Congreso Discurso Oral, Almería 24-26 de noviembre, de 2005



Contextualization in Parliamentary Discourse

Aznar, Iraq and the Pragmatics of Lying


Teun A. van Dijk

Universitat Pompeu Fabra, Barcelona


First draft. November 25, 2005




In this paper it is shown that an adequate theory of discourse also needs a contextual component. Contexts however should not be conceptualized as social situations, but rather as subjective participant definitions of the relevant properties of social situations, that is, as mental models. These context models (or contexts) control the production and reception of discourse. One of the central categories of the schema that organizes such context models is the Knowledge-Device (or K-device), which manages the presupposition and expression of knowledge, defined as shared accepted beliefs. In an analysis of a parliamentary debate held on March 18, 2003, in the Spanish Cortes about military intervention in Iraq, pitching Prime Minister Aznar against opposition leader Zapatero, it is shown how context models in general, and the K-device in particular, control the debate. More specifically it is shown how the accusation of lying, leveled against Aznar (as well as against Bush and Blair in other contexts), is based on criteria for shared knowledge. Thus, whereas conventional approaches provide a semantic approach based on the notion of producing false beliefs, this paper at the same time provides a pragmatic account of lying.


Keywords: context, context model, mental model, parliamentary discourse, Iraq, Aznar, Zapatero, Spanish Parliament (Cortes), knowledge, lying, manipulation.


The relevance of ‘context’


One of the new developments in discourse studies is the growing interest in the analysis of context. This development should be seen against the background of the increasingly multidisciplinary approach to the analysis of text and talk in most of the humanities and social sciences. It is no longer adequate to just examine the ‘linguistic’ structures of discourse ‘itself’, and not even to limit oneself of the autonomous interaction structures of conversation, but to look beyond discourse and examine its cognitive, social, political, cultural and historical environments:


·        In linguistics, and especially in Systemic Linguistics, several proposals have been made to analyze context (Ghadessy, 1999; Leckie-Tarry, 1995; see also Fetzer, 2004; see also the critical analysis of this approach in Van Dijk, 2004).

·        Conversation analysis (CA) itself has extended its scope since the early 1990s to the study of the institutional and organizational context of talk in interaction (Boden & Zimmerman, 1991; Drew & Heritage, 1992; Sarangi & Roberts, 1999).

·        Anthropological linguistics obviously has a disciplinary interest in the study of the cultural aspects of talk, and hence in the situated aspects of face to face interaction (Duranti, 1997, 2001; see also Duranti, & Goodwin, 1992).

·        Interactional sociolinguistics, especially with the seminal work of John Gumperz (see, e.g., Gumperz, 1982a, 1982b, 1992), has paid special attention to the subtle ‘contextualization cues’ of discourse (see also, Auer & Di Luzio, 1992).

·        Critical Discourse Studies (CDS) by definition studies talk and text within their societal contexts, with specific attention for relationships of power, domination, and social inequality (Fairclough, 1995; Wodak, & Meyer, 2001; Wodak & Chilton, 2005).

·        The same is true, more specifically, for feminist discourse studies, interested in the critical study of gender domination and its reproduction in discourse (Lazar, 2005).

·        Even in more formal approaches in discourse studies, for instance in Artificial Intelligence and related fields, it is being recognized that automatic language production and comprehension is impossible without modeling the context of communication (see, e.g., Akman, et al. 2001).

·        Earlier work on situations and episodes in social psychology might be taken as an example for more detailed studies of context (Argyle, Furnham & Graham, 1981; Forgas, 1979; 1985; Scherer & Giles, 1979).


In sum, at least since the 1990s, we witness a diverse but steady attention being paid to the relevance of context analysis in various areas of discourse studies and the humanities and social sciences in general. However, although there seems to be a widespread consensus about the relevance of such an extension of discourse studies, there is hardly any agreement about what the notion of context actually means, or should mean. Generally speaking, and quite vaguely, context is seen as the explanatory ‘environment’ of discourse, but the problem is how to define, that is to de-limit, such an environment, without running the risk of engaging in a theory-of-everything. Thus, in a parliamentary debate, the context may be limited to the setting of the Spanish Cortes, the diputados present, and some other obvious candidates for contextual categories, but what about such ‘contexts’ as the current situation in the country, or the current international situation, or Spanish foreign policy, etc.? Most of such environments may well be relevant to study such discourse, but the question is also whether they were actually relevant for the speakers in parliament themselves. Hence, usually contexts are limited to the relevant aspects of the environment. This, however, begs the question because in that case we need to define what is relevant and what not, and we may end up with a circular definition.

More specifically, for instance in Conversation Analysis, has been the tendency to be very parsimonious with admitting (relevant) ‘context’ and only when somehow aspects of such a context are  “procedurally consequential” for talk (Schegloff, 1987, 1991, 1992, 2003). This seems to restrict contextual influence on talk quite seriously, but even then we do not have a solution, because the influence of the context may be on the assignment of meanings or functions in discourse that are obvious to the participants, but not explicit expressed. That is, such a definition is empiricist and even behaviorist, because it only admits “observable” evidence, and not other evidence or methods of inquiry (experiments, post hoc explanations, introspection, participant protocols, etc.). Still, it provides at least one (superficial) criterion for the elusive notion of ‘relevance’.

In Critical Discourse Studies (CDS) the admission of ‘contextual’ evidence is obviously more liberal, and any property of discourse, at any level, that seems to require a contextual description or explanation may thus be accounted for, for instance in terms of dominance relations between speakers and recipients, or in terms of organizational roles or positions of speakers or authors. But such an account is wide-open to the critique that there is no limit to the analysis of possibly relevant context is terms of social structures (e.g. of domination) that impinge on discourse.


Towards a new theory of context


It is against this general background in discourse studies in various disciplines, and in view of developing a general, multidisciplinary theory of context that integrates these various directions of research, that I have been proposing, also since the 1990s, a new theory of context in terms of a specific kind of mental models of experience: context models (Van Dijk, 1999, 2001, 2004, 2005; forthcoming). This new theory was also premised on the obvious limitations of a more formal approach to context in my earlier book on text and context (Van Dijk, 1977).

The fundamental idea behind this proposal is that contexts are not some kind of objective ‘social situation’ – which is the dominant, informal view of context in linguistics, discourse studies and other disciplines. Rather, for contexts to be relevant for text and talk, as required, they should be characterized as being subjective, namely as definitions of the relevant aspects of the communicative situation by the participants themselves.

This idea, as such, is not new and has been formulated, without much detail, especially in various studies in social psychology and the social sciences (Brown & Fraser, 1979; Duranti & Goodwin, 1992). The notion of ‘the definition of the situation’ is famous in the history of sociology (Thomas, 1966). However, the more subjectivist definitions hardly have been taken seriously, also because they seem to be inconsistent with the social perspective of these various approaches to language use, discourse and communication. Indeed, a subjectivist approach is typically rejected as individualist, and hence incompatible both with microsociological (‘interactional’) approaches, as well as with macrosociological (‘system’) approaches.

Moreover, the fact that the very concept of ‘defining the situation’ by participants also has obvious cognitive aspects, makes such a subjectivist approach even more suspicious in most of the socially oriented approaches to discourse (see, Van Dijk, 2006; special issue of DISCOURSE STUDIES). And yet, it is precisely this cognitive aspect that provides the necessary theoretical and empirical bridge or interface between social situations and social structure, on the one hand, and the necessarily personal nature of each individual discourse, on the other hand. That is, contexts can only be shown to be relevant, if they are relevant-for-the-participant. This can only be shown when it is being shown for individual participants, without abstractions and generalizations towards higher level interactions or groups or social structures.

Such context definitions are relevant for participants in interaction, and hence they are construed under the influence of other participants and the interaction itself, but they still remain definitions of individual participants. Indeed, the participants of an interaction, as well as writers and readers in written communication, may not have the same definition of the communicative situation in the first place, as is typically the case, and actually shown, in communicative conflicts.

In sum, although some of the ideas behind my concept of context as a subjective definition of the communicative situation can be found in the literature, no systematic theory has been formulated about the precise nature and influence of such definitions.



Contexts as mental models


In the broader perspective of contemporary cognitive science, and more specifically within the framework of cognitive model theory (Johnson-Laird, 1983; Van Dijk & Kintsch, 1983), I thus have proposed to formulate the notion of context in terms of a specific kind of memory representations of participants, namely as mental models (Van Dijk, 1999). Like all our personal experiences, these mental models are stored in the personal (autobiographical) memory, episodic memory, of language users (Neisser & Fivush, 1994; Tulving, 1983). Thus, ‘experiencing’, ‘being aware of’ a communicative event in which one participates is what we call a context model, or simply a context. This notion of mental model includes exactly the kind of properties we usually ascribe to contexts: they are subjective, they are definitions of the situation, and they may be incomplete, biased, prejudiced, etc. and hence nothing like an ‘objective situation’. That is, they are participant constructs. In this sense, my proposal is consistent with a constructionist approach to discourse in social psychology (Edwards & Potter 1992), but unlike such an approach the only explicit way such constructs can be formulated is in terms of cognitive representations of some kind, and without reduction to other kinds of objects, such as discourse or interaction.

Thus, contexts are subjective episodic models (experiences) of participants dynamically construed (and updated) during interaction. As we know from more informal context studies and experiences, such context models have a profound influence on discourse, and vice versa, also discourse has a profound influence on the definitions of the situation by the participants. That is, there is ongoing, dynamic mutual influence between talk or text and  its production or comprehension on the one hand, and the way the participants see, interpret and construe the other, ‘environmental’ aspects of such discourse, such as the setting, the participants, the ongoing action, as well as the goals and knowledge of the participants.

In terms of the psychology of discourse production this means that context models control this production process, and thus guarantee that what it said and especially also how it is said, are adequate or appropriate in the current situation. This would mean control over much of the (variable) sound structure, the syntax, lexical choice and any other discourse structures that may vary in the situation.

The precise cognitive process model of this kind of control is beyond the scope of this article (see Van Dijk, 1999). Suffice it to say that the kind of theoretical approach is perfectly compatible with contemporary studies of discourse processing (Van Dijk & Kintsch, 1983; Van Oostendorp & Goldman, 1999), and in fact supplies an important missing link in this approach, namely the fundamental context dependency of discourse processing. Such context dependency was usually ignored or dealt with in an ad hoc way in cognitive psychology, for instance in terms of independent variable control in laboratory experiments (age, gender, et. of subjects), which obviously are only a very imperfect simulation of ‘real’ communicative contexts.

One of the many attractive aspects of the theory proposed is that the ‘pragmatic’ kind of context model here proposed, combines very well with the kind of ‘semantic’ (situation) models previously proposed as the basis for the meaning and reference production and comprehension of discourse (Johnson-Laird, 1983; Van Dijk & Kintsch, 1983; Van Oostendorp & Goldman,. 1999). That is, both pragmatic and semantic ‘understanding’ of discourse are based on mental models in episodic memory, that is, on the ways participants subjectively understand the situation they talk or read about, on the one hand, and the situation in which they are now communicating – two kinds of representations that are obviously related, and that also define the boundaries between semantic and pragmatics.

By definition context models have the same structure as other kinds or models, such as the (semantic) models participants construct of the events they write/speak or read/hear about, or refer to.

Also, in order to be able to function adequately in fractions of seconds, that is, realistically and online, context models obviously cannot have dozens, let alone hundreds of possible categories representing communicative situations. Rather, the number of fundamental categories, as is usual in such case, is probably around seven (Miller, 1956), although this might be multiplied by seven if each main category has seven subcategories. For instance, if we have a main category Participants, subcategories may be: communicative roles (Speaker, Recipient, Overhearer, etc.), Social identities (gender, ethnicity, class, etc), social roles (father, friend, etc.), Relations between participants (competitors, etc.), etc.

Of course, empirical research needs to establish which are the categories that are relevant in each culture, assuming that such cultural variation is to be expected – even when some categories may be universal (there is always a Speaker role, and always speakers assume that Recipients share knowledge with them, etc.). In this way, the same general frame of context models may be strategically – and hence very rapidly – construed and dynamically adapted during the understanding of the social situation of a discourse.




Micro and macro contexts


Most studies of context focus on the immediate, face-to-face situation of interaction, that is, on the micro-context. Yet, there are reasons to assume that language users also construe some kind of ‘macro’ context (Van Dijk, 2006). Thus, when teaching a class, I am not only aware of the students present, of myself as a teacher, and some other typical context characteristics of teaching.

In line with the usual (but problematical) distinction in sociology between micro and macro accounts of society (Knorr-Cetina & Cicourel, 1981), I may also represent myself as a member of a profession and as a member of an institution, the University, and hence as performing locally something that also can be represented globally: The university educating students, that is, teaching as a macro notion.

Such macro contexts need not be a permanent awareness during my actual (micro) teaching at the level of interaction, but the overall functionality or ‘sense’ of what I am now doing sometimes need to be construed or activated, and sometimes also will be made explicit. Indeed, we thus also have an elegant way to link micro and macro levels of society, not as levels or categories of the analyst, but as participant categories.

As is the case for semantic and pragmatic micro and macrostructures (Van Dijk, 1980), also macro contexts have the same structure as micro contexts: Settings, Participants, Actions, Aims, etc., but only at another level, e.g., that of cities or countries, groups or organizations, collective and repetitive actions, overall goals, and so, as when it is represented that the university in Spain teaches the students such and such, or when the Spanish Cortes voted for military action in Iraq.

Note though that because of the usual processing constraints also macro contexts cannot consist of a vast amount of categories and presupposed knowledge: it will only be a partial construction of now relevant aspects of social structure, and much of the time such knowledge will remain backgrounded in some kind of long term ‘working’ memory.




The usual properties of social situations construed to be relevant in both traditional concepts of contexts, as well as in the sociocognitive approach to context defined as context models, are especially Setting (Time, Place), Participants and their various Identities, Roles or Relations, the ongoing Action, and the Goals of the participants.

Usually forgotten as a crucial category of contexts, however, is the knowledge of the participants, and especially the mutual knowledge about each others’ knowledge (Van Dijk, 2005). Yet, obviously such an epistemic component is necessary in order to describe and explain how speakers or writers are able to manage the very complex task of adapting their talk and text to the (assumed) knowledge of the recipients. In each sentence, and for each word, they need to know not only that the recipients know the words used by them, but also they need to know what the recipients already know about the events spoken or written about. Thus, if it is assumed that the recipients already know some ‘fact’ (proposition, etc.), then such a fact must be presupposed if it is assumed that it is readily accessible in memory (inferrable), or it must be reminded if such a fact, usually a concrete event, might have been forgotten even when it was probably known earlier.

Language users can only do this in the few (fractions of) seconds they have in the production of each word or sentence, when their context models keeps tracks of the kind of knowledge which kind of recipients have. This does not mean that speakers need to have hypotheses about the hundreds of thousands of facts recipients know – something which would be totally impossible within the very limited size of context models.

Rather, speakers use handy strategies (Van Dijk & Kintsch, 1983) to calculate what knowledge recipients have. For instance, for communication among members of the same social group or community, the easiest strategy is simply to assume that other members have the same general sociocultural knowledge as I have. For personal communication among friends, family members or acquaintances, such shared knowledge is not the sociocultural shared knowledge of groups or communities, but the kind of knowledge that has been shared in previous interactions, and that hence is stored in ‘old’ context models. In this case, speakers only need to activate old context models and check whether some proposition has been communicated before. If in doubt, they will make sure to repeat or remind the ‘old’ information. More difficulties exist for the management of knowledge among people from different cultures, that is, from different epistemic communities, although here too much ‘universal human’ knowledge may be presupposed.

The K-device of the context models consists of these kinds of strategy for the activation, expression, presupposition, reminding or repeating of shared or new knowledge. At each moment in the context model, it strategically ‘calculates’ what of the information of the speaker, as stored in mental models about public or private events or in sociocultural knowledge, is probably shared by the recipient or not, and hence what may or must be presupposed, reminded, repeated or newly asserted.

My approach to knowledge is rather pragmatic and discursive than semantic and logical. That is, unlike in epistemology, I do not define knowledge as “justified true beliefs”, which is an abstract, semantic definition in terms of the relations between beliefs and the ‘world’ (see, among a vast number of books, e.g., the reader of Bernecker & Dretske, 2000). Truth in my view is not a property of beliefs, but a property of discourse about the world. The crucial condition of human knowledge is not whether or not beliefs abstractly correspond with (refer to, etc.) some state of affairs in some world or situation, but whether such a belief is shared by other members of a community, the knowledge-community. That is, we only say that someone knows something when we as observers or speakers know the same thing, that is, share a belief with that person. In other words, saying that X knows p, presupposes that the speaker also knows p. This is true both for (inter) personal knowledge as well as for socioculturally shared knowledge. In the latter case, it is presupposed that all and any member of a community share some belief.

Moreover, and contrary to the shared beliefs we call opinions, or indeed (mere) ‘beliefs,’ knowledge is shared belief that has been or can be certified by the knowledge criteria of some community. These may be the common-sense criteria of everyday life (observation, inference or reliable sources), but also those of specialized communities, such as those of the mass media, science or social movements. Obviously, since these criteria may change historically, and vary culturally, knowledge is by definition relative, as it should be. But note that also this relativity is relative – as it should be – because within an epistemic community beliefs that have been accepted or certified according to the criteria of that community, are hold considered to be knowledge, even when from outside the community (or in a later stage of the community) we may deem such beliefs to be mere beliefs, opinions or superstitions. Thus, we may say that in the Middle Ages people from their own perspective knew that the earth was flat, and hence presupposed it in public discourse. It also follows that all that we claim to know today, is our knowledge, and may well be defined differently elsewhere or in the future. Such relativism is nothing to worry about, because for all practical purposes we act and speak in terms of, or on the basis of this (relative) knowledge, and it is this knowledge that is socioculturally shared with others.




Lying is a verbal act that involves the illegitimate manipulation of knowledge in interaction and communication. It is not a speech act in the classical sense, because it does not satisfy the usual appropriateness conditions: There are no systematic conditions that must be fulfilled in order to lie ‘appropriately.’

 Rather, lying should be defined as a violation of the specific pragmatic conditions of appropriate affirmation, and at the same time, more generally, as a violation of the general ethical norms of truthfulness that are at the basis of all human interaction. Note though that if in some situation speaking the truth violates other norms or values (e.g., those of politeness, discretion, face-keeping, legitimate secrecy, and so on) lying may well be ethically permitted. For instance, lying to the enemy in an interrogation may be considered ethically correct. Obviously, each of these situations needs its own definition, and hence its own context model.

In this sense, lying is generally illegitimate if it hurts the interests of the recipient or others. In a broader sense of social and political legitimacy, this is especially the case if the liar (as a person or institution) is thus able to establish or confirm its domination, and thus abuses of its power, for instance through its control over the means of communication and public discourse, and hence over the public access to knowledge (see, e.g., Barnes, 1994; Lewis & Saarni, 1993; Wortham & Locher, 1999).

We see that lying is a complex phenomenon that may be dealt with in philosophical (ethical), semantic, pragmatic, psychological, social, political, and cultural terms. In the limited framework of this paper, however, we shall merely deal with it in terms of context-based pragmatics, although I shall later also include more fundamental criteria of legitimacy. In those terms, lying is a communicative act controlled by a context model in which the speaker knows that p is not the case, but has the purpose to make the recipient believe that p is the case. In other words, if a lie is successful, speaker and hearer have a different mental model of the events referred to.

If appropriate assertions presuppose that the speakers know what they assert, then lies are not appropriate assertions. However, recipients may not know that this condition is not satisfied, and interpret a lie as an appropriate assertion, and change their knowledge accordingly. We see that in order to describe what lies are we need context models that are able to represent speakers and recipients and their knowledge and goals.

Lies are members of a larger family of communicative acts that violate the conditions of previous knowledge, such as deception, self-deception, misleading, errors, mistakes, confabulations, fiction, and so on, that shall not further be analyzed here.

Thus, the crucial (ethical) difference is that when lying speakers know that what they assert is false, whereas in other communicative acts (errors, self-deception) they may not know so, or not be aware of it.

There are various intermediate cases, for instance when the speaker believes what is asserted but is not sure, which also suggest that the knowledge component of the context model should be scalar, making lies more or less ‘blatant.’ Similarly, what may be a genuine assertion from the point of view of the speaker may be a lie for the recipient when the recipient believes that what the speaker asserts is false, and knows it.

Note that all these conditions can easily be formulated in terms of context models, which for each participant spell out not only her or his knowledge, but also the beliefs about the knowledge, intentions and goals of the other participants.


Parliamentary discourse


As is the case for all discourse genres, also parliamentary discourse (and its lies) are largely defined by its context properties (Van Dijk, 2000, 2004; see also the other studies in Bayley, 2004; Steiner, 2004).

Although there are several kinds of linguistic or discursive structures that characterize parliamentary discourse, such speeches delivered in a parliamentary debate, these are seldom unique. Rather, parliamentary debates share with other formal genres a number of characteristics of style and interaction, such as speaker and turn control by a Chair, time allocation for turns, a formal lexicon, elaborate syntax, and the usual structures of argumentation and persuasion characteristic of debates. Only some of these, such as forms of address (“my honorable friend”), may be quite typical and unique, for instance in the UK House of Commons, or “Sus señorías” in the Spanish Cortes. Similarly, contents may be constrained in terms of topics that have to do with publish affairs, the country, and so on – but also these, for instance “immigration” may also be topics in news reports, editorials or everyday conversation. And formal grammatical style will be similar to the grammatical style of any formal meeting, e.g., the board meeting of a company. Note though that although, in isolation, none of the these categories are exclusive for parliamentary discourse, their combination may well be quite (proto)typical of parliamentary discourse.

What is exclusive of parliamentary debates as a genre, however, are such obvious context categories as the Setting (House of Parliament), the Participants (MPs, opposition, etc.), the Aims (policies, etc.), and the political knowledge and ideologies of the participants. In other words, although content and even style of what is said in parliament may be shared by other types of communicative events, the function of such structures should be established in relation to the specific political situation: The speakers-MPs are ‘doing’ politics, legislate, represent the voters, govern the country, and so on. In the rest of this paper I shall examine such categories in some more detail, and apply the results in a study of a specific parliamentary debate.

Thus, the question is what speakers in parliamentary debates need to know about the specific communicative situation in order to be able to speak appropriately. And secondly, how does such (presupposed) knowledge influence their talk. The latter question is important not only as an empirical check on what otherwise might be pure theoretical speculation, but also as part of a discovery procedure, because we have few other methods to have access to the knowledge of participants. Note though that the general criterion of relevance or of procedural consequentiality for the recognition of context elements should not be limited to directly “observable” features of talk, lest we adhere to a behaviorist conception of talk and text. Knowledge is generally defined as shared factual belief of a community, and by definition presupposed. This means that also in interaction, such knowledge is seldom formulated, expressed or manifest. Thus, in a parliamentary debate, all participants know they are MPs, and hence seldom actually need to say so, unless in very specific situations, in which such identity needs to be affirmed, displayed and made explicit. Similarly, the aim of parliamentary debates may be to persuade the recipients, but again, such aims will seldom be made explicit. Thus, the participants presuppose many if not all of the properties of the communicative situation, as subjectively represented in their context models, but the analyst only can infer these from indirect expressions or manifestations, for instance to make properties of talk accountable.


Contextual analysis of a fragment of a parliamentary debate


In the rest of this paper, I shall examine some context categories through a partial analysis of a fragment of a debate in the Spanish Cortes on March 12, 2003, on the war in Iraq. In this debate Prime Minister José María Aznar defends his decision to support U.S. intervention in Iraq, against the opposition led by PSOE leader José Luís Rodríguez Zapatero, currently Prime Minister. From this debate I have selected a fragment in which Aznar and Zapatero directly confront each other (see the full text of this fragment in the Appendix).

Note that this analysis is not like any other analysis of discourse or talk in interaction, which would require a vast account of all the properties of this fragment of parliamentary interaction. Rather, we focus on those elements of this talk that are understandable only in light of the putative context models of the two participants. We may call this a contextual analysis. But since also context models are complex, also these can only be analyzed partially. I shall therefore focus on the point highlighted above: the management of knowledge in such debate, and especially what some participants, outside observers or analysts call lying (see also Van Dijk, 2003).

It should also be stressed that we only have the written transcript as made available in the official Diario of the Spanish Cortes, and not the usual, much more detailed, transcript of spoken interaction.


Forms of address and presentation


Before we deal with the crucial aspect of knowledge, we should briefly deal with other aspects of context models and their control of discourse. Thus the forms address used in this debate and in this fragment, presuppose that the participants know whom they are talking to. Thus, when the President of the Cortes invites Zapatero, she addresses him by his full first and last name, as well as by his current function of ‘diputado’ (MP) and honorary title (don). Note that the address is not direct, in the form of a second person pronoun, but indirect, in a third person pronoun, which makes the invitation or permission to speak rather a form of presentation of the next speaker. But since she has the official right to distribute turns, this presentation of the next speaker is at the same time directed at Zapatero, namely as an invitation to take the floor. The Diario itself describes the position of the President of the Cortes as such (Presidenta), but such is not part of the debate itself, but rather a property of its transcript, as is the case for the identification-descriptions of all speakers.

Finally, as part of her presentation of the next speaker, the President also mentions the type of institutional speech act (a question), as well as its number. In many other forms of interaction, this would be anomalous, because speakers hardly can know what the next speaker is going to say or do. So, this presupposes an institutional condition, in which the action of the next speaker has already been announced earlier, or that interventions or at least the beginning of interventions, by rule need to be in the form of questions, for instance, as in this case, after the intervention of the Prime Minister, Aznar. All this, obviously, already presupposes quite some contextual knowledge, namely about the sequence of speakers, their functions (MP, President), the nature of their speech acts, the conditions of such speech acts, the formal relations between the speakers, as well as the formality of the event, requiring honorary titles such as ‘don’. Since the function of the President is only described by the Diario, but not expressed or otherwise signaled in the debate until the next move by Zapatero, this function is implicit by the very action of presenting and inviting the next speaker. The consequentiality of this context feature (Function of Participant) in this case is thus not formulated but only enacted. That Zapatero may speak next, similarly presupposes that he is an MP, but in this case such a function is explicitly described by the President.

Similar forms of interaction may be found in the next turns, when Aznar is not named, but only described as ‘Señor presidente del Gobierno’. On later presentations in this fragment, Zapatero is just presented by his name, and without the description of his function (‘diputado’) or his honorary title (‘don’).

When Zapatero takes the floor, he not only addresses the President of the Cortes in polite terms, that is, in terms of her function, and acknowledging being invited to speak next with a routine speech act of thanking (‘Muchas gracias’), but in the same turn addresses his ‘real’ addressee in this debate, namely Aznar, but only with his last name, and the minimally polite form of address ‘señor’, without any further honorifics, or even his function.


Political roles and relations


We see that the simple interaction in the first few turns in this debate is premised on shared general knowledge about parliamentary interaction rules, on the one hand, and on the specific context models of the three participants. That is, each is able to play its specific role in this exchange, the President of the Cortes presenting next speakers, allocating turns, and Aznar and Zapatero appropriately addressing her and each other.

In addition to this rather general routine, which is very similar in other formal encounters, the context models need to specify the specific political roles and relations of the participants, so that Aznar can speak first, and that Zapatero, as leader of the opposition speaks first after him. This is crucial to interpret Zapatero’s question in line 5-7, because this is not merely a question, but also intended and interpreted as a form of political opposition. This may also be seen by the reaction of Aznar, who not merely replies with “Yes”, but reformulates and hence implicitly disagrees with the proposition presupposed by the question: Instead of defining it as a ‘military intervention,’ he defines it ‘as disarming the regime of Saddam Hussein,’ that is, in more positive terms. Politically speaking, this implies that he rejects the implicit critique of the opposition by defending government policy as a positive action. That is, in order to be able to engage in political interaction, the participants need to represent themselves and each other in terms of their institutional roles and relationships, and each contribution to the debate needs to be interpreted in this framework. A question that literally (in another context) may be interpreted as a request for information or for an opinion, is thus intended, heard and reacted to as a form of political opposition in parliament. In the same framework it is then expected that the leader of the opposition won’t be satisfied by that simple reply of the Prime Minister, and in a next turn will challenge that reply, as indeed happens as from line 20. Such political inferences are only possible when the participants have context models in which such political inferences can be made (Van Dijk, 2005).


Knowing and lying


In Zapatero’s next turn (line 20-39), in which he details his question and opposition directed at Aznar, he accuses Aznar of hiding his real position and decision regarding the intervention in Iraq and of being each time less credible. Again, in general interactional terms an affirmation directed at a recipient implying that the recipient does not respond (to questions) and does not tell what he really wants, may be heard as an accusation, especially if general norms or rules require people to be honest and tell the truth.

This would be the analysis of a similar exchange in a mundane conversation (but formulated in less formal style, of course). However, in this specific political situation, such an indirect accusation, formulated by the head of the Opposition and directed against the head of the government, is a fundamental form of opposition. If the Prime Minister is accused of lying, this implies he has lied to the nation, and especially in case of military intervention, this may be a breach of crucial norms of a democratic state. In other words, by indirectly accusing Aznar of lying, Zapatero at the same time accuses him of political misconduct, of misleading the nation and the people, and as a Prime Minister who cannot be trusted. These political implicatures are not derivable from the meaning of what Zapatero says, nor from the meaning of the interaction (an accusation) but from the contents of the context models of the participants, who interpret this exchange as an interaction between Prime Minister and Head of the Opposition (Van Dijk, 2005).

Note also that the style of this intervention indexes the formality of the encounter in general, and the formality of parliamentary interaction in particular. Apart from formal lexical items (“auténticas posiciones”, etc.), thus, Zapatero has recourse to a polite euphemism. Instead of telling Aznar in so many words that he is a liar, he says that Aznar is “cada día menos creible”, a euphemism that has the usual combined function of saving face and expressing politeness and respect, as is due to the Prime Minister. Again, such a way of speaking presupposes that Zapatero’s context model at that model represents such conditions (the definition of the situation, the parliamentary session, his own role as leader of the Opposition, the recipient’s role of Prime Minister, etc.).

In the rest of his intervention, Zapatero continues his accusation against Aznar and his government, as may be expected from the head of the Opposition. The overall strategy in this accusation is to show the inconsistency between the findings (or rather the lack of findings) of the experts (Blix, Baradei) of the United Nations, and the decisions of Bush and Aznar not to allow more time for the inspections, and wanting to intervene militarily right away. At the same time Zapatero emphasizes that what is being given as reasons for the interaction (weapons of mass destruction, links with terrorism) are actually false. In line 32-33 the accusation is formulated in terms of what Aznar’s government (and specifically the Minister of Foreign Affairs) has done, namely sign a resolution that does not give more time to the Inspectors. That is, the phrases “usted ha firmado… usted acaba a travès de su gobierno” show that Zapatero defines and addresses Aznar not as any person or politician, but as head of the government. We see how the (semantic) model of the situation talked about (what happens at the U.N. and Iraq) is combined with the context model representing the respective definitions of the participants of the current situation. Aznar and Zapatero share their definitions of the current (parliamentary) Setting, and the roles of the participants, but as soon as Zapatero accuses Aznar, the definition of the situation of course is no longer identical: Zapatero accuses Aznar as a liar and as deceiving the nation, and obviously Aznar has another definition of himself.

Zapatero’s rhetorical questions in lines 35ff imply that according to him Aznar cannot know more than the experts who have done local research in Iraq – an expertise that is rhetorically enhanced by emphasizing the time of their work (years of investigation). Since Aznar cannot possibly know more than the experts, the tacit conclusion is that either he must have secret information, or that he, Bush and Blair are misleading the world about the intervention in Iraq by inventing a pretext. Opposition elsewhere, for instance in the USA and the UK, engaged in the same strategy, namely of focusing on what later turned out to be the lies of their leaders about the situation in Iraq (Allman, 2004, Boyle, 2004; Fossà & Berenghi, 2003; O’Shaughnessy, 2004; Stothard, 2003).

Contextually relevant here is not just the accusation (and its formulation), but also the management of knowledge.  We here reach the heart of the matter and this debate, both in Spain as well as elsewhere, pitching one camp (Bush, Blair, Aznar and their advisors) against another doubting or challenging the official reasons of the war: the alleged Weapons of Mass Destruction and the link between the Iraqi regime and the terrorists.

In this specific debate this means that Zapatero voices the doubts about the opposition about the veracity of the official grounds for the intervention. He thus not only indirectly accuses Aznar of lying, but also, politically, of misleading the nation, and hence as a ‘bad’ leader of a ‘bad’ government, as is his task as the leader of the Opposition. That is, for all those participants in this session, this is a normal and legitimate act, according to their context models.

Note that as we have seen above much of the interaction in this debate is related to this issue of (mutual) knowledge and lying. Thus, the negation of the concepts expressed by the lexical items ‘auténticas’ and ‘creibles’ already provides an expression of the context model of Zapatero about Aznar. That is, he does not merely say that what is false, but that the speaker, Aznar, is becoming less credible – i.e., is probably lying, or at least mistaken. Secondly, this opinion is argumentatively supported by the next propositions in his intervention (lines 23 ff), which features the usual authority move, citing the U. N. experts Blix and Baradei, rhetorically emphasizing their expertise (‘informe exhaustivo’,’ maxima autoridad’, ‘afirmaciones muy claras,’ etc.).

Similarly, the rhetoric question (line 35 ff) – ‘Por qué sabe Usted más que el señor Blix’? – also focuses on and challenges the knowledge of Aznar, presupposing that Aznar cannot know more than an expert who was in Iraq, and hence implying again that either Aznar is lying or having secret knowledge he does not want to divulge. In either case, Aznar is mis-leading the nation, and hence a bad Prime Minister. Obviously, such arguments, in a public debate in parliament at the same time presuppose that Zapatero knows that all or most MPs know that Aznar cannot know more than the experts, and hence must be lying or have been lied to himself. Thus, pragmatically, Zapatero accuses Aznar of lying (or of being manipulated), which is already serious, but much more important is the political implicature of the accusation, namely that Aznar is either incompetent, or being manipulated by Bush & Blair, or misleading the nation. Such inferences can be shared in parliament or in the public sphere only when the context models of the participants construe Aznar as PM, Zapatero as leader of the Opposition, and Aznar as head of a conservative government colluding with (also conservative) President Bush, and so on. If we want to understand what goes on in this parliamentary debate, we at least need to simulate these context models of the participants: It is their definition of the communicative situation that controls what they say and how they say it.

We also see that with such definitions of the communicative event, Zapatero can legitimately accuse Aznar of lying, although he does not indirectly. Also in the rest of his intervention he challenges the veracity of the claims of Aznar, Blair and Bush, for instance by referring in lines 47ff to lacking proof and the falsity of some claims. This means that Zapatero makes explicit some of the knowledge criteria of our culture, e.g. empirical proof, which again does double duty: delegitimizing Aznar as PM, but at the same time legitimizing his own opposition by requiring legitimate proof for a policy that affects the whole country.

In a next move (lines 54ff), Zapatero further emphasizes that his doubts are not merely his, but general. That is, he thus moves from personal opinions and context models, to generally shared beliefs, and hence to knowledge. Thus, he again both emphasizes that he is right and Aznar is wrong, a political polarization that is in line with the opinions and acts of the leader of the opposition. By referring to public opinion and the Spanish people, this move of evidentiality (Chafe & Nichols, 1986), not only supports his own judgment, but also emphasizes his support by the people, thus politically legitimizing himself. On the basis of this accumulated evidence, and concluding that Aznar’s position is untenable, Zapatero finally and rhetorically asks Aznar to vote against the war in Iraq.

Since Zapatero focused his opposition on Aznar’s lack of the proven necessity to intervene in Iraq, it may be expected that Aznar in his reaction (lines 74ff) does so to. After trying to delegitimize Zapatero and his discourse by degrading it to that of demonstration slogans and banners, he immediately focuses on the question of ‘proofs’, and asserts that none are needed – for which he claims the relevance of U.N. resolution 1441. In order to interactionally avoid the burden of proof, Aznar thus changes the topic by emphasizing that Saddam Hussein needs to provide proofs. And once focusing on Saddam Hussein as the dictator and emphasizing that Zapatero does not require anything from him, Aznar associates his opponent to the one who is generally considered to be the bad guy, thereby delegitimizing Zapatero by the suggestion of collusion or association. When talking about Saddam Hussein, Aznar activates the usual, ideologically based mental models and general attitudes about the dictator, thus taking distance from the current, contextually defined, accusation of Zapatero. But by then counter-accusing Zapatero of not requiring proof from Saddam Hussein, he again brings back the focus to what is politically relevant in this debate, namely defending himself against the opposition (and public opinion) and accusing them of colluding with the enemy.

We see that in order to be able to make such inferences from the debate we need to assume that the participants construct and continuously update context models. These context need to feature general sociocultural and political knowledge as well as specific knowledge about specific events, on the one hand, and a representation of the political identities (PSOE, PP), the political roles (PM, leader of the opposition), the political relations (political antagonists), and political goals (delegitimize, etc.) of the participants. Only then are we able to understand what the participants are really doing (politically) and especially how they are able to do it. Without such an approach, that is, in pure interactional terms, we would be limited to a rather superficial account of questions and answers, accusations and defenses, and so on. Even we would acknowledge that the debate takes place in parliament, we would be able to account for the formal style, and perhaps the specific turn allocation and distribution rules of the institution. But the political point of this debate would remain unanalyzed or underanalyzed if we would not also construe the political situation as being subjectively and personally, as well as collectively, construed in the context models of the participants.




The study of discourse should not be limited to an ‘autonomous’ analysis of text and talk, but also develop a theory of context. It has been assumed in this paper that such a context should not be formulated in terms of objective social situations or even their relevant properties, but rather in terms of mental models: context models. Such models represent the subjective definitions of the communicative situation, and control each participant’s contribution to interaction.

Besides the usual categories of social situations, such as Setting and Participants (and their identities, roles, relations, etc.), context models also should feature the ‘cognitive’ aspects of the situation, such as the goals and the knowledge of the participants. Especially the knowledge of the participants is crucial in the management of what information remains implicit and what is being explicitly expressed, reminded or presupposed in discourse. The K-device of the context model does just that: strategically controlling knowledge in interaction.

Such strategies are also crucial for the management of lies. In order to be able to lie successfully speakers must know what recipients know or do not know. And vice versa, when recipients know that what speakers assert is false, they can accuse them of lying. That is, lies not so much need a semantic account in terms of the truth of propositions, but rather a pragmatic approach in terms of how knowledge is being managed by context models.

In political debates such management of lies is crucial for the legitimacy of the participants. I therefore analyzed one of the debates in the Spanish Cortes, on the occasion of the looming war in Iraq, pitching Prime Minister Aznar, against Opposition leader Zapatero. As was the case in the USA, the UK and elsewhere, also here the crucial point was whether Aznar was lying about the real reasons to go to war, amidst serious doubts about the alleged Saddam Hussein’s possession of weapons of mass destruction and links with international terrorism. In this analysis it is shown that in addition to the usual structures and strategies of discourse and interaction, a contextual approach accounts for the constraints of institutional interaction, such as forms of address and presentation, speaker order and turn allocation. More specifically it is shown that context models feature a knowledge component that controls the strategies of lying, accusations of lying and defending oneself against such accusations. Moreover, it is shown that in this way not only the debate can be analyzed in general contextual terms, but more specifically as a form of political interaction.








Akman, V., Bouquet, P., Thomason, R., & Young, R. A. (Eds.). (2001). Modeling and using context Third International and Interdisciplinary Conference, CONTEXT 2001, Dundee, UK, July 27-30, 2001 : Proceedings. Berlin New York: Springer-Verlag.


Allman, T. D. (2004). Rogue state. America and the world under George W. Bush. New York.


Argyle, M., Furnham, A., & Graham, J. A. (1981). Social situations. Cambridge: Cambridge University Press.


Auer, P., & di Luzio, A. (Eds.) (1992). The Contextualization of Language. Amsterdam: John Benjamins.


Barnes, J. A. (1994). A pack of lies:  Towards a sociology of lying. New York: Cambridge University Press.


Bayley, P. (Ed.). (2004). Cross-cultural perspectives on parliamentary discourse. Amsterdam Philadelphia: J. Benjamins Pub. Co.


Bernecker, S., & Dretske, F. I. (Eds.). (2000). Knowledge: Readings in contemporary epistemology. Oxford: Oxford University Press.


Brown, P., & Fraser, C. (1979). Speech as a marker of situation. In K. R. Scherer, & H. Giles (Eds.). Social markers in speech, (pp. 33-62). Cambridge: Cambridge University Press.


Drew, P., & Heritage, J. (Eds.). (1992). Talk at work: Interaction in institutional settings. Cambridge (England New York: Cambridge University Press.


Duranti, A. (1997). Linguistic anthropology. Cambridge New York, NY, USA: Cambridge University Press.


Duranti, A. (Ed.) (2001). Linguistic Anthropology: A Reader. Malden, MA: Blackwell.


Duranti, A., & Goodwin, C., (Eds.). (1992). Rethinking Context: Language as an Interactive Phenomenon Cambridge: Cambridge University Press.


Edwards, D., & Potter, J. (1992). Discursive psychology. London: Sage Publications, Inc.


Fairclough, N. (1995). Critical discourse analysis. The critical study of language. London: Longman.


Fetzer, A. (2004). Recontextualizing context. Amsterdam: Benjamins.


Forgas, J. P. (1979). Social episodes: The study of interaction routines. London New York: Published in cooperation with the European Association of Experimental Social Psychology by Academic Press.


Fossà, G., & Barenghi, R. (2003). The Bush show. Verità e bugie della guerra infinita. San Lazzaro di Savena (Bologna): Nuovi mondi media.


Ghadessy, M. (Ed.). (1999). Text and context in functional linguistics. Amsterdam Philadelphia: J. Benjamins.


Gumperz, J. J. (1982a). Discourse Strategies. Cambridge: Cambridge University Press.


Gumperz, J. J. (1982b). Language and social identity. Cambridge ( England New York: Cambridge University Press.


Gumperz, J. J. (1992). Contextualization and Understanding. In A. Duranti and C. Goodwin (Eds.), Rethinking Context: Language as an Interactive Phenomenon (pp. 229–252). Cambridge: Cambridge University Press.


Johnson-Laird, P. N. (1983). Mental models. Cambridge: Cambridge University Press. [«PR[b]»]



Knorr-Cetina, K., & Cicourel, A. V. (Eds.). (1981). Advances in social theory and methodology. Towards an integration of micro- and macrosociologies. London: Routledge & Kegan Paul.


Lazar, M. (Ed.). (2005). Feminist Critical Discourse Analysis. Gender, Power and Ideology in Discourse. Houndsmills, UK: Palgrave MacMillan.


Leckie-Tarry, H. (1995). Language & Context. A functional linguistic theory of register. Edited by David Birch. London: Pinter.



Lewis, M., & Saarni, C. (Eds.). (1993). Lying and deception in everyday life. New York, NY: Guilford Press.


Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81-97.


Neisser, U., & Fivush, R. (Eds.). (1994). The Remembering self. Construction and accuracy in the self-narrative. Cambridge: Cambridge University Press.


O'Shaughnessy, N. J. (2004). Politics and propaganda. Weapons of mass seduction. Ann Arbor: University of Michigan Press.


Sarangi, S., & Roberts, C. (Eds.). (1999). Talk, work, and institutional order: Discourse in medical, mediation, and management settings. Berlin New York: Mouton de Gruyter.


Schegloff, E. A. (1987). Between Macro and Micro: Contexts and Other Connections. In J. Alexander, R. M. B. Giesen and N. Smelser (Eds.), The Micro-Macro Link (pp. 207-234). Berkeley: University of California Press.


Schegloff, E. A. (1991). Reflections on talk and social structure. In: Boden, D., D. H. Zimmerman (Eds.), Talk and social structure: studies in ethnomethodology and conversation analysis. (pp. 44-71). Cambridge: Polity Press.


Schegloff, E. A. (1992). In Another Context. In Duranti, Alessandro, & Goodwin, Charles (Eds.), Rethinking Context: Language as an Interactive Phenomenon. (pp 191-227). Cambridge, England: Cambridge University Press.


Scherer, K. R., & Giles, H. (Eds.). (1979). Social markers in speech. Cambridge, (Eng.) New York: Cambridge University Press, Paris, Éditions de la Maison des sciences de l'homme.


Steiner, J. (2004). Deliberative politics in action. Analysing parliamentary discourse. New York: Cambridge University Press.


Thomas, W. I. (1966). Situational analysis: The behavior pattern and the situation. (1928). In M. Janovitz (Ed.), W. I. Thomas on social organization and social personality. Chicago: Chicago University Press.


Tulving, E. (1983). Elements of episodic memory. Oxford: Oxford University Press.


Van Dijk, T. A. (1977). Text and context. Explorations in the semantics and pragmatics of discourse. London: Longman.


Van Dijk, T. A. (1980). Macrostructures: An interdisciplinary study of global structures in discourse, interaction, and cognition. Hillsdale, N.J.: L. Erlbaum Associates.


Van Dijk, T. A. (1999). Context models in discourse processing. In: van Oostendorp, Herre, & Goldman, Susan R. (Eds.), The construction of mental representations during reading. (pp. 123-148). Mahwah, NJ, USA: Lawrence Erlbaum Associates.


Van Dijk, T. A. (2003). Knowledge in parliamentary debates. Journal of Language and Politics, 2(1), 93-129.


Van Dijk, T. A. (2004). Text and context of parliamentary debates. In Paul Bayley (Ed.), Cross-Cultural Perspectives on Parliamentary Discourse. (pp. 339-372). Amsterdam: Benjamins.


Van Dijk, T. A. (2005). Contextual knowledge management in discourse production. A CDA perspective. In Ruth Wodak and  Paul Chilton (Eds.), A New Agenda in (Critical) Discourse Analysis. (pp. 71-100) Amsterdam: Benjamins, 2005.



Van Dijk, T. A. (Ed.). (2006). Discourse, interaction and cognition. Special issue of Discourse Studies, 8(1).


Van Dijk, T. A., & Kintsch, W. (1983). Strategies of discourse comprehension. New York: Academic Press.


Wodak, R., & Chilton, P. (Eds.). (2005). A New Agenda in (Critical) Discourse Analysis. Amsterdam: Benjamins.


Wodak, R., & Meyer, M. (Eds.). (2001). Methods of critical discourse analysis. London: Sage


Wortham, S., & Locher, M. (1999). Embedded metapragmatics and lying politicians. Language and Communication, 19 (2), 109-125.




La señora PRESIDENTA: Pregunta número 17 que formula el diputado don José Luis Rodríguez Zapatero.


El señor RODRÍGUEZ ZAPATERO: Muchas gracias, señora presidenta. Señor Aznar, ¿cree necesaria una intervención militar en Irak?


La señora PRESIDENTA: Muchas gracias, señor Rodríguez Zapatero. Señor presidente del Gobierno.


El señor PRESIDENTE DEL GOBIERNO (Aznar López): Creo que es necesario el desarme del régimen de Sadam Husein.


La señora PRESIDENTA: Muchas gracias, señor presidente. Señor Rodríguez Zapatero.


El señor RODRÍGUEZ ZAPATERO: Señor Aznar, a fuerza de no responder y no decir las auténticas posiciones y las auténticas decisiones que usted ha tomado, va a resultar cada día menos creíble. En la última reunión del Consejo de Seguridad de Naciones Unidas hemos visto un informe exhaustivo, supongo que respetable de los señores Blix y El Baradei, los inspectores que tienen la máxima autoridad en todo este debate y que hicieron afirmaciones muy claras. La frase, creo, más importante es: ¿cuánto tiempo necesitamos para verificar el desarme de Irak? Textualmente, dijeron: ni años ni semanas, meses. Y usted ha firmado una resolución que da apenas unos días, y usted acaba, a través de su Gobierno, de su ministra de Asuntos Exteriores, de decir que 45 días es inaceptable y está en contra. Pero, ¿por qué sabe usted más que el señor Blix? ¿Por qué suplanta la autoridad del señor Blix, que lleva años y meses trabajando, buscando lo que hace y lo que tiene que cumplir, que es el mandato de Naciones Unidas? (Rumores.)


La señora PRESIDENTA: Señor Mancha.


El señor RODRÍGUEZ ZAPATERO: Mire, señor Aznar, hay que defender la justicia y la legalidad internacionales, y en este juicio que todos estamos viendo ustedes han perdido la razón; no han presentado pruebas y algunas han sido falsas. Los peritos o los inspectores han dicho lo que han dicho, que acabo de referirle, no han demostrado ninguna vinculación con el terrorismo internacional, ninguna, ni con Al Qaeda ni con el terrorismo mal llamado islámico y, claro, cuando las razones no existen, la diplomacia hemos visto que se ha convertido no en más pruebas sino en dólares, no en poner razones y argumentos sino en hacer ofertas. Por eso, prácticamente ya nadie defiende sus tesis, ni las de su Gobierno, ni en el Consejo de Seguridad, ni en la opinión pública, ni en la ciudadanía española. Señor Aznar, debe de tener un momento de reflexión y de no llevar a este país a una posición indefendible, a la luz del derecho internacional, de la razón y de autoridad de Naciones Unidas, que no respeta ni el embajador que usted ha nombrado allí en la persona de su secretario general. Abandone su posición, señor Aznar. Vote el próximo día, si es que llega a haber votación en el Consejo, a favor de las tesis razonables que ha presentado Blix y no de Bush, vote con la Europa cercana, vote con la legalidad internacional, vote por el tiempo para la paz y no para… (Aplausos.)


La señora PRESIDENTA: Muchas gracias, señor Rodríguez Zapatero. Señor presidente del Gobierno.


El señor PRESIDENTE DEL GOBIERNO (Aznar López): Ya sabíamos que S.S. saldría con pancarta otra vez, pero no hace falta que nos adelante su discurso (Rumores.), ya comprendo que llevaba 15 días sin ir detrás de una pancarta y debe ser una cosa bastante insoportable de aguantar, señoría. Quiero decirle, para que S.S. comprenda, que la Resolución 1441 no obliga a nadie a presentar pruebas nada más que a Sadam Husein, que es al único al que usted no le exige que cumpla nada, al único. Al único que las resoluciones de Naciones Unidas obliga a presentar pruebas de desarme desde hace 12 años, señor Rodríguez Zapatero, es a Sadam Husein, que es al único al que usted, cuando habla de graves consecuencias no le advierte de ninguna grave consecuencia. Usted admite graves consecuencias para el Gobierno, para todo aquel que no está de acuerdo con usted, pues el único que está advertido por la comunidad internacional de la cual nosotros nos tenemos que alejar, de que si no le respeta y no prueba su desarme está sometido a serias consecuencias, es exactamente aquel que usted olvida en todos sus discursos, en todas sus actuaciones y en todas sus pancartas sistemáticamente, señoría, el único que está obligado a hacerlo, el único. (Aplausos.) Desde hace 12 años, señoría, lo lleva haciendo, y desde hace 12 años lo lleva incumpliendo, y las resoluciones de Naciones Unidas dicen exactamente lo que yo le he dicho, y los inspectores dicen que si hubiese cooperación podrían terminar el trabajo rápidamente. ¿Cuál es el problema? ¿Por qué hay armas de destrucción masiva? Porque no se quieren deshacer de las armas de destrucción masiva. Usted lo que nos pide, señoría, es que no hagamos nada o que aliviemos la presión, lo cual es el mejor mensaje para todos los dictadores que quieren tener armas de destrucción masiva, incluido también Sadam Husein, y eso, señoría, nosotros no lo vamos a hacer. Y eso significa respetar la legalidad y eso significa respetar el Consejo de Seguridad de Naciones Unidas. Sin duda que si eso no ocurre así, habrá graves consecuencias, señoría, para el pueblo iraquí, porque seguirá estando bajo una tiranía, y para el pueblo kurdo, que seguirá siendo también atacado por ese dictador, y agrava las consecuencias para la seguridad del mundo. No habrá un mundo más inseguro que un mundo en el que no se respete la ley, pero vamos a intentar que eso no sea así. Si lo conseguimos, señoría, no le deberemos nada a su actuación. Muchas gracias, señora presidenta. (Prolongados aplausos.)