1. On the procedural analysis of political inquiries

The literature on the operation of inquiry committees is extensive. In political science and constitutional law, political inquiries are discussed with regard to the functions they fulfil within the democratic polity. Accordingly, in this perspective, inquiries serve such purposes as: controlling the work of government (Claveloux Reference Claveloux1983; Aberbach Reference Aberbach1986; Alba Reference Alba1995; Baumgarten Reference Baumgarten2000; Höpfner Reference Höpfner2004); containing scandals (Germis Reference Germis1988); binding public attention (Willems et al. Reference Willems, Wolf and Eckert1993); or re-establishing tarnished reputations (Ziemske Reference Ziemske1991; Rozell and Wilcox Reference Rozell and Wilcox2000; Blaney Reference Blaney2001). On the whole, there are few sociological works on inquiry committees, exemplified by analyses such as Alexander’s (Reference Alexander and Alexander1989) of the “Watergate” inquiry and Lynch and Bogen’s (Reference Lynch and Bogen1996) work on the “Iran-Contra” hearings. Alexander describes the Watergate hearings’ social function in terms of a ritual renewal and purification of US society. Based on the TV coverage from the “Iran-Contra hearings”, Lynch and Bogen analyse the pragmatics of memory work in the inquiry, including its systematic limitations. The “making of history” thus acquires a dual character: on the one hand, it emerges as the practical and public reconstruction of the history of a secret and illegal weapons deal; on the other, as a historically charged spectacle of recent US history itself.

In this respect, research on committees of inquiry proceeds in an analogous manner to research on procedure more generally. A “procedure” is defined and evaluated in legal terms. An important frame of reference in current legal and political scholarship on procedure is the difference between adversarial and inquisitorial legal systems. In this line of thinking, procedural designs are evaluated according to the norms of decision and standards of evidence they imply: in adversarial procedure, the search for and submission of evidence, examination of witnesses, and the assertion of claims are dominated by the parties, and decisions are generally reached by way of a “common sense” logic. Inquisitorial procedural designs, on the other hand, rely on an authoritative figure, such as a judge or government official, guiding the “search for the truth” “from above”, and decisions are made according to an administrative logic. Thus, in the adversarial procedure “authority is fragmented and […] hierarchical control is relatively weak”, while in inquisitorial procedure “legal authority and decision-making is hierarchically organised and disputants […] play a more restrained role” (Kagan Reference Kagan2006, p. 5).

Considering the prevalence of the adversarial/inquisitorial distinction, it is not surprising that many recent works on the design and operation of procedure focus on the relative advantages and disadvantages of these two procedural types (Kritzer Reference Kritzer2004, p. 367, pp. 376-77). Similarly, evaluations of a given procedure often rely on the adversarial/inquisitorial distinction and proceed from there to look at the respective “other” type as a source of innovation (Kagan Reference Kagan2006; Bräcklein 2006). Politico-legal scholarship liberally applies those same criteria to committees of inquiry to argue for a need for reform (Schröder Reference Schröder1988; Ziemske Reference Ziemske1991; Morlok Reference Morlok2000; Weisgerber Reference Weisgerber2003) or to assess the effectiveness and desirability of amendments (Schröder Reference Schröder, Schneider and Zeh1989, Reference Schröder2000). There is also much concern with the legal limits of inquiries, e.g., vis-à-vis the executive (Baumgarten Reference Baumgarten2000; Rozell Reference Rozell2002), the media (Mangan Reference Mangan1994), lawyer-client interactions (Rich Reference Rich1988; Chud Reference Chud1999), or private persons (O’Neill Reference O’Neill2002). The common denominator of these works is that the analysis of procedure proceeds from the rules of procedure.

This “norm-bias” (Vollmer Reference Vollmer1996) is not confined to legal and political scholarship; it also pervades sociological or socio-legal thinking about procedure. Commonly, a procedure is evaluated according to normative criteria such as those of discourse ethics (Apel and Kettner Reference Apel and Kettner1992; Bal Reference Bal1994), procedural justice (Epp Reference Epp1998), public access (Bora Reference Bora1999), or the acceptance of outcomes (Machura Reference Machura2001). However, a procedure and its rules of procedure are not one and the same. The former is much more: a “social system” (Luhmann Reference Luhmann1969, p. 41), a “partially self-governing, complexity-reducing framework of action” (ibid.); a “matrix of possible events that can only occur with their specific meanings in this procedure” (ibid. p. 45, translations are my own, JS). Here, Luhmann lays the foundations for a sociological concept of procedure emancipated from law.

Besides separating procedure from legal norms, Luhmann also redefined the relation between the two. Legitimation durch Verfahren [Legitimisation by Procedure] offers a sociological understanding of legal norms next to a dynamic model of decision. Legal norms are integrated into the operation of procedure as “social systems serving a specific function, which is to arrive at a singular binding decision, and which are therefore restricted in their duration from the beginning” (Luhmann, Reference Luhmann1969, p. 41, translation JS). During this limited time, a complex, self-referential microcosm unfolds disengaged from everyday life. Here, nuances and details gain currency which seem irrelevant from the “outside”. The general critique of Legitimation durch Verfahren (Treiber Reference Treiber1986; Machura Reference Machura1993; Röhl and Machura Reference Röhl and Machura1997) was rather focussed on Luhmann’s concept of legitimisation. By contrast, we consider the crucial innovation to be Luhmann’s sociological concept of procedure (Rosenfield Reference Rosenfield1998; Ziegert Reference Ziegert, Banakar and Travers2002). In this paper, we push this sociological turn by describing procedure as the interplay of self-referential and contingent processes of decision-making, communication and knowledge production. This serves to establish a notion of procedure which does not function as the antonym of informality, but rather as the opposite of ritual (i.e., as contingent). This also serves to qualify other common interpretative frames (Luhmann Reference Luhmann1969, p. 175): the procedure is not the same as a single session of the inquiry, or interaction among co-present actors, even if performances do matter a great deal; it is not identical to the rules of procedure, even if these are followed, called upon, or established in the procedure itself; neither is it identical to parliament or the committee dealing with the issue, even if the procedure involves political organisations. Thus, our starting hypothesis reads: participants are only able to act issue-oriented, and their doings can only be reconstructed adequately with reference to the specific procedure in question and its current state. We aim to show that and in how far the political inquiries in question are actually conducted as a (contingent, self-referential and fenced-in) procedure, which in turn unfold their own retrospective and prospective references, roles and demands, as well as “good reasons” and powers of judgement.

We connect Luhmann’s notion of procedure as a social system to three further strands in sociological research:

1. a) to the ethnomethodological conversation analysis of pre-structured verbal exchanges in institutional settings, such as law courts, doctors’ offices, or parliaments;

2. b) to ethnographic laboratory studies on the socio-material and discourse-practical organisation of knowledge production and epistemic cultures;

3. c) to theories of decision-making from the sociology of organisations, explaining the sequential generation of previously contingent, currently binding and prospectively definitive decisions.

Each of these research traditions provides empirical criteria to assess self-reference and autonomy. We ask (a) how participants communicate in the hearings (b) to establish the facts (c) which then provide the basis for a definitive judgment.

2. Do three processes make a procedure?

Starting from a micro-sociological analysis of question/answer exchanges, we are going to address the inquiries’ degrees of self-reference and integration with regard to their respective (self-) enacted knowledges and their (self-) founded decisions. We measure “procedurality” according to the following considerations.

a) Communicative connections. Conversation analysis emphasises the sequential structuring of hearings and testimonies. According to this view, meaningful contributions result from successive, coordinated and methodical production. Studies on “talk at work” (Drew and Heritage Reference Drew and Heritage1992) in general and, more particularly, “courtroom studies” (Atkinson and Drew Reference Atkinson and Drew1979) have developed further relevant pre-structuring elements shaping the local course of conversation. Conditions such as “multi-party” figurations, “overhearing audiences”, or the “pre-allocation” of turns are important concepts here. In general, conversation analysis requires restraint in interpretation: the aim is to reconstruct the successive construction of meaning by the participants; roles and rules only gain relevance in the course of conversation; even seemingly simple definitions, such as the beginning and end of a hearing are produced by the participants themselves. Luhmann assumes a similar self-reference for procedure in general: “The beginning and the end are markers that are produced by the proceedings, that is, which are identified recursively in the management of the proceedings” (Luhmann Reference Luhmann2004, p. 206). With a view to procedurality, this poses the question, whether and in how far participants’ retrospective and prospective contributions integrate one procedure, and how participants render their contributions as contributions to the procedure. Conversely, one could ask whether the interrogations are merely disconnected events. This leads us to a first necessary (but not sufficient) condition in our definition of procedure – an integrated communicative process.

b) Foundation of knowledge. Ethnographic laboratory studies (Knorr-Cetina Reference Knorr-Cetina1981, Reference Knorr-Cetina1999) open the perspective on knowledge processes, which are fed by local, pragmatic tests and examinations. In the context of legal procedure, knowledge is generated by way of eye witnesses, expertise, or “outsourced” investigations (Valverde Reference Valverde2003; Sarat et al. Reference Sarat, Douglas and Umphrey2007). Contributions are documented, certified as facts, and made available for future reference (Latour Reference Latour2002). The review and fixation of contributions can include the use of notes (Scheffer Reference Scheffer2006), protocols (Smith Reference Smith1985), files (Vismann Reference Vismann2001; Scheffer 2007b), or archives (Lynch Reference Lynch1999), depending on the scope of the inquiry. In this way, the inquiry solidifies its own “case history” (Luhmann Reference Luhmann1969, p. 44) and keeps it available for future occasions. Speakers must respond to the current standard of knowledge. In the case of public inquiries, the additional element of an interested audience turns into a knowledgeable and verifying authority. In this sense, the publicity of knowledge processes reminds us of early modern scientific experiments (Shapin and Schaffer Reference Shapin and Schaffer1985; Shapin Reference Shapin1994), the results of which had to pass before an audience of “gentlemen”. Knowledge processes exhibit procedurality to the degree they rely on facts that have been produced and processed internally. The question is whether the inquiries proceed autonomously, i.e., from the facts they generate themselves, or, conversely, whether they follow external postulates. In the latter case, a necessary, but not sufficient condition of procedurality would remain unfulfilled.

c) Decision-making ability. The concepts borrowed from the sociology of organisations to analyse decision-making processes focus on a “history of the decision” (Luhmann Reference Luhmann1969, p. 40), i.e. a space of possibilities that is sequentially narrowed down to a definitive and final decision. Luhmann adds further sociological qualities to this stepwise selection from pre-selections. The self-referential character of the decision-making process immunises the procedure against social differences of rank and status. The decision-makers judge cases according to the participants’ “definitive” and “binding” statements (Wolff and Müller Reference Wolff and Müller1997). Participants are held to account for their performances and representations. The contingent character of decisions, paired with the internally-produced criteria for those decisions, motivates involvement and participation (Luhmann Reference Luhmann1969, p. 114). A decision observes behaviour as choice and thus as attributable (ibid. p. 44). This formulates another necessary, but not sufficient, condition for procedurality. We will speak of procedure if one binding decision emerges at the end of an internal decision-making history. Conversely, procedurality can be disputed if the decision is fixed per se or if it is missing altogether.

The processes of communication, knowledge production and decision-making prove to be more or less self-referential. Furthermore, they do not proceed separately, but refer to each other in the course of the inquiries. How can we imagine their concurrence? Communication, knowledge production and decision-making are mutually structuring. Conversational interaction provides knowledge; protocols keep this knowledge available to provide subsequent hearings with new issues and questions; decision-making again utilises communication and knowledge in the form of criteria for judgement and contrasts. These three processual perspectives also indicate in how far the inquiries actually function as an integrated, self-referential, and autonomous procedure. In the following, we attempt to show that and how the inquiries succeed to detach and emancipate themselves – more or less –Footnote ¹ from their environment as a procedure.

3. Methodical Maxims

The synopsis of processes of communication, knowledge production and decision-making leads us to some methodical ground rules. Firstly, we take a distanced and close-up look at political inquiries. We take a distanced stance towards the exciting political debates and investigations; on the other hand, we close in on the inquiries by looking at their details in a partially microscopic analysis. Both these movements can be formulated as the basic rules of our own procedure (of data analysis): our conclusions grow from process dataFootnote ², which are produced and made available in the course of the inquiries. Analogous to the minimal requirements for an “open” procedure, we do not interpret the data from their results. Thus, the question of procedurality only emerges after our empirical exploration. As a first step, we are merely concerned to use the interrogations in the inquiries as a starting point to determine how they operate with regard to knowledge production and decision-making.

We will hone this stipulation by way of a contrastive method. The following considerations provide the basis for our selection of cases: a) at first glance, the inquiries are concerned with similar issues. They both deal with matters of the “war against terror” and its legitimacy. b) Both inquiries exhibit a considerable lack of procedural rules. This factor propels the question about procedurality, i.e. the degree of autonomy and integration exhibited by the communicative, knowledge-producing and decision-making processes. c) Thus, both inquiries become didactic pieces on the programming of these processes by their operators. d) The inquiries already proved fundamentally different in the first data sightings. Thus, quick exchanges of questions and answers in the Hutton Inquiry provide a stark contrast to the markedly long speeches in the CIA committee.

Further analysis showed persistent differences in the progression, methods, and approach of the respective inquiries. We integrated these differences as features of political inquiries. We identified, ordered and contrasted the respective peculiar elements according to general criteria: inquiry methods, objects of inquiry, integration of the inquiry, impositions and their limits. In this way, we actively produced comparability between our cases (Niewöhner and Scheffer Reference Niewöhner and Scheffer2008). This results in a typification of strong and weak procedure. The confrontation of the cases served to focus some single characteristics: the questions “How is this there? How is this here?” stimulated empirical-analytical “explication” (Hirschauer Reference Hirschauer2001). Contrasting the political inquiries serves not only to thicken and explicate; at the same time, it stimulates the “sociological imagination” (Mills Reference Mills and Mills1959), i.e., it served to imagine a certain scope of variation. The contrasts invite us to apply the procedural analysis to further cases (e.g., the German “Visa” committee) and to extend the bases of comparison by adding further facets (e.g., the role of the political public, which is only sketched here). Finally, the contrasts are to be merged and interpreted according to the three criteria of procedurality.

A further methodical ground rule relates to the empirical use of theoretical concepts. We are concerned with using the contrasts in order to operationalise the notion of procedure as a matter of varying magnitudes, or a sum of variables. With respect to our contrasting cases, this does not culminate in a distinction between procedure and non-procedure, but rather in a relativisation of procedurality. “Strong procedure” and “weak procedure” do not denote ideal types, but rather points in an empirical spectrum. In this way, our analysis also serves as an example of how very fine-grained methodological approaches can contribute to the specification of grand theoretical blueprints. Thus, the empirical variation highlighted by the contrasts strongly invites further empirical work leading to a sociological explication of the notion of procedure.

4. The cases: Hutton Inquiry and CIA Committee

The Hutton Inquiry was constituted only days after the weapons expert and former UN inspector in Iraq, Dr David Kelly, was found dead near his home. The inquiry was set up on the intiative of Prime Minister Tony Blair in July 2003 at a time of heated debate: anti-war demonstrations were spreading, individual ministers were protesting and resigning, and there was open dissent between EU member states. Immediately before his death, David Kelly was at the centre of a controversy between the Government and the BBC about a dossier on the threat of Iraqi “weapons of mass destruction”. On 29 May 2003, BBC journalist Andrew Gilligan quoted an “unnamed source”, who had claimed that the dossier was changed against the wishes of the intelligence services shortly before its publication. The allegation was that the “sexed up” version of the dossier was meant to legitimise Britain’s decision to go to war vis-à-vis a critical public. The Government denied this. In June 2003, the parliamentary Foreign Affairs Select Committee (FAC) concerned itself with “The Decision to go to War in Iraq”. In the course of this inquiry, clues emerged that Dr Kelly may have been Gilligan’s “anonymous source”. On 18 July, the day after the Ministry of Defence confirmed Dr Kelly was the likely source, he was found dead.

The Prime Minister immediately announced an independent inquiry. The Secretary of State for Constitutional Affairs commissioned the High Court judge Lord Hutton to preside over the inquiry. A “Letter of Appointment”, dated 24 July 2003, defined the terms of the inquiry: “Urgently to conduct an investigation into the circumstances surrounding the death of Dr David Kelly”. In principle, British ministers can establish political inquiries without prior debate or division in Parliament (Ziemske Reference Ziemske1991). Such an extra-parliamentary inquiry does not have judicial powers to summon or swear in witnesses.Footnote ³ At the beginning of his inquiry on 1 August 2003, Lord Hutton announced a number of self-imposed procedural standards. He announced a two-stage progression: a first phase “will be devoted to obtaining, in a neutral way, by counsel to the Inquiry, an account of the events which took place by those who took part in them”; a second phase is concerned only with “persons whom I have notified of possible criticisms”. The latter are recalled and cross-examined by counsel for the various parties (BBC, Government, Kelly family) (Hearing transcript Ht., 01/08/03, 21-22).

The inquiry encountered extensive media interest. In Luhmann’s terminology: as a public issue, it was a “breakthrough” (1971, p.18), i.e. it quickly became “part of public opinion in the sense of our concept; it appears in daily press reporting in a way which presupposes that everybody knows the history of the issue” (ibid.; translation JS). In the course of the inquiry, the press quoted extensively from the protocols, constructed preliminary scores, and speculated about possible outcomes. A surge of critical voices about the inquiry can only be found on the day after Lord Hutton published his final report. Criticism focused on fixation on detail and political blindness:

We compare the Hutton Inquiry to the “Temporary Committee on the alleged use of European countries by the CIA for the transport and illegal detention of prisoners” (TDIP). It was set up by the European Parliament (EP) as a “temporary committee” according to Article 175 of the EP’s rules of procedure on 18 January 2006 (P6_TA-Prov(2006)0012). Parliament was unable to agree on the competence of regular committees of inquiry (Kipke Reference Kipke1989; Shackleton Reference Shackleton1998; Dreischer Reference Dreischer, Holtmann and Patzelt2004) – as, for example, in the case of an “alleged violation of community law”. The terms of reference defined tasks (“to collect and analyse information to find out whether the CIA”), the general course of action (“will liaise and cooperate as closely as possible with the Council of Europe”), time frame (four months plus a possible extension) and composition (“46 members of the European Parliament”) of the inquiry. By comparison, the EP’s rules of procedure (II, Ch. 2, Art 42-48) specify the proceedings and operation of inquiries only generally. Implementation is left to the committees themselves. The CIA Committee discussed its proceedings and mode of operation at its second meeting on 13 February 2006.Footnote ⁴ It agreed on “general principles” (Working document n°1, 14/02/06). The work programme contains a “provisional calendar of meetings” until the first preliminary report, as well as suggestions for invitations (a list of persons and institutions), delegations and special reports.

The appointment of the inquiry was preceded by various judicial inquiries in member states and by unsuccessful queries to the European Commission and the Council of Europe (E-2203/05 and E-2204/05). After the Washington Post published an “investigative report” on 2 November 2005 and the Council of Europe initiated an investigation, the EP applied for the appointment of a special committee. This committee began its work on 26 January 2006, without much press coverage. Ten months later, the committee published a final report, which focused, next to a political assessment of the results, on the work accomplished and on assessing its cooperation with various governments, international institutions and organisations.Footnote ⁵ The EP agreed on the committee report with slight changes in wording on 14 February 2007 (P6_TA-PROV (2007)0032).

As determined in its first work programme, the committee did not only conduct hearings, but also sent delegations and analysed flight and satellite data. The committee cooperated with further public institutions (e.g. the Council of Europe) and with journalists called as more or less knowledgeable informants. The running press reporting was limited to a few highlights: the constituent meeting, the interim reports, important witnesses’ testimonies, and the final report. Public debate would not spark. As a “public issue” (Luhmann Reference Luhmann and Luhmann1971, p. 18), the inquiry (unlike its object: the CIA flights) remained in its “latent phase” (ibid.).

Press observers, like some committee members, doubted the relevance and probative force of the running investigations. The final report, published on 28 November 2006, on the other hand, was showered with praise: the press wrote about a “new confirmation” (The New York Times, 29/11/06) and “serious circumstantial evidence” (Washington Post, 29/11/06). Some commentators called for “a forceful EU response” (The Times, 29/11/06) or “continued investigations” (Taz, 24/01/07, own translation). According to The Independent (29/11/06), however, the inquiry showed “little new”.

In the following, we turn to the goings-on in the committees. We probe and contrast the inquiries’ unique characteristics with reference to shared problems: How do they process their terms of reference? How do the inquiries produce knowledge? How do they reach decisions? We relate our contrasting empirical answers to the problem of (kind and intensity) of procedurality. In the conclusion, we will at once strengthen and weaken the contrast of strong and weak procedure through the notion of interplay. It can be shown that participants, as well as audiences and sponsors, do at times invoke the respective other procedural register – and thus pit formal requirements against each other.

1. Lists of questions and series of questions: strong and weak modes of questioning

Question: How do committees of inquiry investigate?

Answer: by way of questions. While the EU-committee works off lists of questions, the Hutton Inquiry displays dynamic question/answer exchanges. The following excerpt is from the interrogation of journalist Andrew Gilligan on the second day of the inquiry. Queens Counsel (QC) James Dingemans, counsel for the inquiry, is questioning Gilligan on the notes he produced after his informal interviews with his informant, the weapons inspector Kelly:

1. 1 Q. You wanted to read into that your questions. Was there any question that provoked the first note on BBC/7/57?

   A. Yes. We started by talking about other things and then we got on to the dossier; and I said: What happened to it? When we last met

2. 5 you were saying it was not very exciting. He said: Yes, that is right, until the last week it was just as I told you. It was transformed in the week before publication. I said: To make it sexier? And he said:

   Yes, to make it sexier. […]

   [6 lines omitted]

3. 10 Q. Then there is the entry that is just a single word, “Campbell”. Was there any question that gave rise to that entry?

   A. Yes, it was something like: how did this transformation happen?

   Q. Right.

   A. And then the answer was that, one word.

4. 15 Q. And he just said “Campbell”?

   A. Yes.

   Q. And what question led to the next entry?

   A. Well I was surprised and I said: What, you know, Campbell made it up? They made it up? And he said: No, it was real information but

5. 20 it was unreliable and it was included against our wishes.

   Lord Hutton: May I just ask you, Mr Gilligan, looking at the first paragraph, you put the question: Was it to make it sexier? And Dr Kelly replied: Yes, to make it sexier?

   A. Yes, to make it sexier, yes, so he adopted my words.

6. 25 Lord Hutton: Now are you clear in your recollection that you asked how was it transformed, and that the name Campbell was first spoken by Dr Kelly?

   A. Yes, absolutely.

   Lord Hutton: It was not a question by you: Was Campbell

7. 30 involved in this?

   A. No, it was him. He raised the subject of the 45 minutes and he raised the subject of Campbell.

   (Ht, 12/08/03, 25-27)

Formally, the following course of interaction can be identified: a request (1-2) elicits a short narration (3-8). After a short digression, the QC invites Gilligan to continue his reconstruction of the course of the interview (10-11). The respondent then expands on his previous answer and the interrogator specifies the response (14-15). This is followed by another invitation to continue the narration (17), before the chair intervenes and has Gilligan confirm his answer to the initial question by repeating it (21-23).

Lord Hutton’s “May I just ask you” (21) suggests to the witness that this issue was of relatively minor importance. The respondent takes up his own repeated answer and delivers a “telltale” reformulation (“He adopted my words”, 24). Lord Hutton now has the answer to the second request confirmed (25-27). The next enquiry suggests expanding Gilligan’s “admission” in response to the previous suggestive question to further parts of the interview (29-30).Footnote ⁶ This sets up the conclusion: If Andrew Gilligan suggested a specific formulation to his informant once, why not a second or third time?

The CIA Committee does not display those “exciting” question/answer-dynamics. This kind of dynamic is unlikely because questions are commonly asked in rounds. In such a round of questions, every committee member has the opportunity to ask (a block of) questions in their allotted time. On the chair’s initiative, questions can also be collected.

We can observe the effect of this organisation in the following example. A member is questioning Mrs Mariner, the representative of the organisation Human Rights Watch:

The questioning results in heterogeneous blocks of questions, which the respondent then works off. The collection of questions does not aim to evoke discrepancies by repetition or to dramatise by “zooming in” on details. Questions like “Can you tell us something more about those facilities [offshore prisons]? We are they or were they and what proof, evidence or clues are there as to their existence?” (VR, 21/03/06, 5) seek to elicit further information.

Next to their lists of questions, members also deliver assessments and evaluations on the factual situation:

Confronted with a variety of questions and evaluations, respondents have a wide margin to structure their answers according to their own requirements. In this, they are not required to follow a certain order or even to respond to every question. Omissions, sometimes with reference to time constraints, are the rule. Some questions remain unanswered.

The committee receives answers that connect facts, indications, and assumptions. At the same time, the respondents position themselves politically and with regard to the contents of the question. Thus, journalist Stephen Grey attempts to use, e.g., reservations (“we do not know”, “my understanding is”) and extensions (speculation and generalisation: “it knew”), as well as external evidence, to match his information to the question at hand:

The journalist’s talk encompasses facts, indirect knowledge, suspicions and nescience. This variance is not the only contrast to the Hutton Inquiry. The CIA Committee further allows for respondents to assist in specifying problems and questions. Respondents themselves ask questions, plead for information, or formulate accusations.

This provides a first impression of how “asking questions” is done in the respective inquiries. Accordingly, the EU Committee conducts it hearings as negotiations between relatively co-equal colleagues. Respondents experience relatively large degrees of freedom to shape their own contributions. In the Hutton Inquiry, on the other hand, the logic of interrogation with decisive default answers dominates the questioning. Those different modes of participation relate closely to the question of how the individual hearings are integrated into the overall inquiry and how the former work to integrate the latter by way of prospective and retrospective connections.

2. Statements or issues: on the operation of memory and internal connections

The hearings involve different forms of remembering and connecting. Besides hearing protocols, the EU-committee also produces interim statements and “working documents” containing topical and situational reports. The protocols do not feed back into the hearings. They do not serve to “pin down” respondents. Inconsistencies (e.g., incorrect quotes or contradictory dates) are neither registered nor addressed. The rapporteur complains about the committee members “forgetfulness”: “It seems almost as if each hearing is our first” (VR, 21/03/06,16). However, this way of utilising documents does not only imply forgetfulness, but also a specific way to remember.

In the EU-committee, memory is not focused on statements but rather on issues. It provides issue-oriented connections within a heterogeneous field of indications. Connections are not only made to issues that arise in the committee itself. E.g., a committee member asks Mrs Mariner for ulterior investigative results: “I wonder whether you have reached a different conclusion from that drawn by the investigations carried out in Sweden and whether you are familiar with their contents” (VR, 23/02/06, 9).

These issue-oriented connections do not impose a binding force on respondents; they do not confront the latter with telltale inconsistencies. Instead, the connections serve to establish indications or leads, which are then followed further. We can observe this relation to procedural history in the way that members mark “wrong answers”. Rather than representing contradictions, “wrong answers” are marked generally ex ante:

In this line, the rare instances when witnesses are recalled do not serve to test earlier answers, but rather to collect further information and to provide updates on relevant issue areas. The prosecutor Spataro formulates this in his second appearance:

The British inquiry also establishes connections and continuities across several hearings, in line with one of our criteria for procedurality. However, in the Hutton Inquiry, memory is not structured topically. Here, connections are made between statements rather than between issues. The Hutton Inquiry interlocks speakers and utterances; it proceeds by way of authorised and attributable statements. This is precisely the purpose of the verbatim protocols, produced by stenographers during the hearings. The protocols serve as the authoritative medium to fix utterances as statements “en detail” and to keep those statements available for further hearings (Antaki and Leudar Reference Antaki and Leudar2001).

In the Hutton Inquiry, repeated questioning makes connections to earlier statements possible. Witnesses are confronted with statements from the first round of hearings. The inquiry culminates in cross-examinations, in which witnesses have to relate to “their own statements”. See this extract from the Government lawyer’s cross-examination of Gilligan:

Connections to earlier answers emphasise the Hutton Inquiry’s self-reference. The inquiry itself produces – in public – the facts and contrasts meant to substantiate and legitimise the final judgement. At the same time, the close retrospective connection to statements marks sharp boundaries to external “hearsay” and “speculations”. Only what emerges from the inquiry itself counts as evidence. By contrast, the issue-oriented structuring of the EU Committee aims at exploration. It is open to more or less vague and indirect indications to a complex set of issues. It collects and sifts findings from a conglomerate of inquiries.

3. Culminations and extensions: what “drives” the inquiries

In the questioning of Gilligan, the interrogator attempts to entangle the witness in his own statements by use of “cunning questions”. This style of questioning relies on the verbatim protocols of earlier hearings. These provide the basis to reveal inconsistencies, or, alternatively, to strengthen a favoured version. The following interrogation by the Inquiry QC, Mr Dingemans, deals with the newspaper article in which Gilligan directly accused Alastair Campbell, the Prime Minister’s director of communications and strategy, for the “transformation of the dossier”:

1. 1 Q. Did you have any contact after your Today broadcast with Dr Kelly before you wrote this article?

   A. No, I did not. I tried to speak to him but had not been able to get through, although not specifically to talk about this but to see

2. 5 how the thing had gone down, the broadcast.

   Lord Hutton You say you had not got through. Did you try his home number?

   A. Yes.

   Lord Hutton And what, there was no reply?

3. 10 A. There was an answer machine.

   Lord Hutton Did you also try his mobile?

   A. I cannot remember if I tried his mobile or not, I am afraid. I may have done.

   Mr. Dingemans When did you do this?

4. 15 A. I am not precisely sure when, but not very long after the broadcast.

   [5 lines omitted]

   Q. Did you make any other attempt to contact Dr Kelly before his death?

5. 20 A. No. I mean, in the later stages I very badly wanted to speak to him; but I knew – you see, after the furore blew up I knew that the risk might be that I would compromise him by trying to phone him. In fact, I did try to phone him once from a phone box and again I just got the answerphone and I did not leave a message again.

6. 25 Q. So your evidence is that you did try to contact him?

   A. Yes, I did try to contact him twice. The – as I say, once before the main sort of fuss about my story blew up and then once after.

   [8 lines omitted]

   Q. But you have already mentioned phoning from a public phone

7. 30 box, you tried that once. Did you try it again?

   A. No, I only tried it the once.

   Q. And why, if, as you say, you were keen to talk to him, did you not try again?

   A. Because I was worried that any attempt to contact him might

8. 35 be – you know, might compromise him.

   (12/08/03, 116-118).

Instead of focussing on the content of the article, the questions concern Gilligan’s contacts with his informant. The enquiries are concerned to establish what became of Gilligan’s call to Kelly (9); when exactly he called (14); which numbers he dialled (6,11); and whether he had tried again (32-33). The questions investigate more and more details. Incidentally, they render “vague” answers more precise (e.g. 10/11). The facts of the matter in question become increasingly detailed, and their pertinence is hardly discernible for observers and witnesses alike. The following confrontation finally reveals the full force of the questions. Accordingly, the “inscrutable” questions circle around an inconsistency between Gilligan’s current testimony and his own previous comments to the FAC. The QC quotes extensively from the latter:

1. Q. Can I take you, in that respect, to part of the evidence you gave to the Foreign Affairs Committee? It is FAC/3/35. This was the second time you gave evidence. […] Can I just take you to the middle answer:

2. 40 “Let me just make this absolutely clear. The source did not say either that Mr Campbell did not insert or that he did insert it, I have never claimed otherwise. The claim was that the 45 minute claim was inserted ‘against our wishes’, against the wishes of the source, and that the claim had been transformed in the week before it waspublished at

3. 45 the behest of Alastair Campbell (I imagine you meant the dossier rather than the claim there). That is entirely consistent with everything I have said, it is entirely consistent. There is no difference between what I said to you before and what I have said now. It has been interpreted in the media to say that we reported that

4. 50 Mr Campbell inserted the 45 minute claim, that may be the case but we reported it neither way. We said, or the source said, that the transformation had occurred in the week before it was published at the behest of Alastair Campbell. That is the claim we have always made and that is the claim that the source has always made and that is the

5. 55 claim the source continues to make.” Had you contacted Dr Kelly?

   A. No, and I think that was a mistake saying “continues”. I mean, it was a – the atmosphere at that session was extremely fraught and I – you know, I – I was not – it was not the best performance I have ever given in front of a committee.

6. 60 (12/08/03, 118-120).

“And why, if, as you say, you were keen to talk to him, did you not try again?” (32-33). This culmination establishes one side of the inconsistency; the flip side is provided by an earlier statement introduced from a previous inquiry. This results in a contrast which seems to take Gilligan (as well as the “casual” observer) by surprise: the statement “That is the claim the source continues to make” (54-55) presupposes further contact between the reporter and his informant. However, Gilligan has denied such further contacts until this point. Gilligan finds himself in a quandary: he is forced to correct at least one of his earlier statements.

The Hutton inquiry culminates in questions of credibility and responsibility. The unfolding communication climaxes in knowledge tests that gain decisive character. At the same time, the decisions connect to witnesses’ “free” representations, i.e. those for which the individual witness is responsible. This strength in bringing seemingly irrelevant details to the fore, however, also entails a highly visible liability: the inquiry risks getting lost in details.Footnote ⁷

Involving witnesses by culminating in details is directly opposed to the operation of the EU Committee, which expands and diversifies its issue-oriented investigation. Rather than inducing respondents to repeat their statements and contrasting those statements, the committee sifts new information and attaches new leads to these. The leads are meant to specify “nebulous speculations” and transform them into justified suspicions. By “collect[ing] and analys[ing] information” (terms of reference), the committee investigates “whether” specific legal infractions happened. The collected indications are combined to provide scenarios. The goal is to establish an overall picture which can instruct further, more precisely directed inquiries. The rapporteur, Claudio Fava, provides an exemplary expression of this “drive” in his question to Khaled al-Masri:

These “what else” type of questions indicate the procedural “drive”: what else happened, when, where, etc. The investigation does not bracket hearsay or speculation, but mobilises these. In this line, a “victim”, Mr Kurnaz, is also asked for “second-hand knowledge”:

Indications can ground further research. The corresponding request “It would be very useful if you could help us ask the right questions” (VR, 23/02/06, 11) would indeed never be asked in the Hutton Inquiry. The CIA inquiry performs extensive searches. It does not falsify the submissions made by “guests”. The openness of the inquiry makes it difficult to integrate diverse epistemic standards, to establish stable and sustainable selections, and thus to provide for a systematic conclusion. Rather than concluding by “answering the question” posed by the terms of reference, the inquiry simply concludes within the time frame set by parliament.

4. Actions and conditions: the objects of inquiry

Differences with regard to the integration, self-reference and delineation of the two procedures do not only correspond to different methods of inquiry, but also to different objects. The EU Committee asks whether something possibly happened. To formulate the problem in this way, a general initial suspicion is sufficient. The suspicion is fed by “superficial knowledge” – hearsay, individual fates and fortunes, indications, etc. The reverse of this superficial knowledge is not to be seen in (false) statements that can be attributed, as in the Hutton Inquiry, but rather in nescience with all its side effects, such as mere suspicion and secretiveness. Appearing before the Committee, the European Council’s investigator, Dick Marty, qualifies the object and knowledge of the inquiry as follows:

In order to initiate an inquiry, parliament must first recongise a deficit in information (Kipke Reference Kipke1989; Höpfner Reference Höpfner2004), which cannot be remedied by queries to other public bodies. As a consequence, committee members are often left in the dark. They speculate on an object that is still gaining shape. Thus, Dick Marty responds to a member’s question “From your vantage point, where do you see things which we as a committee should be doing here?” with the comment: “It seems to me that you have already received certain pieces of information today which could be reasons to put a whole series of questions to representatives of particular countries” (VR, 23/02/06, 49).

The “incidents” the inquiry addresses operate as indications for “deficiencies”. Instead of (attributed) actions, the inquiry is concerned with (deplorable) circumstances. The investigations aim to define and illuminate these circumstances and to evaluate them politically. Such an inquiry is rather far removed from determining facts. The CIA Committee aims at general clarification, political and moral judgement, and preparing general recommendations (EP Rules of Procedure, Art. 175 f., App. VIII). The rapporteur expresses these aims: “It is our job to evaluate the data we have gathered […] and to form our political opinion on that basis” (VR, 21/03/06, 16).

The questions in the Hutton Inquiry investigate the “circumstances surrounding the death of Dr David Kelly”. It is to be established who is (generally and to what extent) responsible for his death. The inquiry assumes and reconstructs causality. Actions have an immediate or indirect effect on subsequent events. The inquiry is tasked to attribute occurrences as actions and actions as causes. On this dimension, the Hutton Inquiry can be likened to investigations in a penal court. Testimony is taken from eye witnesses of the events in question. Statements – e.g., “I was slightly after, you know maybe 10 or 15 minutes late. He was waiting when I got there.” (Ht, 12/08/03, 16) – do position a witness relative to the event in question. The witness expresses that there was indeed a meeting on which he can provide inside information. Additionally, beyond witnessing, actual participation is evaluated according to norms such as reasonableness and accuracy (see final report, p. 3). In this way, the inquiry constructs and evaluates the past as a chain of actions and consequences of actions. The following questions (Ht, 12/08/03) establish a chain of actions in Gilligan’s work as a reporter:

These questions do more than inter-link actions; they also serve to evaluate those actions in light of the established links. They establish a connection between Gilligan’s “careless” reporting and the “unnecessary” escalation of the conflict:

The questions assemble actions like chain links. This produces a causal chain, starting with Gilligan’s interview with Kelly and ending in Kelly’s act of desperation. The conclusion attributes a partial liability for Kelly’s death to Gilligan, due to his “misinterpretations”, his “hasty reporting”, and his “failure to issue a correction”. The conclusion suggests: Had Gilligan (among others) acted differently, Dr Kelly would possibly still be alive. This substantive reasoning on personal guilt provides a sharp contrast to the broad strokes used to draw a picture of “alarming” or “deplorable” conditions in the CIA Committee.

5. Informants and eye witnesses: the epistemic subjects

Committees of inquiry investigate antecedent conditions or occurrences by way of interrogation. The selection of witnesses varies, depending on the object of the inquiry. Thus, in order to close a chain of actions, including possible attributions of responsibility, the Hutton Inquiry mobilises eye witnesses, i.e., mainly persons who observed incidents immediately. Expert witnesses, who provide classification and evaluation of these observations, are of secondary relevance. The category of eye witnesses becomes double-edged when statements feed into hardly-straightforward truth tests and moral assessments.

Lord Hutton’s witness list grows out of a hypothetical chronology of the events leading up to Dr Kelly’s death. This chronology provides a basis for the order by which witnesses are called in.Footnote ⁸ It provides clues as to the pertinence of individual witnesses. The eye witnesses selected by Lord Hutton can be arranged in concentric circles in relation to Kelly’s last actions as a weapons expert (not as a private person): the innermost circle includes those who had contact with Kelly shortly before his death (e.g., Gilligan or Kelly’s immediate superiors). The next, wider circle contains persons who are expected to give information on political decisions “in the Kelly affair” (e.g., government ministers or advisers). The outer circles are constituted by witnesses (e.g., his widow or a psychiatrist) who have only indirect knowledge of his professional activities and can give information on his behaviour, the state of his health, or his personal plans and ambitions.

A close “causal” proximity to the weapons expert Kelly also implies a high potential for a witness to be deemed responsible for the investigated “circumstances”. With its logic of causal and moral attribution, the inquiry procedure anticipates that witnesses might attempt to hide or downplay their responsibility. Thus, the investigators’ task is to elicit information on the witnesses’ own involvement. The approach used in the Hutton Inquiry resembles the testing of hypotheses. In the course of the inquiry, the preliminary chronology – including its causal assumptions – is tested and/or modified.

By contrast, the CIA Committee relies on cooperation. It invites “Guests” (cf. work programme, protocols) for “discussion” or an “exchange of views”. The “guests” are often interested in clarification on their own part; they want to contribute and to bring their own interests to bear on the investigations. Thus, the following question to the European Council’s rapporteur, Mr Marty, is symptomatic:

The committee expects “guests” to help in reconstructing a general perspective. They appear as informants, similar to natives in relation to “their culture”. The hearings operate as discussions, exchanging and synchronising both factual knowledge and assessments. In this logic, “victims” are also asked for “second-hand” assessments and information beyond their own immediate experiences. Their contribution to clarification is crucial to the inquiry; personal involvement is rather distracting. The informants are not chosen according to their proximity to some act, but rather according to possible insights in possible occurrences. The “work programme” lists a selection of informants and assigns those to general categories (“representatives of institutions”, “journalists”, “witnesses”, “victims”, etc.). This list is completed in the course of the hearings. The incomplete and vague “guest list” corresponds to a certain lack of previous knowledge.

The chair introduces each guest to the committee. The protocol notes each guest’s name and function. At the beginning of the hearing, the chair assigns general competencies. Thus, e.g., Mr Kurnaz is also questioned as “a witness of what was going on in Guantánamo” (VR, 22/11/06, 5), while the journalist Grey appears as an expert: “He [Mr al-Masri] also said he had passed this information on to the press, so I imagine that you too know about it, as a specialist in the matter. My question is: what do you think of his version of events?” (VR, 21/03/06, 12). The invitation is based on an ascribed inside or expert knowledge, which in turn frames and stimulates the questioning. The relatively relaxed setting in the committee allows guests (here, e.g. the journalist Grey) to position themselves by way of evasive (“I do not want to respond to”), qualifying (“I want to try”), or definitive (“I am in no doubt”) statements:

Sometimes, ascribed knowledge and actual answers diverge. Guests tend to disappoint expectations or to justify their invitation. In the EP’s CIA committee, there is a tacit, general imperative for cooperation. Appeals are made to conscience, integrity, and character. Answers “provoke” moral reprobation if and when they “cover up” or “conceal”. In contrast to the Hutton Inquiry, reprobation does not establish personal but rather political responsibility, which is “attached” to states, institutions, or interest groups.

6. Impositions and rejections

Both inquiries establish particular procedural roles (witnesses and guests) with their respective limited degrees of freedom. These roles impose “presumed participant or member knowledge” (Scheffer Reference Scheffer and Oltmer2002) and a demand for cooperation. Thus, the EU Committee oftentimes does more than ask for information, demanding additional assessments and judgements. The demand to have an opinion can be experienced as imposing, particularly if it is to be announced in public. In this perspective, Grey’s refusal in the previous example appears as marking an imposition: as a journalist, he refuses to position himself “politically” in public. The following statements deal with the impositions that accompany public presentations. Each statement mobilises the respondent’s external role demands – either to devalue the information given by the respondent, or to enhance its value:

Another committee member, obviously a political opponent of the first, reacts with an “upvaluing” question:

A further imposition relates to the fact that the CIA Committee relies on cooperation with other institutions and inquiries. Many “guests” are invited as representatives of public institutions and are thus bound to secrecy and loyalty. External membership obligations (qua public service law or professional ethics) come to conflict with the internal role of “guest”. In this line, the protection of sources forms a “hard” professional-ethical boundary for journalists. In the following example, this boundary stirs up fantasies on behalf of the questioner about what the respondent might have concealed, or might have to conceal:

The prosecutor Spataro delimits another boundary. He resists the double imposition of “merely” having to comment on speculations in his role as prosecutor:

Analogously, invited public servants refer to public service law and official secrets, members of parliament to a parliamentary code of honour, and members of other inquiry committees to ongoing proceedings. They all confront the committee with external boundaries of acceptability. The same holds for the “High Representative for the CFSP” (Common Foreign and Security Policy, of the EU). Mr Solana cannot cooperate:

In competing with other membership roles and loyalties, the committee encounters narrow limits. It cannot presume cooperation – and yet it must, since it lacks other powers.

In the Hutton Inquiry, it is relatively more difficult for witnesses to elude the dynamics of questioning with reference to imposing questions. Nevertheless, questions can be marked as imposing and rejected. Sometimes, this move can succeed when witnesses refer to the procedural “drive”. Thus, the committee’s attention to very minute details strains witnesses’ mnemonic capacities (possibly excessively). Witnesses regularly point out the limits to their memories (Lynch and Bogen Reference Lynch and Bogen1996). Answers such as “I cannot remember the exact words I used” (Ht, 12/08/03, 56) effectively obstruct follow-up questions:

Rejecting a question as “imposing” sometimes requires justification. Thus, Gilligan refers to a lapse of his short-term memory: “I cannot remember exactly what I told her because it was a mobile phone conversation and I did not take notes of that conversation” (Ht, 12/08/03, 56).

Another imposition relates to the procedural aim of assessing responsibility. Witnesses can repudiate ascriptions of responsibility by claiming limits due to their inferior organisational status: “For discussions on the sort of practice of the Today Programme, then you would need to talk to higher level people than me. I am only a reporter” (Ht, 12/08/03, 140).

Furthermore, witnesses can themselves appear as impositions to the British inquiry when they provide speculations rather than first-hand knowledge. They elude the imposing operation of the procedure by changing to another register of knowledge. Thus, Gilligan scantily fills in gaps in his memory with general considerations. His speculations culminate in counter-accusations. Here, Gilligan discusses when the government might have disclosed Dr Kelly’s identity:

Gilligan bypasses the preference for eye witness accounts with expert assumptions. The call to order “You have no knowledge of that” relates to the Hutton Inquiry’s requirement that statements must be founded in immediate cognisance, i.e. in participant knowledge. In contrast, Gilligan here “merely” delivers assumptions instead of an eye witness account.