1 - Nation

Summary

This chapter traces how the concept of ethnicity emerged as a depoliticised alternative to nationality. By the end of the nineteenth century, the triumph of nationalism as the hegemonic source of state legitimacy had resulted in the politicisation of the nation concept. This conceptual linkage of ‘nation’ with ‘state’ opened up a terminological vacuum: If nationhood implied statehood, what label should be given to those stateless nations and national minorities that had neither a state of their own nor the political capacity to acquire one? Against this backdrop, the chapter traces how an embryonic concept of ethnicity was articulated to fill in the terminological void. The chapter’s empirical focus is on the early twentieth-century academic literature on nationalism and the establishment of the world’s first international minority rights regime after the First World War. The argument also has significant implications for debates surrounding the conceptual distinction between ‘civic’ and ‘ethnic’ nationalism.

The word ‘nation’ today has two contrasting meanings, signifying both a community of people and a sovereign state. This dual meaning also informs IR’s conceptualisation of nations and nationalism.¹ On the one hand, there is a widespread tendency among IR scholars to equate the nation with the state and to use terms like ‘national interest’ and ‘national security’ with reference to the interest and security of states. Nationalism, likewise, has been depicted as a ‘centripetal force’ that binds the state together: ‘the better the state’, Kenneth Waltz asserts, ‘the more nationalistic’ it is.² This conception of nationalism as the primary source of state legitimacy is rarely explicitly theorised, but instead serves as the unthinking background assumption that allows IR scholars to focus on what they are really most interested in: relations between pre-constituted nation-states. On the other hand, in the relatively few cases where IR scholars actually subject nations and nationalism to critical scrutiny, they tend to be pitted against the state in an antagonistic relationship. In the literature on nationalist conflict, for example, the ‘state-to-nation balance’ or ‘nation-to-state ratio’ is seen as a key variable in determining the likelihood of conflict in a given region.³ In this antagonistic framing, nationalism is no longer the centripetal force that holds the state together, but a centrifugal one that threatens to pull it apart.

The contradictory uses of ‘nation’ and ‘nationalism’ have not gone unnoticed by scholars. The most common response has been to treat this as an unfortunate terminological muddle that could easily be avoided if political scientists just took more care to define their terms.⁴ While terminological clarity is important, reducing the contradictions of the nation concept to a problem of definition tells us nothing about the sources or functions of these contradictions. In fact, the contradictory meanings of the word ‘nation’ are a reflection of the contradictory structure of the modern state, which is torn between universalism and particularism. The universalistic aspect of the state is evident in its claim to sovereignty, whereby the state effectively takes on the role of a secularised God. ‘The state’, as Hegel famously wrote, ‘consists in the march of God in the world.’⁵ Insofar as the state claims to be the sovereign guarantor of order and justice, it need only exist in the singular; the idea of an all-encompassing world-state is by no means conceptually incoherent.⁶ At the same time, however, the state is also ‘a bordered power-container’ ruling over a finite portion of the earth’s surface and population.⁷ The universalistic claim of the state to be the sovereign guarantor of order and justice thus stands in fundamental contradiction with the inescapable territorial particularity of every actually existing state.⁸

It is the contradictory structure of the modern state that gives rise to nationalism as an ideology of state legitimacy.⁹ Simply put, the nationalist solution to this contradiction is to insist that every state should represent a nation and that every nation should have a state of its own: ‘let all nations have their own political roofs, and let all of them also refrain from including non-nationals under it’, as Ernest Gellner memorably put it.¹⁰ Or, in the more succinct formulation of Eric Hobsbawm, ‘nation = state = people’.¹¹ By positing the existence of a pre-political nation as the foundation of the state, nationalism is able to conceal the arbitrariness of international borders and justify the territorial particularity of the state. Rather than stemming from a failure in conceptual precision, the dual meaning of the word ‘nation’ is a consequence of those specific historical circumstances that saw the triumph of the territorial state as the hegemonic form of political organisation.

Some obvious empirical barriers stand in the way of the nationalist solution to the problem of state legitimacy. To begin with, there are far more nations than states: the number of ethnically or culturally distinct human populations that exist on this planet is orders of magnitude greater than the number of independent political units.¹² Granting statehood to every distinct group of people would necessitate a radical redrawing of international political boundaries and pave the way for a potentially endless process of political fragmentation. As Hans Morgenthau presciently observed in 1957, the only thing that can halt the proliferation of competing nationalist claims is state power.¹³ The result is the ‘A-B-C paradox’ where ‘nation B invokes the principles of nationalism against nation A, and denies them to nation C – in each case for the sake of its own survival’.¹⁴ This institutionalised state of hypocrisy is further exacerbated by the continual intermingling of populations through migration and intermarriage, which makes it impossible to apply the principle of national self-determination consistently.

Yet, crucially, the shortcomings of the nationalist solution are not limited to these empirical obstacles. Even if all human migration were to cease and the political map were to be redrawn from scratch, the nationalist solution would still fall short of its aspirations. No amount of ethnic cleansing can ever ensure the congruence of national and political boundaries, for the limit of the nationalist solution is internal rather than external, logical rather than empirical. This is because, contrary to what nationalist ideologues claim, the nation is not a natural organism but a social construct that has to be continually reproduced through daily rituals and cultural practices.¹⁵ And, as Aamir Mufti points out, this process of constructing a majoritarian national culture necessarily produces national minorities.¹⁶ Counterintuitively, it is not the ethnic diversity of the world that frustrates the pursuit of nation-state congruency, but the pursuit of nation-state congruency that produces the ethnic diversity of the world. Due to this constitutive impasse, nationalism ultimately proves to be both cure and poison to the state: curative because it justifies the boundedness of the state with reference to a pre-political nation, yet poisonous because it opens the door to a never-ending series of secessionist and irredentist claims.

The central claim of this chapter is that the emergence of ethnicity in the twentieth century was, in part, an attempt to solve the contradictions of nationalism by articulating an alternative vocabulary to describe stateless nations and national minorities. Simply put, safeguarding the international order from the destabilising force of nationalism requires the hierarchisation of two kinds of nation: those that possess or deserve statehood and those that do not. This, precisely, is the basic difference between a nation and an ethnic group: ‘An ethnic group is distinguished from a nation, including an ethnic nation, by being a group with a common culture that does not seek to be a political community, does not seek self-governance, and certainly does not seek to constitute themselves into a state.’¹⁷ If an ethnic movement claims the right to statehood, that movement ‘by definition becomes a nationalist movement’.¹⁸ It is this ‘apolitical concept’¹⁹ of ethnicity that underpins the ontology of the international order and the legitimacy of the nation-state. By serving as a residual or ‘filler’ category, the concept of ethnicity absorbs the surplus of nations that violate the principle of nation-state congruency. In contrast to the politically explosive vocabulary of nationality, which today is inextricably intertwined with notions of sovereignty and statehood, the vocabulary of ethnicity offers a depoliticised medium through which minority rights can be addressed without placing into question the existing arrangement of international borders.

To develop this argument, the remainder of the chapter proceeds in four sections. The first section provides a brief history of modern nationalism, focusing on the perceived transformation of nationalism from a constructive to a destructive force around the turn of the twentieth century. In Hegelian terms, the nationalisation of European politics produced a dialectical reversal that turned the relationship between nationalism and international order on its head: at the very moment that an international order of nation-states coalesced in Europe, nationalism turned against its own creation. Building on this analysis, the second and third sections show how the politicisation of the nation concept – the sublation of the nation into the state – opened up the conceptual space for ethnicity in academic and political discourse, respectively: the second section focuses on the conceptual frameworks deployed in academic studies of nationalism and international relations around the turn of the century, while the third section looks at the conceptualisation of national minorities in political negotiations and international treaties, focusing on the minority rights regime set up at the end of the First World War. The fourth section wraps up the chapter by demonstrating how the conceptual hierarchy between nationality and ethnicity has been projected onto the international plane through the elaboration of a contrast between ‘Western’ or ‘civic’ nationalism on the one hand and ‘non-Western’ or ‘ethnic’ nationalism on the other.

Nationalism and International Order

The conceptual entanglement of ‘nation’ and ‘state’ was a long and drawn-out process that can be traced back to the late medieval period.²⁰ However, it was not until the French Revolution that nationalist ideas were first put into practice and the concept of the nation was redefined ‘from a diffuse sentiment to a specific program for political and constitutional action’.²¹ The Declaration of the Rights of Man and of the Citizen, promulgated by the French National Constituent Assembly in 1789, loudly proclaimed to the world that ‘the source of all sovereignty resides essentially in the nation’.²² In addition to this reconfiguration of domestic authority structures, the ensuing Revolutionary and Napoleonic wars also spurred the reconfiguration of external relations among European polities from a dynastic to a national model.²³ Indeed, the world-historical significance of these events lies precisely in their transnational scope. What the French Revolution inaugurated was not only a politicised concept of the nation, which had arguably existed for a century or two, but a modular conception of the nation that could be propagated around the world.²⁴

During and after the French Revolution, the concept of the nation was not primarily defined in opposition to other nations or foreigners, but in opposition to the ancien régime and the conservative doctrine of divine right.²⁵ The concept of the nation thus acquired a distinctly liberal and progressive orientation that would remain with it for most of the nineteenth century. What made the ‘principle of nationality’ so attractive to nineteenth-century liberals was precisely the fact that the political concept of the nation was historically novel and that it was opposed by the conservative segments of society. So long as nationalist discourse remained the preserve of a relatively small liberal-bourgeois elite, those ‘ethnic’ elements that would later become key features of nationalist discourse did not matter very much politically.²⁶ Even in the Germany of the Romantics, traditionally seen as the wellspring of ethnic nationalism, ‘it was a liberal fusion of progress and cultural nationality that dominated nationalist discourse for much of the century’.²⁷

Alongside its universalistic orientation and liberal character, another defining feature of nineteenth-century nationalism was its adherence to what Eric Hobsbawm calls the ‘threshold principle’. By and large, nineteenth-century commentators assumed that the principle of self-determination was only applicable to nations that were culturally and economically of a ‘viable’ size. The construction of nations was generally seen as a process of unification into larger entities, typified by the experiences of Italy and Germany, rather than a process of fragmentation into smaller ones.²⁸ The threshold principle was politically significant because it ensured that nationalism did not threaten the universalistic narrative of progress that characterised nineteenth-century liberalism. An evolutionary understanding of human development toward ever-larger communities – from families to clans, from clans to tribes, and from tribes to nations – reinforced this alliance of nationalism and liberalism.²⁹ When the preeminent English liberal thinker John Stuart Mill discussed Irish nationalism in 1861, for example, he acknowledged that the Irish ‘are sufficiently numerous to be capable of constituting a respectable nationality by themselves’.³⁰ In contrast, he believed that smaller ‘half-savage’ groups such as the Bretons and the Welsh should assimilate into the French and English nations, respectively.³¹ Mill’s characterisation of the smaller European nationalities as ‘half-savage’ also underlines how the requirement of size was coupled to a requirement of civilisation. As Mill put it in 1859, ‘barbarians have no right as a nation, except a right to such treatment as may, at the earliest possible period, fit them for becoming one’.³²

The Transformation of Nationalism

The period between 1871 and 1914, known as ‘la belle époque’, marked the high point of the alliance between nationalism, liberalism, and imperialism. In 1871, the great European nation-building projects of the nineteenth century culminated in the national unifications of Italy and Germany. Over the following decades, the nationalisation of European politics reached a point where every state was cultivating for itself a national base from which it could draw its legitimacy.³³ Even autocratic multinational empires such as Austria-Hungary and Russia began articulating ‘official’ nationalisms to prop up their regimes.³⁴ Formerly just one political position among many, nationalism now established itself as the universal ground for all political ideologies. In parallel, a new wave of European imperial expansion overseas meant that virtually the entirety of the earth’s surface was divided up among a handful of imperialistic nation-states. As the blank spaces on European maps were filled in, the number of independent polities in the world sunk to an all-time low.³⁵ By the turn of the century, the ardent proponent of British imperialism Cecil Rhodes observed with equal pride and sadness that the ‘world is nearly all parcelled out, and what there is left of it is being divided up, conquered and colonised’.³⁶ The sense of culmination and finality that defined the turn-of-the-century zeitgeist was eloquently captured by the American political scientist Paul S. Reinsch in 1900: ‘The nations, having passed through their historical evolution, stand now, with fully developed individualities, face to face.’³⁷

Precisely at its moment of triumph, nationalism shed its liberal-universalistic orientation and took on an increasingly ethnic-particularistic guise. In many ways, this dialectical reversal was the logical corollary of nationalism’s own success. As national consciousness penetrated wider and deeper into European society, it became increasingly important for elites to understand how ordinary people felt about the nation in order to mobilise public opinion behind political decisions. Language and other ‘ethnic’ criteria of nationhood thus acquired greater salience.³⁸ At the same time, the intensification of imperial rivalry among the great powers catalysed the division of Europe into competing transnational blocs centred on racialised ideologies such as pan-Germanism and pan-Slavism.³⁹ The combined effect of these developments was the rise of an ethnocentric cult of the nation-state, exemplified by the establishment of the French right-wing political movement Action française in 1899. Charles Maurras, a leading figure of the movement, defined the new ‘integral nationalism’ as ‘the exclusive pursuit of national policies, the absolute maintenance of national integrity, and the steady increase of national power’.⁴⁰ Numerous turn-of-the-century commentators noted this ‘exaggeration’⁴¹ or ‘perversion’⁴² of nationalism with trepidation. In the midst of the First World War, which seemed to validate these fears, one British author lamented that the principle of nationality had ‘changed in character with its success’. Nationalism was no longer ‘the cry of an oppressed people’ but had instead ‘become allied with national pride, and with the wish to acquire power and territory’.⁴³ The word ‘nationalism’ itself was popularised in the last decade of the nineteenth century precisely as a term of abuse to denounce the new ethnocentric and power-hungry cult of the nation-state.⁴⁴

The exaggeration of nationalism in international politics was accompanied by a second closely related shift: the transformation of nationalism from a unifying into a disintegrating force. At the very moment that the European nation-building projects seemed complete, nationalism turned against its own creations. With an ever-growing number of ‘new’ or ‘unhistorical’ nations in Central and Eastern Europe asserting their right to self-determination, the ‘threshold principle’ of the nineteenth century was effectively abandoned.⁴⁵ E. H. Carr captured this transformation in his 1945 book Nationalism and After, where he divided the history of nationalism into three stages. The first stage comprised the gradual dissolution of the feudal system in the early modern period, while the second stage stretched from the end of the Napoleonic Wars to the outbreak of the First World War.⁴⁶ Carr was sympathetic toward this early form of nationalism, describing the nineteenth-century balance between nationalism and internationalism as ‘the work of art rather than of nature’.⁴⁷ By the last quarter of the century, however, ‘the first subterranean rumblings began to shake this splendid edifice’ as nationalism entered its third and most recent phase: ‘After 1870 the constructive work of nation-building seemed complete. Nationalism came to be associated with “the Balkans” and with all that the ominous term implied.’⁴⁸ Over the following years, the exemplary case of nationalism would increasingly shift from the national unifications of Italy and Germany to the fragmentation associated with the Balkans.

The dual transformation of nationalism in the last quarter of the twentieth century reflected the underlying contradiction between universality and particularity that cuts through the modern territorial state. On the one hand, nationalism turned out to be not particular enough: the legitimacy issues of the state can never be fully relieved through recognition by its national citizenry. As Ayşe Zarakol observes, the legitimacy of the state derives from it being sovereign for its citizens as well as over its citizens. This claim to universality cannot be assuaged by the essentially domestic and particularistic operation of nationalism. As a result, the state’s search for ontological security must turn outward into the international realm: ‘It must be sovereign in the world.’⁴⁹ The inadequacy of nationalistic particularism as the state’s foundation thus generates a nationalistic universalism or nationalistic imperialism that ‘glows with the animus of greed and self-aggrandisement at the expense of others’.⁵⁰ This ‘brutal egotism’ and ‘narrow chauvinism’ among nation-states, as one commentator remarked in 1923, is ‘the basic fact of international society’.⁵¹

On the other hand, nationalism also turned out to be too particular: by grounding the legitimacy of the state in particular ethnocultural characteristics, nationalism always-already – also threatens the unity of the state by paving the way for secessionist claims. If nationalistic universalism constitutes an international source of conflict, then nationalistic particularism constitutes a domestic one. This particularistic threat is ineradicable because, contrary to the claims of nationalists themselves, the nation is not given by nature but a historically contingent ideological construct. By the end of the First World War, it was becoming increasingly evident that nationalist claims for independence would not exhaust themselves. ‘If the right of every group, however small, which may happen to be ethnically and linguistically distinct from the rest of the population, to separate and organize itself into a new state, were admitted and exercised in practice, it would lead to chaos and anarchy’, the American political scientist James Wilford Garner wrote in 1928.⁵² An important corollary to this was the foregrounding of the problem of national minorities as a pressing international concern. ‘In the last resort there must always be minorities that suffer’, the British historian Arnold Toynbee concluded in 1915. ‘We can only secure that the minorities are as small and the suffering as mild as possible.’⁵³

The dual threat that nationalism posed to international order was at the forefront of Hans Morgenthau’s pioneering work on international politics. The particularistic or domestic aspect of the new nationalism was the focus of his 1957 article on ‘The Paradoxes of Nationalism’, where he described nationalism as ‘a principle of disintegration and fragmentation’ that culminated in ‘anarchy’.⁵⁴ For Morgenthau, the disintegrative force of twentieth-century nationalism represented a stark departure from the unificatory nation-building projects of the previous century: ‘No longer are national minorities to be protected against the state; it is now the state which must be protected against the minorities.’⁵⁵ Echoing Carr’s aforementioned linkage of twentieth-century nationalism with the Balkans, Morgenthau feared that the proliferation of nationalist claims would lead to ‘Balkanization, demoralization, and barbarization on a world-wide scale’.⁵⁶ Given that the logic of nationalist fragmentation had ‘no inherent limits’, the only thing that could put a stop to it was state power: ‘the process of national liberation must stop at some point, and that point is determined not by the logic of nationalism, but by the configurations of interest and power between the rulers and the ruled and between competing nations’.⁵⁷

State power may have been seen as the antidote to nationalism’s drive toward fragmentation, but in solving one problem it also created another: the assertion of state power transformed nationalism into a political religion that claimed for one nation-state the right to impose its will upon others. This was the universalistic or international dimension of the new nationalism. ‘Traditional nationalism sought to free the nation from alien domination and give it a state of its own’, Morgenthau explained in Politics among Nations. ‘Once a nation had united its members in one state, national aspirations were satisfied, and there was room for as many nationalisms as there were nations which wanted to establish or preserve a state of their own.’ Nationalist conflicts in the nineteenth century had been either conflicts between a subject nationality and its alien master or conflicts between two nations over the delimitation of their respective territories.⁵⁸ In contrast, nationalism in the twentieth century took the form of a ‘nationalistic universalism’ where ‘the nation is but the starting point of a universal mission whose ultimate goal reaches to the confines of the political world’.⁵⁹ The emergence of the new nationalism was foreshadowed by the Napoleonic Wars and the First World War, but Morgenthau saw the 1930s and 1940s as the decisive break.⁶⁰ ‘While the old nationalism seeks one nation in one state and nothing else’, he wrote, ‘the new one claims for one nation and one state the right to impose its own values and standards of action upon all other nations.’⁶¹ The dangers of nationalistic universalism were multiplied by the advent of the atomic age, which heralded the possibility of mutually assured destruction and rendered the protective functions of the nation-state obsolete.⁶² ‘If the West cannot think of something better than nationalism’, rang Morgenthau’s sombre conclusion, ‘it may well lose the opportunity to think at all.’⁶³

The Neutralisation of Nationalism

Insofar as the twin threats that nationalism poses to international order stem from the contradictory structure of the modern territorial state, they can never be fully eradicated. Yet their worst destabilising effects can be alleviated through a double move. The first part of this double move is well-known to IR theorists and entails displacing the ‘problem of difference’⁶⁴ from the domestic to the international realm. This is achieved through the construction of a spatial distinction between the inside and the outside of the state, whereby the sovereign national identity is located within the bounds of the state territory and difference is projected out onto the international plane.⁶⁵ ‘This demarcation and policing of the boundary between the “inside” and the “outside” of the political community’, as David Blaney and Naeem Inayatullah explain, ‘defines the problem of difference as between and among states; difference is marked and contained as international difference.’⁶⁶ By constructing an idealised global grid of internally homogeneous nations, the inside/outside framework avoids the overlap of ‘self’ and ‘other’. The contradiction between universality and particularity is thus seemingly resolved: the identity of the nation appears universal domestically, but particular internationally. This is the culmination of the nationalist fantasy where, as Ernest Gellner famously put it, ‘all nations have their own political roofs’.⁶⁷

Of course, the ideal of the homogeneous nation-state remains forever an aspiration rather than a reality. In 1972, Walker Connor noted that only twelve of the world’s states could be considered ethnically homogeneous to any significant degree, and the situation has hardly ‘improved’ since then – if anything, the international migration fuelled by globalisation has led to even greater heterogeneity.⁶⁸ Every state contains within its borders national minorities that governments must seek to manage through policies ranging from assimilation and toleration to expulsion and extermination.⁶⁹ Crucially, however, such incongruencies are not reducible to a purely empirical issue: it is not simply the case that the ineradicable diversity of humankind makes it impossible to segregate national communities into distinct political units, or that realities of migration and intermarriage undercut attempts to achieve ethnic homogeneity within the state’s territory. In fact, the situation is precisely the opposite: it is not the purported ethnocultural diversity of the world that prevents the construction of homogeneous nation-states, but the attempts to construct homogeneous nation-states that produce the ethnocultural diversity of the world.⁷⁰ The nationalist fantasy of the congruent nation-state is thus a fantasy in the strictest sense of this word: an impossible project that is not only destined to fail, but that can only exist through its very failure.⁷¹ In the final analysis, the ‘excess’ of nations over states is a necessary consequence of nationalism itself.

While the discursive construction of the inside/outside dichotomy has become a familiar trope of IR theory, the second part of the double move has received less attention: the problem of difference has to be erased not only from the domestic realm, but also from the international realm. Otherwise, international politics would appear as a chaotic realm of pure difference and particularity that universal reason could not tame. For many of IR’s pioneers, the experience of two world wars and mass genocide certainly seemed to validate such a pessimistic assessment. Prominent figures such as Hans Morgenthau and John Herz had emigrated to the United States from Germany in the 1930s and were acutely aware of nationalism’s dangers.⁷² Morgenthau, for example, worried that the exaggerated nationalism of the Second World War had dealt ‘the final, fatal blow’ to international restraints on destructive power politics.⁷³ At the same time, these traumatic experiences underlined the vital importance of bringing the chaotic international realm within the grasp of reason. Thus, it was against a backdrop of extreme nationalist violence that the first attempts to develop ‘scientific’ theories of international relations emerged in the late 1940s and 1950s, superseding the rather eclectic IR scholarship of the early twentieth century.⁷⁴ If there was any chance of formulating a rational theory of politics among nations, Morgenthau believed, such a theory could not begin from the particularistic logic of nationalism, but had to take as its starting point a universal understanding of power grounded in human nature: ‘the struggle for power is universal in time and space’.⁷⁵

If the first part of the double move exorcises the problem of difference from the domestic realm, then the second exorcises the problem of difference from the international realm. To achieve this, the nation can appear on the international plane only in its universalistic capacity and not in its particularistic capacity: only the abstract and transposable form of the nation can be present; the concrete ethnocultural contents that make each individual nation unique must be discarded. Strictly speaking, therefore, a nation can no longer appear on the international plane as a nation at all, but must instead appear as a state. The difference between nation and state thus collapses as the former is sublated into the latter. Pushed to its logical conclusion, this conceptual short circuit extinguishes all qualitative differences between nations and transforms them into those ‘black boxes’ or ‘billiard balls’ that IR theorists are so fond of: nations become ‘like units’ devoid of qualitative differences. All that remains of the particularity of nations is their spatial aspect – territorial boundedness – as well as quantitative differences that can be captured through abstract and universally applicable categories such as ‘power’ or ‘capabilities’.⁷⁶ Through this process of abstraction, which subordinates quality to quantity and difference to identity, the international realm is made accessible to reason.⁷⁷ This glorious triumph of universal reason reaches its climax when the term ‘anarchy’ ceases to indicate ‘disorder and chaos’ and transforms instead into an ‘ordering principle’ that allows the IR theorist to ‘explain important aspects of social and political behavior’.⁷⁸

International Relations’s resolution of the contradictions of nationalism is thus more intricate than usually recognised. It is not simply a case of displacing the problem of difference from the domestic realm to the international plane, but also of then erasing the problem of difference from the international plane as well. Through this double move, which collapses the nation into the state, the domestic political system and the international political system are made accessible to the universalistic pretences of political theory and international theory, respectively. Meanwhile, nationalism comes to be seen as a political pathology and banished to the discipline’s peripheries.⁷⁹ All that remains of the nation’s particularistic constitution is an ethereal spectre that makes itself felt through banal phrases such as ‘national interest’, ‘national security’, ‘national defence’, and, of course, the term ‘international’ itself. This haunting presence of the nation is an inexorable remainder and reminder of the nationalist forces that constructed the international order and that may at any moment erupt once again to obliterate it.⁸⁰

But what happens to difference? What happens to the ineradicable kernel of ethnocultural particularity that fractures the nationalist fantasy from within? As the remainder of this chapter demonstrates, the consolidation of a system of nation-states in Europe was paralleled by the invention of a new concept that helped to neutralise the internal contradictions of nationalism: the concept of ethnicity. Through the elaboration of ethnicity as a depoliticised alternative to nationality, the particularistic dimension of the nation was given a separate conceptual existence from those universal categories that were applied to domestic and international politics. As the nation was sublated into the state and international anarchy metamorphosed into international order, the element of particularity inhabiting the nation ‘dropped out’ – was excreted, vomited, expelled – in the form of ethnicity. Although this conceptual separation of the universal and the particular components of the nation can never eliminate the contradictions of nationalism, it nevertheless provides these contradictions with a form in which they have ‘room to move’.⁸¹ To substantiate this argument, the next two sections trace how a conceptual space for ethnicity opened up in academic discourse, and international legal and political practice, respectively.

Conceptualising the Non-political Nation

By the dawn of the twentieth century, the nationalisation of European politics had made the words ‘nation’ and ‘state’ interchangeable. The obvious problem that this posed to political commentators was how to refer to those stateless national groups that did not have a state of their own and perhaps did not even seek to acquire one. This section traces how turn-of-the-century scholars grappled with this conceptual puzzle, and how these definitional dilemmas eventually paved the way for a new conceptual distinction between nations and ethnic groups. In terms of source material, the focus is primarily on British, French, and American scholars who engaged with questions of international order and nationalism in the late nineteenth and early twentieth centuries. Together with encyclopaedia entries, two bibliographical compilations provided a helpful starting point: Parker Thomas Moon’s Syllabus on International Relations from 1925 and Koppel S. Pinson’s A Bibliographical Introduction to Nationalism from 1935.⁸² Additional texts were then added through a ‘snowball’ approach (consulting the references of known works to find new works) to arrive at a corpus of well over fifty texts spanning the period from 1885 to 1945, although not all have been cited. Many of the texts were written by political scientists, but the authors also include historians, sociologists, lawyers, and philosophers, among others. This disciplinary heterogeneity is hardly surprising, given IR’s interstitial and embryonic state. At the turn of the century, as Nicolas Guilhot observes, ‘IR was generally considered to be an interdisciplinary field located on the margins of political science, with no method of its own – a sort of commons, as it were, plowed by various disciplines ranging from economics to geography’.⁸³

State versus Nation

Perhaps the most straightforward way to deal with the conceptual issues generated by the conflation of nation and state was simply to reject this conflation. While this had become a minority position by the beginning of the twentieth century, a handful of important scholars continued to oppose the interchangeable use of the two terms.⁸⁴ Foremost among them was Alfred Zimmern, a pioneer in the study of international affairs who became the first Woodrow Wilson Chair of International Politics at Aberystwyth in 1919 and the first Montague Burton Professor of International Relations at Oxford in 1930.⁸⁵ In Zimmern’s view, nationality and statehood belonged to categorically different realms:

Nationality, like religion, is subjective; Statehood is objective. Nationality is psychological; Statehood is political. Nationality is a condition of mind; Statehood is a condition in law. Nationality is a spiritual possession; Statehood is an enforceable obligation. Nationality is a way of feeling, thinking and living; Statehood is a condition inseparable from all civilized ways of living.⁸⁶

For Zimmern, therefore, internationalism was not about relations between states or nation-states, not even about the interactions of diplomats, but about relations between nations as cultural or spiritual entities.⁸⁷ ‘The true contact between the West European national triangle’, he explained in 1923, ‘must be a contact, not between trust-magnates or labor-leaders or even statesmen from the three countries, but, so to speak, between Shakespeare, Molière and Goethe.’⁸⁸ Zimmern’s argument was partly motivated by a desire for terminological clarity, but political and ethical considerations were also central. During the First World War, he drew a distinction between ‘true’ and ‘false’ nationalism, aligning the former with a benign cultural or spiritual conception of nationhood and the latter with the political ideal of the nation-state.⁸⁹ Restraining international competition and conflict required disentangling ‘the problems of nationality’ on the one hand from ‘the problems of statehood and citizenship’ on the other. ‘It is from their century-old confusion that so much mischief and bloodshed have arisen’, he concluded.⁹⁰

Unsurprisingly, protestations such as Zimmern’s were unable to reverse the growing tendency to substitute nation for state, epitomised by the establishment of the League of Nations – in fact, an inter-state organisation – in 1919. James Wilford Garner, Professor of Political Science at Illinois, was among a slew of prominent authors to point out this trend during the inter-war years: ‘the term “nation” as used to-day by most writers connotes a political organization; that is, a nation is not only an association of which the bonds of union are cultural and spiritual, but it is also a politically organized aggregation. In short, it is a state.’⁹¹ For many scholars, the equivalence of nation and state was especially apposite when it came to international politics. Thus, one early study of the relationship between nationalism and war explained that the word ‘nation’ could refer specifically to ‘a state in which there is one nationality, a national-state’, before specifying that ‘in international relations’ even a multinational polity such as Austria-Hungary ‘is considered a nation like every other’.⁹² Another commentator noted that both ‘nation’ and ‘state’ could be used ‘to signify politically organized communities which enter into international relations’.⁹³ Stephen Haley Allen’s International Relations, among the earliest books explicitly dedicated to the study of international politics, unapologetically used ‘nation’ and ‘state’ as synonyms.⁹⁴ For all of these scholars and the innumerable others who accepted the equivalence of nation and state, there was a need to coin another term to describe those national minorities and stateless nations that did not partake in international relations – entities that contemporaries sometimes described as ‘repressed’, ‘oppressed’, or ‘submerged’ nations.⁹⁵

Nation versus Nationality

The first candidate to occupy the terminological vacuum created by the politicisation of the nation concept was ‘nationality’. As one scholar noted in 1916, it was due to the increasing use of the term ‘nation’ in a ‘political sense’ in the nineteenth century that ‘nationality’ came to be employed as a concrete noun with reference to ethnocultural groups.⁹⁶ In the French language, too, the word ‘nationalité’ was introduced in the early nineteenth century to describe various forms of spiritual, religious, or ethnocultural (but never legal or political) attachment.⁹⁷ Among the earliest works to explicitly discuss the distinction between ‘nation’ and ‘nationality’ was Henry Sidgwick’s The Elements of Politics in 1891:

by ‘a nationality’ we usually mean a body of human beings united by the kind of sentiment of unity or fellow citizenship that is required to constitute a nation, but not possessing in common an independent government which they alone permanently obey: being either divided among several governments, or united under one government along with persons of a different nationality.⁹⁸

The key factor that differentiated a nation from a nationality was thus political unity. This point was echoed by other influential scholars of the time. In 1912, the British historian and politician James Bryce wrote that ‘a Nation is a nationality, or a subdivision of a nationality, which has organized itself into a political body, either independent or desiring to be independent’.⁹⁹ Two decades later, the American pioneer of nationalism studies Carlton Hayes suggested that a nationality became a nation ‘by acquiring political unity’.¹⁰⁰

Recalling the ‘threshold principle’ that had regulated the recognition of new nations in the nineteenth century, the requirement of political organisation was usually coupled to notions of rank and size: a nationality was typically believed to be smaller and less accomplished than a nation proper, and thus equated with minorities rather than majorities. For example, commenting on groups such as the Scots in Britain and the Slovenes in Yugoslavia, the American political scientist James Wilford Garner claimed that it ‘would be excessive flattery to their pride to call them “nations”; the term “nationality” more nearly corresponds to their importance’.¹⁰¹ Echoing this sentiment, Pablo de Azcárate, a Spanish diplomat who worked for the League of Nations Minorities Section in the 1920s, opined that a nationality and a minority were ‘in the last resort … one and the same’.¹⁰²

Not all scholars resorted to the word ‘nationality’ to defuse the contradictions of the nation concept. The American historian Harry Elmer Barnes, for example, preferred to differentiate between a ‘nation’ and a ‘national state’.¹⁰³ So did Arthur Holcombe, Professor of Government at Harvard.¹⁰⁴ In such instances, it was ‘nation’ that was the ethnic or cultural category and ‘national state’ that was the legal or political category. Yet the legal or political dimension was not so easy to isolate from the nation concept. Thus, even as Holcombe insisted that a nation was a cultural rather than a political unit, he conceded that there was a ‘tendency on the part of members of a nation to wish to dominate the state of which they happen to be a part or, failing that, to organize a state of their own’.¹⁰⁵ The same line of argument can be found in the work of John William Burgess, Professor of Political Science and International Law at Columbia and a key figure behind the establishment of the American Political Science Association in 1903.¹⁰⁶ Echoing Barnes and Holcombe, Burgess defined a nation principally in ethnocultural terms: not all nations were ‘endowed with political capacity or great political impulse’ and that it was ‘therefore not to be assumed that every nation must become a state’.¹⁰⁷ Specifically, he believed that it was ‘the Teutonic nations’ that were ‘particularly endowed with the capacity for establishing national states’.¹⁰⁸ Nonetheless, like Holcombe, he felt compelled to add a caveat: ‘Where the geographic and ethnic unities coincide, or very nearly coincide, the nation is almost sure to organize itself politically, – to become a state.’¹⁰⁹ Having acknowledged the tendency of nations to seek statehood, Burgess was obliged to refer to nations that lacked this political capacity as ‘unpolitical nations’.¹¹⁰ This contrast between ‘unpolitical’ and ‘political’ nations achieved the same purpose as Barnes and Holcombe’s distinction between ‘nations’ and ‘national states’ or the more widespread distinction between ‘nationalities’ and ‘nations’. In each pairing, the first term designated an ethnocultural community, while the second referred to a similar community that was also politically organised.

Some of the binary frameworks elaborated by early twentieth-century scholars of nationalism and IR are summarised in Table 1.1 above. By making it possible to conceptually differentiate nation-states from stateless nations and national minorities, these distinctions enabled a partial resolution of the definitional dilemmas posed by the dual meaning of the word ‘nation’. In parts of Central Europe, these conceptual distinctions were even developed into legal categories that justified the subordination of lesser nationalities under a dominant national identity.¹¹¹ The Hungarian Nationality Law of 1868, for example, granted some language rights to non-Magyar groups but reserved the term ‘nation’ or ‘nemzet’ to Hungary alone. Non-Magyar communities within the Hungarian part of the Austro-Hungarian Dual Monarchy were instead described as ‘nationalities’ or ‘nemzetiség’, a term adopted from Austrian legal terminology. In political terms, all nationalities belonged to the unitary Hungarian nation.¹¹²

Table 1.1 Splitting the nation concept

Author	Date	Non-political category	Political category
James Bryce	1912	Nationality	Nation
John Holland Rose	1916	Nationality	Nation
Théodore Ruyssen	1917	Nationality	Nation
Harry Elmer Barnes	1919	Nation	National state
Arthur Holcombe	1923	Nation	National state
James Wilford Garner	1928	Nationality	Nation

The splitting of the political and non-political halves of the nation concept did not necessarily require a dichotomy; the same effect could also be achieved through the articulation of multiple neighbouring categories. In 1929, for example, Bernard Joseph proposed a tripartite framework consisting of national groups, nationalities, and nations. Each step along the sequence was also a step toward statehood, culminating in the nation, which was defined as ‘a group of persons who constitute the population of a single state’.¹¹³ What distinguished a nationality from a national group was a ‘will to live’ as a nation.¹¹⁴ A national group was thus a potential nationality and a nationality was a potential nation. In a similar vein, the influential French jurist Louis Le Fur developed a four-tiered framework consisting of ‘race’, ‘nation’, ‘patrie’, and ‘état’. The first of these was of no political importance, but merely one possible contributing factor to a sense of nationality. The second tier was ‘nation’ or ‘nationalité’ (Le Fur used these terms interchangeably) and referred to an ‘entité moral’ with significant political potential.¹¹⁵ The third tier represented the fulfilment of this political potential: ‘La patrie, c’est la nation ayant pris conscience d’elle-même et devenue, de la part de ses membres, l’objet d’une sorte de culte, d’un sentiment spécial, le patriotisme, à base de reconnaissance et d’amour.’¹¹⁶ Both ‘patrie’ and ‘état’ entailed political authority over a territory, but the latter was distinguished from the former by the possession of absolute sovereign authority over its lands.¹¹⁷ For Le Fur, this fourth and final stage was the logical culmination of the lower terms in the framework.¹¹⁸ Every nation or nationality was thus a potential state: ‘la nationalité, c’est avant tout une virtualité, un État en germe, – comme inversement l’État c’est la nation juridiquement organisée’.¹¹⁹

Le Fur’s characterisation of statehood as the natural culmination of nationality was widely shared by his contemporaries. In 1916, the English historian John Holland Rose defined a nation as ‘a people which has attained to state organization’ and a nationality as a people which had ‘not yet attained’ such organisation.¹²⁰ In 1928, James Wilford Garner approvingly cited Le Fur’s characterisation of a nationality as ‘a state en germe’ and described political independence as ‘the natural fruit of nationality where the population is sufficiently numerous and capable of maintaining a separate state existence’.¹²¹ In the same vein, the French pacifist philosopher Théodore Ruyssen argued that ‘the nation is the complete form – or, as we should say in philosophy, the idea or final cause – which the nationality desires to realize’.¹²² Ruyssen sketched out a three-tiered framework whereby an ethnic group could develop, via a nationality, to a full member of international society: ‘La nationalité, c’est le groupe ethnique privé de l’indépendance politique et qui aspire à la conquérir … ; c’est, si l’on veut, la nation en puissance, mais assez consciente de cette puissance pour tendre de toutes ses forces au droit de prendre rang, en pleine égalité, dans la Société des Nations libres.’¹²³ All in all, there was a widespread sense that a nationality was somehow unfulfilled without development into a nation-state. Nationalities were seen as ‘des êtres jeunes, voire même enfants’ that would blossom into nations once they reached ‘pleine maturité’.¹²⁴ In this unilinear temporal sequence, the ‘possible’ was understood as an ‘unrealized actual’ that remained somehow ‘lacking’ or ‘incomplete’ until it had been fully actualised.¹²⁵

An especially influential German-language text is also worth mentioning here: Johann Kaspar Bluntschli’s Lehre vom modernen Staat. Originally published in German in 1851, the book was translated into French in 1871 and into English in 1885. The English translation, titled The Theory of the State, was widely read, reissued in numerous editions, and served as the main political theory textbook in Cambridge and Oxford in the late nineteenth century.¹²⁶ By the 1910s, Bluntschli’s works were circulating as far as East Asia, informing the development of Japanese and Chinese conceptions of the nation.¹²⁷ The book is structured around familiar conceptual distinctions, with the German ‘Volk’ possessing a political connotation that the word ‘Nation’ lacked. Thus, as Bluntschli himself noted, it was ‘Volk’ and not ‘Nation’ that corresponded to the English and French ‘nation’.¹²⁸ The translators heeded Bluntschli’s recommendation, rendering ‘Volk’ as ‘nation’ and ‘Nation’ as ‘people’ or ‘peuple’.¹²⁹ The latter category was defined as ‘a union of masses of men of different occupations and social strata in a hereditary society of common spirit, feeling and race, bound together, especially by common language and customs, in a common civilisation which gives them a sense of unity and distinction from all foreigners, quite apart from the bond of the State’.¹³⁰ For Bluntschli, a people was therefore ‘not a political society; but if it is really conscious of its community of spirit and civilisation, it is natural that it should ask to develop this into a full personality with a common will which can express itself in act; in fact, to become a State’.¹³¹ At the same time, Bluntschli sought to maintain some kind of threshold principle to such development, asserting that ‘only a people of political capacity can claim to become an independent nation’.¹³² It was precisely the acquisition of an independent state that marked the transformation of a people into a nation: ‘By a Nation (Volk) we generally understand a society of all the members of a State as united and organised in the State. The Nation comes into being with the creation of the State. It is the consciousness, more or less developed, of political connection and unity which lifts the Nation above the People.’¹³³

Nationality versus Ethnic Group

In their efforts to distinguish between the political and non-political meanings of the word ‘nation’, turn-of-the-century commentators produced a cacophony of conceptual frameworks. These frameworks were mostly populated by nation-based words such as ‘nation’, ‘nationality’, ‘national group’, and ‘national state’, complemented by a smattering of other terms such as ‘race’ and ‘people’. The most popular distinction was no doubt the binary contrast between nations and nationalities, but other conceptual schemas could serve the same purpose equally well. Through such distinctions, it became possible for authors to acknowledge the ‘excess’ of nations over states while also maintaining a threshold principle that limited the number of nations destined for statehood.

The gatekeeping function of the nation/nationality distinction was soon undermined by two interrelated factors. First, the presupposed temporal link between nationalities and nations presented a problem: given that nationalities were widely expected to mature into politically independent nations, the nationalist threat to the international order was not so much neutralised as merely deferred into the future. Recognising this issue, James Bryce lamented the failure of existing terminology ‘to distinguish a Nationality which, like the Scottish, does not seek to be politically independent from a Nationality which, like the Lithuanian, does so desire’.¹³⁴ The second and related problem was that the terms ‘nationality’ and ‘national’ were increasingly equated with ‘citizenship’ and ‘citizen’, respectively.¹³⁵ Carlton Hayes explained the situation in 1933 as follows:

It was in part to atone for the abuse of the word ‘nation’ that the word ‘nationality’ was coined in the early part of the nineteenth century and speedily incorporated into most European languages. Thenceforth, while ‘nation’ continued to denote the citizens of a sovereign political state, nationality was more exactly used in reference to a group of persons speaking the same language and observing the same customs. The jurists have done their best to corrupt the new word ‘nationality,’ just as they had corrupted the old word ‘nation’; they have utilized ‘nationality’ to indicate citizenship.¹³⁶

In this way, much like the word ‘nation’ before it, the word ‘nationality’ also acquired a ‘political’ meaning alongside its ‘ethnic and cultural’ meaning.¹³⁷ By the beginning of the twentieth century, there was fast emerging a conceptual lacuna that these nation-based words were struggling to fill. Ultimately, it was the language of ethnicity that was inserted into this void. Faced with the politicising and temporalising thrust of modern nationalism, the popularisation of ethnos-based terms represented an attempt to depoliticise and detemporalise nationality and thus freeze the political map.

In turn-of-the-century discussions of nationalism and international relations, ethnos-based words make sporadic appearances but receive little conceptual development. As early as 1890, John William Burgess explained that ‘the word nation is a term of ethnology, and the concept expressed by it is an ethnological concept’. He accordingly defined the nation as a ‘population of an ethnic unity, inhabiting a territory of a geographic unity’.¹³⁸ When clarifying what he meant by ‘ethnic unity’, Burgess explained that this referred to ‘a population having a common language and literature, a common tradition and history, a common custom and a common consciousness of rights and wrongs’.¹³⁹ During the inter-war years, Louis Le Fur referred to ‘minorités ethniques’ and Raymond Leslie Buell to ‘ethnic minorities’, but neither offered a definition.¹⁴⁰ Others noted in passing that the term ‘nation’ was etymologically an ‘ethnic’¹⁴¹ or ‘ethnical’¹⁴² concept. Another brief but illuminating distinction between ethnicity and nationality can be found in a two-volume study on civilisation and nationhood by the French philosopher and sociologist Joseph Thomas Delos from 1944. Delos equated a ‘communauté ethnique’ or ‘groupe ethnique’ with a ‘communauté de conscience’ and described this as a preliminary stage to the emergence of a ‘communauté nationale’:

Le passage de la communauté de conscience à la conscience de former une communauté est une transformation de la plus haute importance. Au moment où s’éveille la conscience de son unité et de son individualité et où s’affirme la volonté de continuer cette vie commune, le groupe ethnique franchit une étape, et il serait souhaitable, croyons-nous, de lui réserver alors le nom de communauté nationale. Grâce à cet élément subjectif, – conscience et vouloir-vivre commun, – la nation apparaît distincte du milieu ethnique, au sens strict du mot, tout en lui restant liée comme un stade postérieur est lié au stade antérieur.¹⁴³

Delos thus considered a national community to be a higher or more developed form of an ethnic group, with the passage from the latter to the former entailing the emergence of a shared desire to live together as an independent community.

The general impression that emerges from these early twentieth-century texts is an understanding of ethnicity as a foundation of nationhood, but without the subjective or political dimension of the latter. In this sense, the concept of ethnicity was almost indistinguishable from the non-political meaning of nationality. Indeed, the French author René Johannet pointed out in 1918 that most contemporary definitions of ‘nationalité’ were identical to the French anthropologist Joseph Deniker’s definition of ‘groupe ethnique’.¹⁴⁴ In an influential study published in both French and English in 1900, Deniker had proposed a new conceptual distinction between races and ethnic groups: whereas races were theoretical abstractions based on physical traits, ethnic groups were ‘real and palpable groupings … formed by virtue of community of language, religion, social institutions, etc.’¹⁴⁵ Among some anthropologists, then, the concept of ethnicity was already emerging as a relatively coherent and well-defined alternative to the concept of race (see Chapter 2). In the early twentieth-century literature on nations and nationalism, by contrast, the embryonic concept of ethnicity still lacked a clear definition and remained jumbled up with older conceptions of race and nationality. In the relatively few cases where ethnos-based terms such as ‘ethnic’ and ‘ethnical’ do make an appearance, they did not yet possess the status of a distinct concept, but functioned instead as a supplementary vocabulary that helped to qualify or specify other (typically nation-based) terms. It was not until the second half of the twentieth century – when the concept of race was pushed aside – that the ethnos-based vocabulary would move from the margins of social and political discourse to centre stage.

Minority Rights

In international political and legal practice, the mismatch between national and political boundaries is managed through minority rights provisions. The first time that minorities were described as ‘national’ rather than religious communities was in 1815 at the Congress of Vienna. Although several subsequent nineteenth-century treaties also contained clauses pertaining to minority rights, it was not until after the First World War that an international regime of minority protection was put into practice. Set up under the auspices of the newly formed League of Nations, the international minority rights regime was chiefly a response to the proliferation of nationalist claims in Central and Eastern Europe after the disintegration of the Ottoman, Habsburg, and Romanov multinational empires. Against this backdrop, minority rights essentially functioned as a ‘substitute’ for national self-determination in those instances where statehood was considered either unfeasible or undesirable by the great powers.¹⁴⁶

There are substantial connections between the scholarly debates discussed in the previous section and the political negotiations that produced the minority rights regime at the end of the First World War. Not only was national self-determination a central issue in both settings, but the Allied Powers’ reliance on expert commissions to produce empirical data for the peace talks also meant that there were significant overlaps in personnel. The British Political Intelligence Department included historians Arnold Toynbee and Alfred Zimmern, both of whom had produced influential scholarly works on nationalism, while the French Comité d’Études featured renowned geographers such as Jean Brunhes and Emmanuel de Martonne.¹⁴⁷ The American expert commission, known as ‘the Inquiry’, was the largest of all. Set up by Woodrow Wilson in September 1917, it numbered some 150 scholars and collected or produced nearly 2,000 separate reports and documents, plus at least 1,200 maps.¹⁴⁸ At the end of the war, numerous members of the Inquiry served as advisors to the American plenipotentiaries attending the peace conference and as negotiators on international commissions.¹⁴⁹ Although nearly every major international conference since 1815 had provided some role for experts and advisers, this was the first time that the major powers sought to formulate a clear and systematic approach ahead of time.¹⁵⁰ The emphasis on calculation and classification in the reordering of the international system was a significant break with the tradition of rule by right or warfare, signalling the triumph of ‘population politics’ on the international plane.¹⁵¹ Informed by a wealth of empirical data, the work of the Allied preparatory commissions reconceptualised political and ethnographical boundaries as measurable and manipulable objects that could be rationally arranged to create ethnically homogeneous nation-states.¹⁵² In the words of Hannah Arendt, ‘the nation had conquered the state’.¹⁵³

The bulk of this section is concerned with the negotiations over the wording of the minority rights treaties that were imposed on several Eastern European states at the conclusion of the First World War. In addition to the treaty documents themselves, the key primary sources include the minutes of the Committee on New States and the Protection of Minorities, which was responsible for the drafting of the minorities treaties; the minutes of the Greco-Bulgarian Mixed Commission, which oversaw the subsequent exchange of populations between Greece and Bulgaria; the diaries of David Hunter Miller, an American lawyer who served on the Inquiry and the Committee on New States and the Protection of Minorities; and the published records of the United States Department of State. The section concludes with a brief discussion of the development of minority rights since 1945.

The Problem of Minorities after the First World War

The so-called ‘problem of minorities’ was among the central issues facing the peacemakers at the end of the First World War. For the advocates of national self-determination, the solution to this problem was territorial readjustment: the state should be made to fit the nation by redrawing existing political boundaries. By contrast, the advocates of minority protection prioritised maintaining the territorial status quo even if this entailed the co-presence of multiple national groups within the same state. As C. A. Macartney explained in 1934, the idea of minority protection was premised on the assumption ‘that it is possible to put an end to the whole movement towards so-called national self-determination’.¹⁵⁴

The development of Woodrow Wilson’s drafts for the Covenant of the League of Nations during the peace process reveals a shift in emphasis from territorial readjustment to minority protection. His first draft made no reference to minority rights, embracing instead the principle of national self-determination. Thus, Article III of the draft made provisions for ‘territorial readjustments’ if these were to become necessary ‘by reasons of changes in present racial conditions and aspirations or present social and political relationships, pursuant to the principle of self-determination’. In a revolutionary proposal, contracting parties were required to ‘accept without reservation the principle that the peace of the world is superior in importance to every question of political jurisdiction or boundary’.¹⁵⁵ Unsurprisingly, Wilson’s draft met with heavy protest. In his commentary on the proposal, David Hunter Miller underlined the practical difficulties of drawing political boundaries in accordance with ‘racial or social conditions’ and noted that the provisions suggested by Wilson would merely ‘legalize irredentist agitation’.¹⁵⁶ The idea of territorial readjustment was thus swiftly abandoned. Instead, Wilson’s subsequent drafts required all new states ‘to accord to all racial or national minorities within their several jurisdictions exactly the same treatment and security, both in law and in fact, that is accorded the racial or national majority of their people’.¹⁵⁷ Miller applauded the shift in favour of minority protection, noting that ‘protection of the rights of minorities and acceptance of such protection by the minorities constitute the only basis of enduring peace’.¹⁵⁸

Despite featuring prominently in the drafting process, all clauses pertaining to minority protection were dropped from the final text of the Covenant. This was done at the behest of the British delegation, which preferred to settle the provisions of the territorial treaties before considering the minorities question.¹⁵⁹ As a result, the inter-war minority protection regime comprised a motley collection of international instruments, eighteen in all: five minorities treaties concluded with Poland, Yugoslavia, Romania, Czechoslovakia, and Greece; special minorities provisions in the four peace treaties imposed on Austria, Bulgaria, Hungary, and Turkey; four subsequent conventions pertaining to Danzig, Memel, Upper Silesia, and the Åland Islands; and five unilateral declarations by Albania, Estonia, Latvia, Lithuania, and Iraq upon their entry into the League of Nations. The wording of these instruments was schematic, with the Polish treaty serving as a model for the others.¹⁶⁰ The proceedings of the Committee on New States and the Protection of Minorities, hastily created on 1 May 1919 to undertake the drafting of the minority rights provisions, offers valuable insights into the negotiations behind the wording of these instruments. The Committee initially consisted of Philippe Berthelot of France, David Hunter Miller of the United States, and James Headlam-Morley of the United Kingdom, with E. H. Carr as secretary. It was later enlarged to include five more representatives from France, the United States, Italy, and Japan; Robert Cecil also attended several meetings as a representative of the League of Nations.¹⁶¹ In total, the Committee held sixty-four meetings between May and November 1919.¹⁶²

National Minorities versus Racial, Religious, or Linguistic Minorities

The starting point of the deliberations of the Committee on New States and the Protection of Minorities was two statements drafted by Woodrow Wilson. One of these was a general statement on religious liberties and will not be discussed here. The other concerned ‘racial or national minorities’ and read as follows:

The State of                covenants and agrees that it will accord to all racial or national minorities within its jurisdiction exactly the same treatment and security, alike in law and in fact, that is accorded the racial or national majority of its people.¹⁶³

It is important to note the reference to ‘national minorities’ here, given that the phrase was subsequently dropped from all official documents. Several references to ‘national minorities’ can also be found in early drafts of the minority rights clauses and in the correspondence of the Committee. For example, the initial draft clauses for the protection of minorities in Poland, put forth by David Hunter Miller at the Committee’s first meeting, made reference to ‘the several national minorities’ in Poland and stated that ‘the Jewish population of Poland shall constitute a national minority’.¹⁶⁴ These clauses were based on Jewish proposals for the protection of minorities that had been given to Miller by Julian Mack and Louis Marshall of the American Jewish Congress.¹⁶⁵ In the end, however, the term ‘national minority’ is nowhere to be found in the League of Nations treaties or declarations.¹⁶⁶ Instead, the minority rights instruments referred to ‘racial, religious or linguistic minorities’.

Why was the term ‘national minorities’ dropped? Pablo de Azcárate, a Spanish diplomat and the third director of the minorities section of the League of Nations, ventured the following explanation in 1945:

[T]he expression ‘national minority’ refers to a more or less considerable proportion of the citizens of a state who are of a different ‘nationality’ from that of the majority. The objection to this definition of a minority is that it involves such an indefinite, and probably indefinable, concept as that of nationality. It was doubtless in order to overcome this objection that the treaties ending the 1914–1918 war, in their provisions relating to the protection of minorities by the League of Nations, did not speak of ‘national’ minorities, but of minorities of ‘race, language and religion.’¹⁶⁷

De Azcárate is no doubt correct when he describes the concept of nationality as ‘indefinite’ and ‘probably undefinable’. However, an analysis of the correspondence and documents of the Committee on New States and the Protection of Minorities suggests that the word ‘national’ was dropped for more specific reasons than. In an illuminating report to the Council of Four – composed of Georges Clemenceau of France, David Lloyd George of the United Kingdom, Vittorio Emanuele Orlando of Italy, and Woodrow Wilson of the United States – on 14 May 1919, the Committee made some observations on the status of the Jews and concluded that they were a ‘racial’ and ‘religious’ minority but not a ‘national’ minority: ‘The other minorities differ from the Jews in that they are national minorities inhabiting in more or less compact bodies certain specified areas. […] The Jews are both a religious and a racial minority, and special questions therefore arise in their case which do not arise in the case of other minorities.’¹⁶⁸ The report then went on to state that Jewish demands to be ‘recognized as a definite nationality which would have separate electoral curias in the Diet and other electoral bodies’ had been unanimously rejected on the grounds that this would amount to ‘setting up a State within a State, and would very seriously undermine the authority of the Polish government’.¹⁶⁹ Even James Headlam-Morley, a vocal advocate of Jewish rights, resisted Jewish demands for national autonomy, noting in his diary in May 1919 that he ‘could not support any claim to “national” rights’ on the part of the Jews.¹⁷⁰

The views of the Committee were reflected in the Council of Four, where both Woodrow Wilson and David Lloyd George expressed the opinion that nothing could be more dangerous than the creation of a Jewish state within Poland.¹⁷¹ Paul Mantoux, the interpreter of the French Prime Minister Georges Clemenceau at the peace conference, also recorded the following comment by Arthur Balfour: ‘Nous ne devrions rien stipuler pour les Juifs, mais seulement pour les personnes de religion juive. Il est dangereux de paraître légiférer en faveur d’une race.’¹⁷² Unsurprisingly, similar protestations were raised by the Polish delegate Ignacy Jan Paderewski, who expressed concern that meeting the Jewish demands ‘would transform the Jews into an autonomous nation’.¹⁷³ To allay these fears, the final draft of the Polish treaty was accompanied by a letter from Clemenceau to Paderewski explicitly stating that the clauses relating to the Jews ‘do not constitute any recognition of the Jews as a separate political community within the Polish State’.¹⁷⁴

In sum, the decision to adopt the phrase ‘racial, religious or linguistic minorities’ instead of ‘national minorities’ was motivated by two interrelated considerations. First, due to the association of nationhood with statehood, the term ‘national minorities’ seemed to justify the creation of states within states – something that the peacemakers desperately wished to avoid. The ‘trinity’ of race, religion, and language was seen to encompass the same content as the term ‘nationality’, minus the dangerous political component.¹⁷⁵ The desire to prevent minority groups from becoming states within states was also reflected in the individualistic wording of the treaties, which referred to ‘members of minorities’ rather than ‘minorities’.¹⁷⁶ The second and related factor behind the decision to avoid the phrase ‘national minorities’ was the lack of clarity about whether the Jews could be considered a ‘national’ minority, given their geographical dispersion and the centrality of religion to their common identity. As a result, the peacemakers were concerned that the term ‘national minorities’ could be exploited by governments to exclude the Jews from protection.¹⁷⁷

Racial Minorities versus Ethnic Minorities

The word ‘ethnic’ is nowhere to be found in any of the English language versions of the treaties. However, it was not entirely absent from deliberations. For example, the records of the United States Department of State use phrases such as ‘ethnic composition’¹⁷⁸ and ‘ethnographic map’¹⁷⁹ on multiple occasions. The French ‘ethnique’ is also widely used in relevant documents, and early French plans for the organisation of the peace conference referred to the rights of ‘ethnical and religious minorities’.¹⁸⁰ Moreover, the records of the Committee on New States and the Protection of Minorities show that correspondence translated from French into English usually rendered the French ‘ethnique’ as ‘ethnic’ or ‘ethnical’. In the final version of the Polish treaty, there is some inconsistency: the adjective ‘racial’ in Articles 8 and 9 was translated into French as ‘ethnique’, but ‘racial minorities’ in Article 12 was translated as ‘minorités de race’. It is unclear why the French text varies between the two formulations while the English text only uses the term ‘racial’, but the most likely explanation is simply that the French language had lacked a direct equivalent of ‘racial’ until just a few years prior. Whereas in the English language the word ‘racial’ had been in use since the mid nineteenth century, the first definition of ‘racial’ in the French language did not appear until 1911.¹⁸¹ In practice, this meant that the French ‘ethnique’ served as the equivalent of the English ‘racial’ until the early decades of the twentieth century. The subsequent separation of race and ethnicity can be seen in Louis Le Fur’s coinage of ‘racique’ as an alternative to ‘ethnique’ in 1921: ‘J’emploi ce néologisme car le terme “ethnique,” qui vient de ethnos (peuple), est ambigu, la notion de peuple se rapprochant plus de celle de nation que de celle de race.’¹⁸²

Some insight into the meaning of the word ‘race’ in the English language is provided by the proceedings of a General Conference held by the Inquiry on 2 August 1918, which included fifteen of the commission’s senior members. When the definition of ‘race’ was brought up, Isaiah Bowman stated that the term should be used ‘only in its ethnological meaning’. The conference subsequently adopted a formulation for ‘boundaries’ that included ‘linguistic, religious, racial, historical, strategic, etc.’ factors.¹⁸³ Among scholars, there is some disagreement over how these statements should be interpreted. According to Volker Prott, the separation of linguistic and religious criteria from racial ones in the formulation for ‘boundaries’ suggests a hereditary understanding of race among members of the Inquiry.¹⁸⁴ By contrast, Jeremy Crampton claims that the deliberations exemplify a ‘socio-cultural’ rather than a ‘hereditary’ conception of race.¹⁸⁵ Both interpretations can be justified and the best explanation is quite simply that a clear distinction between the biological and the sociocultural spheres did not yet exist (see Chapter 2). The question of whether ‘race’ was used in a sociocultural or hereditary sense is thus somewhat misleading.

The uses of the term ‘ethnic’ in these discussions were even more varied than the uses of ‘race’. In a letter to the Greek delegation, for example, the French delegate Berthelot referred to ‘ethnic minorities, such as Mussulmans, Albanians, Bulgarians, Koutzo-Valachs, the Jews of Salonika, and the monks of Mount Athos’.¹⁸⁶ Similarly, the Greek delegate Eleftherios Venizelos wrote of ‘the scholastic liberty of the ethnic minorities’ including ‘Jews and Mussulmans’.¹⁸⁷ From these examples, it is clear that the meaning of ‘ethnic minorities’ was not limited to racial characteristics (however defined) but could also encompass linguistic and religious differences. Insofar as race, religion, and language were viewed as the key components of national identity, this suggests that the phrase ‘ethnic minorities’ effectively served as the depoliticised equivalent of ‘national minorities’.

The language of ethnicity also appears in the documents concerning the reciprocal exchange of populations between Bulgaria and Greece after the war. The idea for the ‘racial adjustment’ of populations originated with the Greek delegate Venizelos.¹⁸⁸ At the peace conference, Venizelos circulated draft clauses proposing the establishment of a mixed commission to oversee the population exchange. The matter was eventually referred to the Committee on New States and the Protection of Minorities, which accepted Venizelos’s proposal to insert a clause into the Treaty of Neuilly to bind Bulgaria to accept forthcoming provisions for a voluntary exchange of populations.¹⁸⁹ Thus, the second paragraph of Article 56 of the Treaty required Bulgaria ‘to recognise such provisions as the Principal Allied and Associated Powers may consider opportune with respect to the reciprocal and voluntary emigration of persons belonging to racial minorities’. In the French version, ‘racial minorities’ was rendered ‘minorités ethniques’. Some of the correspondence of the Committee on New States on this matter also refers to ‘ethnic minorities’, usually in translations of French-language documents.¹⁹⁰

The Treaty of Neuilly was later supplemented by the Convention of Neuilly, which contained the specific provisions for the population exchange. The first draft of the Convention, prepared in French by the Greek delegation and translated into English, referred to the rights of ‘ethnic minorities’.¹⁹¹ The Italian delegation noted the discrepancy between the Greek draft and the other minorities treaties and proposed the following amendment: ‘In order to avoid all ambiguity, it would be preferable to retain, for all alien minorities, the same expression that we find in the treaties for the protection of minorities. It would thus be preferable, instead of speaking simply of ethnic minorities, to say: Minorities of race, religion, or language.’¹⁹² The Committee on New States and the Protection of Minorities promptly accepted the Italian delegation’s suggestion and brought the text of the Convention in line with the other minorities treaties. Article 1 of the final draft accordingly read as follows: ‘The High Contracting Parties recognise in favour of their nationals belonging to racial, religious, and linguistic minorities the right to emigrate freely into their respective territories.’¹⁹³

The Convention of Neuilly was drawn up rather hastily and left many questions to be decided by the Mixed Commission that would oversee the exchange of populations. In 1922, the Commission drew up the Rules on the Reciprocal and Voluntary Emigration of Greek and Bulgarian Minorities.¹⁹⁴ The working language of the Commission was French and the Rules were accordingly written in French.¹⁹⁵ The generic phrase for a minority in the Rules was ‘minorité ethnique’. For example, Article 34 of the Rules required that persons seeking to emigrate acquire ‘un certificat de minorité ethnique’.¹⁹⁶ This ethnic minority certificate, provided by the mayor of the locality where the applicant was domiciled, would serve as evidence that the applicant belonged ‘ethniquement’ to the nationality of the country to which they sought to emigrate. On the model template of the ethnic minority certificate, the Mixed Commission included the following explanatory note: ‘Indiquer la nationalité à laquelle le requérant appartient au point de vue ethnique, religieux, linguistique.’¹⁹⁷ What is interesting here is the reappearance of the two other elements of the trinity (religion and language) alongside ethnicity. This suggests that the meaning of ethnicity oscillated between a broad and a narrow interpretation. Ethnicity in the broad sense encompassed all three qualities of national minorities (race, religion, and language), whereas ethnicity in the narrow sense effectively functioned as a synonym of race, possibly excluding religion and language. However, given the looseness with which the terms ‘race’ and ‘racial’ were used, this distinction was by no means clear-cut. The Mixed Commission appears to have resolved the issue in favour of a broad interpretation when it determined that whenever there was doubt as to whether a person was akin by their race, religion, or language to the people of a country, this doubt should be resolved in favour of the person in question.¹⁹⁸ This flexibility could be taken to mean that the Commission sought to encourage the greatest possible amount of migration so as to achieve homogeneous nation-states and a definitive resolution of the problem of minorities in this region. However, it can also be seen to reflect the importance of the ‘subjective criterion’ in ethnic belonging, thus distancing ethnicity from a biological understanding of race.¹⁹⁹

To sum up, the conceptualisation of minority rights in the aftermath of the First World War was dominated by the trinity of race, religion, and language. Use of the word ‘national’ was almost entirely avoided due to its association with statehood. In this sense, the end of the First World War signalled the moment when the concepts of nationhood and statehood were decisively collapsed into one another in international political discourse – a conceptual union consecrated by the establishment of the League of Nations as an organisation of territorial states. This discursive institutionalisation of the legal-political definition of the nation created a pressing need for an alternate term to designate those stateless nations and national minorities that disrupted the territorial grid of purportedly congruent nation-states. The rather unwieldly phrase ‘racial, religious or linguistic minorities’ was the official name given to the category of non-political nations in the inter-war minority rights treaties, but references to ‘minorités ethniques’ were commonplace in French and the term ‘ethnic minorities’ also makes occasional appearances in English during the drafting process. In this context, ethnicity functioned as a ‘filler’ category, operating in the interstices of race, religion, and language to provide a depoliticised alternative to the language of nationhood and nationality. By allowing statesmen and their advisors to conceptualise minority groups without evoking the spectres of irredentism or secession, the concept of ethnicity was inaugurated as the guarantor of the nation-state and the gatekeeper of international order.

The Problem of Minorities after the Second World War

At the end of the Second World War, the defunct League of Nations was supplanted by the United Nations. Although the language of self-determination was repeatedly invoked during the First World War and the inter-war years, it was never actually incorporated into positive international law. In 1945, by contrast, the principle of self-determination was expressly enshrined in the Charter of the United Nations.²⁰⁰ Thus, Article 1 of the Charter proclaimed that a principal aim of the organisation was to ‘develop friendly relations among nations based on respect for the principle of equal rights and self-determination of peoples’. Unsurprisingly, the United Nations has run into many of the same conceptual conundrums that plagued the architects of the League of Nations minority rights regime. In particular, the references to ‘nation’ and ‘people’ in the Charter provoked controversy due to concerns that they might legitimate secession.²⁰¹ At the time, the use of these terms was justified on the grounds that they encompassed colonies, mandates, and protectorates that did not qualify as states but nevertheless fell within the remit of the United Nations.²⁰²

Like the Covenant of the League of Nations, the Charter of the United Nations does not contain any specific provisions for the protection of minorities. The question of minority rights was instead passed to the Sub-Commission on the Prevention of Discrimination and Protection of Minorities, which published a memorandum on the definition and classification of minorities in 1949. The memorandum emphasised that the term ‘minority’ should not be interpreted in its broad or literal sense to include any social class or cultural group that was dominated by another class or group, but should be applied ‘especially to a national or similar community’.²⁰³ Invoking Ferdinand Tönnies’s influential distinction between Gemeinschaft and Gesellschaft, the memorandum defined a nation as a ‘community’ united by affective factors such as culture or descent, and a state as a ‘society’ or ‘organisation’ united by interest.²⁰⁴ While asserting that ‘most nations have their own State’, the memorandum recognised that the boundaries of nations and states did not always coincide.²⁰⁵ It was precisely to these incongruences between national communities and state boundaries that the category of the minority was to be applied.

Echoing the interwar debates surveyed above, there has been repeated controversy over the use of the phrase ‘national’ minorities’ in international treaties since the Second World War. A very clear example is the drafting of Article 27 of the International Covenant on Civil and Political Rights (ICCPR). The debate that erupted during the drafting process in 1953 centred on the use of the word ‘national’ to describe minorities, with delegates coalescing into three camps: those who favoured the phrase ‘ethnic, religious or linguistic groups within States’; those who favoured ‘national minorities’; and those who proposed ‘national, ethnic, religious and linguistic minorities’ as a compromise.²⁰⁶ The term ‘national minorities’ was backed by the Soviet delegate, who defined a nation as ‘an historically formed community of people characterised by a common language, a common territory, a common economic life and a common psychological structure manifesting itself in a common culture’.²⁰⁷ According to the Soviet delegate, ‘an ethnic or linguistic group could form a national minority, but a group could be called an ethnic or linguistic group long before it had reached the stage of becoming a national minority’.²⁰⁸ Ethnicity was thus understood conceived as something broader than nationality – or, to put it the other way round, nationality added an extra (political) layer to ethnicity.²⁰⁹ The Soviet proposal was met with strong objections from delegates of states that refused to recognise the existence of rival national groups within their territories. According to the French delegate, for example, the Soviet proposal ‘affected only countries where the minorities possessed national characteristics; such cases were not commonly met with in other countries’.²¹⁰ Likewise, the Indian delegate claimed that ‘the Soviet Union proposal created certain difficulties for her country which, while composed of a number of different linguistic groups, had no national minorities’.²¹¹ Representing the compromise position, the delegate from the Philippines was willing to accept the inclusion of the word ‘national’ into the text, but ‘only on the understanding that it would not prejudge the application of the principle of self-determination to the new article’.²¹² In the end, recalling the outcome of the debates in 1919, the controversial term was left out of the ICCPR and the less politicised expression ‘ethnic, religious or linguistic minorities’ was used instead. The ICCPR was adopted by the United Nations General Assembly in 1966.²¹³

Although it was left out of the ICCPR, the term ‘national minorities’ can be found in a plethora of other international agreements relating to minority issues. These include the Convention against Discrimination in Education, adopted by the United Nations Educational, Scientific and Cultural Organization (UNESCO) in 1960; the Helsinki Final Act, adopted by the Conference on Security and Co-operation in Europe (CSCE) in 1975; the Copenhagen Document, adopted by the Organization for Security and Co-operation in Europe (OSCE) in 1990; and the Framework Convention for the Protection of National Minorities, adopted by the Council of Europe in 1994.²¹⁴ Significantly, the word ‘national’ was also inserted into the first international instrument exclusively devoted to minority rights: the Declaration on the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities, adopted by the United Nations General Assembly in 1992.²¹⁵

The idea of a United Nations declaration on minority rights had initially been floated by Special Rapporteur Francesco Capotorti’s Study on the Rights of Persons Belonging to Ethnic, Religious and Linguistic Minorities in the 1970s.²¹⁶ In response, the United Nations Human Rights Commission set up a working group to draft a declaration and Jules Deschênes, a Canadian member of the Sub-Commission on the Prevention of Discrimination and Protection of Minorities, was asked to prepare a report on the definition of ‘minority’.²¹⁷ The Deschênes report, submitted in 1985, recommended that the word ‘national’ be left from any declaration out due to its lack of clarity.²¹⁸ The controversial nature of this term was also reflected in the fact that it was suspended in square brackets for most of the drafting process. In the end, however, the working group decided to ignore Deschênes’s advice, drop the brackets, and include the word ‘national’ in the declaration.²¹⁹ In 1991, the working group published the following summary of the discussions:

Concern was voiced about the addition of national minorities to those listed in article 27 of the International Covenant on Civil and Political Rights. On the one hand, a preference was expressed for focusing on guarantees for national minorities only, because members of ethnic, religious and linguistic minorities should as a matter of existing principles enjoy equality with other citizens of States. It was also stated that there was need to expand article 27. On the other hand, it was said that it would be difficult or even impossible to set up legal distinctions between national and ethnic groups, that the term ‘ethnic’ probably encompassed ‘national’ and that, in order to avoid confusion in different jurisdictions, a formulation including all these elements should be prepared by the Working Group.²²⁰

Further clarity on the choice of terminology came in 2005, when the working group published a commentary on the declaration. Notably, the commentary emphasised that the addition of ‘national’ minorities to the list of minorities to be protected ‘does not extend the overall scope of application beyond the groups already covered by article 27. There is hardly any national minority, however defined, that is not also an ethnic or linguistic minority’.²²¹ The commentary also took care to point out that minority rights were individual rather than collective rights.²²²

All in all, the protection of minority rights accomplishes in international political and legal practice what ‘ethnicity’ accomplishes in conceptual terms. Taken together, the articulation of the concept of ethnicity and the institutionalisation of international minority protection absorb the ‘excess’ of nations that cannot be accommodated within the ontological gridwork of the states system. The international minority rights regime thus functions as a ‘safety valve’ that helps to minimise the threat of secession.²²³ However, precisely by foregrounding the existence of subordinated national communities within states, the discourse of minority rights also serves as a reminder of the nationalist violence that forged the present boundaries of the international order and that may at any moment return to pulverise them again. This ominous underside of minority protection is manifest in the controversies that have surrounded the phrase ‘national minorities’ throughout the twentieth century. In many ways, these terminological difficulties would be simplified if the term ‘national’ were dropped completely – and yet, stubbornly, this term ‘refuses to fade away’.²²⁴ A symptom of nationalism’s internal contradictions, the spectre of the nation haunts the discourse of minority rights.²²⁵

From the Standard of Civilisation to the Standard of Congruency

The preceding sections have shown how the concept of ethnicity and the international minority rights regime emerged in tandem as a means of neutralising the threat that nationalism poses to the international order. As discussed in more detail earlier in the chapter, this neutralisation entails a double move: first, the problem of difference is displaced from the domestic to the international plane such that difference becomes located between (rather than within) nation-states; second, any qualitative differences between nation-states are erased as the concept of the nation is sublated into the concept of the state. Through this double move, nation-states come to be seen as congruent ‘like units’ differentiated from one another only by quantitative factors such as their territorial size and material capabilities. The concept of ethnicity emerges as the particularistic residue of this dialectical process, an undialecticisable kernel excreted by the process of sublation. For the international order, however, this residual difference presents an intractable problem: given that an ethnic group is in essence a depoliticised nation, the nationalist threat is not so much eliminated as merely deferred into the future. There is always the possibility that an ethnic group might become politicised at some later date, leading to new secessionist or irredentist claims. Due to this lingering threat, the double move outlined above has to be supplemented with a third gesture: the displacement of the concept of ethnicity from the ‘self’ to the ‘other’, historically from the West to the non-West. In this way, the domestic hierarchy between the majority nation and ethnic minorities is transferred onto the international plane, producing an international hierarchy between a civic West and an ethnic non-West.

The contrast between the civic West and the ethnic non-West can be traced back to nineteenth-century distinctions between civilised and uncivilised peoples. Among the most influential of these was the contrast between ‘historical’ and ‘non-historical’ peoples that can be found in the works of G. W. F. Hegel, Karl Marx, and Friedrich Engels, among others. In this context, ‘historical’ peoples were conceived as civilised nations that either already possessed a state of their own or had the political capacity to acquire one, and that were destined to play a major role in world history. By contrast, ‘non-historical’ peoples were smaller ethnic units that lacked the capacity to develop into civilised states and were consequently destined to be colonised, assimilated, eradicated, or reformed. Engels notoriously described these uncivilised populations as ‘Völkerabfälle’, loosely translated as ‘ethnic trash’.²²⁶ Analogous distinctions can be found in the works of many other nineteenth-century scholars. John Stuart Mill contrasted ‘civilized’ and ‘backward’ peoples, asserting that only the former were capable of becoming political nations.²²⁷ Similarly, Johann Kaspar Bluntschli argued that only those peoples with ‘manly qualities’ had the capacity to form national states of their own: ‘The incapable need the guidance of other and more gifted nations; the weak must combine with others or submit to the protection of stronger powers.’²²⁸ Despite their individual nuances, what all of these nineteenth-century frameworks had in common was their grounding in a unilinear metanarrative of civilisational progress where nationhood represented the highest stage of civilisation. In the words of Prasenjit Duara, ‘to be a nation was to be civilised and vice versa’.²²⁹ The operative distinction was not (yet) between two different kinds of nation, but between civilised nations and various uncivilised peoples that were denied the status of nations proper.

The link between nationalism and civilisation was broken around the turn of the twentieth century. On the one hand, the dark side of nationalism was made plain by the upsurge of nationalist violence between 1914 and 1945 that has been labelled the ‘European civil war’.²³⁰ Against the backdrop of total war and genocide, it became clear that nationalism did not necessarily go hand in hand with civilisational progress, but could also run counter to it. As Western commentators sought to make sense of these contradictions, binary distinctions between ‘good’ and ‘bad’ nationalisms began to proliferate: Alfred Zimmern distinguished ‘true’ from ‘false’ nationalism in 1918, while Carlton Hayes contrasted ‘original’ to ‘derived’ nationalism in 1928.²³¹ For both scholars, the first type of nationalism remained compatible with the universal values of Western civilisation, while the second entailed its degeneration into mysticism and violence. On the other hand, at the same time that the civilised European nations appeared to descend into a state of primitive anarchy, the supposedly ‘non-historical’ or ‘backward’ peoples were asserting their right to national self-determination ever more forcefully. At the end of the First World War, the disintegration of the Ottoman, Habsburg, and Romanov multinational empires signalled the definitive collapse of the ‘threshold principle’ that had limited the recognition of new nations during the nineteenth century. Henceforth almost any group of people, regardless of its size or perceived civilisational standing, could claim for itself the status of nationhood.²³² The non-European world, too, saw the proliferation of new anti-imperial national and transnational movements at the turn of the twentieth century. By challenging prevalent narratives of racial hierarchy and foregrounding the civilisational achievements of non-European peoples, these movements dismantled the Eurocentric standard of civilisation that had restricted the concept of the nation to a select group of white Europeans.²³³ It was also in reaction to this anti-imperial groundswell that there also emerged, for the first time, a distinct notion of ‘the West’ as a distinct geocultural entity.²³⁴

The turn-of-the-century crisis spurred a fundamental rethinking of the relationship between nationalism and civilisation. With regard to the typologisation of nations, the most important upshot of this crisis was a rotation of the primary axis of comparison by ninety degrees: the vertical distinction between civilised nations and their various uncivilised ‘others’ was now supplanted by a horizontal distinction between Western and non-Western nations. To be clear, this conceptual reorientation did not entail a wholesale erasure of the civilisational hierarchy so much as its internalisation: as the non-historical or backward peoples were belatedly granted the status of nations, these newcomers were also cast as developmentally behind the established nations of the West. One of the most salient manifestations of their developmental backwardness was the supposed mismatch between national and political boundaries. ‘The superiority of Western culture arises from the fact that Western Europe has larger compact ethnological masses, while the East is the classic soil for the fragments of nations’, the eminent German historian and politician Heinrich von Treitschke explained in 1916.²³⁵ Similar distinctions between an ethnopolitically congruent West and a fractured or incongruent non-West can be found in the writings of British, French, and Italian scholars from the same period, although ongoing geopolitical rivalries ensured that they typically excluded Germany from the West.²³⁶ During the First World War, for example, the French philosopher Théodore Ruyssen described Germany, Austria-Hungary, and Turkey as ‘conglomerations of imperfectly absorbed and unequally treated nationalities’.²³⁷

The incipient contrast between Western and non-Western nationalism was systematised and popularised by Hans Kohn’s The Idea of Nationalism in 1944. In chapter VII of this landmark work, Kohn distinguished nationalism ‘in the Western world’ from nationalism ‘outside the Western world, in Central and Eastern Europe and in Asia’. Western nationalism, Kohn explained, was ‘a predominantly political occurrence’ that either preceded or coincided with the formation of the state. As a political project led by a strong bourgeoisie, Western nationalism remained aligned with the liberal – universal values of the Enlightenment. In the non-Western world, by contrast, nationalism emerged at a later stage of development and ‘found its first expression in the cultural field’. Non-Western nationalism therefore ‘grew in protest against and in conflict with the existing state pattern – not primarily to transform it into a people’s state, but to redraw the political boundaries in conformity with ethnographic demands’.²³⁸ Because it was rooted in traditional ties of kinship and status rather than a rational social contract, non-Western nationalism also ‘lent itself more easily to the embroideries of imagination and the excitations of emotion’.²³⁹ Kohn’s distinction between the civic West and the ethnic non-West was subsequently taken up by numerous influential scholars and still serves as a touchstone for nationalism studies today. In testimony to Kohn’s lasting influence, the framework is widely known as the ‘Kohn dichotomy’.²⁴⁰

Significantly, the distinction between the civic West and the ethnic non-West is not merely an academic abstraction, but has also informed international political and legal practice. During the inter-war years, this can be seen in the limitation of the League of Nations minority rights instruments to the new states of Eastern Europe. The exclusion of Western states from the burden of minority protection was based on the assumption that they were sufficiently ‘civilised’ to be able to integrate any existing minorities into their national cultures.²⁴¹ Proposals by Latvia, Finland, and Lithuania in the 1920s to generalise the minority protection regime to include all member states were met with stiff opposition and never made any headway.²⁴² The eastward displacement of ethnicity during the inter-war years is even more explicit in the advisory opinion of the Permanent Court of International Justice (PCIJ) on the Greco-Bulgarian Communities Case in 1930. Although most of the inter-war minority rights instruments were written in individualistic language, the Convention of Neuilly that regulated the exchange of populations between Greece and Bulgaria exceptionally referred to ‘communities’. When Greece and Bulgaria were unable to agree on the meaning of this term, the Mixed Commission overseeing the population exchange sought an advisory opinion from the PCIJ.²⁴³ The PCIJ’s opinion was given in both English and French, with French as the authoritative language.²⁴⁴ In its definition of communities, the PCIJ established an explicit link between the concept of ethnicity and the concept of minority: ‘les communautés ont un caractère exclusivement minoritaire et ethnique’. The English version of the advisory opinion rendered ‘ethnique’ as ‘racial’ throughout. Thus, the above sentence was translated as: ‘communities are of a character exclusively minority and racial’.²⁴⁵ Significantly, the PCIJ’s opinion also referred to a ‘tradition’ of collective identity ‘which plays so important a part in Eastern countries’.²⁴⁶ In this way, the PCIJ projected the concept of ethnicity onto Eastern Europe and excused Western European states from the burden of protecting ethnic minority communities within their territories.

When nationalism returned to European frontpages at the end of the Cold War, so did the practice of imposing unequal responsibilities for minority protection upon Western and Eastern European states. Thus, the universal justice-based track championing individual rights was supplemented by a security-based track that places special obligations on Eastern European states on the grounds that minorities in those states constitute a security threat to the continent. Security was also understood differently for Western and Eastern Europe, with a narrow interpretation of war between states applying to the former and a much broader conception applying to the latter, legitimating Western intervention in Eastern European countries even when there was little or no prospect of outright war. The resultant contrast between an ‘ethnic’ Eastern Europe and a ‘non-ethnic’ or ‘post-ethnic’ Western Europe can be seen in the tendency of some Western European states – notably France – to deny the existence of any ethnic minorities on their national territory.²⁴⁷ ‘Although it has no national minorities on its territory, France, conscious of the importance which this question has for many participating States and of many populations, is ready to participate in the elaboration of conclusions which would be inspired by these ideas and to give them its accord’, the French delegate announced at the CSCE Meeting of Experts on National Minorities in 1991. More specifically, the French delegate distinguished between those states ‘which have been constructed, founded, assembled through a slow economic, social, cultural, and political process’ and those ‘where the entanglement of peoples remains extreme and is the sometimes recent reminder of tumultuous upheavals’. Minority rights, the French delegate insisted, were only relevant for the latter.²⁴⁸ More generally, the externalisation of ethnicity beyond the West is manifest in how phenomena such as ‘ethnic cleansing’ and ‘ethnic conflict’ are primarily associated with Eastern Europe and the Global South.²⁴⁹

The construction of an international hierarchy between the civic West and the ethnic East is the final step through which the phantasmatic ideal of the congruent nation-state is made present in the West.²⁵⁰ Thus, even when Western nations are described as multicultural societies that encompass a plurality of ethnic groups, they are simultaneously presented as well-integrated communities bound together by an overarching civic culture. Meanwhile, non-Western nations are said to be plagued by a perennial mismatch between the boundaries of the nation and the boundaries of the state. The most salient manifestation of this imputed incongruence is the tendency to characterise many non-Western countries – especially those of postcolonial Africa – as ‘artificial states’ without ‘natural boundaries’.²⁵¹ The dangers and shortcomings of non-Western nationalism are thus perceived to stem not simply from its backward-looking attachment to an ethnic or organic conception of the nation, but more specifically, from the alleged mismatch between political and ethnographical boundaries. It is this lack of congruence – the gap between the nation and the state – that is said to produce the emotional and violent tendencies of non-Western nationalism by pushing non-Western nations to ‘compensate by overemphasis and overconfidence’ for their developmental backwardness.²⁵² In the final analysis, the ‘ethnic’ quality of non-Western nationalism does not refer to a positive presence, but to a constitutive gap, a traumatic fissure, that fractures the nationalist project from within: the gap between the universalistic conception of the nation as a legal or political unit, on the one hand, and the particularistic conception of the nation as an ethnic or cultural community, on the other. The international hierarchy between the civic West and the ethnic non-West emerges when this inner contradiction of the nation concept is captured, reified, and projected onto the international plane.