELICIT: confirm   1            1                       14    17.9%

TOTAL             1   7        1     5      2   9      78     100%

Table 6: Type of questions posed by the interviewer

Type of question      No.   %

No necessary threat   17    21.8%
Face threat           61    78.2%

TOTAL                 78     100%

TABLE 7: Categories of face threat in questions posed by the

Categories of face threat                           No.   %


1. Negative impression of personal
competence/judgment/decision/policy                 26     43%
2. Losing credibility                               15     25%
3. Personal difficulties in future                   8     13%
4. Contradicting personal past statements/policie    4    6.5%
5. Negative impression of public persona             2    3.3%
6. Failure to present a positive image of self       3      5%


7. Creating or confirming a negative impression
about the party or its policies, actions,            1      1%
principles, etc.


8. Not supporting positively valued
people/institutions                                  2    3.3%

TOTAL No. OF QUESTIONS                              61    100%
COPYRIGHT 2009 Universidad de Chile, Facultad de Filosofia y Humanidades
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Palacios, Carlos Aspee
Publication:Lenguas Modernas
Article Type:Report
Date:Jan 1, 2009
Previous Article:La presencia del espanol en Canada en el marco de la ensenanza de lenguas extranjeras.
Next Article:Analisis sistemico-funcional de la estructura de conversaciones en ingles y en espanol.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters