¹ In particular, the study of international institutions. See Keohane, Robert O., ‘International Institutions: Two Approaches’, International Studies Quarterly, 32, no. 4 (1988), pp. 379–96CrossRefGoogle Scholar. It should be noted that what in this essay are termed the ‘interpretive’ and ‘positivist’ approaches, Keohane refers to as ‘reflective’ and ‘rationalist’ approaches, respectively.

² This essay will draw most specifically on the contribution of Charles Taylor whose germinal piece, ‘Interpretation and the Sciences of Man’, served to inspire the title and much of the content of this paper. Taylor's piece has been reprinted in Rabinow, P. and Sullivan, W. M. (eds.), Interpretive Social Science: A Second Look (Berkeley, 1987), pp. 33–81Google Scholar. Taylor's work is ideally suited to the main objective of this paper-i.e., to underscore what is unique to an interpretive approach to the study of the social world-in that it is both accessible to the uninitiated as well as representative of interpretive social science more generally.

³ Also sometimes referred to as ‘hermeneutics’.

⁴ On this point, see Dallmayr, Fred R. and McCarthy, Thomas A. (eds.), Understanding and Social Inquiry (London, 1977), pp. 77–80Google Scholar. The positivist philosopher Otto Neurath has equated Verstehentechniques with a ‘cup of coffee’: something which might increase the ‘serendipity’ of the social scientist, but which has no place in empirical work. Dallymayr and McCarthy, Understanding and Social Inquiry, p. 6.

⁵ For example, in designing questionnaires, guiding the researcher in the interview process, etc.

⁶ McCarthy, Thomas A., The Critical Theory of Jiirgen Habermas (Cambridge, MA, 1978), p. 153Google Scholar.

⁷ It can, of course, be argued that ‘meaning-oriented behaviouralists’ have brought ‘subjective meanings’ into the public realm only by operationalizing those meanings in terms of specific forms of behaviour-i.e., a specific opinion on an issue is operationalized in terms of a specific response to a question in an opinion survey. From this perspective, what is being correlated is not a ‘subjective meaning’ with a specific behaviour, but rather one type of behaviour (e.g., the response to an opinion survey) with another (e.g., a vote in an election).

⁸ Paul Rabinow and William M. Sullivan, ‘The Interpretive Turn’, in Rabinow and Sullivan, Interpretive Social Science, p. 7.

⁹ Charles Taylor, ‘Interpretation and the Sciences of Man, ’ p. 46. This is not to say that human beings are always fully cognizant of their participation in this on-going self-definition and self-interpretation process. Rather, in ‘Giddens’ terms, the activity of self-definition and self- interpretation often takes place at the level of ‘practical consciousness’ (and not the more explicitly self-consciousness level of ‘discursive consciousness’). See Giddens, Anthony, The Constitution of Society: Outline of a Theory of Structuration (Cambridge, 1984), p. 375Google Scholar.

¹⁰ For example, the regularities in the interaction of molecules which serves as the focus of chemistry need not be considered in relations to any ‘self-interpretations’ or ‘self-definitions’ of those molecules.

¹¹ Taylor, ‘Interpretation and the Sciences of Man’, p. 48.

¹² ‘Interpretation and the Sciences of Man’, p. 58.

¹³ ‘Interpretation and the Sciences of Man’, p. 59.

¹⁴ ‘Interpretation and the Sciences of Man’, p. 59.

¹⁵ ‘Interpretation and the Sciences of Man’, p. 59. As Taylor notes, one particularly promising way of conceptualizing ‘intersubjective meanings’, the full implications of which are beyond the scope of this paper, is as ‘rules’ having both normative and constitutive effect. Once again, Anthony Giddens has made an important contribution in this regard. See his discussion of ‘structuration theory’ in New Rules of Sociological Method: A Positive Critique of Interpretive Sociologies (London, 1976)Google Scholar. For a useful discussion and critique of Giddens’ contribution, see Thompson, John B., ‘The Theory of Structuration’, in Held, D. and Thompson, J. B. (eds.), Social Theory of Modern Societies: Anthony Giddens and His Critics (Cambridge, 1989).Google Scholar

¹⁶ ‘Interpretation and the Sciences of Man’, p. 57.

¹⁷ ‘Interpretation and the Sciences of Man’, p. 52.

¹⁸ ‘Interpretation and the Sciences of Man’, p. 53.

¹⁹ Rabinow and Sullivan, ‘The Interpretive Turn’, p. 6.

²⁰ Rabinow and Sullivan, ‘The Interpretive Turn’, p. 6.

²¹ Rabinow and Sullivan, ‘The Interpretive Turn’, p. 6. The work of Hans-Georg Gadamer has particular relevance to this point. For a good introduction to his work in this regard, see Georgia Warlike, Gadamer: Hermeneutics, Tradition, and Reason (Stanford, California: 1987).

²² ‘The Interpretive Turn’, p. 7 Rabinow and Sullivan are quoting Taylor. Or again, as Taylo r notes, ‘It is not just that all or most people in our society have a given set of ideas in their heads and subscribe to a given set of goals. The meanings and norms implicit in these practices are not just in the minds of the actors but are out there in the practices themselves, practices which cannot be conceived as as set of individual actions, but which are essentially modes of social relation, of mutual action’ (‘Interpretation and the Sciences of Man’, pp. 56–7).

²³ Taylor, ‘Interpretation and the Sciences of Man’, p. 43.

²⁴ It should, of course, be emphasized that to ‘make sense’ of some behaviour in no way implies that the behaviour is rational. On the other hand, as Taylor notes, ‘even contradictory irrational action is ‘made sense of, when we understand why it was engaged in’. (‘Interpretation and the Sciences of Man’, p. 43).

²⁵ From the perspective of the ‘hermeneutic circle’, then, the proper analogy for the methodology of an interpretive approach is not the method of physics (the subsumption of empirical regularities under covering laws, nor that of Dilthey's ‘romantic hermeneutics’ (empathic identification), but rather the learning of a second language (the ‘web of meaning’ which constitutes observed social practices) and then the translation of that language into one's mother tongue (the concepts of the social scientist). For a discussion of hermeneutics not as ‘empathy’ but as a form of ‘translation’, see Gadamer, Hans-Georg, Truth and Method (New York, 1988)Google Scholar.

²⁶ Though it might be argued that the formulations of some interpretive theorists-starting with Dilthey-have tended in this direction. For a critique of this more limited notion of an interpretive approach to the social world, see Fay, Brian, Social Theory and Political Practice (London, 1975)Google Scholar, ch. 4. For a response which argues that Fay's critique does not apply to all forms of interpretive social science-in particular, to that proposed by Taylor-see, CharlesGibbons, Michael, ‘Introduction: The Politics of Interpretation’, in Gibbons, Michael (ed.), Interpreting Politics (Oxford, 1987)Google Scholar.

²⁷ I have placed the word ‘recovering’ in inverted commas to underscore the problematic nature of the ‘recovery’ process. Specifically, the notion of ‘recovering’ the intersubjective meanings particular to a community of human agents is problematic for at least two reasons. First, though such meanings are always the product of an on-going process of self-interpretation and self-definition, as was noted above, that process is rarely one about which human agents are reflexively self-conscious. As a consequence, the human agents themselves may be incapable of articulating the ‘intersubjective meanings’ which constitute their practices. To ‘recover’ unarticulated ‘intersubjective meanings’ is then no simple and straightforward activity.

Secondly, the notion of ‘recovery’ must also be nuanced through the recognition that all acts of ‘recovery’ involve interpretation. That is, the ‘reading‘ of a specific ‘web of meaning’ and social practice is, in keeping with the reality of the ‘double hermeneutic’, expressed in the language and terms of the social scientist. And because there is no way to escape the ‘hermeneutic circle’-because there is never any way to establish the validity of a particular reading beyond any doubt-every ‘reading’, no matter how plausible or sophisticated, remains potentially contestable.

In short, an interpretive approach does not alter in any way the sense in which a ‘science’ of international politics is ‘interpretive’ in the way that all scientific activity is ‘interpretive’-i.e., all scientific activity involves the ‘interpretation’ of data in terms of paradigm-specific conventions about what constitutes ‘valid knowledge’. In other words, the focus on the ‘intersubjective meanings’ which constitute social practices is no escape from the ‘Cartesian anxiety’; i.e., that should objective standards for truth and knowledge not exist, we are left with chaos and madness. On the nature of the ‘Cartesian anxiety’, see Bernstein, Richard, Beyond Objectivism and Relativism (Philadelphia, 1985)Google Scholar. For an argument which does advocate the incorporation of ‘intersubjective meanings’ into international relations theory as a means of getting around the problems posed by the paradigm-determined nature of all knowledge, see Roger Tooze, ‘Economic Belief Systems and Understanding International Relations’, in Little, R. and Smith, S. (eds.), Belief Systems and International Relations (Oxford, 1988), pp. 132–3Google Scholar.

²⁸ ‘Interpretation and the Sciences of Man’, p. 59 (my emphasis).

²⁹ ‘Interpretation and the Sciences of Man’, p. 46 (my emphasis).

³⁰ Because an interpretive approach focuses, first and foremost, on ‘intersubjective meanings’, there is no difficulty in accommodating the need to examine, for example, unintended consequences of human behaviour or structural dimensions of human interaction (as there would be were ‘subjective meanings’ the exclusive focus).

It should be noted that although Taylor's work is being privileged in this article, contemporary interpretive social science draws on and has been influenced by at least five distinct traditions, including (i) the tradition of phenomenology/ethnomethodology as developed by Husserl and Schutz; (ii) the ‘linguistic tradition’ as developed by the later Wittgenstein and Winch (and into which Taylor's work falls); (iii) the hermeneutic tradition as developed by Heidegger and Gadamer; (iv) the tradition of critical theory as developed by Marx and Habermas; and (v) the tradition of genealogy as developed by Nietzsche and Foucault. Despite their differences, thefivetraditions do share the commonality of seeing the dimension of ‘meaning’ as both ‘intersubjective’ in nature and ‘constitutive’ of social reality. For useful introductions to the various traditions represented in contemporary interpretive social science, see Gibbons, Interpreting Politics; Dallmayr and McCarthy, Understanding and Social Science; and Rabinow and Sullivan, Interpretive Social Science: A Second Look.

For a critical discussion of the distinctive characteristics of Taylor's approach-an approach Gibbons terms ‘Critical-Expressivism’-see Gibbons, ‘Introduction: The Politics of Interpretation’, in Gibbons (ed.), Interpreting Politics. For an interesting comparison of this approach with that of genealogy, see Gibbons, Michael T., ‘Interpretation, Genealogy and Human Agency’, in Ball, T. (ed.), Idioms of Inquiry: Critique and Renewal in Political Science (Albany, 1987).Google Scholar

³¹ Taylor, ‘Interpretation and the Sciences of Man’, p. 47. For a good discussion of the place of the ‘hermeneutics of recovery’ in interpretive social science, see Michael Gibbons, ‘Introduction’, in Gibbons (ed.), Interpreting Politics.

³² Politics Among Nations, 5th edn (New York, 1973), p. 4Google Scholar.

³³ Politics Among Nations, p. 5.

³⁴ ‘The Study of International Relations’, in Higgott, Richard (ed.), New Directions in International Relations? Australian Perspectives (Canberra, 1988), p. 92Google Scholar.

³⁵ George, ‘The Study of International Relations’, p. 92.

³⁶ It should be noted that George is not the only one to identify a tension in Morgenthau. For a parallel interpretation which sees the same kind of tension in Morgenthau's work-this time expressed as a conflict between a hermeneutically inspired ‘practical realism’ and a positivist ‘technical realism’-see Ashley, Richard, ‘Political Realism and Human Interests’, International Studies Quarterly, 25, no. 2 (1981), pp. 204–35CrossRefGoogle Scholar.

³⁷ George, ‘The Study of International Relations’, p. 93.

³⁸ This is not to say that Morgenthau is consistently positivistic in his approach to the study of international relations. I would agree with Ashley (see note 36 above) that Morgenthau's positivist ‘technical realism’ is somewhat mediated by his hermeneutic ‘practical realism’. However, contrary to Ashley (and George), I do not think that Morgenthau's incorporation of Verstehen into his work is evidence for the ‘practical’, non-positivist dimension of his approach.

³⁹ Morgenthau, Politics Among Nations, p. 4.

⁴⁰ See Vasquez, John, The Power of Power Politics: A Critique (New Brunswick, 1983)Google Scholar.

⁴¹ For an overview of the literature of the early wave of systems theory, see Sullivan, Michael P., International Relations: Theories and Evidence (Englewood Cliffs, 1976)Google Scholar, part II. The most sophisticated and elegant contemporary contribution to ‘systems theory’ is, of course, Waltz's, KennethTheory of International Politics (New York, 1979)Google Scholar.

⁴² Steve Smith, ‘Belief Systems and the Study of International Relations’, in Little and Smith (eds.), Belief Systems and International Relations, p. 16.

⁴³ Morgenthau, Politics Among Nations, p. 5.

⁴⁴ ‘Strategic Beliefs, Mythology and Imagery’, in Little and Smith (eds.), Belief Systems and International Relations, p. 140.

⁴⁵ As Smith has noted, the dual focus of the discipline was manifest in Waltz's analysis of the causes of war, as well as serving as the subject of a classic article by Singer, in which he spoke of the ‘level of analysis problem’ in international relations theory (Smith, ‘Belief Systems and the Study of International Relations’, p. 16). See Waltz, Kenneth, Man, The State, and War: A Theoretical Analysis (New York, 1959)Google Scholar, and Singer, J. D., ‘The Level of Analysis Problem in International Relations’, in Knorr, K. and Verba, S. (eds.), The International System: Theoretical Essays (Princeton, 1961), pp. 77–92Google Scholar.

⁴⁶ See Boulding, Kenneth, The Image (Ann Arbor, 1956)CrossRefGoogle Scholar.

⁴⁷ See Holsti, Ole, ‘The Belief System and National Images: A Case Study’, Journal of Conflict Resolution, 6 (1962), pp. 244–52CrossRefGoogle Scholar.

⁴⁸ See George, Alexander, ‘The ‘Operational Code’: A Neglected Approach to the Study of Political Leaders and Decision-Making’, International Studies Quarterly, 13 (1969), pp. 190–222CrossRefGoogle Scholar.

⁴⁹ See Axelrod, Robert (ed.), Structure of Decision (Princeton, 1976)Google Scholar.

⁵⁰ See Ernest May, ‘Lessons’ of the Past (New York, 1973). See also, May, Ernest and Neustadt, Richard, Thinking in Time (New York, 1986).Google Scholar

⁵¹ As Steve Smith notes, the ‘Brecher Research Design’ involves a research framework based on the relationship between beliefs and information not dissimilar from the other approaches noted. It is based on the concept of an analytical foreign policy system, the core of which is the postulated ‘linkage between the psychological environment of the decision-makers and the operational environment in which decisions are implemented’ (‘Belief Systems and the Study of International Relations’, p. 26). The operative assumption which underlies this approach is that: To the extent that decision-makers perceive the operational environment accurately their foreign policy acts may be said to be rooted in reality and are thus likely to be ‘successful’. To the extent that their images are inaccurate policy choices will be ‘unsuccessful’. See Brecher, Michael, Steinberg, Blema and Stein, Janice, ‘A Framework for Research on Foreign Policy Behaviour’, Journal of Conflict Resolution, 13 (1969), p. 81CrossRefGoogle Scholar. See also S. Smith, ‘Belief Systems and the Study of International Relations,’ pp. 25–7.

⁵² For an excellent overview of the distinctive characteristics of these different approaches and techniques, as well as references, see Smith, ‘Belief Systems and the Study of International Relations’, pp. 17–27.

⁵³ The term is A. N. Oppenheim's. See his ‘Psychological Processes in World Society’, in Banks, M. (ed.), Conflict in World Society (Brighton, 1984), p. 114Google Scholar.

⁵⁴ Smith, ‘Belief Systems and the Study of International Relations’, p. 11.

⁵⁵ ‘Belief Systems and the Study oflnternational Relations’, p. 16.

⁵⁶ ‘International Institutions’, p. 381. In addition to the interpretive efforts undertaken by Ruggie, Kratochwil, Alker, Ashley, Haas and Cox, one might also note those of Nicholas Onuf, Alexander Wendt, and Raymond Duvall. See Onuf, Nicholas G., World of Our Making: Rules and Rule in Social Theory and International Relations (Columbia, 1989)Google Scholar; Wendt, Alexander, ‘The Agent-Structure Problem in International Relations Theory’, International Organization, 41, no. 3 (1987), pp. 335–70CrossRefGoogle Scholar; as well as Wendt, Alexander and Duvall, Raymond, ‘Institutions and International Order’, in Rosenau, J. and Czempiel, E. (eds.), Global Changes and Theoretical Challenges (Toronto, 1989)Google Scholar.

⁵⁷ ‘International Institutions’, p. 379.

⁵⁸ Krasner, Stephen D., ‘Structural Causes and Regime Consequences: Regimes as Intervening Variables’, in Krasner, S. (ed.), International Regimes (Ithaca, 1983), p. 2Google Scholar. The definition continues: Principles are beliefs of fact, causation, and rectitude. Norms are standards of behaviour defined in terms of rights and obligations. Rules are specific prescriptions for action. Decision-making procedures are prevailing practices for making and implementing collective choice.

⁵⁹ For a representative sampling, see Krasner, International Regimes, as well as Oye, Kenneth A. (ed.), Cooperation Under Anarchy (Princeton, 1985)Google Scholar.

⁶⁰ Kratochwil, Friedrich and Ruggie, John Gerard, ‘International Organization: A State of the Art on the Art of the State’, International Organization, 40 (1986), p. 764CrossRefGoogle Scholar (my emphasis).

⁶¹ ‘International Organization’, p. 771. Nor are they daunted by predictable attempts by at least some members of the international relations community to discredit interpretive approaches: Interpretive epistemologies that stress the intimate relationship between validation and the uncovering of intersubjective meanings are simply too well developed today to be easily dismissed by charges of subjectivism-or, more likely in the arena of international relations theory, of idealism, (p. 765)

⁶² For example, Kratochwil and Ruggie argue that by treating the norms which are an integral part of regimes as intersubjective elements of a ‘web of meaning’ which form the ‘constitutive basis of regimes’-instead of the more common practice of treating them as ‘causal variables’ determining behaviour-regimes can be seen to be vibrant and robust even when norms are ignored in specific instances. See the discussion by Kratochwil, in ‘Regimes, Interpretation and the ‘Science’ of Politics: A Reappraisal’, Millennium 17, no. 2 (1988), pp. 277–8CrossRefGoogle Scholar.

Similarly, they argue that recognizing the intersubjective nature of the ‘principles’, ‘norms’, etc. which constitute regimes as social practices (and not as timeless behavioural regularities) also provides a means of theorizing the change within, and of, regimes. In short, regimes are subject to change because they are the product of an on-going process of community self-interpretation and self-definition in response to changing context. As such, an interpretive approach to the study of regimes avoids positivism's problematic assumption that ‘once the machinery is in place, actors merely remain programmed by it’ (‘International Organization’, p. 770).

It is lamentable-if not surprising given the traditional antipathy in the discipline to meta-theoretical questions-that in a recent re-printing of Kratochwil and Ruggie's ‘International Organization’ the discussion of the need for interpretive methodologies has been deleted. See Diehl, Paul F. (ed.), The Politics of International Organizations (Chicago, 1989), pp. 17–27Google Scholar.

⁶³ Kratochwil and Ruggie, ‘International Organization’, p. 774.

⁶⁴ Nor is this the only difficulty in their treatment of interpretive social science. Even as they are promoting interpretive social science as a distinct alternative to positivism, for example, the formulation of regimes as intersubjective in nature because they involve ‘converging expectations’ comes dangerously close to confusing the important distinction, noted earlier by Taylor, between intersubjective meanings and consensus in terms of subjective meanings. Similarly, in an earlier piece, Kratochwil seems to link an interpretive approach to an analysis of the ‘background of intentions’, rather than the intersubjective meanings which make subjective intentions possible in the first place. See Kratochwil, Friedrich, ‘Errors have Their Advantage’, International Organization, 38, no. 2, (1984), p. 319CrossRefGoogle Scholar. It should be noted that Kratochwil has revised his position in a fashion more in line with interpretive social science in more recent contributions.

⁶⁵ E.g., liberal trade and monetary regimes.

⁶⁶ Hollis, Martin and Smith, Steve, Explaining and Understanding International Relations (Clarendon Press, 1990)Google Scholar.

⁶⁷ A formulation with strong affinities to that of Charles Taylor.

⁶⁸ Explaining and Understanding International Relations, p. 179.

⁶⁹ Explaining and Understanding International Relations, p. 180.

⁷⁰ Explaining and Understanding International Relations, p. 180.

⁷¹ It is even possible that with this latest turn in mainstream interpretive theory, there may be hope for a re-discovery and re-valuation of the contributions of those classical realists who-in contrast to neo-realists-saw even conflictual, coercive forms of state interaction such as the ‘balance of power’ as ‘social institutions’, comprised of rules and roles, and serving to regulate the conflict-prone ‘anarchical society’ of states. The contribution of Hedley Bull-in particular, his The Anarchical Society (New York, 1977)Google Scholar is perhaps the prime example here, though one might read Morgenthau's comments on balance of power as a moral consensus (Politics Among Nations, ch. 14) as an attempt to draw attention to the intersubjective meanings underlying the practices which together comprise the ‘balance of power’.

⁷² Wight, Martin, ‘Why is there no International Theory?’ in Butterfield, H. and Wight, M. (eds.), Diplomatic Investigations (London, 1966), p. 26Google Scholar (my emphasis).

⁷³ Young, Oran R., International Cooperation (Ithaca, 1989), p. 14Google Scholar.

⁷⁴ Cox, , ‘Social Forces, States and World Orders: Beyond International Relations Theory’, in Keohane, Robert O. (ed.), Neorealism ami Its Critics (New York, 1986), p. 218Google Scholar. Radical interpretive theorists have traditionally worked to show how the intersubjective meanings which underpin the global order have a fundamental class content. Most recently, however, feminist theorists have shown how these intersubjective meanings-and the social practices they constitute-are also inherently gendered in nature. See, for example, Enloe, Cynthia, Bananas, Beaches, & Bases: Making Feminist Sense of International Politics (London, 1989Google Scholar, as well as worth, Sandra Whit, ‘Gender in the Inter-Paradigm Debate’, Millennium, 18, no. 2 (1989), pp. 265–72CrossRefGoogle Scholar.

⁷⁵ Of course, while the possibility of fundamental change may serve as an antidote to positivist/realist-inspired pessimism about the possibility of progress-a pessimism grounded in the assumption of the essentially unchanging (and unchangeable) nature of international politics-it is not sufficient, in and of itself, to dispel all forms of pessimism. For example, it is quite possible to accept the potential for changing existing social practices while maintaining that efforts to effect progress through change of social practices inevitably make things worse. Similarly, one can argue that the recognition of the possibility of change does not, on its own, necessarily lead to emancipation, in that the latter also requires an independent, normative judgement about the way the world should be. For these two points I am indebted to Ruth Abbey and Nicholas Onuf respectively.

⁷⁶ Habermas, Jurgen, Knowledge and Human Interests, trans. Shapiro, Jeremy J. (Boston, 1971), p. 310Google Scholar.

⁷⁷ ‘International Institutions’, p. 381.

⁷⁸ On the distinction between ‘traditional’ and ‘critical’ theory, see Horkheimer, Max, ‘Traditional and Critical Theory’, in Horkheimer, , Critical Theory: Selected Essays (New York, 1989), pp. 188–243Google Scholar.

⁷⁹ It is ironic, in this regard, that one of the most insightful analyses of the way intersubjective meanings constitute not only regulatory institutions within the global order but the global order itself is to be found in Ruggie's discussion of the shift from the medieval to the modern world system. See John Gerard Ruggie, ‘Continuity and Transformation in the World Polity: Toward a Neorealist Synthesis’, in Keohane (ed.), Neorealism ami its Critics, pp. 131–57. It is not clear how Ruggie reconciles his historicist sensitivity with the reification of the state/inter-state system that his realism would seem to require, and that his meta-theoretical pronouncements on the appropriate subject matter of interpretively oriented analysis reinforce.

⁸⁰ This concern can be seen quite clearly, for example, in Cox's Gramscian-inspired efforts to develop an historicist approach which focuses on the intersubjective meanings which predominate in a given context, and how changes in those meanings give rise to changes in global order. Although a full discussion of Cox's conception of hegemony in international politics is beyond the scope of this study, it is significant that what distinguishes it from the positivist-inspired neo-realist conception of hegemony is that Cox's Gramscian notion of hegemony ‘joins an ideological and intersubjective element to the brute power relationship’. (‘Postscript’, in Keohane (ed.), Neorealism ami Its Critics, p. 246). See also Cox, , ‘Gramsci, Hegemony and International Relations: An Essay in Method’, Millennium, 12, no. 2 (1983), pp. 162–75CrossRefGoogle Scholar.

⁸¹ It can also be seen in R. B. J. Walker's call for analysis of the global order which engages in a critique of the reification of that order. See Walker, R. B. J., ‘History and Structure in the Theory of International Studies’, Millennium, 18, no. 2 (1989), pp. 163–183CrossRefGoogle Scholar.

It is important to recognize, moreover, that the change in intersubjective meanings which is the product of the on-going process of collective self-interpretation and self-definition cannot be equated with a change in subjective ‘preferences’, as is suggested by Keohane. Subjective preferences may indeed change, but such a change would not result in a fundamental change as would be the case of change in intersubjective meanings of which subjective preferences are derivative. As such, Keohane's formulation of intersubjectivity as ‘preferences’ indicates again the difficulties experienced by mainstream, positivist-inspired theorists in comprehending the nature of the interpretivist challenge. See Keohane, ‘International Institutions’, p. 391.

⁸² As opposed to the ‘instrumental’ form of practice associated with ‘traditional’ theory. For a good discussion of the distinction between the two forms, see Fay, Brian, Critical Social Science: Liberation and Its Limits (Cambridge, 1987)Google Scholar, ch. 5.

⁸³ For example, Robert Keohane, operating from a Lakatosian positivist position, faults interpretive approaches for their failure to develop an adequate ‘research program that could be employed by students of world polities'.

For an ‘interpretivist’ response to and criticism of Keohane's positivistically informed discussion of interpretive approaches to international politics, see Walker, ‘History and Structure’.