nigel lawson's tent: discourse analysis, attribution theory and the social psychology of fact

European Journal of Social Psychology, Vol. 20,405424 (1 990)

Nigel Lawson’s tent: Discourse analysis, attribution theory and the social

psychology of fact

JONATHAN POTTER AND DEREK EDWARDS Discourse and Rhetoric Group, Department of Social Sciences, University of Loughborough, Loughborough, Leicestershire, LE7 7 3TU, UK

Abstract

The general aim of this paper is to show some of the limitations of the attribution theory approach to ordinary reasoning when compared to a discourse analytic alternative. Three central shortcomings with attribution theory are documented, each stemming from the method of presenting subjects with factual vignettes from which they are required to draw inferences: (a) its asocial and unexplicated notion of information; ( b ) its realist view of linguistic description; ( c ) its constrained account of participants ’ activity. These paws are illustrated in practice through a discourse analytic study of the management of factual versions in a political dispute (over a controversial briefing between a British politician, Nigel Lawson, and a group of journalists). SpeciJically, it focuses on ‘consensus information’, examining the way notions of consensus are used when warranting and undermining versions. Two features of consensus accounts are examined: ( a ) consensus across a group of observers of an event; (b) corroboration between independent individuals. In each case, the rhetorical organization of factual accounts is documented by analysing both the way the consensus is constructed and the way it is undermined or discounted. The analysis explores how the tfacts of the matter’, rather than existing as criteria for the resolution of disputation, were themselves part andparcel of the disputation itself: In attribution theory terms, the clear distinction between ‘consensus information’and the attributions whichflow from it becomes unwork- able. It is suggested that the analysis provides an exemplar for a discourse orientated social psychology of fact.

We would like to thank Michael Billig, Teun van Dijk, Paul Drew and Margaret Wetherell for making helpful comments on earlier versions of this paper. We were also helped by audience comments after presentations at the BPS Social Psychology Section Conference, Canterbury, September 1989; the Conver- sation, Discourse, Conflict conference, Dublin, April 1989; and after talks to the psychology departments at Nottingham University and Southampton University, the Sociology Department at York and the Human Sciences Department at Loughborough University.

004&2772/90/050405-20$10.00 0 1990 by John Wiley & Sons, Ltd.

Received 2 February 1990 Accepted 26 April 1990

406 J. Potter and D. Edwards

INTRODUCTION

Nobody can say the chancellor of the exchequer is not a bold man. Who else would invite 10 senior journalists to his home for a briefing and then state that when his guests departed they went into a tent to concoct a ‘farrago of invention?’ (Susan Crosland, Sunday Times, 13 November, 1988).

In the last few years discourse analysis’ has been established as an alternative perspective on major psychological and social psychological phenomena. For example, particular attention has been paid to the way a detailed study of the action orientation (Heritage, 1984) of discourse forms the basis for reconceptualizing five fundamental topics of enquiry: attitudes (Billig, 1989; Potter and Wetherell, 1988a; Smith, 1987); categories (Billig, 1985; Condor, 1988; Widdicomb and Wooffitt, forthcoming); gender (Billig, et al., 1988; Hollway, 1989; Wetherell, et al., 1987); memory (Billig, 1990; Edwards and Middleton, 1986; Wooffitt, 1989) and social representations (Billig, 1988; McKinlay et al., forthcoming; Potter and Litton, 1985). In each case, these studies have shown the importance of examining the discursive practices and rhetorical constructions which lie at the heart of social life and that as we come to grips with these features major theoretical reorientations become necessary.

The aim of this paper is twofold. We shall extend the discourse analytic critique into the area of attribution theory and develop a new theme in discourse analysis itself - the social psychology of factual discourse. Attribution theory is surely one of the most fundamental social psychological perspectives of the last two decades. Significant pointers to the importance of rhetoric in this area have been laid down by Billig (1982, 1987); this paper extends these critiques and provides a more fully realized analytic exemplar of the alternative. While we could not hope to simply refute a complex and heterogeneous perspective such as attribution theory, we hope to highlight certain limitations and show the heuristic value of the discourse alternative.

DISCOURSE AND ATTRIBUTION THEORY Attribution theory (or, more precisely, the cluster of related ideas from Heider, 1958; Jones and Davis, 1965; Kelly, 1967; Hilton and Slugoski, 1986 that are usually referred to as ‘attribution theory’) has suffered a number of conceptual critiques (Buss, 1978; Harre, 1981; Parker, 1989; Sabini and Silver, 1980; Semin, 1980; Shotter, 1981). While recognizing the force of some of these criticisms, our concern is specifically with the way attribution theorists have dealt with discourse in their research practice. In particular, there are three closely related areas of difficulty, all of which stem from the use of ready-made factual vignettes from which subjects are supposed to draw attributional inferences: (1) the notion of information; (2) the view of language; (3) the limits on participants’ agency. (Our points here are directed at work broadly within the Kelley (1967, 1973) and McArthur (1972) tradition).

’ There are a number of different theories and approaches which have been advanced under the rubric of discourse analysis which is an area where there are important inconsistencies of theory and definition. Our own approach is outlined more fully and systematically with respect to alternatives in Potter and Wetherell (1987) and Potter et al. (forthcoming).

Nigel Lawson’s tent 407

The notion of information as a natural phenomenon

Attribution studies embody a basic assumption about the nature of world; namely, that it contains information which is a given starting point for attributional reasoning. Specifically, vignettes and other ‘stimulus materials’ are assumed to embody information, and the question at stake in an attribution study is what the participant infers from it. The idea of information itself remains unexplicated.

From a discourse analytic standpoint, the facts or information on which the attributional reasoner supposedly relies are not ‘naturally’ given but are as much a part of the social process as the inferences themselves. Indeed this facthnference distinction becomes an analytic topic in its own right rather than a research presupposition. For example, the distinction between what is fact and what is inference is often central to participants’ concerns. It is a rhetorically useful dichotomy for constructing or undermining others’ factual accounts. For discourse analysis, as we will show, the focus of interest moves to the way particular facts or states of affairs are constructed in discourse and, in turn, how these are invoked in expressions of practical reasoning, or inference-making.

Words and the world

For attribution theorists, language - both the given vignettes and the expressed inferences made by subjects - is treated as mere description. The vignettes used as the ‘stimulus materials’ are taken as straightforward stand-ins for the world. That is, they are taken as descriptions or summaries of actions and events, and the notion of description or summary is taken to require no further explication. For discourse analysts, however, things are not so simple; ‘describing’ and ‘summarizing’ are themselves complex social/discursive processes (see e.g. Edwards and Mercer, 1989; Edwards and Potter, forthcoming; Heritage and Watson, 1979), with the emphasis on descriptive discourse as a form of situated activity rather than decontextualized cognitive representation. This brings to the fore questions of how a description is constructed per se - as description, rather than as value, belief or view, and the question of what the description is being used to do.

We are by no means the first to offer a critique of presuppositions about language in attribution theory or to develop a language-based understanding of the process of attribution. Abraham (1988), Hilton (1990), Turnbull and Slugoski (1988) and Sermin and Fiedler (1989) all do this in various ways. However, none of them explores the rhetorical and action orientated qualities of discourse which we maintain are crucial.

Some attribution workers have moved to study more naturalistic materials, either on their own or as a supplement to more traditional materials (e.g. Harvey et al., 1988; Fletcher et al., 1987). This work, however, also presupposes that language is an essentially descriptive medium; indeed, at times, it offers a neo-introspectionism where accounts are a clear and direct path to cognitions. What they fail to address is the action orientation of discourse. This deficiency is sustained by the use of three sorts of procedures: partly through focusing on extended non-contextualized accounts where activities may be less apparent than in conversational materials; partly through failing to address the distinction between attribution as a discursive act and as a cognitive process; and partly through deploying the type of content


analysis methodology which can easily obscure rhetorical/functional aspects of discourse (see Abraham, 1984; Potter and Wetherell, 1987 for critiques of this methodology). In the course of the paper it should become clear that we are proposing a thoroughgoing theoretical reorientation rather than merely a move to more naturalistic sources of data.

The activity of the attributional reasoner

Given that informational accounts are constructed phenomena and that discourse is functionally orientated we can start to see a whole dimension of attribution that has been ignored. This focuses on how people construct versions specifically in order to make certain conclusions inferable. Of course, attribution workers depend on this when they marshal discourse into vignettes in the conduct of their research; however, it has not been theorised as a more general feature of people’s (subjects’) own discursive practices. As Antaki (1988a) notes, in natural conversation recipients may question factual presuppositions in attributional questions or ‘deconstruct’ their assumptions in a variety of other ways.

In contrast to this, discourse analysis takes the constructed nature of such accounts, and the way these constructions provide a basis for inferences of different kinds, as a central focus of study. We will argue below that construction and inference are part and parcel of the same process, and we suggest that attribution theory is only able to isolate attribution as a cognitive, inferential process in individuals via a kind of epistemological gerrymandering, in which a factual, given world is reflected in a simple descriptive language that is controlled by the experimenter, and rendered unavailable for constructive work by the subject.

It is important to distinguish what we are suggesting here from recent ‘functional extensions’ of attribution theory (Forsyth, 1980; Harvey and Weary, 1984; Hewstone, 1983). Our interest is in discursive practices; for example, those through which blame or praise is assigned. That is, the function is not tied to the motivation of individuals but is situated within activities, and identifiable in terms of conversational pragmatics. Indeed, the whole focus of the research moves away from the cognition of the individual to the discursive practices of the collectivity. This focus also separates discourse work from recent work on the structure of explanation (Antaki, 1988b; Antaki and Naji, 1987), although at times their concerns converge (e.g. see Antaki, 1985 on the rhetorical function of certain causal structures).

As we have indicated, our three critical themes are not intended to show that attribution theory is straightforwardly wrong. Rather, what we have tried to do is to sharpen up some concerns which will separate the attribution perspective from a discourse analytic one. Having done that we turn, more constructively, to sketching the positive alternative which we are proposing: a social psychology of factual discourse.

SOCIAL PSYCHOLOGY OF FACTUAL DISCOURSE

In virtually any situation appeal to the facts, to what really happened and what is only invention, can be a powerful device. Factual accounting is the stuff of arcane scientific disputes over whether neutrinos have been detected, and also of mundane

Nigel Lawson S tent 409

domestic conflict over who last washed the dishes. By invoking the notion of ‘factual accounting’ we are expressly avoiding the central attribution theory distinction between fact and inference. We shall illustrate our approach by examining a dispute about the facts in a political controversy. Of course, in politics the successful establish- ment of one’s version of events can be crucial: elections, policies and careers can hang upon it. We shall look at how, in a specific case, the stable, neutral, out-there- ness of a version is constructed and undermined, and at the activities performed by these constructions and critiques.

If discourse analysis is understood in its widest sense it is possible to identify a concern with factual discourse as an emerging theme of interest. Factual discourse is here understood broadly as discourse which provides a purportedly accurate description of events or of states of affairs. Such work runs from the conversation analytic (e.g. Pomerantz, 1986; Schegloff, 1988; Smith, 1983) through work on science discourse (e.g. Gilbert and Mulkay, 1984; Lynch, 1985; Lynch and Woolgar, 1989) and the discursive study of cognition, shared knowledge and memory (e.g. Coulter, 1979; Edwards and Mercer, 1987; Suchman, 1987), to work in the broad field of literary theory and post-structuralism on mimesis and narrative (e.g. Barthes, 1974; Culler, 1983; Genette, 1980). All these researchers have, in one way or another, eschewed the simple realist view held by attribution theorists that descriptive language is a mirror on reality (however distorted), or a snapshot (however grainy) of the world, in favour of examining facticity as a discursive construction. They share the view that the constructed nature of discourse is available for study through examination of the discourse itself, rather than through some other sort of study of how it come to be said or written, or of its ‘effects’ on readers (Potter et al., 1984).

Work on factual discourse has been concerned with two key questions:

(1) How is discourse organized to warrant a factual case? (2) How is factual discourse used to accomplish specific activites?

That is, on the one hand, it has examined the various devices and procedures that are drawn upon by people to offer a specific version of the world as real, actual or factual. For example, Woolgar (1 980) has examined the use of ‘externalizing devices’ in a Nobel Prize acceptance speech which work to characterize the phenomenon as out-there to be discovered rather than a construction of research instru- ments or of theoretical developments. On the other hand, it has explored some of the ways in which these versions can be used to accomplish activities. For example, Wowk (1984) has examined the way a murder confession may construct a factual version of the victim in such a way as to mitigate the act of killing.

As we have indicated, attribution theory has made limited attempts to address the second of these questions when it has raised the issue of the function of attribution (Hewstone, 1983; Tedeschi and Reiss, 1981), although it has not adopted the thoroughgoing functional or action orientation of discourse analysis. However, the first of these questions has not been posed at all within the theory, and it is perhaps not difficult to see why this is. As soon as we question how factual versions are constructed in discourse the very phenomenon of attribution starts to dissolve into more general issues about how accounts are constructed and managed, and into the rhetoric of everyday and institutional interaction.

Our empirical examination of these issues focuses on the notion of ‘consensus


information’; that is, one of the three types of information deemed crucial in the Kelley/McArthur variant of attribution theory (Hilton et al., 1988). By concentrating on the way notions of consensus are used in a particular dispute we shall illustrate in practice the three critical themes we have developed with attribution theory.

CONSENSUS AS A RHETORICAL CONSTRUCTION

We shall address here two major types of devices through which consensus is accomplished, or evoked (although this should not be taken as suggesting that these are the only ones available). One way of warranting a claim to fact is to maintain that it has the assent of reliable witnesses. This form of warranting must be dis- tinguished from the superficially similar consensus warrants that assert that some belief is true or justified because it is endorsed by many people, or the claim that some activity is acceptable because many people do it (cf. Hilton et al., 1988). The crucial feature of the accounts we are interested in is that consensus in reports provides corroboration about the factuality of a version - if one witness to a car accident claims the driver was going too fast they may be discounted; however, if most or all witnesses claim this they are likely to be more convincing.

The second form of consensus warranting we shall examine attends to a potential problem with the first form. Witnesses may agree because they all saw the same thing; however, they may agree because they have cooked up a story together or because, more innocently, through interacting with one another they have developed a common, but flawed, understanding. Thus there is value in finding witnesses who have not communicated or are independent. Their versions can be seen to be uncon- taminated through contact. Thus this second form of consensus warranting stresses the independence of the holders of the consensual view.

In this paper we shall examine how consensus was formulated and deployed in a drawn-out dispute between a leading politician and a group of journalists, concerning what had been said at a press briefing. In particular, we shall be concerned with the ways accounts of consensus were used when marshalling a case and, further- more, how they could be undermined. That is, we shall focus on the rhetorical or argumentative features of these kinds of accounts (cf. Billig, 1987, 1989b).

ANALYTIC MATERIALS AND CONTEXT

The materials that we shall be working with concern a press briefing between Nigel Lawson, the then British Chancellor of the Exchequer (chief finance minister), and ten Sunday newspaper journalists, that took place on the Friday 4th of November 1988. We chose to look at this body of materials because it promised conflict between versions of a particular event with fairly clearly identifiable interests at stake as well as a moderately clearcut time sequence of responses and counter responses. It was also a body of naturalistic materials where there was ample scope for what would traditionally be called attributional reasoning in a variety of different ways and levels.

This meeting was one of a series of ‘off-the-record’ briefings in which senior politicians ‘float ideas in the press’ to gain responses to plans which they may not yet

Nigel Lawson s tent 4 1 1

wish to commit themselves publicly. These ‘unattributable’ briefings are called ‘lobby journalism’, the ‘lobby’ being the group of journalists who are invited to attend. As will emerge more fully below, these meetings have been a source of substantial disputation. For example, some newspapers have criticised the lobby system for being a way of manipulating the press.

This particular lobby meeting was the focus of considerable controversy and was the subject of accounts in the press for several weeks afterwards. The most contentious issue concerned a proposal to alter benefits payable to old-aged pensioners. The Sunday press reports had it that a major and controversial alteration was afoot, in which benefits currently payable universally to all pensioners, might in future be ‘targeted’ upon the more needy, through a process of ‘means testing’ (income assessment), such that some would receive reduced allowances or none at all, while others might receive increases.

During the Sunday following the briefing the policy suggestion was fiercely attacked by both Labour (left wing) politicians as well as some members of the Chancellor’s own Conservative (right wing) party. Subsequently (on the Monday, on radio and in Parliament debate) Lawson made a startling claim. He denied that he had said any such thing; he had not proposed the policy the papers had suggested. And he went on to claim that the journalists had got together and their ‘fevered imagina- tion’ had produced ‘a farrago of invention’ whose accounts ‘bear no relation whatever to what I said’ (Hansard, 7 November: 22,24,26 - Hansard is the official published record of speeches and debates in the House of Commons). Thus instead of the debate being focused on the specific policy its axis came to revolve around the question of what had actually gone on at the briefing, and it is here that accounts involving the consensus of independent witnesses become a crucial feature of argument.

In the following days the press produced long articles to defend and justify their accounts against Lawson’s accusation, and made counter accusations that questioned Lawson’s version and depicted him as trying to escape from the embarrassing controversy he had produced. Some three weeks after the original briefing an increase to pensioners’ benefits was announced which provided &200 million of additional (rather than re-distributed) money. One newspaper described this as ‘Extra benefits for 2.6 million pensioners rushed through to cover Lawson’s means test gaffe’ (Guar- dian, 25 November).

The analytic materials we used were derived mainly from six newspapers: The Guardian, The Times, the Sun, the Observer, the Sunday Mirror and the Sunday Times. These particular papers were chosen because they are for the most part (except- ing the Sun and the Sunday Mirror) considered to be ‘quality’ papers and they cover different shades of political opinion, with the Guardian, the Sunday Mirror and the Observer usually being thought of as rather less right winglconservative than the Times, the Sun and the Sunday Times. These materials were supplemented by Hansard. The analytic issues at stake in studies of this kind are described in more detail in Potter and Wetherell (1987) and Wetherell and Potter (1988).

EXPLICIT CONSENSUS: WHAT EVERYONE SAW

Our analysis starts with an examination of the way consensus may be constructed to warrant a case, and how it may be subsequently undermined through being recast


as collusion. The following three extracts all invoke consensus as a warrant for truth (one and two are contributions to parliamentary debate by MPs whose consti- tuencies are recorded in brackets).

(1) Mr A. J. Beith (Berwick-on-Tweed): How on earth did the Chancellor, as a former journalist, manage to mislead so many journalists at once about his intentions? (Hansard, 7 November: 23).

(2) Mr David Winnick (Walsall, North): As all the Sunday newspapers carried virtually the same story, is the Chancellor saying that every journalist who came to the briefing - he has not denied that there was one - misunderstood what he said (Hansard, 7 November: 26).

(3) The reporters, it seemed, had unanimously got it wrong. Could so many messengers really be so much in error? It seems doubtful (Guardian, 8 November).

In attribution studies, the sorts of factual construction offered in Extracts 1 to 3 would be replaced by experimenter-constructed vignettes, of the following sort:

Sally bought something on her visit to the supermarket. Almost everyone else bought something on their visits to the supermarket. (Hilton and Slugoski, 1986, p. 78).

There is nothing participants can do about these statements except to take them as given, and make the appropriate inferences - Sally is unreal, and the constructor of the vignette is not in conversation with its recipient. Our ‘vignettes’ offer a very marked contrast; they are posed with irony and as questions - indeed, as rhetorical questions: they are dialogical, with their rhetorical structure inherent in the presen- tation of the facts of the matter, such that the desired inferences concerning Lawson’s veracity are rendered unavoidable. Also, because they are taken from contexts of real discourse, we can go on to analyse how their construction is contextually occasioned. The Sally vignette assumes the attributional epistemology of given reality (given by the experimenter)followed by causal inference (on the part of the subject). Crucially, it is an epistemology that is built into method, rather than something empirically discovered about how people think.

Let us replace our ‘vignettes’ in their context. In the sequence of events, Extracts 1 to 3 follow Lawson’s claim that the reporters were wrong. That is, he has questioned the factual status of the reports. Using the idea of witnesses corroborating versions, we take the rhetorical force of these accounts to be something like this; it is reasonable to imagine that some of the journalists might be misled in a briefing of this kind but not that they all should. If a number of observers report the same thing, that encourages us to treat the status of that thing as factual. The consensuality of the reporter’s accounts is offered as a basis for skepticism about the Chancellor’s.

Given that the accomplishment of a consensus across versions is one of the broad rhetorical tasks of Extracts 1 to 3, we can understand some of their detail as flowing from it. That is, the passages do not merely state that consensus is present, but provide the basis for a rhetorical appeal to the reader to construct it herself. For example, the extracts work on the quality or adequacy of the consensus, emphasizing the large scale of the consensus and its unanimity. The large size of the consensus is worked up using the description ‘so many’*journalists, which picks out the number


of journalists as exceptional or notable. In Extract 3, the effect is further enhanced by the contrast between the number of the ‘messengers’ and the size of the ‘error’:

(3b) (A) Could so many messengers really be (B) so much in error (Guardian, 8 November)

In Extract 2 unanimity is emphasized with the description ‘every journalist’ and this is reiterated in the description ‘all the Sunday newspapers’. That is, the newspaper reports are split from the journalist’s reports, as if to provide descriptions of two sorts of consensus each of which confirms the other. This presents consensus on consensus, doubling up as it were to make the most of what is available. A similar reiteration of consensus with a more marked rhetorical format appears in Extract 4:

(4) Mr Lawson held a Press Conference for ten political journalists.

universal benefits for old people irrespective of need.

at the time by the Treasury.

ten were all wallies (a colloquial term of abuse implying foolishness).

1 -,

2 -+

3 -,

(Sun, 9 November)

All ten got the impression that he had floated the idea of abandoning

All ten wrote stories for Sunday papers which were not questioned

But the next day the Chancellor himself told the Commons the

In Extract 4, not only is the consensus reiterated to repeat its effect but the repeated formulations make up the first two parts of a three-part list - a construction which has been shown to be rhetorically important in a number of different types of discourse (Heritage and Greatbatch, 1986; Jefferson, forthcoming; Wooffit, 1989).

In Extract 5, the quantity of the observers is emphasized in a different and rather neat way.

(5) The Government’s case now seems to consist of one unlikely tale piled on top of another: first, 10 journalists all fall victim to the same mass-delusion; then.. . ( Observer, 13 November).

Lawson’s claim that the journalists are mistaken is re-characterized as the claim that they all fell victim to a ‘mass delusion’. This description is effective in two ways. First, it presents Lawson as claiming something highly implausible. The description has strong connotations of the paranormal, occult, hypnosis and irrationality; so by attributing this description to Lawson his claims are made to appear weak or unlikely. Second, the phrase is often used to deal with large numbers of people, as implied by the term ‘mass’ (one thinks of mass media, mass society, the masses, etc.), more apposite for large crowds than small groups of journalists; so this description again works to heighten the perceived quantity of the observers.

In Extracts 1 to 5, then, the journalistic accounts are formulated as a consensus. The mere fact that this set of observers agreed over what went on is treated as making Lawson’s criticisms doubtful. However, consensus is not so straightforward an indication of truth, and neither was it treated as such in our data.

One of the analytic strategies we have been using when dealing with these materials is to identify and explore variability across versions. Such variability can be used as an analytic lever because it is indicative of the action or rhetorical orientation of the versions. This analytic point is explored in more detail in Potter and Mulkay


(1985), Potter and Wetherell (1987) and Wetherell and Potter (1988). It is an approach which is particularly valuable when dealing with materials, such as newspaper articles, where the sorts of sequencing information that is used by conversation analysts (e.g. Atkinson and Heritage, 1984) is limited.

In the materials we have examined there is variability both between Lawson’s versions and those of the journalists, as we have shown, and of course, it is variability within Lawson’s own talk that the dispute is all about. However, it is also possible to identify variability within the accounts of the journalists themselves. We shall start by looking at Lawson’s version(s) of events and examine the way in which they both differ from and orientate to the version offered in Extracts 1 to 5 . Clearly, Lawson’s integrity is threatened, both as an individual and as a member of the Government, if the newspaper version holds sway in the face of his protestations. He provides his own alternative versions in the course of the Parliamentary debate.

Extract 6 is taken from Hansard. In sequential terms, it follows the publication of the original articles in the Sunday papers, the criticisms of various politicians on the Sunday, and within the Parliamentary debate it follows shortly after an attack on the Chancellor (Extract 1) which specifically used the argument about consensus.

Mr Lawson: Let me say that the only announcement I have to make - the only change I have to inform the House of - is the one that I informed the House of today. This is a matter which I hinted to certain journalists on Friday.

1 4 They misunderstood what I was saying- 2 4 [Laughter.]

There is no [Interruption.] Mr Speaker*: Order. This kind of laughter and these interruptions take up a great deal of time. Many hon. Members want to ask questions about this matter. Mr Lawson: and as a result went in for a farrago of invention but that is no reason for the hon. Member to take it out (Hansard, 7 November: 24).

(6)

3 4

4 4

* The speaker is the parliamentary chairperson who calls members to take turns at speaking, attempts to regulate interruptions, and so on. Parliamen- tary etiquette requires that members refer to one another as Hon(ourab1e) Members.

Extract 6 is particularly interesting for the sequential unfolding of Lawson’s account of the briefing. In the course of debate Lawson had repeatedly denied having an intention to introduce means-tested benefits and claimed instead to be preparing to introduce an extra benefit for pensioners. He reiterates this point at the start of Extract 6 and suggests that it was this latter policy, the extra benefits, that was indicated to journalists.

He then characterizes the journalists as having misunderstood him (arrow 1). This provokes ironic laughter (arrow 2) and then Lawson is interrupted entirely and the Speaker has to recover order (arrow 3). At this point Lawson produced another gloss on the reporting. Departing from the rather weakly blaming idea of ‘misunder- standing’ he suggest that the stories are a ‘farrago of invention’ (arrow 4 - the

Nigel Lawson’s tent 41 5

Oxford English Dictionary, 2nd Ed., defines ‘farrago’ as a ‘medley’ or ‘confused mix- ture’).

To help make sense of this sequence we have drawn on an analysis of idiomatic expressions provided by Drew and Holt (1989). Drew and Holt suggest that idiomatic expressions - that is, cliched or proverbial expressions - tend to crop up at specific junctures in conversations; for example, where a speaker is making a complaint of some kind and the recipients are withholding affiliation. The idea is that because these expressions are largely figurative or formulaic they have a robustness that makes them hard to challenge with specific facts or information. That means they are suited to inauspicious environments, where a speaker’s version of events is not being well received. Lawson’s environment is certainly inauspicious. In nearly 15 minutes of debate he has been asked a number of hostile questions and has been interrupted on seven occasions during his replies. There is further interruption following his complaint about the press in Extract 6. It is following this highly disaffiliative response to his complaint that he offers the idiomatic formulation that the reporters ‘went in for a farrago of invention’.’

Lawson’s phrase ‘farrago of invention’ was heavily quoted in the newspapers on the following days. In at least one case the quotation appears to be ironically drawing attention to the persuasive work attempted by the phrase:

(7) Nobody can say the Chancellor of the Exchequer is not a bold man. Who else would invite 10 senior journalists to his home for a briefing and then state that when his guests departed they went into a tent to concoct a ‘farrago of invention?’ (Sunday Times, 13 November).

Despite its idiomiatic form, with any rhetorical payoff that may bring, Extract 6 deals in only a weak manner with the consensual nature of the press stories. Specifi- cally, it does not attend to potential questions of motivation - why should the journalists make up this story? If Lawson can establish a motive his case will be more convincing.

Extract 8, again from Lawson in Parliament, provides both a more elaborate and a rather different account for the journalists’ stories than does Extract 6.

(8) Mr Lawson: I am grateful to my hon. Friend. In fact, the statements, as I sad, - the statements that appeared in the press on Sunday bore no relation whatever to what I in fact said. What I have said to them [the reporters] is that, while we were absolutely, totally committed to maintaining

* There are some interesting complexities with this phrase. The Random House Dictionary of the English Language offers the idiom: ‘a farrago of doubts, fears, hopes and wishes’. Not only is this idiom, and indeed the term ‘farrago’ itself, obscure but it is not precisely quoted by Lawson. However, even if, as is likely, the majority of the audience would have been unable to define the term or recall the idiom, this would have only been to exaggerate the formal properties of idioms that Drew and Holt identity. It certainly makes any challenge difficult and, in context, the phrase ‘farrago of invention’ allows readers to infer that some sort of suspect and possibly foreign sounding kind of invention was underway. In this respect it has properties similar to ‘mass-delusion’ used in Extract Five. For those familiar with the etymology of the term it would also have served perfectly for Lawson’s needs. The Oxford English Dictionary cites Canning (1927, Poetical Works): ‘No longer do we want this farrago of cowardice, cunning and cant’ which provides a suitably perjorative frame for understanding the reporter’s actions. Moreover, the idea of a ‘farrago of doubts, fears, hopes and wishes’ applied to the journalists presents the image of a confused group which was motivated to forge a single consensus.


Ms Clare Short (Birmingham, Ladywood): They will have their shorthand notes.

Mr Lawson: Oh yes, they will have their shorthand notes and they will know it, and they will know they went behind afterwards and they thought there was not a good enough story and so they produced that. They will know that I said that, while.. . (Hansard, 7 November: 26).

Extract 8 reiterates the wrongness of the journalists’ story; and this time uses the strong formulation that the press reports ‘bore no relation whatever’ to what he had said in the briefing. However, in this passage for the first time Lawson accounts for the difference between the stories and ‘the facts’. The account is that the facts did not make a good enough story so the reporters colluded together to produce a better one. It is this collusion that explains the consensus. The stories tally not because that is how the facts were but because the journalists got together and invented a single story which they all used, and the motive offered is that they needed to invent such a story because the real story was not good enough to print. ! One of the common features of the construction of factual accounts is the manage-

rhent of the issue of interests and motivation. That is, one of the ways of warranting accounts is to make them appear disinterested or unmotivated (Potter and Wetherell, 1988a; Smith, 1978) while one way to undermine them is to invoke interests or motivations (Mulkay and Gilbert, 1982; Yearley, 1982, 1987). It is this latter strategy that is deployed by Lawson in Extract 8.

Our discussion so far has focused on the variability between Lawson and the journalists in their accounts and explanations of the consensus in press reports. However, things are both more complicated and more theoretically interesting than that. There are other accounts which deviate from the consensus argument in the newspapers. Here we shall focus on a single example. The Guardian, which had strongly criticized the Chancellor and his version of events in its editorials and reports, provides the following description.

(9) Because they assumed that the comments were unattributable, the correspondents present did not attempt their usual assiduous verbatim note. Instead they gathered afterwards to ask each other the traditional lobby question: ‘What’s the story?’ The question cannot on this occasion have taken long

to answer. (Guardian, 11 November)

In this version instead of the consensus being a product of independent observers fitting the words to the facts, it is a product of negotiation as the reporters later decide how to characterize the event.

Why should the Guardian produce this version, seeming, as it might, to side more with Lawson on the matter than with its own and other versions in the press? One notable feature of the Guardian is that it has withdrawn from the system of lobby journalism. Extract 9 occurs as part of an editorial critique of the system - one of the reasons the lobby is bad is because such meetings are normally ‘off the record’, and so journalists do not take ‘proper notes’ and have to get together to sort the story out afterwards. The continuation of Extract 9 in the following passage makes this clear.

(10) The reports duly appeared, with that impressive unanimity which is the

Niger Lawson s tent 4 17

dangerous fruit of mass lobby briefings, on Sunday morning. (Guardian, 11 November).

So both the emphasis on consensual validation and the ostensibly contradictory notion of consensus as collusion, are both sustainable in that they are contextually occasioned to do different kinds of rhetorical work: critising Lawson, critising the lobby.

By analysing our extracted ‘vignettes’ for their rhetorical organization, and by replacing them within their discursive context, we have shown how these formulations of given fact, of words and the world, are not merely ‘given’ bases for inference, but are themselves constructed and de-constructible by the participants themselves, a rhetorical process inseparable from the drawing of inferences. As a further demon- stration of the interesting and analysable complexity of these discursive constructions, we turn now to one of the critical aspects upon which the consensus-versus-collusion dichotomy is argued: the notion that a joint account is produced by independent individuals. Once again, we are able to reveal the constructed complexity, and the rhetorical organization, in real discourses, of what attribution theorists take to be ‘consensus information’.

CORROBORATION INDEPENDENT WITNESSES AGREE

One of the features of the use of different accounts to corroborate one another is that they should be independent. This idea is important in courts of law, of course, where there is the notion of collusion between witnesses. Indeed, is part of the reason for which witnesses are kept out of the courtroom, so that they remain uncontamin- ated by the stories of others. The issue of independence is also drawn on in Lawson’s ‘they went behind and produced’ the story account, reproduced in Extract 8: the point being that different inferences are available if the consensus is arrived at independently than when it is contrived together.

On the weekend following the Lawson briefing the Sunday papers introduced a new reason for doubting Lawson’s version. The reason was that new information had been obtained by a television news editor which provided independent corroboration. The claimed virtue of this was that it agreed with the reports from the newspaper journalists - it fitted in to the consensus - but it was obtained independently. The value for the papers of such an independent account was that it could be used to rebut Lawson’s accusations of collusion. And indeed, it was used in exactly this way. For example, the following extract uses the independence of the television story to cast doubt on Lawson’s explanation in terms of misreporting.

( 1 1) It emerged yesterday that ITN [Independent Television News] - which ran a similar story in its 8.55 p.m. bulletin on Saturday 5 November - based it on separate conversations with Treasury officials. Until now it had been assumed that the ITN report was based on early editions of last Sunday’s papers.

According to Michael Brunson, political editor of ITN, he reached the same conclusion by independent means, without attending the briefing. So far, the Chancellor has claimed the only source for the means-testing


stories was deliberate misreporting by the journalists he briefed the previous Friday.

Mr Brunson said yesterday that he spoke to the Treasury last Saturday, having already acquired ‘a certain amount of information’ and ‘came away with the even stronger impression that there would be cuts in the Christmas bonus’ (Observer, 13 November).

The rhetorical effectiveness of this passage is similar to that of Extracts 1 through to 5. It warrants the original version produced by the journalists by offering a further degree of consensus in reportings, but this time one which is garnered independently of the journalists’ lobby. The same argument was used in a rather more compact form on the front page of another paper:

(12) ITN says the story was obtained and checked independently of the Lawson briefing, which it did not attend, and was written before the first editions of last Sunday’s newspapers appeared (Sunday Times, 13 November).

In Extracts 11 and 12, then, independence is stressed to warrant the factuality of the version. However, following our analytic heuristic we can examine the materials for alternative versions (Potter and Wetherell, 1987). There is an analogy here with Smith’s (1978) study of a mental illness account. Smith noted that although the narrative warranting the existence of the illness offered various independent witnesses to its factual reality, a close reading of the narrative suggested another version in which the witnesses were not as independent as they appeared at first sight. That is, her close reading reveals the constructed, discursively accomplished nature of the independence.

In fact, we do not have to read very closely to find just such variation. Even in Extract 11, Brunson is quoted as saying that he went to the Treasury ‘having already acquired a certain amount of information’. This element is elaborated on in Extract 13:

(13) Shortly after lunchtime last Saturday, Michael Brunson [. . .] came across a story that was to dominate the week’s news.

Robin Cook, Labour’s social security spokesman, told him that something was going on in the Treasury over pensioner’s benefits. Cook had been alerted by one of the Sunday paper lobby correspondents briefed by Nigel Lawson [. . .] at 11 Downing Street the day before.

Brunson had not been to the briefing. But his own Treasury contacts soon confirmed what Cook said the lobby correspondents had been told. Brunson knew he had a story to lead last Saturday evening’s main ITN News at 8.55 ( Sunday Times, 13 November).

This extract departs from ideal model of independent corroboration in two ways. First it suggests that the origin of the television editor’s information was one of the journalists from the meeting, and that the channel for this information was an Opposition spokesman who was heavily critical of Lawson in the following week’s papers. This seems to offer a much weaker idea of independence than is implied in Extracts 11 and 12. Instead of Brunson approaching the Treasury naively and finding the same thing as the reporters - the sort of naivety we might expect from independent witnesses to a robbery - in this version he is already approaching


the situation with a ready-made and potentially hostile version of it, which could provide a frame for interpreting any claims then made by Treasury spokespersons. Second, the reference to the television editor’s need for ‘a story to lead [the] main ITN news’ echoes Lawson’s suggestion concerning journalist’s motivations for mis- reading the facts (Extract 8 above). In Extract 13, then, the ITN editor can be read as being directly influenced by the putatively collusive lobby and party to exactly the same truth - corrosive motivation as the reporters.

What are we to make of this? Why is there this variability in versions of independence in the newspapers? We are certainly not suggesting that this latter version (Extract 13) should be taken as the ‘truth of events’ seeping through the accounts in some way. Instead, we suggest that both versions can be understood as pragmati- cally occasioned. The pragmatic force is clear enough for the independent corroboration version, this being used to undermine Lawson’s accusations of collusion and confabulation. But why should this paper then present this detailed description which only weakens that version? We suggest that there are two considerations here.

One clue to the matter is that over this following weekend several of the papers published lengthy pieces presenting what purported to be moment-by-moment accounts of the disputed briefing and the events that followed it. These narratives can be seen as a form of ‘empiricist warranting’ (Gilbert and Mulkay, 1984; Mulkay and Gilbert, 1983) of the conclusions that had previously been drawn. That is, they presented conclusions as if these were constrained by a neutral and available record of events. The account of the Labour politician’s involvement with the television editor is an integral part of this record because it explains how they both became involved at such an early stage. In effect, the narrative form used requires the detail of connections to maintain ;ts coherence (cf. Gergen, 1988; Gergen and Gergen, 1987).

A further feature of these detailed descriptive accounts is that they contained

(14) Mr Lawson sat in an armchair in one corner, next to a window looking out over the garden of No 1 1 Downing Street. The Press Secretary, Mr John Gieve, hovered by the door. The rest of us, notebooks on our laps, perched on chairs and sofas in a circle around the Chancellor. It was 10.15 on the morning of Friday, 4 November. . . (Observer, 13 November).

Narrative description of this sort creates a graphic, vivid and believable world, described in terms suggesting direct perception or fresh visual memory (Rimmon- Kenan, 1983). This provides a contextual warrant for what is said - a point we explore elsewhere with respect to this set of materials (Edwards and Potter, 1989). In effect, then, we can map some potential tensions which arise when the same claim - that Lawson’s version of events is flawed - is warranted in different ways. The apparent conflict of versions arises from the deployment of two different kinds of warranting: independent witnessing, and narrative-empiricist description, working against each other. The conflict arises from our de-contextualizing them, and placing them in opposition. In the contexts in which they each occur, they are separate from each other and sensible, each deployed to bolster the press’s case, that its story was not confabulation (independence), and that it belonged to a believably graphic, richly detailed narrative.

These detailed narrative descriptions demonstrate the inescapably social-pragmatic

much collateral information. Take Extract 14, for example:


status of descriptive vignettes. It might be assumed that the more detailed a factual description is, the more graphic its picturing of scene and circumstance, the more careful its construction of a fine-grained sequence of events, then, provided no obvious confabulation is being offered, the less scope there would be for pragmatic work to be done in constructing truths about the world. On the contrary, however, the provision of more and more ‘facts’ merely provide further warrant for the narrator’s status as truth-teller, such that the more contentious part of the story is camouflaged amidst a mass of ‘realistic’ detail. Indeed, one might say it is the novelist’s art, in creating a believable world for the purveyance of fiction.

DISCUSSION FROM ATTRIBUTION TO FACTUAL DISCOURSE

In this study we have examined the way consensus-in-accounts can be used as a technique for warranting the factuality of particular versions and how this warranting can be contested in turn. In section One we showed the way a specified version is warranted as factual or correct because different observers are formulated as provid- ing the same version in their discourse. And we examined how such an account could be undermined by reference to some sort of collusion. It was notable that the collusion was itself made accountable, either by being depicted as a consequence of motivation to produce a ‘good story’ or as a structural consequence of the lobby system of journalism.

The evidential significance of consensus can be buttressed by documenting the independence of its constituents. In Section Two we showed the way newspapers responded to Lawson’s collusion accounts by offering a further version that fitted with the consensus but was acquired independently of the others. Such an account can be undermined by showing that there are other reasons for the similarity in accounts (shared interest, perhaps) or that the independence was faulty. In this case, no direct attempt to undermine the independence arguments was found; nevertheless, an examination of the way in which detailed narratives of those events were produced in very different rhetorical contexts revealed a more restricted notion of independence than where it was itself used as the crucial argument.

More generally, we have tried to illustrate the utility of using variability between versions (Potter and Wetherell, 1987) as an analytic heuristic to identify some of the rhetorical processes at work in a particular political controversy. Variability is a useful guide to how discursive devices are flexibly deployed. For example, just as consensus can be worked up or textually strengthened to legitimate a position, so it can be placed in a new context which engenders skepticism. Thus, instead of consensus being offered as if derived from the objectively given nature of the world, it can be characterized as a product of negotiation or collusion.

Warrants of the truth of versions seek to place such versions out in the world, as things that everyone can independently see, things with a vividly describable texture, whose patent reality leaves no room for doubt or disagreement. Conversely, versions can be undermined by orienting to the perceivers of the world, to the possibility of error, of observers’ interestedness in their particular versions, to their inconsis- tency, or the possibility of mass delusion, collusion or confabulation.

It is important to note the rhetorical nature of these discursive devices. They do not themselves speak unequivocally for truth or for error; they are available


for rhetorical work, for the bolstering of a case or else for its refutation (Billig et al., 1988). Thus, while consensus is usable as a warrant for truth, it also offers a basis for the accusation of collusion. Similarly, while detailed description of action and circumstance supports the factuality of versions, it leaves itself vulnerable to the construction of alternative narratives; for example, it may be used to undermine another case which needed to gloss superficially the independence of observers. Con- sensus and independence, then, are not merely as analysts’ categories for theorizing about what people say, but may be studied as categories that speakers themselves flexibly deploy in the course of rhetorical work.

At this point an attribution theorist might argue the following. This study has thrown up an interesting range of phenomena about the constructed nature of accounts; however, what is needed now are some experiments on account production which will mesh it in with the existing attribution theory. For all we know such experiments might appear in the new few years. Yet we see three crucial problems with this kind of response. First, it presupposes that the facthnference distinction can be validly maintained. What we have argued, and illustrated with the Lawson materials, is that this distinction is an artifact of the methods of attribution research and that when we look at attribution in natural contexts it dissolves into a complex web of rhetoric. Second, it presupposes that inference is an individual, cognitive process. We have shown the way that inferences and upshots are repeatedly provided in the relevant passages of discourse. Third, it continues with the assumption that attribution is a category of abstract reasoning. In contrast, we have suggested that phenomena previously understood in this way are more coherently seen as constitu- tive parts of activities such as blamings and rebuttals.

More generally, the study of discourse leads us td examine the entire epistemological framework of the attributional approach, from the investigator’s role in defining the given information or facts of the matter, through the way such versions of reality are constructed precisely in order to encourage particular inferences and upshots, to the re-working of those inferences and upshots via re-formulations of the given information. Inference and given reality are each simultaneously the business of constructive work done by participants as part of the rhetoric of ordinary discourse; our suggestion is that the most effective research strategy will be one that does not attempt artificially to split them apart and which studies these processes in natural contexts.

REFERENCES

Abraham, C. (1984). ‘Problems in categorizing content: A discussion of the limits of content analysis’. Paper presented to the British Psychological Society London Conference, December.

Abraham, C. (1988). ‘Seeing the connection in lay causal comprehension: A return to Heider’. In: Hilton, D. (Ed.) Contemporary Science and Natural Explanation, Harvestermeatsheaf, Brighton.

Antaki, C. (1985). ‘Ordinary explanation in conversation: Causal structures and their defence’, European Journal of Social Psychology, 15: 21 3-230.

Antaki, C. (1988a). ‘Explanations, communication and social cognition’. In: Antaki, C. (Ed.) Analysing Lay Explanation: A Case Book, Sage, London.

Antaki, C. (1988b). ‘Structures of belief and justification’. In: Antaki, C. (Ed.) Analysing Lay Explanation: A Case Book, Sage, London.


Antaki, C. and Naji, S. (1987). ‘Events explained in conversational “because” statements’,

Atkinson, J. M. and Drew, P. (1979). Order in Court: The Organization of Verbal Interaction

Atkinson, J. M. and Heritage, J. (Eds) (1984). Structures of Social Action: Studies in Conver-

Barthes, R. (1974). S/Z, Jonathan Cape, London. Billig, M. (1982). Zdeology and Social Psychology, Blackwell, Oxford. Billig, M. (1985). ‘Prejudice, categorization and particularization: from a perceptual to a

Billig, M. (1987). Arguing and Thinking: A Rhetorical Approach to Social Psychology, Cam-

Billig, M. (1988). ‘Social representation, objectification and anchoring: a rhetorical analysis’,

Billig, M. (1989a). ‘The argumentative nature of holding strong views: A case study’, European

Billig, M. (1989b). ‘Psychology, rhetoric, and cognition’, History of the Human Sciences, 2:

Billig, M. (1990). ‘Collective memory, ideology and the British royal family’. In: Middleton,

Billig, M., Condor, S . , Edwards, D., Gane, M., Middleton, D. and Radley, A. (1988). Zdeologi-

Buss, A. R. (1978). ‘Causes and reasons in attribution theory: a conceptual critique’, Journal

Condor, S. (1988). ‘ “Race stereotypes” and racist discourse’, Text, 8: 69-90. Coulter, J. (1979). The Social Construction ofMind, Macmillan, London. Culler, J. (1983). On Deconstruction, Routledge and Kegan Paul, London. Denzin, N. K. (1978). The Research Act, Aldine, Chicago. Drew, P. and Holt, E. (1989). ‘Complainable matters: The use of idiomatic expressions in

Edwards, D. and Mercer, N. M. (1989). ‘Reconstructing context: the conventionalization of

Edwards, D. and Middleton, D. (1986). ‘Joint remembering: constructing an account of shared

Edwards, D. and Potter, J. (forthcoming). ‘The Chancellor’s memory: Language and truth

Fletcher, G., Fincham, F., Cramer, L. and Heron, N. (1987). ‘The role of attributions in

British Journal of Social Psychology, 26: 119-26.

in Judicial Settings, Macmillan, London.

sation analysis, Cambridge University Press, Cambridge.

rhetorical approach’, European Journal of Social Psychology, 15: 79-103.

bridge University Press, Cambridge.

Social Behaviour, 3: 1-16.

Journal of Social Psychology, 19: 203-23.

289-307.

D. and Edwards, D. (Eds) Collective Remembering, Sage, London.

cal Dilemmas: A Social Psychology of Everyday Thinking, Sage, London.

of Personality and Social Psychology, 36: 13 1 1-1 32 1.

making complaints’, Social Problems, 35: 501-20.

classroom knowledge’, Discourse Processes, 12: 91-104.

experience through conversational discourse’, Discourse Processes, 9 (4): 423-459.

in discursive remembering’, Applied Cognitive Psychology.

the development of dating relationships’, Journal of Personality and Social Psychology, 53: 481-89.

Forsyth, D. R. (1980). ‘The functions of attributions’, Social Psychology Quarterly, 43: 184-

Genette, G. (1980). Narrative Discourse, Cornell University Press, Ithaca, NY. Gergen, M. (1988). ‘Narrative structures and social explanations’. In: Antaki, C. (Ed.) Analys-

ing Lay Explanation: A Case Book, Sage, London. Gergen, K. and Gergen, M. (1987). ‘Narratives of relationship’. In: Burnett, R., McGee,

P. and Clarke, D. (Eds) Accounting for Relationships: Explanation, Representation and Know- ledge, Methuen, London.

Gilbert, G. N. and Mulkay, M. (1984). Opening Pandora’s Box: A Sociological Analysis of Scientists ’ Discourse, Cambridge University Press, Cambridge.

Harre, R. (1981). ‘Expressive aspects of descriptions of others’. In: Antaki, C. (Ed.) The Psychology of Ordinary Explanations of Social Behaviour, Academic Press, London.

Harvey, J. and Weary, G. (1984). ‘Current issues in attribution theory and research’, Annual Review of Psychology, 35: 427459.

Harvey, J., Turnquist, D. and Agostinelli, G. (1988). ‘Identifying attributions in oral and written explanations. In: Antaki, C. (Ed.) Analysing Lay Explanation: A Case Book, Sage, London.

189.

Heider, F. (1958). The Psychology of Interpersonal Relations, Wiley, New York.

Nigel Lawson S tent 423

Heritage, J. (1984). Garfnkel and Ethnomethodology, Polity Press, Cambridge. Heritage, J. and Watson, R. (1979). ‘Formulations as conversational objects’. In: Psathas,

G. (Ed.) Everyday Language: Studies in Ethnomethodology, Irvington, New York. Heritage, J. and Greatbatch, D. (1986). ‘Generating applause: A study of rhetoric and response

at party political conferences’, American Sociological Review, 92: 1 10-1 57. Hewstone, (Ed.) (1983). Attribution Theory: Social and Functional Extensions, Blackwell,

Oxford. Hilton, D. J. (1990) ‘Conversational processes and causal attribution,’ Psychological Bulletin,

107: 65-81. Hilton, D. J. and Slugoski, B. R. (1986). ‘Knowledge-based causal attribution: the abnormal

conditions focus model’, Psychological Review, 93: 75-88. Hilton, D. J., Smith, R. H. and Alicke, M. D. (1988). ‘Knowledge-based information acqui-

sition: Norms and the functions of consensus information’, Journal of Personality and Social Psychology, 55: 530-540.

Hollway, W. (1989). Subjectivity and Method in Psychology: Gender, Meaning and Science, Sage, London.

Jones, E. E. and Davis, K. E. (1965) ‘From acts to dispositions: the attribution process in social perception’. In: Berkowitz, L. (Ed.), Advances in Experimental Social Psychology (vol. 2). New York: Academic Press.

Jefferson, G. (forthcoming). ‘List construction as a task and interactional resource’. In: Psathas, G. and Frankel, R. (Eds.) Interactional Competence. Ablex, Norwood, N J.

Kelley, H. H. (1967) ‘Attribution theory in social psychology’. In: Levine, D. (Ed.), Nebraska Symposium on Motivation. University of Nebraska Press, Lincoln, Neb.

Kelley, H. H. (1973) ‘The processes of causal attribution’, American Psychologist, 28: 107-128. Levinson, S. (1983). Pragmatics, Cambridge University Press, Cambridge. Lynch, M. and Woolgar, S. (1989). ‘Introduction: Sociological orientations to representational

practice in science’, Human Studies, 11: 99-1 16. McArthur, L. A. (1972). ‘The how and what of why: some determinants and consequences

of causal attribution’, Journal of Personality and Social Psychology, 2: 171-193. McKinlay, A., Potter, J. and Wetherell, M. (forthcoming). ‘Discourse analysis and social

representations’. In: Breakwell, G. and Canter, D. (Eds) Empirical Approaches to Social Representations, Oxford University Press, Oxford.

Mulkay, M. (1985). The Word and the World: Explorations in the Form of Sociological Analysis, Allen Unwin, London.

Mulkay, M. and Gilbert, G. N. (1982). ‘Accounting for error: how scientists construct their social world when they account for correct and incorrect belief‘, Sociology, 16: 165-183.

Mulkay, M. and Gilbert, G. N. (1983). ‘Scientists’ theory talk’, Canadian Journal of Sociology,

Parker, I. (1989). The Crisis in Social Psychology and How to End it, Routledge and Kegan Paul, London.

Pomerantz, A. (1979). ‘Telling my side: “Limited access” as a “fishing device”’, Sociological Inquiry, 50: 186196.

Pomerantz, A. (1986). ‘Extreme case formulations: a new way of legitimating claims’, Human Studies, 9: 219-230.

Pomerantz, A. (1989). ‘Constructing skepticism: Four devices used to engender the audience’s skepticism’, Research on Language and Social Interaction, 22.

Potter, J. and I. Litton (1985). ‘Some problems underlying the theory of social representations’, British Journal of Social Psychology, 24: 81-90.

Potter, J. and Mulkay, M. ( I 985). ‘Scientists’ interview talk: Interviews as a technique for revealing participants’ interpretative practices’. In: Brenner, M., Brown, J. and Canter, D. (Eds) The Research Interview: Uses and Approaches, Academic Press, London.

Potter, J. and Wetherell, M. (1987). Discourse and Social Psychology: Beyond Attitudes and Behaviour, Sage, London.

Potter, J. and Wetherell, M. (1988a). ‘Accomplishing attitudes: fact and evaluation in racist discourse’, Text, 8: 5148.

Potter, J. and Wetherell, M. (1988b). ‘The politics of hypocrisy: notes on the discrediting of apartheid’s opponents’, Social Psychology Section Newsletter, 19: 3042.

8: 179-197.


Potter, J. , Stringer, P. and Wetherell, M. (1984). Social Texts and Context: Literature and

Potter, J., Wetherell, M., Gill, R. and Edwards, D. (forthcoming) Discourse-Noun, verb

Rimon-Kenan, S. (1983). Narrative Fiction: Contemporary Poetics, Methuen, London. Sabini J. and Silver, M. (1980). ‘Baseball and hot sauce: A critique of some attributional

Schegloff, E. A. (1988). ‘Description in the social sciences I; Talk-in-interaction’, Papers in

Searle, J. (1983). Intentionality: An Essay in the Philosophy of Mind, Cambridge University

Semin, G. (1980). ‘A gloss on attribution theory’, British Journal of Social and Clinical Psy-

Shotter, J. (1981). ‘Telling and reporting: Prospective and retrospective uses of self-ascription’. In: Antaki, C. (Ed.) The Psychology of Ordinary Explanations of Social Behaviour, Academic Press, London.

Social Psychology, Routledge and Kegan Paul, London.

or social practice, Philosophical Psychology.

treatments of evaluation’, Journal for the Theory of Social Behaviour, 10,83-95.

Pragmatics, 2: 1-24.

Press, Cambridge.

chology, 19: 291-300.

Smith, D. (1978). ‘K is mentally ill: the anatomy of a factual account’, Sociology, 12: 23-53. Smith, D. (1983). ‘No one commits suicide: textual analysis of ideological practices’, Human

Studies, 6: 309-359. Smith, J. (1987). ‘Making people offers they can’t refuse: A social psychological analysis

of attitude change’. In: Hawthorn, J. (Ed.) Propaganda, Persuasion and Power, Edward Arnold, London.

Suchman, L. (1987). Plans and Situated Action: The Problem of Human Machine Communica- tion, Cambridge University Press, Cambridge.

Tedeschi, J. T. and Reiss, M. (1981). ‘Verbal strategies in impression management’. In: Antaki, C. (Ed.) The Psychology of Ordinary Explanations of Social Behaviour, Academic Press, London.

Thompson, J. (1984). Studies in the Theory of Ideology, Polity, Cambridge. Turnbull, W. and Slugoski, B. R. (1988). ‘Conversational and linguistic processes in causal

attribution’. In: Hilton, D. (Ed.) Contemporary Science and Natural Explanation, Harvester/ Wheatsheaf, Brighton.

Weiner, B. (1985). ‘ “Spontaneous” causal thinking’, Psychological Bulletin, 97: 74-84. Wetherell, M. and Potter, J. (1988). ‘Discourse analysis and the identification of interpretative

repertoires’. In: Antaki, C. (Ed.) Analysing Lay Explanation: A Case Book, Sage, London. Wetherell, M., Stiven, H. and Potter, J. (1987). ‘Unequal egalitarianism: a preliminary study

of discourses concerning gender and employment opportunities’, British Journal of Social

Widdicomb, S . and Wooffitt, R. (forthcoming). “‘Being” versus “doing”’ punk (etc.): on achieving authenticity as a member’. Journal of Language and Social Psychology.

Wooffitt, R. (1988). ‘On the analysis of paranormal phenomena’, Journal of the Society for Psychical Research, 55: 139-149.

Wooffitt, R. (1989). ‘(Telling) tales of the unexpected: A sociological analysis of accounts of the paranormal’. Unpublished DPhil., University of York.

Woolgar, S. (1980). ‘Discovery: Logic and sequence in a scientific text’. In: Krohn, R., Knorr, K. and Whitley, R. D. (Eds) The Social Process of ScientiJic Investigation, Reidel, Dordrecht.

Wowk, M. (1984). ‘Blame allocation, sex and gender in a murder interrogation’, Women’s Studies International Forum, 7: 75-82.

Yearley, S. (1981). ‘Textual persuasion: the role of social accounting the construction of scientific arguments’, Philosophy of the Social Sciences, 11: 409435.

Yearley, S. (1982). ‘The relationship between epistemological and sociological interests: some ambiguities underlying the use of interest theory in the study of scientific knowledge’, Studies in the History and Philosophy of Science, 13: 353-388.

Yearley, S. (1984). ‘Proofs and reputations: Sir James Hall and the use of classification devices in scientific arguments’, Earth Sciences History, 3: 2543.

Yearley, S. (1987). ‘Demotic logic; causal discourse and the structure of explanations’, Text, 7,181-203.

Psychology, 26 59-7 1.

nigel lawson's tent: discourse analysis, attribution theory and the social psychology of fact

Documents