It has become commonplace to talk about world politics in terms of “identity.”  Click-and-open your way through your favourite journals or press catalogues and you will find that IR is experiencing no shortage of it.  Spend more time on this exercise, and you will also find that the concept refuses to boil down to even a single set of meanings.  Given that identity can and does serve theories based on different assumptions, dealing with very different cultural, social, and political issues, and at different levels of analysis, this is not surprising. But for some, this remains a concern: with so many cooks and cookouts, the conceptual and analytical broth gets spoiled, leading to “overproduction” and the “devaluation of meaning” [1].

Concerns of this type are overblown, both in the case of identity and in general.  Concepts need no special consensus or orderly coordination to be helpful in the scholarly attempts to explain how the world operates.  And yet, it cannot hurt to pause over a concept and think about how it is used in a given academic discourse.

What, then, is with identity and IR?

If we consider “national character” as one of its predecessor, then the concept of (national? state? ethnic? etc.) identity has a very long lineage in the study of international relations.  Consider the following paragraph about The Hellenic kingdom and the Greek Nation, from a 1837 pamphlet by the Scottish historian George Finlay:

The most prominent features in the character of the Greek under every varied change in his lot, are, we think, activity of mind, general intelligence, and aptitude to comprehend and receive the mental impressions of others, inquisitiveness and a love of knowledge, joined to a strong desire for personal independence and equality. These feelings, we think, may be traced in all the provinces where the Greek language is spoken, and seem constantly to have exerted their influence on the nation. We do not pretend to deny, that many of these feelings may, and that some are often misled to evil, but still we doubt not, that every candid inquirer will be convinced, that possessing these feelings, the Greek must have a national character capable of leading him to the highest pitch of mental improvement, and the power of so modelling his institutions, that he may not only ensure his moral progress where he has already gained political independence, but must obtain also the amelioration of his moral and social condition, even where he remains subjected to a foreign yoke [2].

What made the newborn Greek state eminently viable was the fact that good governance had always been in the collective personality of “the Greek.”  This argument has few supporters at the time of this writing (what’s the G in “PIGS”?), but it must have been at least somewhat convincing at the time, otherwise Austria, Britain, France, Prussia, and Russia would have not, “talked Greece into Europe”, as Jennifer Mitzen once put it [3].

National character-based explanations of politics were used well into late twentieth century; even today, in fact, the concept (albeit in much more scientific forms) is used in ways that are both fruitful and illuminating [4].  The replacement of national character with identity began in the 1950s, catching on rather quickly in the 1970s.  Behind this shift was a combination of a number of intellectual and political forces, the main being the racist freight carried by the older concept [5].

The Work-Horse

In contemporary IR, identity arrived on the double wings of socio-linguistics and socio-political psychology sometime in the 1980s.  The so-called third debate brought the discipline closer to the concept (and to nearby discourses of subject positions, subjectivity, subjectivation and so on), as post-positivist after post-positivist went on to show how identities profoundly shape the ways in which actors construct both themselves and their interests.  What catapulted identity to the top of the agenda, however, was constructivism, a meta-theory with an “almost frightening potential,” as Ole Wæver described it at the time [6].  Identity, observed Ted Hopf in his review of Alexander Wendt’s Social Theory of International Politics, is one of the “theoretical work-horses of constructivist accounts of global politics” [7].  Within the span of a few years, one could say, the concept was thoroughly domesticated, colonized, and even commodified [8].

It must be noted that not every approach to identity is constructivist.  What drives much rationalist scholarship dealing with identity (usually reduced to “revealed preferences”) and at least some psychological theories of “identification,” is a combination of materialism and empiricism, not constructivism.  Within IR, however, the dominant definition of identity is congenial to ontologies that posit the existence of social and inter-subjective phenomena. But what causal power or at least enabling force might these phenomena have?  On this question, there is much debate even among constructivists.  As a matter of (some) constructivist ontological teachings about agents and structures, identity cannot be sliced up into different levels of analysis – a venerable IR referent for the domains of “causal” action.  Identities of states and nations, for example, are constructed both on ‘inside’ and ‘outside’ and identity-formation is irreducible even to the sum of heretofore separately analyzed ‘unit’ and ‘systematic’ variables. State/national identity, put differently, emerges from state-society relations, international institutions and transnational norms, as well as from collectively constituted notions of selfhood at the systemic level – all at once. Further, the construction is subject to various feedback loops, which is the reason why constructivists (or at least structurationists among them) underline that agents and structures such as elites and nations, societies and empires, states and the international system are always mutually constituted.

For some constructivists – let’s call them poststructuralists – the ontological message is all too clear: mutual constitution implies that social entities cannot be separated, stabilized, or bracketed, especially not for the purposes of causal analysis.  Moreover, Foucauldian purists would add, the goal of social and political inquiry is not causation anyway, but the so-called power analysis (the study of ways subject positions are constituted, including IR and its claim to scientism through causation-talk).  Identity structures action and it is in turn structured by power.

There are two relevant points to consider here: first, the concept of causation in IR is today much broader and deeper than a decade or two ago.  The Humean model of causation, still the positivist gold standard, now sits next to Aristotelian, Cartesian, Gallilean, Kantian, and Weberian models.  Most analytical pragmatists and most critical realists would probably assert that anything that produces a certain reality is causation [7].  Second and related, in a social-scientific enterprise like IR, ontology must yield to analysis at some point because no account of the social and political world can vary everything at the same time. The main reason lies in the nature of the language we use: it is successive, not simultaneous. Also, even the most anti-epistemological analysis must value some intellectually evaluative criteria based on language – spelling and grammar in the exposition, consistency of the narrative and such.

Virtually all IR studies stabilize one phenomenon (e.g., the state, nation, networks) at the expense of some other phenomenon or phenomena (foreign policy, drive for recognition, or security community-building institutions and practices).  In poststructuralist IR, such stabilizations usually appear as ‘arrestations’, ‘temporary reifications’, ‘strategic or provisional essentializations’ and other seemingly practical, pragmatic and pedagogical moves, but never as causal ones (denying that two social phenomena are causally related requires many such stabilizations, ironically). But while poststructuralists are right to argue that explaining causation cannot be divorced from understanding meaning, it may also be that any understanding of the meaning-making processes cannot proceed without some causal inferences, broadly defined [8].

Multiplicity and Intersectionality

Related to stabilization is the so-called multiplicity or the idea that identities are constituted as overlapping, flexible, plural, fluid and so on (groups and individuals change over time and place). In his analysis of Greek national character, for instance, Finlay talks about this when noting:

The extreme difficulty of pourtraying, at full length, the Greek character, must be immediately felt by anyone who reflects on the varied fate of the different portions of this singular people. Let us examine the commonest accusation current against the Greeks in the seaports of the Mediterranean. The dealers in figs and raisins generally describe the Greeks as a race of the rankest cowards. Nay, the whole Frank population of the Levant unite at least in this accusation amidst all the warlike tribes who march to battle under the eye of the predestinated Turk, the Roumeliot Greek has ever enjoyed the very highest reputation for valour. His services are sought for by the Pashas of Europe and Asia, and he is placed on a footing with the Arnaout as his equal in courage. What, too, but a respect for the courage, as well as a confidence in the fidelity of the Greek armatoli, could have induced the Turks to preserve this Christian militia for nearly four centuries?

Read from the perspective of contemporary identity theory, what the passage inimates is that national identities are emergent and dependent on the ‘Other’: the Roumeliot Greek is meaningful only in relation to the Frank, the Arnaout, the Turk, as well as to some other historical or geographical Greek.  It also implies that identities vary across a population: the Greek character, much like constructivist IR, cannot be easily generalized about.  Next, it might also suggest that people live in the concentric circles of identity, ranging from very narrow (one fig & raisin stand, one armatoli band) to very broad (men, the Levant, the Christendom).  These circles are heuristic, not hierarchic: the meaningful behaviours associated with each of these identities depend on situational constructions. In fact, following feminist theorists, the circles are intersectional: gender, professional, regional and religious identities – to name the aforementioned four – form not in isolation, but in multidimensional interaction with each other [10].

One of the many analytical challenges involved is to try and account for (the “content and contestation” of) those multiple affiliations without sneaking in a prefab theory on the constitutive essence or salience of one identity over all others.  Deciding which identities predominate in a given context (when, how, and in what proportion, under what conditions etc.) is a challenge for theory and methodology.  A major question animating constructivist IR, for example, is how identities that originate ‘inside’ state–society relations are reproduced at the level of international, transnational, intermestic, and other levels ‘outside’ the unit, where, in turn, they are consolidated, challenged, or diminished.  The reverse process is just as relevant: how identities coming from ‘abroad’ are adopted and adapted ‘at home’.  The current generation of empirical constructivist scholarship in IR tends to approach these questions from the perspective of political contestation, analyzing how ideas, policies, institutions, or practices are legitimated in terms of identity.  Here, too, there is usually much stabilization and bracketing since most solo-authored studies cannot theorize and analyze all of these dimensions of identity-formation, certainly not with the same depth.


So how might we go about finding out which identity predominates in a given context?  Once again Finlay’s analysis of the state of the Greeks cca 1830s (“a just estimate of their national character”) provides an answer.  To tell his tale of the Greek national character, Finlay says he relied on techniques which we today might describe as a mix of participant observation (“long residence in the East, and long intercourse with the Greeks”), interviews (“It has been remarked by all travellers…”), and secondary-source historical contextualization:

No rural population in Europe has ever arrived at a higher degree of civil organization, arranged their local governments better, or displayed more energy and judgment in the conduct of their municipal concerns, than the Greeks…Even the collection of the public taxes was generally transacted by their means, and the amount was thus remitted to the Turkish authorities, without that oppression which usually marked the direct communication of Turks with Greeks. Much of that strong spirit of nationality, which has ever formed a leading feature in the Greek character, and has enabled the people to transmit to the present generation some of the institutions and usages, as well as the language of the ancient republics, is to be ascribed to this system of local governments.

The “identity as a variable” project by Abdelal et al has identified half-dozen methodologies as “best suited for identity research”: discourse and content analysis, surveys, ethnography, experiments, agent-based modeling, and cognitive mapping.  One could probably add a few more categories in the spirit of methodologically pluralist revelry, but the main lesson of this collection is this: methodology always follows from theory which always follows from ontology. Put slightly differently, one must determine what exists in the world (and how minds are connected to the world) before one decides what methodologies and methods to use.  So if by identity we mean the subjective perception of individuals, then surveys and experiments could be our methods of choice.  If we define identity in terms of the stories individuals and groups tell to, and about, each other, then some form of textual analysis is expected.  And if identity is an intersubjective social structure, interpretivist methodologies like ethnography, discourse analysis, “practice-tracing” are needed.  Once the underlying assumptions and causal (or, if you still prefer, constitutive) processes are specified, the concept of identity no longer appears so mercurial and so woolly [11].

IR has found the concept identity compelling because it can be used to provide novel and relatively coherent maps of global politics, whether the focus is on nations, states, diasporas, empires, regions, or civilizations. The overall exercise, however, remains a collective effort that goes ‘beyond identity’ by default: new stories about the world always begin and end in dialogue with old stories.  Writing years before identity become widely used in the discipline, Gearóid Ó Tuathail made this point memorably:  “Questions of identity may insinuate their way into all forms of politics, but all forms of politics are not about questions of identity” [12].

Srdjan Vucetic is Assistant Professor at the Graduate School of Public and International Affairs, University of Ottawa. Parts of the ensuing discussion are adapted from his book, The Anglosphere: a Genealogy of a Racialized Identity in International Relations (Stanford, 2011).


[1] The quotations are from that bête noire of the identity-based social-scientific scholarship, the 2000 piece by Rogers Brubaker and Frederick Cooper that called for a moratorium on the use of what they judged to be a hopeless concept. Brubaker and Cooper, “Beyond ‘Identity’,” Theory and Society, 29 (2000), 1–47. There is no such field as the philosophy of identity (yet), by there is a large body of philosophical thought on the concept.  For quick overviews of the past and current definitional anarchies see Akeel Bilgrami, “Notes Toward the Definition of ‘Identity,’” Daedalus 135 (2006), 5–14. For a longer discussion, see the relevant entries in Stanford Encyclopedia of Philosophy:

[2] I am fond of nineteenth century British writing on the Balkans, I admit. Finlay embarked on his field research during the Greek War of Independence, in which he partook alongside other European philhellenes (contrary to the majority, he insisted on the discontinuity between ancient and modern Greeks, however).  This essay was prelude to The History of Greece (completed in 1861, published in the book-length form in 1877), which become a standard history on the country in the Victorian era (and not just in the UK).  Findlay’s work is still appreciated today, especially by Albanian and Macedonian nationalists, who in it manage to find impartial evidence of Alexander the Great’s non-Greek identity.

[3] Jennifer Mitzen, “Reading Habermas in Anarchy,” American Political Science Review 99 (2005), 401-417, p. 412. On the discourse of PIGS and how Greece is being talked out of Europe these days, see Panagiotis Roilos, “Economic Crisis, the ‘PIGS’ and the European ‘Animal Farm’, 12 May 2012.

[4] Peter Mandler, The English National Character: The History of an Idea from Edmund Burke to Tony Blair (Yale, 2006).

[5] See Ibid., p. 81-12, for Walter Badgehot’s conceptualization of national character, for example.

[6] Wæver “Figures of International Thought: Introducing Persons Instead of Paradigms,” Iver B. Neumann and Ole Wæver, eds, Masters in the Making? (Routledge, 1997), p. 25. Constructivist IR can’t be easily generalized about, but it builds on the core idea that reality and knowledge are socially and politically constructed, contingent, and (in principle) contestable.

[7] Ted Hopf, “Constructivism All The Way Down,” International Politics (2000) 37, 369–378, p. 369. In building his international theory, Wendt broke down the concept (“a property of intentional actors that generates motivational and behavioral dispositions”) into collective, corporate, role and type identities. Wendt, Social Theory, Cambridge, 1999, pp. 224-33). Hopf’s theory, societal constructivism, sets out to endogenize identity by bringing together its social, structural, and cognitive dimensions. For a fresh-off-the-press statement, see Hopf, Reconstructing the Cold War: The Early Years, 1945-1958 (Oxford, 2012).

[8] Compare, say, David Campbell, Writing Security: United States Foreign Policy and the Politics of Identity (Minnesota, 1992) and Jutta Weldes, Constructing National Interests: The United States and the Cuban Missile Crisis (Minnesota, 1999), to, say, Henry R. Nau, At Home Abroad: Identity and Power in American Foreign Policy (Cornell, 2002). For a recent argument on how and why identity was appropriated in IR, see Nik Hynek and Andrea Teti, “Saving Identity from Postmodernism? The Normalisation of Constructivism in IR,” Contemporary Political Theory,” Contemporary Political Theory 9 (2010), 171-199.

[9] On the problem of social-scientific knowledge and its basis, and why no meta-theory can prevail in IR, see Patrick T. Jackson, “What the Philosophy of Science is Not Good For,” short editorial for e-IR, February 23, 2009, Retrieved on 18 June from I use Jackson’s terminology to differentiate among philosophical positions.

[10] I am simplifying once again. There is much social-theoretic disagreement over the extent to which “identity” is independent from social or cognitive structures. Identity is not the same as identification or a discourse of identity, for example. There is also much debate on whether identity is always a product of the repression of difference, power relationships, social hierarchies and such.

[11] Advocates of “eclectic” scholarship in IR, note, contend against setting methodological decisions in ontological stone. The main reference in this paragraph is to the four hundred pages-long handbook by Rawi Abdelal, Yoshiko M. Herrera, Alastair Iain Johnston, and Rose McDermott, eds., Measuring Identity: A Guide for Social Scientists (Cambridge, 2009). This is mainly a collection of expert essays on methodology, but an introductory chapter offers a useful discussion of (collective) identity as a “social category that varies along two dimensions-content and contestation,” with content further broken down into four components (pp. 18-29). Another useful how-to collection on identity is Patricia Goff and Kevin Dunn, eds., Identity and Global Politics: Empirical and Theoretical Elaborations (Palgrave Macmillan, 2004). Practice-tracing is Vincent Pouliot’s latest brainwave (Paper prepared for the workshop on “Process Tracing in the Social Sciences: From Metaphor to Analytic Tools”, Georgetown University, March 2012; on file with the author).

[12] The next sentence narrows down this critique: “Preoccupation with the politics of identity can create a history without materialism, a history without economic exploitation, capital accumulation and power applied for the instrumental purpose of economic gain.” Ó Tuathail, “Dissident IR and the Identity Politics Narrative: A Sympathetically Skeptical Perspective (Review Essay),” Political Geography Quarterly 15 (1996), 647-653, p. 652.

Tags: , , ,


Please Consider Donating

Before you download your free e-book, please consider donating to support open access publishing.

E-IR is an independent non-profit publisher run by an all volunteer team. Your donations allow us to invest in new open access titles and pay our bandwidth bills to ensure we keep our existing titles free to view. Any amount, in any currency, is appreciated. Many thanks!

Donations are voluntary and not required to download the e-book - your link to download is below.