Volume 3, No. 3, Art. 13 – September 2002

Reflexivity and Subjectivity: A Dialogical Perspective for and on International Relations Theory

Xavier Guillaume

Abstract: The aim of this essay is to discuss the places of the notions of subjectivity and reflexivity within International Relations (IR) theory by contrasting the classical perspective inspired by positivism with a dialogical perspective which tries to fully integrate these notions in its examination of the specific nexus between identity and alterity. This dialogical perspective proposes a reflexive tool, the hermeneutical locus, which is intended to provide a way to assess the subjectivity of the cognised without falling into reification and a reflexive archaeology of the discipline of IR itself. In this sense, a dialogical approach wishes to offer a reflexive tool for and on IR theory.

Key words: subjectivity, reflexivity, dialogism, Mikhail M. Bakhtin, International Relations theory, identity-alterity nexus, reification, archaeology

Table of Contents

1. Introduction

2. International Relations Theory, the Question of the Identity-Alterity Nexus, and the Dialogical Approach of International Relations

3. The Hermeneutical Locus as a Reflexive Tool for and on International Relations Theory

4. Conclusion







1. Introduction

The general theme of this FQS special issue on reflexivity and subjectivity is of great interest and significance in social sciences today and more particularly in the field of International Relations (IR). IR is the field of study that attempts to understand and/or explain interactions between state-based and/or non state-based actors on a level that is not limited to the domestic realm and which is often termed "international politics". In IR the traditional positivist perspective has proved either to be a failure in its pretensions to provide the only theoretical and scientific framework for social sciences or it has shown its limits as an epistemic research programme (see LAKATOS 1970). In IR theory, the struggle during the late eighties and early nineties against the dominant positivist research programmes resulted in the emergence, among other currents, of a "middle ground" meta-theory known as constructivism. This emergence lead to a reflection in the discipline which specifically tackled not only the ontological, but more specifically the epistemological and axiomatical dimensions of the different IR research programmes (ADLER 1997; LAPID 1989). [1]

The main aim of this article is to provide a reflection on the meaning to give to reflexivity and subjectivity when one discuss IR theory according to a specific positive heuristic of the constructivist programme, that of dialogism (GUILLAUME 2002). Throughout the text, except when specified, I will adopt Mats ALVESSON and Kaj SKÖLDBERG's (2000, p.5) definition of reflexivity as a conscious and continuous attention to "the way different kinds of linguistic, social, political and theoretical elements are woven together in the process of knowledge development, during which empirical material is constructed, interpreted and written". The interconnected notion of subjectivity designates, in turn, the quality of either the cognised or the cognisant agents within the scientific framework in each dimensions of their interrelations. A dialogical perspective, inspired by Mikhail M. BAKHTIN's work, thus wishes to provide an epistemic framework to the relations cognisant and cognised subjects entertain together in each level of their interactions. [2]

Reflexivity and subjectivity are in effect fundamental problematics to those wishing to design social sciences beyond the positivist standpoint as far as the latter only recognised reflexivity in the sole consciousness of the cognisant to his work. Furthermore, the positivist standpoint categorically refused to fully consider the problem of subjectivity, especially that of the cognised. The real challenge, however, resides in the articulation of a coherent relation between the two elements within a research programme, trying thus to integrate them in a scientific framework (i.e. a project aimed at constituting and constructing knowledge) without falling into the recurrent problems of a "relational theory of reference", such as postmodernism, which, in the words of Alexander WENDT (1999, pp.56-57, original emphasis),

"cannot account for the resistance of the world to certain representations, and thus for representational failure or misinterpretations. [...] The external world to which we ostensibly lack access, in other words, often frustrates or penalizes representations. Postmodernism gives us no insight into why this is so, and indeed, rejects the question altogether." [3]

Dialogism represents an answer to WENDT's challenge for "a theory that takes account of the contribution of mind and language yet is anchored to external reality", as it proposes a form of relational perspective which soundly integrates this "external reality". This type of perspective, as underlined by Mustafa EMIRBAYER (1997, pp.287-288, my emphasis), effectively

"derive[s] the meaning, significance, and identity [of the cognised] from the (changing) functional roles they play within that transaction [i.e. relationality]. The latter, seen as a dynamic, unfolding process [e.g. a dialogue], becomes the primary unit of analysis rather than the constituent elements themselves." [4]

These dimensions are furthermore "inseparable from the [relational] contexts within which they are embedded", or, in other words, in the external reality missing in WENDT's account of relational perspectives. By anchoring an understanding of international relations through a relational perspective of the cognised, dialogism provides a way to articulate the notions of reflexivity and subjectivity that are necessarily linked to a sound consideration of the cognised self-understanding/perception and its relationality to other cognised (see BRUBAKER & COOPER 2000). Moreover, reflexivity and subjectivity are an actual problematic as they are related to both cognised and cognisant agents (whether individually or collectively) in the three different levels constituting the scientific framework. This essay therefore wishes to address these three following levels: (i) the level of the cognised "objects" (we will see that this term itself is questionable) in their subjective interrelations as agents within political, social, economical, and historical structures; (ii) the level of the scientific research itself, the reflexive and subjective relations the cognisant "subject" entertains with the cognised "object"; (iii) the level of the reflexive and intersubjective interactions between cognisant "subjects". Dialogism, as we will see, precisely offers a model to understand or, at least, problematise each levels either within their own dimensions or in their interrelating ones. [5]

In order to bring about this threefold connection, I will first discuss the traditional positivist frame adopted by most IR scholars and situate the notions of reflexivity and subjectivity in relation to it. This will lead me to tackle the first level mentioned above and call attention to the subjective quality of the cognised, especially when the latter is problematised as an identity. I will then introduce the foundation of a dialogical approach in IR which specifically attempts to integrate the notion of identity as a specific factor in international relations while embodying the notions of reflexivity and subjectivity at the core of its epistemological concerns notably by underlining the transgredient character between identity and alterity. In the second section, I will introduce the two other levels of the scientific inquiry by stressing the need to develop a processual/relational approach of IR theory in order to avoid as much as possible the risk of reification existing in the passage from the first to the second level and to conduct a reflexive assessment of one's own discipline, such as IR theory, by considering it as a site of cultural practice. Dialogism is thus presented as a reflexive tool for and on IR theory, through the hermeneutical locus it creates, with all its epistemological consequences either in the relations the cognisant entertains toward the cognised or the cognisant toward other cognisant. [6]

2. International Relations Theory, the Question of the Identity-Alterity Nexus, and the Dialogical Approach of International Relations

Traditionally, IR theory only focalised on the two main factors (interest and power) that were deemed to causally explain what was considered to be at the core of the discipline: war and peace. Up to the end of the 1980s, the main trend within IR theory was to address these factors within a positivist frame regarding natural sciences as the way to follow if one wanted to do "real" science. Contrary to most social sciences, IR only lately fully integrated another possible "explanatory" factor—identity—to its general theoretical framework, finally avoiding it and dismissing it as secondary as it used to within traditional perspectives. This emergence, as illustrated in several contributions (see for instance LAPID & KRATOCHWIL 1996; KATZENSTEIN 1996), partly enabled IR scholars to rethink the epistemological and methodological dimensions of their field by stressing out more saliently the question of identity, without actually threatening the more traditional approaches, thus reconsidering a usually unproblematised element of IR theory: the ways to assess the object of study. This section will first present the traditional developments in IR concerning the notions of subjectivity and reflexivity and then introduce dialogism as an alternative perspective for IR to approach both these notions in function of a specific factor, identity, and the three dimensions characterising the scientific inquiry. [7]

As mentioned, IR theory for long has been based upon a perspective that only interest and power were at the heart of any explanation, or even understanding, of the international world. This foundation is the result of a disciplinary "debate" in IR between the so-called "idealists" and the emerging "realists" during the interwar period. As the (realist) naming suggests (see CARR 1939), the formers were deemed to provide a normative, utopian, ought-to-be perspective on world politics, based on the ground that ideas/norms were at the heart of international relations (e.g. WILSON's fourteen points1)). The latter, however, were depicted, in a debate that was mainly constructed by them (see THIES 2002), as providing a "realist" and "objective" perspective on world politics, taken the international world as it is in its anarchic and conflictual dimensions and not as the "idealists" wished it could be, a non "zero-sum" game in which conflicting national interests can be resolve through the rule of law (see BROWN 2001, pp.21-30). In his definition of political realism, Hans J. MORGENTHAU (1967, p.4, my emphasis), the founding father of post-war realism, synthesised what was at stake in the development of the discipline at that time and the subsequent basic epistemological driving lines which were to be adopted by most IR scholars since then, whether "realist" or not.

"Political realism believes that politics, like society in general, is governed by objective laws that have their roots in human nature. In order to improve society it is first necessary to understand the laws by which society lives. The operation of these laws being impervious to our preferences, men will challenge them only at the risk of failure. Realism [...] must also believe in the possibility of developing a rational theory that reflects, however imperfectly and one-sidedly, these objective laws. It believes also, then, in the possibility of distinguishing in politics between truth and opinion—between what is true objectively and rationally, supported by evidence and illuminated by reason, and what is only a subjective judgment, divorced from the facts as they are and informed by prejudice and wishful thinking." [8]

According to MORGENTHAU, then, all IR scholar should follow a positivist-empiricist perspective according to which the objects cognised are metaphysically detached from their linguistic and theoretical assessment by the cognisant. What matters for such a perspective is the correspondence between the signifiers used by the cognisant and the referents, objectively rendered by the signifiers, deemed to be existing independently from the formers. Observation, experimentation, falsidical testing and the judgement of history are among the many ways by which theories can be measured as correspondent to the truth of external reality. This monistic metaphysic implies that as there is one world "out there", there can only be one true (scientific) perspective corresponding to this sole reality (see SPEGELE 1996, pp.48-49). The notion of interest provides, in turn, the conceptual link between both worlds. To refer to MORGENTHAU (1967, p.5) again,

"The main signpost that helps political realism to find its way through the landscape of international politics is the concept of interest defined in terms of power. This concept provides the link between reason trying to understand international politics and the facts to be understood. [...] We assume that statesmen think and act in terms of interests defined as power, and the evidence of history bears that assumption out. [...] Thinking in terms of interest defined as power, we think as [the statesman] does, and as disinterested observers we understand his thoughts and actions perhaps better that he [...] does himself. The concept of interest defined as power imposes intellectual discipline upon the observer, infuses rational order into the subject matter of politics, and this makes the theoretical understanding of politics possible." [9]

From its realist birth then, the discipline of IR has been trying to reach a "scientific", and thus honourable, status within social sciences, a drive that was mainly and until quite recently embodied in the attempt to do and be as "real science" does and is. By "real science" it was usually and effortlessly meant natural sciences and their positivist/rationalist approach in addition to a tendency to only adopt quantitative and/or formal methods. Far from rejecting a priori any "positivist" approach in IR, I support that IR theory should not bluntly strive to attain a "positivist" and thus "scientific" status by focalising on what might well be an aporia for many aspects of international relations. Social sciences cannot become like the natural sciences even though they might well be inspired methodologically by them. "Positivist" approaches to international relations have in effect proved to be feasible, necessary and, often, relevant. Recent analysis of fundamental topics such as war or national identity, however, showed that there is a demand for different and/or complementary approaches as Erik RINGMAR (1996) and Rodney B. HALL (1999) underlined in their recent books which focused on this other fundamental factor of international relations: identity. If the positivist quest thus was highly understandable, it missed, however, one simple, yet major, flaw in such a drive: natural sciences do not deal with self-conscious and reflexive subjects.2) [10]

We can now sort out what are the traditional places held by the notions of subjectivity and reflexivity within IR following MORGENTHAU's statements, perfectly reflecting the three tenets of positivism which remains the main conscious or unconscious epistemic framework used by IR scholars (see NEUFELD 1995, pp.22-38). The first tenet is that of "truth as correspondence" which affirms the validity of positive knowledge by its objective correspondence to the external and observable world. What is at stake here is the assumed separation between the cognised (the object, the external reality) and the cognisant (the subject). The notion of objectivity, following MORGENTHAU and most of the "positivist" literature in IR, refers here to one of its possible definitions: the "quality of what exist independently of any knowledge or idea" (NADEAU 1999, p.451). The second tenet is also linked to the idea of objectivity but in its definition as "the quality of one who sees things the way they are without being influenced by its preferences or habits" (NADEAU 1999, p.451). In the idea of the value-free nature of scientific knowledge resides the objective quality of the cognisant in its capability to discipline, to discern and to separate what the facts are from the values, and the realm of the object from its subjective world. These characterisations of the notion of objectivity are, in turn, to be linked to the third tenet of positivism, the idea of the methodological unity of science which consider that the approaches developed to make sense of the natural world are appropriate to address the social world as its fundamental nature, a realm of object, is not essentially different from the former. In sum, within a positivist scenario, the two definitions of objectivity are encompassing the notions of subjectivity and reflexivity. Subjectivity, for instance, is understood as defining the natural relation the cognisant entertains with the cognised, the aim of the former being to transcend his subjectivity to reach an objective mode of relation to the latter. As the cognised is an object, the only subject of science, the cognisant, has to be objective. In this sense, the cognisant has to be reflexive to the cognised to the extent the cognisant has to discipline his relation to the latter in order for it to reach an objective, value-free, status. Subjectivity and reflexivity are thus subsumed into the two definitions of objectivity as mentioned before. [11]

This positivist understanding of the concepts of subjectivity and reflexivity is highly embedded in the epistemological developments within natural sciences and unfortunately is rather poor for the social sciences if bluntly translated. In effect, social sciences, as part of the humanities, study human consciousness as a subject (whether individually or collectively) of a material world within which it constructs a social, intersubjective world (GUZZINI 2000, p.164) by the interweaving of what BAKHTIN (1986b) would call texts, that is all the activities by which human beings are "readable". A text, in the BAKHTINian sense, encompasses both narrative and behavioural "documents", which should not be understood as an exclusive choice, in the humanities, to the sole discourse analysis or hermeneutical perspectives. The aim of social sciences is to read the different texts of human activities, that the cognised produces whether consciously or not, through different lenses (the diverse sub-fields of social sciences) and a diversity of approaches (positivist, structuralist, systemist, dialogical, etc.). To reframe this argument in BAKHTIN's words (1986c, p.161), natural sciences are monological while social sciences are dialogical in regards to their "object" of studies. Natural sciences necessarily and naturally possess what BAKHTIN calls an "epistemological consciousness" which is "a unitary and unique consciousness".

"Everything this consciousness deals with must be determined by itself alone: any determinateness must be derived from itself and any determination of an object must be performed by itself. In this sense, epistemological consciousness cannot have another consciousness outside itself, cannot enter into relation with another consciousness, one that is autonomous and distinct from it. Any unity is its own unity; it cannot admit next to itself any other unity that would be different from it and independent of it (the unity of nature, the unity of another consciousness), that is, any sovereign unity that would stand over against it with its own fate, one not determined by epistemological consciousness. This unitary consciousness creates and forms any matter it deals with solely as an object and not as a subiectum, and even a subiectum is no more than an object for it. The subiectum is known and understood only as an object, for only valuation could render him a subiectum—the bearer of his own autonomous life, experiencing his own fate" (BAKHTIN 1990, pp.88-89). [12]

The humanities on the contrary have to take into account the subiectum because they possess the "specific task of establishing, transmitting and interpreting the words of others" (BAKHTIN 1981, p.351). Natural sciences are the only subject, the only intellect, at work cognising and speaking about an object which is necessarily voiceless. To give a voice to the cognised, it is then important in social sciences to consider the cognised as a subject and to try to adapt our epistemology to this fundamental consideration. In review then, social sciences are oriented towards the reflexive interconnection of at least two subjects, that of the cognisant and that of the cognised. The fact that the cognised is regarded as an "object" in the positivist sense is as much an epistemic choice as it is to regard the cognised as a "subject" in the dialogical sense. However, the dialogical perspective tries to problematise the inherent subjective quality of the cognised. To quote BAKHTIN (1986d, p.144) again

"Sciences of the spirit [i.e. the humanities]; their field of inquiry is not one but two 'spirits' (the studied and the person who studies, which must not be merged into one spirit). The real object of study is the interrelation and interaction of 'spirits'." [13]

In order to address these different elements by focalising on the question of identity and identity formation, I will hereby elaborate a dialogical perspective, grounded on Mikhail Mikhailovitch BAKHTIN's works. The widespread interest in BAKHTIN thinking has been a major source of renewal in several disciplines such as in literary studies, linguistics or the humanities in general3). It has, however, never been seriously taken into consideration within the field of IR except for some authors which used him in a very impressionistic way (see ASHLEY 1989, pp.263 and 281; SHAPIRO 1989, p.81). One has to refer to an article by Iver B. NEUMANN (1996, pp.148-149 and 154-155) to see an outline of the possible uses of BAKHTIN within IR theory and its potential relevance to the identity-alterity nexus. By focalising on identity and identity formation in international relations, the traditional nexus represented by war and peace (conflict and co-operation), underpinning most IR works so far, would not be alone anymore to serve as an underlying characterisation of the interplay between states or other agents in the international system. The identity-alterity nexus is in effect a contending, yet not necessarily competing, characterisation at the heart of the epistemological and methodological renewal witnessed during the past decade. This renewal fundamentally concentrated on the questioning of the definition and use of omnipresent concepts such as anarchy (i.e. the absence of a hierarchy and an order among the main agents of international relations) in the overall understanding and explanation of international relations. As WENDT (1992) showed, anarchy had been reified by traditional approaches and thus became a substantive concept which was used to deterministically construct its objects of study. Anarchy became the structural lenses through which the object of study—most generally the state—was constructed. The problem with "positivist" perspectives in IR was then not so much about the idea that social facts, the object of study and, for some, reality as such were constructed but rather the question of "how" it was done and which consequences it has reflectively on IR theory. This is precisely the kind of questions a dialogical understanding of international relations tries to answer to. However, by using "identity" as a different pattern of understanding we should not fall into the same conceptual trap. Constructivism, as a research programme, has to avoid following the new "logic of identity", as traditional approaches followed the "logic of anarchy", in order to construct a valid and coherent epistemic body (WEBER 2001, p.78). [14]

In using the category "identity" we should therefore be aware of and emphasise the inherent difficulties and ambiguities related to it, as "identity" is both a category of political and social practices and a category of political and social analysis (see BRUBAKER & COOPER 2000). An "identity" might then be a category of practice of a cognised subject who might be taken as an "object" through the lenses of "identity" as a category of analysis. We are thus precisely in the moment where the traditionally defined subjectivity and objectivity are collapsing together, since what is felt as an "identity" by a subject can be interpreted only through its objectivation—its constitution as a category of analysis. As mentioned, the danger of such a necessary objectivational movement lies in its reification, which according to our dialogical framework can be understood in two complementary ways. First, and in a more classical understanding, if "identity" is not a necessary category of analysis to make sense of phenomenon related to "identity" as a category of practice, it should not be used "...in an implicitly or explicitly reifying manner, in a manner that implies or asserts that ... 'identities' exists and that people 'have' ... an 'identity'". This type of reification is however common, as exemplified by the neorealist use of "anarchy", and therefore scholars,

"[...] should seek to account for this process of reification. We should seek to explain the processes and mechanisms through which what has been called the 'political fiction' of the 'nation' [or 'identity'] can crystallize, at certain moments, as a powerful, compelling reality. But we should avoid unintentionally reproducing or reinforcing such reification by uncritically adopting categories of practice as categories of analysis" (BRUBAKER & COOPER 2000, pp.5-6; see also MELUCCI 1995, pp.42, 50-51). [15]

Second, following a dialogical understanding and in an attempt to address BRUBAKER and COOPER's stance, reification also means to treat the cognised as a voiceless "object". It is to say, as we will see, that "identity", as a category of analysis, should not become a monological construction of categories of practices. The cognised is or has been a subject and "identity" as a fictional (narrative), subjective (individual) and intersubjective (collective) phenomenon cannot therefore be studied as a voiceless "object" since it will characterise it as a thing and not as a reflexive self-representation. A dialogical understanding of "identity", and therefore any objectivational movement adopting such a perspective, partially lift up this problem by approaching an "identity" through the conscious mediation of a hermeneutical locus from which an "identity" is set in a dialogical network relative to other "identities" (alterity) defined as relevant by their place in this network (determined by their expression, contextuality and relationality in/to this "identity"). Dialogism then is a sound way to propose an analysis of "identity" without falling into the reification of the concept, as it was the case with "anarchy". It is furthermore a sound assessment of the epistemic relation the cognisant has with the cognised. In effect, the process of objectivation should not become the effacing of the subjective quality of this "object". The cognisant should enter in a dialogue with the cognised. This dialogue, however, shall never be "full" as there will always be a discrepancy in the communication between the two elements. It precisely is in this dissymmetric, but not necessarily unequal, relation that lies the possibility of knowledge as the cognisant situates him/herself outside the cognised, establishing a dissymmetric dialogue by framing this interrelation through a problématique (in the French acception) (AMORIM 1996, pp.22-23, 31, 59). [16]

So let's turn now to the key notion of dialogism to see how it can allow us to tackle in a positive and heuristic perspective the problematic relations the notion of "identity" raises in IR theory. In effect, dialogism can constitute a model which can epistemologically address the need of an adequate framework in relation to the nexus between "identity" and "alterity" while integrating in this framework the notions of subjectivity and reflexivity.4) Dialogism is a concept which mainly has to do with discursive elements and relates more specifically to utterances. BAKHTIN (1986a, p.92, my emphasis) states that:

"The expression of an utterance can never be fully understood or explained if its thematic content is all that is taken into account. The expression of an utterance always responds to a greater or lesser degree, that is, it expresses the speaker's attitude toward others' utterances and not just his attitude toward the object of his utterance. [...] However monological the utterance may be [...], however much it may concentrate on its own object, it cannot but be, in some measure, a response to what has already been said about the given topic, on the given issue, even though this responsiveness may not have assumed a clear-cut external expression. [...] The utterance is filled with dialogic overtones, and they must be taken into account in order to understand fully the style of the utterance. After all, our thought itself [...] is born and shaped in the process of interaction and struggle with others' thought, and this cannot but be reflected in the forms that verbally express our thought as well. [...] The interrelations between inserted others' speech and the rest of the speech (one's own) [...] are analogous (but, of course, not identical) to relations among rejoinders in dialogue". [17]

An utterance is thus a moment within a semantic network which is composed by the sum of all the existing utterances to which the utterance examined is somehow "responding to". This interweaving of utterances is not necessarily an interrelation between actual and active utterances, as it can in fact exist between actual and active utterances and absent and passive ones (TODOROV 1981, p.97). All in all, an utterance is functionally defined by at least one or many utterances which belong to a set of utterances either contemporary and/or pre-existing, active and/or passive. There is another essential quality of the utterance:

"[I]ts quality of being directed to someone, its addressivity. [...] This addressee can be an immediate participant-interlocutor in an everyday dialogue, [...] a more or less differentiated public, ethnic group, contemporaries, like-minded people, opponents and enemies. [...] And it can also be an indefinite, unconcretized other [...]. Both the composition and, particularly, the style of the utterance depend on those to whom the utterance is addressed, how the speaker (or writer) senses and imagines his addressees, and the force of their effect on the utterance" (BAKHTIN 1986a, p.95, emphasis are mine). [18]

Addressivity is therefore another central characteristic of any utterance since it provides an anchorage between dialogism and a theory of "identity". There are two mains reasons for such an affirmation. On the one hand, it is necessary to understand an utterance as an all-embracing category which is not limited only to purely discursive practices. Any utterance is indeed a conception of the world, a Weltanschauung (TODOROV 1981, p.97). As noted by BAKHTIN in his 1926 "Discourse in Life and Discourse in Art" (BAKHTIN 1994, pp.163-164),

"the extraverbal situation is far from being merely the external cause of an utterance—it does not operate on the utterance from outside, as if it were a mechanical force. Rather, the situation enters into the utterance as an essential constitutive part of [its semantic structure]". [19]

On the other hand, in an intertextual relation5), the utterance witnesses a subject. In other words, an utterance in order to be part of a dialogical network has not only to embody a semantic or logical relation to an object but also receive an author, its originator by which he/she expresses his/her position. From language we thus enter another realm of existence, that of subjectivity, as an utterance can be the actual expression of the cognised self-understanding/representation. [20]

In order to understand international relations in a dialogical manner by focalising on identity as a factor, it is then necessary to provide an analytical tool enabling the cognisant to make sense of the cognised without reifying the latter according to the problématique of the former. The hermeneutical locus provides such a tool as it is characterised as the interweaving in a specific place (the utterance) of an identity's expression, its contextuality and its relations to other identities (other utterances). In review then, dialogism is defined as the interweaving of utterances that respond to each other. An utterance is characterised as a hermeneutical locus through its expression, its context and its relation to other utterances whether this relation is present and/or past, active and/or passive. Furthermore, an utterance's addressivity links its discursive dimension to its subjective one by enabling the discerning of its figuration of alterity, hence dialogically of its own self. What is then missing is an understanding of the significance of alterity to identity (and vice versa) within a dialogical framework. BAKHTIN (1990, pp.15-16) characterises this relation between a self and an other as transgredient. This notion, borrowed from nineteenth century German school of aestheticism, states that an element is named transgredient when it designates "elements of our conscience which are exteriors to it but nonetheless essential to the process of its perfection, to its constitution as a totality" (TODOROV 1981, p.146). Transgredience means that a self establishes a necessary relation with a multitude of other selves (alterity); a self alone cannot therefore constitute and be itself within its own realm of existence without the other. According to BAKHTIN's anthropological philosophy, it is therefore impossible to become truly self-conscious, to be one self, if one does not reveals one's self to the other, through the other and with the help of the other (BAKHTIN 1984, p.287). [21]

For BAKHTIN, in order to reach an encompassing view of a self, one has to integrate through dialogue the vision of a multitude of interactions between the self and selves (alterity) and their position in the world. Dialogism is therefore an epistemologically sound way to establish such semantic network (BAKHTIN 1990, p.36). Even monological utterances enter a dialogical network even if they participate in an unsound ethical and epistemological position since they deny "the existence outside itself of another consciousness with equal rights and equal responsibilities, another I with equal rights [...]". Furthermore, in a monological approach, taken at its extreme,

"[a]nother person remains wholly and merely an object of consciousness, and not another consciousness. No response is expected from it that could change everything in the world of my consciousness. Monologue is finalized and deaf to the other's response, does not expect it and does not acknowledge in it any decisive force. Monologue manages without the other, and therefore to some degree materializes all reality. Monologue pretends to be the ultimate world. It closes down the represented world and represented persons" (BAKHTIN 1984, pp.292-293). [22]

Within a monological figuration then, the other becomes an object of the self's own conscience which can be interpreted and modified at will in function of the self's own needs as a self-representation. BAKHTIN's dialogism is thus concerned with both ethics and epistemology. Ethically, the completion and perfection of one's self is determined by the reflexive and dialogical integration of other selves (contrary to a monological perspective). Epistemologically, dialogism enables us to access soundly the identity-alterity nexus by stressing their interrelations in an interweaving of mutually-responsive utterances which can be understood by the delimitation of a hermeneutical locus. This interweaving relation constitutes in turn the transgredient relation between a self and other selves: one cannot be defined without an other. We will see in the next section the epistemic consequences this has on an IR theory aiming at tackling the identity-alterity nexus in regards to the notions of subjectivity and reflexivity. [23]

Some IR scholars have already underlined the necessity to address the identity-alterity nexus by stressing the ethical question "how we should treat the Other" (WENDT 1999, pp.21-22, original emphasis) when we have to make sense of international politics, which fundamentally is a politics of the other, a politics of alterity (GUILLAUME 2002). Otherness should be taken into account in IR theory to react to the positivistic orientation which dismissed the subjective quality of the other by assuming that it is a functional equivalent to an identity (see WALTZ 1979, pp.95-97). Furthermore, traditional approaches assumed the rationality, and thus uniformity, of the multiple identities interacting in the international system. They even dismissed the subjective quality of the identities at stake refusing to acknowledge this "field of possibilities" (DOTY 1996, pp.340-341) that the notions identity and alterity possess in their political, cultural and sometimes vital implications. This specific "field of possibilities" is beginning to be explored by IR scholars such as R.B.J. WALKER (1993) or David L. BLANEY and Naeem INAYATULLAH (1994; INAYATULLAH & BLANEY 1996) whose postmodern or critical approaches essentially focalised on normative and ethical dimensions. These fundamental contributions, however, seem to adopt a "logic of identity" without trying on the first hand to epistemologically integrate the identity-alterity nexus to IR, and by extension the notions of subjectivity and reflexivity. [24]

This section tried to underline dialogism as an epistemological alternative to the traditional positivist perspective underpinning most IR scholarship since the 1950s. While positivism might have a certain relevance for number of topics related to international relations, it remains that positivism is only but one epistemic stance among others. Dialogism represents one of these possible stances, in particular in relation to the identity-alterity nexus problématique which positivism tends to avoid as the idea of the subjectivity of the cognised is dismissed through the veil of objectivity. In effect, we have seen that a way to tackle the identity-alterity nexus according to dialogism is to define identity as an utterance in the sense and with the consequences I have underlined according to a dialogical perspective. This perspective furthermore defines identity, a subjectivity, in its transgredient relation to otherness, that is alternative subjective self-understandings/representations. Dialogism is thus a way to address the first level of the scientific inquiry, the level of the cognised's subjective relations, problematised in this framework as identities, by recognising the "object of study" as a subiectum and not as a voiceless thing. The hermeneutical locus, understood dialogically as the interwoven relations in an utterance of the identity-alterity nexus, thus becomes the actual "dynamic epistemic structure according to which the 'act' or 'phenomenal' event means and functions" (DOP 2000, p.18). The cognised subjectivity is then accessible to the cognisant through an assessment of its expressivity, its contextuality and its relationality vis-à-vis other subjective self-understandings/ representations. This structure is constructed through the problématique set up by the cognisant according to the epistemic limitations the hermeneutical locus imposes on him. The next section will accordingly present a relational assessment of the two levels of the scientific inquiry left. [25]

3. The Hermeneutical Locus as a Reflexive Tool for and on International Relations Theory

To consider the cognised as a subject is, of course, an ontological position in the same sense that those who consider the cognised as an object participate to another ontological position. To recognise this most forgotten evidence does not only participate to a reflexive consideration about any form of approaches, such as dialogism, but also requires us to undertake an epistemological discussion of this approach according to this stance. In effect, any approach should to a minimum extent integrate a reflection on the epistemic consequences one's own ontological choices bear. Such assessments should, in turn, take into consideration the two levels of scientific inquiry remaining to further discuss: i) the relation between the cognisant and the cognised and, ii) the relations between the cognisants. The hermeneutical locus, a dialogical mechanism I introduced in the previous section, provides us with a tool by which both levels can be considered in examining and taking into account the expressivity, the contextuality and the relationality of each levels. Moreover, each level is more particularly concerned with a specific problem or concern, whether that of reification the cognisant might impose on the cognised or of the meaning of the intersubjective quality of the relations existing between the cognisants. In other words, the hermeneutical locus should represent a reflexive tool for and on IR theory. [26]

As mentioned, one of the main problems which faces any approach wishing to recognise the subjective quality of the cognised is that of reification. Dialogism, via the hermeneutical locus, tries to bypass this problem by focusing on a dialogical account of the cognised. In effect, according to this perspective, the object of study is not the cognised itself but the process of their interrelations and interactions. As Pierre BOURDIEU noted once, "il faut penser relationnellement" [one has to think relationally] (1992a, p.200). The cognisant is actually fostering a subjective and reflexive relation to the cognised according to a problématique delineating the cognised's "objectivation" in its interrelations with other subjects. BAKHTIN (1986c, p.161) sustains that a subject cannot, epistemologically, be studied as thing for to do so would make it voiceless while being subiectum. "Cognition of it [therefore] can only be dialogic". This signifies that the process of objectivation, the structuring of a subject into a problématique, is possible as long as this objectivation does not transform a subiectum into a voiceless object. The objectivation should not become an instrumentalisation of the subject to the problématique and the subsequent theoretical framework as was the case with most positivist scholarship. An interesting and simple example, illustrated by a recent article from Richard LEBOW (2001), is the classical interpretation and use of THUCYDIDES' History of the Peloponnesian War as a one of the main realist forebear in their different claims about the structure of the war-peace nexus (WALTZ 1979, p.127). THUCYDIDES has always been seen as one of the first setting the underpinning elements of IR realism such as: (a) the centrality of the states (city-states) as the dominant and functionally equivalent units of action; (b) the depiction of the states as power-seeking entities whether as a means or as an end and (c) the ascription to the states of a rational behaviour assessable in rational terms (KEOHANE 1986, p.7). Furthermore, THUCYDIDES' account of war and his "causal" approach to facts made him an enlightened observer on the mechanisms at work in an anarchic system—a system in which nothing higher than the units can either delimit or reduce the units in their power-seeking drive—, thus reflecting the "state of nature", a zero-sum game like environment, in which these units act. THUCYDIDES is thus deemed to have set up and to reflect the natural and unchanging rules that drove, drive and will drive the states in their relations and then is basically reflecting the fundamental hard core of the realist research programme (see LAKATOS 1970, p.48). [27]

These use and appropriation are, however, and to follow LEBOW (2001, p.559), a "superficial and one-sided portrayal" of THUCYDIDES' book which is typical of a certain perspective of IR theory as a nomothetic approach to the social world (see LEVY 2001, p.41). In effect, realists tend to superimpose their programme's hard core to the articulation of, in this case, THUCYDIDES' Peloponnesian War while ignoring the incompleteness and errors such a superimposition bears (BAGBY 1994, LEBOW 2001). These critics follows the line of the traditional distinction between nomothetic (e.g. IR theory) and idiographic (e.g. history) disciplines and the need for the former to adopt some of the latter's natural impetus to regard evidences, whether pro or contra one's theory, and context as compelling elements since the former tend to "allow their theories to take priority over the evidence, focus on those historical events that confirm their theories, and ignore the larger context in which events occur and in the absence of which those events cannot be fully understood" (LEVY 2001, pp.49-50). The nomothetic drive to construct and test theories is the underlying reason why THUCYDIDES was perceived by the realists as a forebear since they picked in his work what seemed to fulfil their expectations, as in the Melian Dialogue, and to provide a test for their ahistorical theory on how international relations were framed and worked. LEBOW strikingly demonstrates a contrario that if a "realist" THUCYDIDES is indeed present, he is far from being the only, or even the most important, THUCYDIDES speaking in the Peloponnesian War: There is also a constructivist THUCYDIDES voicing in the text. In effect, a close reading of THUCYDIDES' History shows that his underlying inquiry is that of the relationship between nomos (convention) and phusis (nature), and its implications for the constitution of the Greek culture and civilisation, and that his conclusion emphasises the role of nomos in the construction of identities and in the delineation and limitation of the behaviour of both individuals and societies (LEBOW 2001). As this example illustrated, the main problem with non-reflexive nomothetic perspectives is that they consider the way of doing science in what Michel MEYER (1986, p.20) coins an apocritical direction, that is a direction which isolate a certain number of problems (e.g. why states go to war?) and by answering these problems seeks to solve and suppress them as such. In a fashion that is parallel to natural sciences, the aim of "positivist" social sciences is to offer, as Meyer underlined, "an apocritical answer [which] closes the inquiry, stems back and breaks away from the problems at stake" as it provides an answer which is supposed to solve and suppress the question altogether. [28]

Social sciences, however, by answering to a problématique do not actually solve, and thus suppress, the "problem" at the origin the latter. Natural sciences deal with objective questionings that require definite answers in order to suppress them. Of course, once solved these questionings create new ones but they are definitively "behind" in the realm of science. In other words, we do not ask ourselves the question to know why an apple falls from a tree anymore yet we still wonder why states go to war. This recurrent questioning still manage to create wide debates in IR, as in the Democratic Peace debate, despite nearly three thousands years of more or less cunning tries from THUCYDIDES' hubris to HUNTINGTON's civilisational explanation. Maybe the right answer has yet to be found, and many liberals see in the Democratic Peace argument such a definite answer (see RUSSETT 1993), yet none of the answers so far given brought about the suppression of the question "why states go to war?" or the peace-war nexus problématique as such. Anyhow, if Democratic Peace proponents would prove that in order to solve and suppress this question altogether it was necessary to establish global democracy in the world, it remains nonetheless that the peace-war nexus would still remain as a problématique, whether or not it is transformed in a softer or more focused articulation between co-operation and conflict. States, or other agents of international relations, will always interact, perform or behave in a way that will require IR scholars to consider which nexus might be the most relevant to explain or understand some phenomena. [29]

The focus of IR theory should then shift from this traditional apocritical direction, which blindfolds many IR scholars from epistemic and, often, methodological problems, to a more reflexive direction as underlined by ALVESSON and SKÖLDBERG's definition (2000, p.5). As their definition suggests, there are two moments in which reflexivity can take part in IR theory. The first movement is related to the second level of the scientific inquiry while the second movement is related to the third level. On the one hand, reflexivity is thus present in the relations the cognisant has with the cognised, that is how the former reflexively constructs the latter and the process of its interactions and interrelations with other subjects. Dialogism offers such a frame, as presented in the previous section, by delineating the cognisant scope of inquiry according to a reflexive tool, the hermeneutical locus, which specifically deals with processuality, through its focus on relationality, and aims at bridging IR theory nomothetic tradition to a more idiographic conception for dialogism possesses a particular interest in contextuality and expressivity. The three elements of the hermeneutical locus entail that a special attention should be paid to the actual reflexive way by which the cognisant chooses the "identities" at stake in its construction of a problématique. An elegant way to further avoid reification and instrumentalization is thus to employ conscious counterfactual thought experiments which, in the words of Philip TETLOCK and Aaron BELKIN (1996, p.15), provides, by

"asking people to think how things could have worked out differently[,] a means of preventing the world that did occur from blocking our views of the worlds that might well have occurred if some antecedent condition had taken on a different value". [30]

This specification is only one among several, and not the most important to their eyes, that counterfactuals could take according to TETLOCK and BELKIN. Both authors actually favour counterfactuals within a highly positivist perspective on how to deal with social sciences, rejecting altogether the implications of counterfactuals within more interpretative perspectives (whether constructivist or postmodern for instance) in world politics. Hence, TETLOCK and BELKIN implicitly put aside those latter perspectives as unscientific as they are and not being able to predict events and only able to offer subjective, circular and nonfalsifiable standpoints (TETLOCK & BELKIN 1996, pp.27, 30-31, 37-38). This view, as stressed by Steven WEBER (1996) and Richard LEBOW (2000, p.578), is extremely reductive on the possible use of such an interesting tool and one should, on the contrary, expand its scope by focusing more on its use as a way "to accrue evidence and arguments to convince people that some hypotheses are probably weak and others probably stronger" (WEBER 1996, p.272). Furthermore, and as specified by LEBOW (2000, p.551, my emphasis), "the methods of counterfactual experimentation need to be commensurate with the purposes for which they are used [...]". A dialogical understanding of international relations specifically undertakes such experimentation by concentrating on contextuality and expressivity in order to determine which utterances, which competing or alternative self-understanding/ representation(s) are to be considered as dominant in the dialogical constitution and performance of an intersubjective representation defined as national. There are naturally many dialogical interactions constituting a self-understanding/representation, the point being to evaluate which dialogue is dominant for a specific self-understanding/representation within a situated intersubjectivity. The dominant dialogue should henceforth arise as a heuristic momentum only if it is put in perspective with counterfactual dominant and non-dominant dialogues, that is alternative dialogues which allow the cognisant to a situated and processual reference outside the self-understanding/representation constructed as a subject of study. This necessity to evaluate alternative, counterfactual elements arise from the risk that the dialogue constructed by the cognisant has been so to explain or understand events ex post facto, that is it has been chosen in order to explain or understand a certain outcome, or even to fulfil the expectancies of the problématique or the theoretical model chosen by the cognisant (this point always had been a major criticism to nomothetic disciplines from more idiographic ones). Such a construction would not be epistemologically sound since it would then make inferences in the perspective of the outcome. [31]

Counterfactuals constitute therefore a way to reflexively and soundly construct a problématique and its related theoretical developments. Furthermore, counterfactual thought experiments are vital ingredients in the evaluation and testing of the heuristic of the dominant dialogue proposed by the cognisant (see LEBOW 2000, pp.558-564). This counterfactual logic globally follows a double rationale which firstly establishes a dominant dialogue as a "logical imperative", appealing to a more nomothetic dimension, within a certain context by imagining what other dialogues would plausibly be and which relevance they would have in a specified situated intersubjectivity and relationality. This logical exercise is aimed at establishing the sine qua non dialogue in a certain contextuality framed by the problématique. Consequently, however, this dominant dialogue also has to be a "historical imperative", the second component of the counterfactual rationale and which would result in a more idiographic dimension in IR theory, if one wants to reach a sound understanding. This historical imperative seeks to consider the alternatives which were actually offered by the cognised according to contemporary evidence (see FERGUSON 1997, p.79-90).6) The dominant dialogue chosen should therefore be the result of a double rationale—logical and historical—which through an evaluation of the possible alternatives and of the robustness of historical evidences should both alleviate the reification of the "identities" contemplated, rule out a deterministic study of their dialogue and evaluate the historical pertinence of the dialogue chosen by situating them simultaneously within a probabilistic frame and a historical relevance. [32]

Reflexivity, on the other hand, is also present in the relations the cognisant entertains with other cognisant. Reflexivity, following BOURDIEU's "epistemic reflexivity" (BOURDIEU & WACQUANT 1992), means that the cognisant has to consider that him/herself as a "cultural producer" and thus to reflect on the "socio-historical conditions of possibility" of this specific discourse that is science. The point is to subject the cognisant to the same critical inquiry he/she was compelled to apply to the cognised. In BOURDIEU's terminology, any discipline of social sciences has to undertake a work of "objectivation of the subject who objectivize" in which open debates and mutual criticism, dialogical debates BAKHTIN would say, are undertaken in order to institutionalise reflexivity. IR, however, possessed and still possess to a large extent, according to LAPID (1989, pp.249-250), "the dubious honor of being among the least self-reflexive of the Western social sciences". Some IR scholars have nonetheless tried to tackle the notion of reflexivity and to provide an account of it within the discipline. So far, the most thorough treatment of the notion of reflexivity in IR, to my knowledge, is to be found in Mark NEUFELD's The Restructuring of International Relations Theory (1995). NEUFELD (pp.13-21), drawing from KANT, HEGEL and MARX, situates "theoretical reflexivity", in the overall category of critical thinking, as one of the main characteristics of emancipatory theory. This specific positioning stands in sharp opposition to positivist IR theory, and provides a special "attention to the creative role of human consciousness, and an engagement in social criticism in support of practical political activity orientated toward radical social change", theoretical reflexivity therefore being the "theoretical reflection on the process of theorizing itself". This general definition embeds three core elements which are "(i) self-consciousness about underlying premises; (ii) the recognition of the inherently politico-normative dimension of paradigms and the normal science tradition they sustain; and (iii) the affirmation that reasoned judgements about the merits of contending paradigms are possible in the absence of a neutral observation of language". NEUFELD acknowledges that (i) is admittedly recognised by both positivist and post-positivist stances but stresses that (ii) and (iii) are actually what makes "reflexivity a virtual antonym of positivism" (pp.40-41). Both elements, in effect, oppose the positivist standpoint about "truth as correspondence" and science as a "process without a subject". If anyone committed to a post-positivist perspective can readily agree on the second element, one has to be careful on the first. NEUFELD's normative and emancipatory stances, on the one hand, recognise the intersubjective nature of science (p.42), yet, on the other hand, reassess the necessity of a truth value which would not be situated anymore in an empirical reality, however constructed by the cognisant, but through each theories' politico-normative assumptions. In a paradoxical way, NEUFELD, and many other normativists, actually abide to a form of "positivism" as they remain attached to a fundamental criteria linked to the basic principle of "truth as correspondence", "truth" in this case being the normative requirement of emancipation. Consider the following (p.44):

"[...] reflexive theorists accept incommensurability as the necessary consequence of the fact that paradigm-specific knowledge-defining standards are themselves intimately connected to and embedded in competing social and political agendas, the political-normative contents of which are not amenable to any neutral observation language". [33]

In other words, since "empirie" does not fit to be a relevant referent as there is no neutral observation language, contending perspectives cannot be confronted according to this form of referent. Yet, "reflexive theorists do not accept that recognizing contending paradigms as incommensurable means reasoned assessments are impossible". Furthermore (p.46), "judgements about contending paradigms are possible by means of reasoned assessments of the politico-normative content of the projects they serve, of the ways of life to which they correspond". Thus, judgement is possible with an admittedly unneutral observation language which will reasonably evaluate the different "paradigms" according to a normative referent, that of the emancipatory potential of these "paradigms". If I supported the idea that this type of normativism can be equated to "positivism", it was to underline that both perspectives were attached to the idea of truth, whether "empirical" or "normative", in an epistemic assessment of the value of a theory. The notion of truth, however, has long been discarded by most contemporary epistemologists (see AGAZZI 1988) as an aporia alleviated by the simple knowledge that all scientific (or normative) inquiry participate in an open intersubjective discourse. The point of reflexivity in IR theory precisely is to recognise the intersubjective process any "paradigm" went through in order to establish "operative criteria" on the use of certain number of notions (e.g. anarchy or identity) in order to set a scientific (or normative) agenda according to an agreement on these criteria (see AGAZZI 1988, p.19). The danger of NEUFELD's perspective is to reduce IR theory to a mere backup of theories deemed "emancipatory", the new criterion of truth, which will be defined through "a broader debate about which 'purposes', which 'enquiries' and which 'ideologies' merit the support and energy of International Relations scholars" (NEUFELD 1995, p.68, my emphasis). NEUFELD proposes a crude and simplistic opposition between "science" and "politico-normative" which does not take into account the possibility to render and recognise the second while continuing working in the framework of the first. If one can understand NEUFELD's will to replace the idea of episteme, a timeless and unchanging perspective, by that of phronesis, integrating variability and contingency (p.45), one cannot fail to situate NEUFELD in close range to doxa, the realm of opinion, as the validity criterion chosen by him is highly subjective, not even intersubjective. To the credit of NEUFELD, one has nonetheless to recognise with him that IR theory, as underlined WEBER (2001, pp.129-134), is a site of cultural practice.

"It is a site where stories that make sense of our world are spun, where signifying practices about international politics take place, where meanings about international life are produced, reproduced, and exchanged. [...] More importantly, IR theory is a site of cultural practice because it provides a framework for storytelling itself" (pp.129-130, original emphasis). [34]

As it is rightly pointed by many critical theorists, whether or not following NEUFELD's predicaments, IR theory had the tendency to consider its own framework of analysis, alongside many different concepts such as anarchy or identity, as natural, neutral and unbiased as they were considered to participate in the scientific inquiry in the tradition of "real science". Mainstream IR theory further participated in a standardisation of the discipline by expelling alternative explanations, usually based on non-positivist or non-quantitativist positions, in light of their "constructivist", "unneutral" or "biased" grounding. They placed alternative perspectives outside the realm of IR theory, i.e. the realm of science, and rejected them altogether as most of these perspectives tried to provide a more reflexive turn and function, in the sense given by ALVESSON and SKÖLDBERG, to the discipline (see WEBER 2001, p.131). These parochial views on IR theory were due, in my opinion and as we foresaw in the previous section, to a misplaced focus on what was to be understood as "science", "empirical" or "unbiased", as natural sciences were the actual referent to social sciences. In light of the discussion so far, it seems necessary to create new foundations for social sciences, foundations which acknowledge the peculiar nature of the cognised and the specific place of the cognisant in their overall interactions in a system of knowledge aside from the realm of opinion and political partisanship. [35]

Dialogism, by the intermediary of the hermeneutical locus, provides a reflexive account of IR theory as a site of cultural practice. The hermeneutical locus can in effect be used on utterances (e.g. realism, liberalism, constructivism, postmodernism, etc.) which will be examined as specific utterances according to their place in the academic discussion concerning international relations, thus offering an analysis of IR which takes as a starting point the idea that each current within our discipline, as any fields of social sciences, or any specific discursive assessment (a book, an article, a presentation, etc.), can be considered through a reflexive investigation. The aim of this investigation is to link both the critical tradition, represented here by NEUFELD, and a more traditional "disciplinary history" (see HOLDEN 2002). It is, in a way, to ask for an archaeology of IR theory in all its dimensions. As HOLDEN (2002, p.255) rightfully noted, there should not be any "privileged vantage-point from which IR's disciplinary history can be written. This means that 'critical IR' itself needs to be analysed and contextualized". Critical IR, like any other currents of IR theory, has to submit itself to the same scrutiny it used to submit the others. This scrutiny, as I just mentioned, can take the form of an archaeology, a term inspired by Michel FOUCAULT (1969) who gave it to his inquiry of discourses some 30 years ago. [36]

An archaeology (FOUCAULT 1969, pp.182-183) is fundamentally not a way to find a hidden discourse behind the utterance under examination, in a structuralist sense, but a way to define utterances as such, as "practices obeying to rules". An archaeology of IR aims to define an utterance within IR discourse in its specificity vis-à-vis other utterances and in their unity within "the types and rules of discursive practices that go beyond the individuals works". An archaeology is, finally, an exteriority which does not aim at describing the origin of an utterance but at systematically depicting the latter as an "object" of inquiry. We clearly see here that we are at the juncture of both critical tradition and a more traditional intellectual-historical perspective. In effect, by paying attention to practices, FOUCAULT wanted to pay attention to "a preconceptual, anonymous, socially sanctioned body of rules that govern one's manner of perceiving, judging, imagining, and acting" (FLYNN 1994, p.30). This is to be linked to the critical tradition. FOUCAULT's objective (1969, pp.170-173), however, was also to pay attention to the archive, that is "a practice which raise a multiplicity of utterances as many regular events". The archive is fundamentally "the general system of the formation and transformation of the utterances". An archaeology is then the description of "the utterances as specified practices in their modality of archive". This is to be linked to the intellectual-historical tradition. As FLYNN noted (1994, p.30),

"an archive is the locus of the rules and prior practices forming the conditions of inclusion or exclusion that enable certain practices and prevent others from being accepted as 'scientific,' or 'moral,' or whatever other social rubric may be in use at a particular epoch. [...] Reference to 'epoch' is crucial, for these archives are time-bound and factual; they are discovered, not deduced; they are the locus of practices as 'positivities' to be encountered, not as 'documents' to be interpreted. [...] The claim that these practices are to be registered as facts, not read as the result of intentions of some sort, gives [...] [Foucault's] archaeology its 'positivist' tilt [...]". [37]

By considering a precise current of IR theory, from realism to critical theories, as a hermeneutical locus, dialogism can reflexively bring about a refined analysis of IR theory in the overall sense inspired by FOUCAULT's archaeology. Contextually speaking, an utterance, such as that of realism in IR theory, has to be considered as a practice, in the sense given above, taking into account its predicates, assertions and norms as to retrace both the "internal discursive" and the external impetus behind the origin and evolution of this specific utterance within the discourse of IR. Expressively, the many assertions constituting realism—e.g. the books, articles, conferences, interviews of their proponents—are as many elements delineating an archive and thus subject to both "internal" and "external" inquiry. This type of work is to a certain extent already at the heart of the present focus of "the contextualizers", to borrow HOLDEN's terminology (2002). What is missing, however, is this relational dimension under which should also be analysed the contribution of these utterances in an intersubjective discussion which has been one of the focus of this paper: IR theory discourse. Who is "realism" or any "critical theory" answering to? To which extent their utterances reflect monological or dialogical approaches to the cognised or to other cognisant? How are contextually and expressively constructed the approaches' archives within IR theory and, more generally, within social sciences? Are these approaches even taking part in the epistemic reflexivity any discipline has to go through? As these last questions imply, the cognisant should consciously and dialogically participate in IR discourse comprehended as a semantic network. The notion of reflexivity, as presented here, is a way to go beyond the antinomic perspectives of postmodernism, in which is embedded the idea of deconstruction, and modernism, in which is embedded the idea of positivist social sciences. Reflexivity should link the possibility and the necessity of the scientific knowledge with the idea that any knowledge should be deconstructed as a contingent notion grounded in political, social, economic and historical dimensions. As accurately noted by WACQUANT (BOURDIEU & WACQUANT 1992, p.38), "to confuse the politics of science (knowledge) with that of society (power), is [in the end] to reject the historically instituted autonomy of the scientific field". [38]

The aim of this section was to underscore the necessity for social sciences, and by extension IR theory, to recognise the subjective quality of its "object of study". As mentioned alongside BAKHTIN (1981, p.351), social sciences possess "the specific task of establishing, transmitting and interpreting the words of others". The process of objectification should then not be equated to a process of reification in which the cognised becomes a voiceless thing instrumentalised by a research programme. The cognised, as a subjective self-understanding/representation, has to be integrated within IR theory in an epistemological framework, such as dialogism, which structurally tries to avoid the process of reification and thus avoid entering in either the "logic of anarchy" or that of "identity". The use of a counterfactual logic has thus been presented as a complementary way to avoid reification of the cognised within a certain problématique and to reflexively justify the choice of the setting in an hermeneutical locus. Furthermore, dialogism has been presented in opposition to an apocritical perspective and linked to a fully reflexive and ongoing appraisal of (i) the cognisant relations to the cognised, through the problématique and the objectivation; (ii) the cognisant relations to other cognisant within the intersubjective realm of science. In order to achieve this goal, it has been underlined that IR theory has to face the task of recognising itself as a site of cultural practice and thus acknowledge the need of an archaeology of the discipline and its different currents. A dialogical perspective thus wishes to provide a way to account a real dialogue for and on IR theory. [39]

4. Conclusion

The BAKHTINian definition of dialogism—as an interweaving of utterances characterised by their contextuality, expressivity and relationality—thus unveils an onto-epistemic dimension designating a process by which any cognised is accessible to any cognisant as long as it is replaced within its subjective semantic network and its relationality to other cognised. Accordingly dialogism is not only a reflexive way by which we can construct knowledge about the social world we study as cognisant but also a way by which we can analyse and evaluate the social world we construct as producers of knowledge. This is to follow BOURDIEU in his opinion that the models used by the cognisant to make sense of the cognised should also be reflexively applied onto the cognisant him/herself (BOURDIEU & WACQUANT 1992). The hermeneutical locus has been presented as a reflexive tool for IR theory as a way to avoid reification and monologism when the cognisant is constituting his problématique and thus is dealing with the cognised. Furthermore, the hermeneutical locus has been equally portrayed as a reflexive tool on IR theory in order to provide the discipline with an archaeology aiming at reconstituting the dialogical relations different IR currents entertain one with another and to qualify and situate each archive with the IR theory discourse. These basic conclusions concerning a dialogical approach do not limit themselves to the sole field of IR theory. All fields of social sciences might benefit from a more reflexive approach whether to their "subject"/ "object" of study or to their own discipline itself. A greater epistemological self-consciousness and self-assertiveness within any type of problématique or research design might prove to be a heuristic drive to improve and better delineate the definition of some arch-concept such as power or identity, as such concepts should reflexively be connected to the problématique at work. [40]

It is important to note that a dialogical perspective does not seek to provide the only framework by which subjectivity and reflexivity would be better integrated in IR theory. NEUFELD's attempt (1995) is only but one other framework according to which subjectivity and reflexivity can be addressed within IR. NEUFELD, however, is an epitome of the danger of such an integration if one wants to situate reflexivity and subjectivity at the heart of scientific inquiry. To situate such notions within scientific inquiry is admittedly a normative assumption as to know what is the core of the social sciences composed of. Furthermore, within IR theory, it is to normatively reject monological account of the cognised. But the overall aim of a scientific inquiry, contrary to NEUFELD's politico-normative perspective, is to constitute a body of knowledge within an open and intersubjective discussion (AGAZZI 1988) and not to promote any ideological agenda. To integrate the idea of subjectivity and reflexivity precisely is to constitute a body of knowledge that is self-aware, through its own archaeology, of its bias and limitations. Positivist perspectives, as well as constructivist or critical perspectives, within IR have to take account for the specific characterisation of their "object of study" and to focalise on the process of their interactions and interrelations. A dialogical approach, as I showed, offers such an overall perspective and is specifically designed to tackle the specific nexus between identity and alterity. The latter nexus possesses probably far more reaching potentialities for IR theory than the classical war-peace nexus as it creates more common grounds on which cross-field discussions or examinations can take place. For instance, one can think about linking normative theories about citizenship and politics of recognition (see GIANNI 1998) with IR theoretical perspectives on the interrelations between domestic and international issues focalising on the notion of "national identity" (see KRATOCHWIL 1996). A dialogical approach would thus be a common denominator to reflexively compare and combine both fields. Moreover, this essay wished to establish that a dialogical understanding was a potentially feasible approach to cognise world politics and that to address the question of epistemology was neither a mere speciality nor personal choice but a real heuristic plus for any searchers wishing to clarify and to expand their knowledge on their own practice. [41]


This essay is an epistemological extension of an article written in Millennium (GUILLAUME 2002). I want to thank Elisabetta BRIGHI and Harry BAUER, Millennium's editors, for their help and remarks on the previous article which paved the way for this one. I would also like to thank Katja MRUCK for her help and kindness, Peter MANDAVILLE and Véronique MOTTIER for their comments and to especially thank Matteo GIANNI for our constructive discussions.


1) President Woodrow WILSON's fourteen points justified ideologically the US entry in WWI. It is a classical illustration of an ought-to-be perspective in IR. <back>

2) It might be arguable, however, that small parts of the natural sciences (in its most encompassing acceptation) deal with self-conscious and reflexive subjects (such as the primatologists with the pigmy chimps Bonobo for instance). <back>

3) See for instance the different interdisciplinary books edited in the past five years by Amy MANDELKER (1995), Michael BELL and Michael GARDINER (1998) and Peter HITCHCOCK (1998). <back>

4) The following paragraphs are a synthesis of the dialogical approach I presented in Xavier GUILLAUME (2002). <back>

5) Julia KRISTEVA introduced a distinction between intertextuality, the relation of any utterance with other utterances, and dialogism in order to differentiate BAKHTIN's literary theory from his anthropological philosophy (see KRISTEVA 1986a, pp.34-61; 1986b, pp.109-112). <back>

6) Richard LEBOW (2000, pp.567-574) proposes a very strong criticism of FERGUSON's overall use of counterfactuals. These criticisms are justified especially as FERGUSON's historical restrictions would "exclude entire categories of plausible-world counterfactuals" (p.569). The perspective that I propose hereby is nonetheless still consistent with LEBOW's criteria (pp.581-585) as I try to adapt counterfactuals to a use specific to more interpretative perspectives in IR theory and with a precise epistemological articulation. <back>


Adler, Emanuel (1997). Seizing the Middle Ground: Constructivism in International Relations. European Journal of International Relations, 3(3), 319-363.

Agazzi, Evandro (1988). L'objectivité scientifique. In Evandro Agazzi (Ed.), L'objectivité dans les différentes sciences (pp.13-25). Fribourg: Editions Universitaires.

Alvesson, Mats & Sköldberg, Kaj (2000). Reflexive Methodology. New Vistas for Qualitative Research. London: Sage.

Amorim, Marilia (1996). Dialogisme et altérité dans les sciences humaines. Paris: L'Harmattan.

Ashley, Richard K. (1989). Living on Border Lines: Man, Poststructuralism, and War. In James Der Derian & Michael J. Shapiro (Eds.), International/Intertextual Relations (pp.259-321). Lexington: Lexington Books.

Bagby, Laurie M.J. (1994). The Use and Abuse of Thucydides in International Relations. International Organization, 48(1), 131-153.

Bakhtin, Mikhail M. (1981). Discourse in the Novel. In Michael Holquist (Ed.), The Dialogic Imagination (pp.259-422). Austin: University of Texas Press (original text written in 1934-1935).

Bakhtin, Mikhail M. (1984). Toward a Reworking of the Dostoevsky Book. In Caryl Emerson (Ed.), Problems of Dostoevsky's Poetics (pp.283-302). Minneapolis-London: University of Minnesota Press (original text written in 1961).

Bakhtin, Mikhail M. (1986a). The Problem of Speech Genre. In Caryl Emerson & Michael Holquist (Eds.), Speech Genres and Other Late Essays (pp.60-102). Austin: University of Texas Press (original text written in 1952-53).

Bakhtin, Mikhail M. (1986b). The Problem of the Text in Linguistics, Philology, and the Human Sciences: An Experiment in Philosophical Analysis. In Caryl Emerson & Michael Holquist (Eds.), Speech Genres and Other Late Essays (pp.103-131). Austin: University of Texas Press (original text written in 1959-61).

Bakhtin, Mikhail M. (1986c). Toward a Methodology for the Human Sciences. In Caryl Emerson & Michael Holquist (Eds.), Speech Genres and Other Late Essays (pp.159-172). Austin: University of Texas Press (original text written in 1974).

Bakhtin, Mikhail M. (1986d). From Notes made in 1970-71. In Caryl Emerson & Michael Holquist (Eds.), Speech Genres and Other Late Essays (pp.132-158). Austin: University of Texas Press (original text written in 1970-1971).

Bakhtin, Mikhail M. (1990). Author and Hero in Aesthetic Activity. In Michael Holquist & Vadim Liapunov (Eds.), Art and Answerability (pp.4-256). Austin: University of Texas Press (original text written in 1920-23).

Bakhtin, Mikhail M. (1994). Discourse in Life and Discourse in Art. In Pam Morris (Ed.), The Bakhtin Reader (pp.161-174). London: Arnold (original text written in 1926).

Bell, Michael & Gardiner, Michael (Eds.) (1998). Bakhtin and the Human Sciences. London: Sage.

Blaney, David L. & Inayatullah, Naeem (1994). Prelude to a Conversation of Cultures in International Society? Todorov and Nandy on the Possibility of Dialogue. Alternatives, 19(1), 23-51.

Bourdieu, Pierre & Wacquant, Loïc (Eds.) (1992). Réponses. Pour une anthropologie réflexive. Paris: Seuil.

Bourdieu, Pierre (1992a). Penser relationnellement. In Pierre Bourdieu & Loïc Wacquant (Eds.), Réponses. Pour une anthropologie réflexive (pp.196-206). Paris: Seuil.

Brown, Chris (2001). Understanding International Relations. Second Edition. London: Palgrave.

Brubaker, Rogers & Cooper, Frederick (2000). Beyond "identity." Theory and Society, 29(1), 1-47.

Carr, Edward H. (1939/1964). The Twenty Years Crisis 1919-1939. An Introduction to the Study of International Relations. New York: Harper & Row.

Dop, Erik (2000). A Dialogic Epistemology: Bakhtin on Truth and Meaning. Dialogism, 4, 7-33.

Doty, Roxanne L. (1996). The Logic of Différance in International Relations. U.S. Colonization of the Philippines. In Francis A. Beer & Robert Hariman (Eds.), Post-Realism (pp.331-345). East Lansing: Michigan State University Press.

Emirbayer, Mustafa (1997). Manifesto for a relational sociology. American Journal of Sociology, 103(2), 281-317.

Foucault, Michel (1969). L'archéologie du savoir. Paris: Gallimard.

Ferguson, Niall (1997). Virtual History: Towards a "Chaotic" Theory of the Past. In Niall Ferguson (Ed.), Virtual History. Alternatives and Counterfactuals (pp.1-90). London: Papermac.

Flynn, Thomas (1994). Foucault's Mapping of History. In Gary Gutting (Ed.), The Cambridge Companion to Foucault (pp.28-46). Cambridge: Cambridge University Press.

Gianni, Matteo (1998). Taking Multiculturalism Seriously: Political Claims for a Differentiated Citizenship. In Karen Slawner & Mark E. Denham (Eds.), Citizenship after Multiculturalism (pp.33-55). New York: Peter Lang.

Guillaume, Xavier (2002). Foreign Policy and the Politics of Alterity: A Dialogical Understanding of International Relations. Millennium: Journal of International Studies, 31(1), 1-26.

Guzzini, Stefano (2000). A Reconstruction of Constructivism in International Relations. European Journal of International Relations, 6(2), 147-182.

Hall, Rodney B. (1999). National Collective Identity. Social Constructs and International Systems. New York: Columbia University Press.

Hitchcock, Peter (Ed.) (1998). Bakhtin/"Bakhtin": Studies in the Archive and Beyond [Special Issue]. The South Atlantic Quarterly, 97 (3/4).

Holden, Gerard (2002). Who Contextualizes the Contextualizers? Disciplinary History and the Discourse about IR Discourse. Review of International Studies, 28(2), 253-270.

Inayatullah, Naeem & Blaney, David L. (1996). Knowing Encounters: Beyond Parochialism in International Relations Theory. In Yosef Lapid & Friedrich Kratochwil (Eds.), The Return of Culture and Identity in IR Theory (pp.65-84). Boulder: Lynne Rienner Publishers.

Katzenstein, Peter J. (Ed.) (1996). The Culture of National Security. Norms and Identity in World Politics. New York: Columbia University Press.

Keohane, Robert O. (1986). Realism, Neorealism and the Study of World Politics. In Robert O. Keohane (Ed.), Neorealism and its Critics (pp.1-26). New York: Columbia University Press.

Kratochwil, Friedrich (1996). Citizenship: On the Border of Order. In Yosef Lapid & Friedrich Kratochwil (Eds.), The Return of Culture and Identity in IR Theory (pp.201-222). Boulder: Lynne Rienner Publishers.

Kristeva, Julia (1986a). Word, Dialogue and Novel. In Toril Moi (Ed.), The Kristeva Reader (pp.34-61). New York: Columbia University Press (original text written in 1966).

Kristeva, Julia (1986b). Revolution in Poetic Language. In Toril Moi (Ed.), The Kristeva Reader (pp.89-136). New York: Columbia University Press (original text written in 1974).

Lakatos, Imre (1970). Falsification and the Methodology of Scientific Research Programmes. In John Worrall & Gregory Currie (Eds.), The Methodology of Scientific Research Programmes (pp.8-101). Cambridge: Cambridge University Press.

Lapid, Yosef (1989). The Third Debate: On the Prospects of International Theory in a Post-Positivist Era. International Studies Quarterly, 33(3), 235-254.

Lapid, Yosef & Kratochwil, Friedrich (Eds.) (1996). The Return of Culture and Identity in IR Theory. Boulder: Lynne Rienner Publishers.

Lebow, Richard N. (2000). What's so Different About a Counterfactual? World Politics, 52(4), 550-585.

Lebow, Richard N. (2001). Thucydides the Constructivist. American Political Science Review, 95(3), 547-560.

Levy, Jack S. (2001). Explaining Events and Developing Theories: History, Political Science, and the Analysis of International Relations. In Colin Elman & Miriam Fendius Elman (Eds.), Bridges and Boundaries. Historians, Political Scientists, and the Study of International Relations (pp.39-83). Cambridge: The MIT Press.

Mandelker, Amy (Ed.) (1995). Bakhtin in Contexts. Across the Disciplines. Evanston: Northwestern University Press.

Melucci, Alberto (1995). The Process of Collective Identity. In Hank Johnston & Bert Klandermans (Eds.), Social Movements and Culture (pp.41-63). London: UCL Press.

Meyer, Michel (1986). De la problématologie. Philosophie, science et langage. Paris: Pierre Mardaga.

Morgenthau, Hans J. (1967). Politics among Nations. The Struggle for Power and Peace. Fourth Edition, Revised and Reset. New York: Alfred A. Knopf.

Nadeau, Robert (1999). Vocabulaire technique et analytique de l'épistémologie. Paris: Presses Universitaires de France.

Neufeld, Mark (1995). The Restructuring of International Relations Theory. Cambridge: Cambridge University Press.

Neumann, Iver B. (1996). Self and Other in International Relations. European Journal of International Relations, 2(2), 139-174.

Ringmar, Erik (1996). Identity, interest and action. A cultural explanation of Sweden's intervention in the Thirty Years War. Cambridge: Cambridge University Press.

Russett, Bruce with the collaboration of Antholis, William; Ember, Carol R.; Ember, Melvin & Maoz, Zeev (1993). Grasping the Democratic Peace. Principles for a Post-Cold War World. Princeton: Princeton University Press.

Shapiro, Michael J. (1989). Representing World Politics: The Sport/War Intertext. In James Der Derian & Michael J. Shapiro (Eds.), International/Intertextual Relations (pp.69-96). Lexington: Lexington Books.

Spegele, Roger D. (1996). Political Realism in International Theory. Cambridge: Cambridge University Press.

Tetlock, Philip E. & Belkin, Aaron (1996). Couterfactual Thought Experiments in World Politics. Logical, Methodological, and Psychological Perspectives. In Philip E. Tetlock & Aaron Belkin (Eds.), Counterfactual Thought Experiments in World Politics (pp.3-38). Princeton: Princeton University Press.

Thies, Cameron G. (2002). Progress, History and Identity in International Relations Theory: The Case of the Idealist-Realist Debate. European Journal of International Relations, 8(2), 147-185.

Todorov, Tzvetan (1981). Mikhaïl Bakhtine. Le principe dialogique suivi de Écrits du Cercle de Bakhtine. Paris: Seuil.

Todorov, Tzvetan (1982). La conquête de l'Amérique. Paris: Seuil.

Walker, Robert B. J. (1993). Inside/outside: International as Political Theory. Cambridge: Cambridge University Press.

Waltz, Kenneth N. (1979). Theory of International Politics. New York: McGraw-Hill.

Weber, Cynthia (2001). International Relations Theory. A Critical Introduction. London: Routledge.

Weber, Steven (1996). Counterfactuals, Past and Future. In Philip E. Tetlock & Aaron Belkin (Eds.), Counterfactual Thought Experiments in World Politics (pp.268-288). Princeton: Princeton University Press.

Wendt, Alexander (1992). Anarchy is what States Make of it: The Social Construction of Power Politics. International Organization, 46(2), 391-425.

Wendt, Alexander (1999). Social Theory of International Politics. Cambridge: Cambridge University Press.


Xavier GUILLAUME is a teaching-research assistant/Ph.D. student in the Department of Political Science, University of Geneva, Switzerland. He holds a MA in International History and Politics from the Graduate Institute of International Studies (Geneva) and a MA in Political Sciences from the University of Geneva. His research interests include International Relations theory, epistemology, the question of identity and alterity, mythopoietic construction in politics, and Italian fascism.


Xavier Guillaume

Department of Political Science
University of Geneva
40 Bd. du Pont-d'Arve
CH-1211 Genève 4

Phone: (41) 22 705 83 74
Fax: (41) 22 705 83 64

E-mail: Xavier.Guillaume@politic.unige.ch


Guillaume, Xavier (2002) Reflexivity and Subjectivity: A Dialogical Perspective for and on International Relations Theory [41 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 3(3), Art. 13, http://nbn-resolving.de/urn:nbn:de:0114-fqs0203133.

Revised 2/2007

Copyright (c) 2002 Xavier Guillaume

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.