What is race? Epistemic ambiguity and liberal international order

Abstract

There is increasing interest in how anticolonial actors advanced a norm of racial equality in mid-century formations of liberal international order (LIO). Less attention, however, is afforded to simultaneous epistemic conflicts over the scientific object of ‘race’ and their political effects. During postwar order-building and alongside political struggles for racial equality, there was wide and deep scientific debate on the analytical utility of race as a means to categorize human diversity. Race, I demonstrate, was rendered as epistemically ambiguous, caught between social scientists and philosophers who understood it as a social construct akin to ethnicity and natural scientists who maintained a biological basis. This split was not confined to academic debate but shaped political and normative struggles over the institutionalization of racial equality in LIO. Adopting an object-oriented approach, I argue that the epistemic ambiguity of race generated political effects, at once permitting the reproduction of colonial logics in LIO as well as providing latitude for strategies of resistance. Rather than a linear causal effect, I empirically map the work that epistemic ambiguity performed in the creation of mid-century international order.

International Relations (IR) scholars have persistently drawn attention to the role of race and racism in explaining international politics.¹ Recently, amid ‘challenges’ to the liberal international order (LIO) and the revival of questions of institutional racism, some have called for greater attention to the role of race and racism in the formation of the LIO.² Indeed, within global governance and law, managing international problems around race—such as transnational and domestic racism, apartheid and colonial occupation—has a history that overlaps with the mid-century building of the LIO.³

However, the nature of this relationship remains uncertain. For some, struggles for racial equality are constitutive of the LIO. For others, these anti-racist mobilizations constitute an alternative vision of international organization that sits outside liberal forms of political order. However, both positions tend to focus on politico-normative contestation waged by states, explaining the dynamics between anti-racist coalitions and the LIO in terms of the changing balance of ideological interest within institutions such as the United Nations. To be sure, the rise of newly independent African and Asian states did allow for opportunities for anti-racist coalitions to emerge and articulate what I would term democratic demands for equality. However, missing in these accounts is an appreciation of epistemic conflicts over ‘race’ as a scientific object for parsing human diversity and their connections to normative, legal and institutional outcomes. During mid-twentieth century formations of LIO, we see not only actors mobilizing to place the norm of racial equality within key institutional sites but a set of debates problematizing the object of race that shaped the character of these mobilizations. The central contention in this article, then, is that the institutionalization of international norms and law around racial equality cannot be properly grasped without attending to epistemic contestation and contradictions around the object of race. Race and (anti-)racism are linked.

How did questions about the ‘truth’ of race and racial hierarchies shape normative struggles over the institutionalization of racial equality in the LIO? To answer this question, I adopt an object-orientated approach, which foregrounds knowledgeable practices in the constitution of epistemic and global governance objects and highlights the ways in which the specific production of an object shapes political possibilities. However, I conceptually advance this literature by drawing out the productive role of epistemic ambiguity. Typically, object-orientated accounts focus on how knowledgeable practices make objects legible in distinct ways, reducing uncertainty and distributing political possibilities. But what about objects that are rendered as ambiguous, whose epistemic borders are contested or blend into other similar objects? What happens when scientific or expert discourses do not converge on the contours of an object, but diverge? What effects result?

I theorize that the ambiguous epistemic constitution of objects does not (necessarily) mean that such objects are politically irrelevant; rather, ambiguity generates elasticity in how actors struggle over norms and their institutionalization. Specifically, I argue that in mid-century discourses race became a problematic object, due to the rise of novel theories of race and the moral wake of Nazi Germany's genocidal programme, which was premised on state racism. The result was not clarity over the nature of race⁴ but an ambiguous and contentious split among social scientists who viewed it as a social construction (with no real biological referent), those who pushed further to view it as a specious category produced by European imperialism and chattel slavery, and natural scientists who maintained that race remained a useful analytical category. By the middle of the twentieth century, ‘race’ was an ambiguous object without clear distinctions between it and related objects such as ethnicity, culture, colour or national origin. Based on this work, I demonstrate how anti-colonial actors mobilized epistemic concerns linking imperialism to racial hierarchy, repositioning concerns about racial equality into trenchant critiques of colonial domination. However, the ambiguity of race meant that such strategies were constrained and that they proceeded alongside the reproduction of colonial logics within the LIO.

In this article I do not offer an answer to the question ‘what is race?’, but I show how unresolved debates around this question shaped the institutionalization of norms of racial equality in international law and global governance. In the next section, I situate a conceptual concern with epistemic ambiguity in the literature on political objects before making a case that there is insufficient attention paid to the ambiguity of race in mid-century LIO. I then discuss how race became an ambiguous object and chart its effects, taking care to delineate the work that ambiguity performs as opposed to other possible constructions of race. Finally, I conclude by outlining my theoretical contributions and discussing normative implications for policy-makers concerned with the future of international order amid the rise of majority-world powers and alternative demands for how we collectively live in the world.

Knowledge, epistemic objects and international order

This article posits that we can better understand the historical relationship between anti-racism and the formation of LIO via recourse to epistemic questions around race. That knowledge shapes international politics and the production of political order is hardly a novel claim, but further consideration could be afforded to theorizing the political work of epistemic ambiguity, particularly in relation to the connection between race and the formation of the mid-century LIO.

Ambiguous objects and international orders

To think on epistemic ambiguity and its effects on international order, I work from an object-orientated framework, or a set of approaches that foreground how knowledge constitutes political objects.⁵ Rather than start with key actors or discursive formations that produce subjects' identities and interests, object-orientated literature focuses on the constitution, transformation and death of epistemic categories and concrete problems of governance. From this view, knowledgeable practices create objects (such as migration, development and the economy)⁶ that come to orientate diverse actors and stabilize political order. At once ontologically material and ideational, an explicit focus on the constitution of objects that orientate political action places greater attention on the ways in which discourse and materiality co-produce the world, refusing to resolve either as fundamental.⁷ Such objects are always incomplete—only a ‘temporarily stable’ entity that requires ‘ongoing maintenance work’⁸ or fundamentally open and ‘unfolding structures’⁹—but the working presumption is that epistemic practices are important in that they shore up borders for objects for some period of time.

An object-orientated approach is attentive to the problematization of specific issues, but it also clarifies broader effects on the way these problems come to be governed.¹⁰ That is, most accounts share an obvious commitment: knowledge helps to render the world and our identities legible, permitting different kinds of political actions or interventions. For instance, Bentley Allan argues that the climate is an object of governance co-produced by scientists and states that shaped global climate policy.¹¹ He shows that a geophysical model of the climate, with its emphasis on ‘prediction and control’, has pushed governance towards regulating greenhouse gas emissions rather than experimenting with solutions that address broader ecological challenges.¹²

But what happens when objects are ambiguously produced? If epistemic practices are never complete, there may be ambivalences, uncertainties and deliberate silencing in the constitution of objects that inform how actors engage with them over time. This claim is taken up by those interested in epistemological ignorance, or the ways in which strategic silence or a willing unknowing is deployed to reproduce institutional hierarchies. Charles Mills's racial contract, for instance, is premised on an active forgetting of racism, slavery, genocide and imperialism in the representations of our world, making possible the reproduction of white supremacy.¹³

To be sure, there is a growing literature on epistemological ignorance.¹⁴ Within political science and IR, studies of ignorance can run the empirical gamut: from governance of finance and neoliberalism¹⁵ to war.¹⁶ Typically, strategic forms of ignorance are analysed. Jutta Bakonyi argues that ignorance is strategically used in peacekeeping missions,¹⁷ and Jacqueline Best shows how ambiguity and ignorance function together within the World Bank, shaping development discourse.¹⁸ Most recently, Christina Boswell and Elisabeth Badenhoop consider states of ignorance in relation to governance of migration in the United Kingdom and Germany,¹⁹ and Claudia Aradau and Sarah Perret discuss varied modes of ‘non-knowledge’.²⁰

Epistemic practices, then, not only assemble distinct objects of governance, but the silences and exclusions that precipitate are also significant. While epistemological ignorance focuses on strategic absences, there is a related analytical question of ambiguity in the epistemic production of objects. Scientific expertise is not uniform, and heterogeneous practices, tools, disciplinary contexts and technologies may arrive at multiple and competing representations.²¹ Indeed, Marina Pagliarello maintains that boundary objects, or objects that are imbued with multiple and contradictory meanings, provide the very drivers of ambiguity in policy settings.²²

While epistemic debate is inherent to the process of object constitution, such debate often produces distinct or clearly defined objects. But as I discuss below, debate may be interminable or may produce incommensurate meanings among different actors. Of course, many epistemic categories could be considered ambiguous in the sense that they are always exceeded by a hybrid, shifting terrain of multiple intersecting identities.²³ As postcolonial scholars such as Homi Bhabha have argued, clear and stable epistemic categories are always disrupted by a ‘third space’ defined by cultural hybridity and ambivalence.²⁴ Here, however, I present epistemic ambiguity not in a phenomenological sense, in the ambivalence of how we experience scientific objects or the gaps between epistemic objects and a hybrid world in constant flux, but as contestation and ambiguity in the very production of these objects.

I identify two processes that produce ambiguity. First, epistemic contestation around the nature of objects may not lead to consensus but stalemate. In an object-orientated approach, as Ole Jacob Sending argues, expertise is the product of recognition struggles wherein actors compete to define objects in particular ways, with some able to leverage their positions or capital to produce hegemonic understandings.²⁵ However, I contend that persistent disagreement or a plurality of epistemic positions may result when institutional infrastructures support incommensurability. For instance, disciplinary siloing can erode the capacity of powerful actors to advance their framing of an object as universal or uncontested, allowing for competing framings and approaches to persist. Second, there may be several overlapping objects that complicate or otherwise muddy the socio-material borders of an object. The nature of object constitution involves distinguishing an object from other things and processes in the world, but this may not always be clearly delineated. Epistemic objects may blend into or otherwise overlap with different types of objects.

In some circumstances, lack of consensus on the boundaries of an epistemic object may simply mean that it does not get lodged as a problem of governance, remaining outside the agendas of states and other political actors. But the situation for already existing objects that are integral to institutional practices and order will be more complex when problematizations lead to ambiguous borders. By the twentieth century, for instance, race was materially embedded within infrastructures of colonial occupation, capitalist trade relations and intergovernmental organizations. Amid epistemic challenges, such an object may not simply disappear but generate more complex effects on political contestation and global governance.

What precise effects would result? Epistemic ambiguity generates elasticity, allowing for both the reproduction of hierarchical relations and interruptions to such orders. On the one hand, ambiguity permits a wider latitude of governance practices that allow for the reproduction of inequality within changing political orders. Lack of clarity can be productive for actors to reproduce hierarchical relations. As Ann Stoler argues, European empires—or what she terms imperial formations—were not coherent hub-and-spoke systems with clear and unchallenged lines of authority, but uneven, shifting and contradictory messes of practices that depended upon ambiguity for reproducing colonial rule in different parts of the world.²⁶

On the other hand, the same ambiguity of epistemic objects also permits creative resistance to inequalities in political order. Ambiguity loosens the hold that ontological categories have in defining actors' identities and place in the world, providing ‘slack’ to experiment outside established normative frameworks.²⁷ Such uncertainty and ambivalence in epistemic objects allow for latitude in how actors can recombine and reframe strategies of resistance to upend inequality. As Tiffany Lethabo King demonstrates with the concept of the ‘shoal’, an unruly meeting place of water, sand, rock and reef that ‘exceeds full knowability/mappability’ and is always unstable, such ambiguity can be productive for collective encounters that disrupt colonial logics.²⁸

Thinking with these positions, then, I maintain that there is an irreducible complexity in the effects of ambiguous objects—both the reproduction of unequal order and opportunities to resist. The task at hand is not to theorize linearly efficient effects of ambiguous objects but to map the ways in which ambiguity emerges and shapes political struggles to reproduce or interrupt political order, taking care to delineate how these effects are distinct from unambiguous objects. As I elaborate in the next section, the object of race is a fruitful one to unpack this theoretical position.

Mid-century international order: theories of race and norms of racial equality

There are different literatures that consider the relationship among race, racism and the LIO.²⁹ Postcolonial and decolonial accounts have been integral to understanding how post-Cold War global governance, orientated around liberal norms and institutions, is not inimical to but implicated in imperial dynamics and racial hierarchies.³⁰ Other accounts trace the imperial and racial constitution of liberal institutions and norms in the early to middle years of the twentieth century.³¹ All of these approaches highlight the way in which liberal forms of political order are entangled with colonial practices and racial hierarchies.

Recent work attends to the struggles over the place of racial hierarchies and anti-racist norms within order-building. Marcos Tourinho, for example, sees postwar order as co-constituted by the United States and its allies and anti-colonial actors (states and transnational movements).³² Here, Tourinho positions the latter's efforts as being against the interests of western states but logically internal to liberal order. Zoltán Búzás, meanwhile, addresses the evolution of the LIO through ‘racial diversity regimes’,³³ arguing that the settlement of such regimes depended upon the contests between a traditional coalition, made of colonial powers, and a transformative one. For Búzás, these struggles to advance racial equality within international order involved political contestation within and outside emerging institutions of the LIO.

Amitav Acharya departs from both positions to emphasize the extent to which norms around racial equality formed the basis ‘of an alternative conception of world order’, itself built by anti-colonial actors articulating ‘an intrinsic link between colonialism and racism on the one hand, and between decolonization, racial equality and human rights on the other’.³⁴ Echoing Adom Getachew's impressive recovery of anti-colonial thought and practice in the twentieth century,³⁵ these initiatives sat outside the emerging institutional framework of the LIO. Indeed, for Acharya (and others), the building of the LIO following the Second World War did not represent a ‘fundamental break’ with what came before.³⁶ Rather, as postcolonial theorists have long argued, imperial practices persisted into a world of juridically equal states.

This literature demonstrates the role of western states in supporting and imperfectly reproducing imperial relations and white supremacy within nascent institutions of LIO while also attending to simultaneous struggles to upend such projects. Implicit, here, is that there are traditional/colonial and transformative/anti-colonial coalitions of states, and that we can understand the relationship between racism and LIO in terms of the shifting ideological power of newly independent states or the ‘revolt against the West’.³⁷ However, what is missing is sustained attention to the role of epistemic contestation and its connection to order-building.³⁸ Alongside anti-colonial praxis pushing to inscribe the value of racial equality and upend colonialism was epistemic contestation over the very category of ‘race’. Throughout the first half of the twentieth century, both racial hierarchy and racialism, or the scientific legitimacy of characterizing human diversity in terms of race, were called into question. There was deep disagreement among experts over the meaning of race—whether it was a useful analytical category that grasped genetic difference and phenotypical markers or a social construct, limited in utility compared to similar concepts such as ethnicity, and possibly derivative of European colonialism and the transatlantic slave trade. While cultural anthropologists, sociologists and philosophers began to dissociate race from biological referents, presenting it as a social construct, physical anthropologists and geneticists held on to race to parse human diversity. These were not narrowly ‘academic’ concerns detached from broader political issues; rather, epistemic contestation ran through different institutional sites, including the newly created United Nations Educational, Scientific and Cultural Organization (UNESCO). At issue amid postwar order-building was not just its architecture—still rooted in white supremacy—but the socio-material fact of human diversity still ontologically set within racial categories. The relationship between race and racism, then, the discourse around the quality of human difference (i.e. racialism) and whether that difference mattered for human intelligence and achievement (i.e. racial supremacy), were necessarily interwoven.

Below, I work through the argument that epistemic contests were linked to politico-normative struggles over racial equality in two steps. First, I historically chart the shift from a relatively stable object of race in the nineteenth century, noting how biological conceptions of race produced specific effects, to the production of its ambiguous character in the twentieth century. Here, I focus on UNESCO's statements on race, which constituted an attempt both to define race and to dispute the scientific basis for racial hierarchies. Second, I demonstrate the work that epistemic ambiguity performs in the governance of racial discrimination by showing how a lack of clarity over the object of race shaped its making. The political effects of ambiguous objects are not efficient; instead, epistemic ambiguity led to multiple and contending practices. It permitted the reproduction of colonial logics of government within a nascent LIO, but it also offered space through which anti-racist coalitions could articulate the problem of racism as epistemically linked to imperialism. The resulting international legal instrument (which focused on states taking measures to address racism within their own borders) might have been a weak tool that did little to address structural racism within international order.³⁹ But we should understand the mixed outcome of this instrument, the International Convention on the Elimination of All Forms of Racial Discrimination (ICERD), as not simply a result of negotiations between contending coalitions, but constrained by the ambiguity of race.

In focusing on these sites, my intention is not to dismiss other areas in which anti-colonial actors articulated properly decolonial practices of ‘delinking’ from colonial structures, such as at the Bandung Conference of 1955⁴⁰ or the Asian Women's Conference of 1949.⁴¹ Instead, I discuss the ways in which anti-colonial actors made democratic demands against the inequalities of liberal order-building. Contra Tourinho, these demands were not co-constitutive of liberalism, but democratic interruptions that challenged liberal institutions and values.⁴²

Producing an ambiguous object

From biological fact to ‘ethnic fiction’: race prior to the Second World War

There is debate on the origins of ‘race’. Some argue that racial thinking was operative in medieval Europe,⁴³ while others hold that the systematic classification of humanity along racial lines only emerges during the rise of modern European imperialism, capitalism and the transatlantic slave trade.⁴⁴ Widely accepted, however, is that with the institutionalization of biological sciences in the nineteenth century, race in European discourse was figured as a scientific concept that parsed biological differences among people. For Europeans, the world consisted of a handful of races, determined by blood, which marked out capabilities and constraints: suitability for different environments, propensities to cultural practices, and—significantly—degree of intelligence for self-governance. Scientific racism constituted the epistemic foundations of thinking on race, institutionalizing white supremacy in western academia, imperial and domestic governance, public sentiment, and international order.⁴⁵

So deeply ingrained was scientific racism that by the middle of the nineteenth century, as Alexander Barder argues, ‘race as a concept … made the world intelligible’.⁴⁶ Scientific racism posed races as not merely distinct variations of humanity, but fundamentally different and ‘incompatible’ species, a particular construction of ‘race’ that made possible concrete effects.⁴⁷ It gave rise to the political problem of race war, permitting a range of policy practices such as the expansion of imperialism and genocidal violence. Rendered in the language of object-orientated IR, Barder, then, provides insight into how the clear epistemic borders of race as a biologically determined category generated a set of violent political practices.⁴⁸

This epistemic view of race was hegemonic until the period between 1920 and 1940 when there was a ‘Copernican revolution’ in scientific thinking on race.⁴⁹ The idea that human capacities were constrained or otherwise created by blood was scrutinized by social scientists in general and Black scholars in particular. Energized by Franz Boas's interventions in anthropology that collapsed distinctions between race and ethnicity, the object of race as a meaningful category of human difference was problematized.⁵⁰ Rather than biological fact, something static and constraining, human difference became increasingly seen as a matter of heterogeneous and dynamic material and cultural practices. Alain Locke, for instance, not only dissociated race from biological referents but made a clear case for the social, political and historical production of race in 1916 by setting the specious concept of race as derivative of imperial conquest.⁵¹

UNESCO statements on race: epistemic contestation

The problematization of race deepened after the Second World War when—on the recommendation of the Second Session of the Commission on Human Rights⁵²—the UN Economic and Social Council requested UNESCO to study the question of race. UNESCO assembled an international group of scientific experts from the biological, social and philosophical sciences (though it ended up largely consisting of social scientists), to definitively affirm the nature of race, the myth of racial hierarchy and solutions to problems of racial prejudice. The impetus to do so was complex, involving a collective sense of responsibility for the Holocaust and, importantly, the contradictions of liberal invocations of a new international order amid the continuation of European colonialism and US segregation.⁵³ The turn to science to dissolve racist thinking indicated how much faith was placed in this kind of expertise postwar. As Michelle Brattain writes, ‘science’ promised an anti-racist knowledge that would both dismiss scientific racism and highlight a more plastic understanding of how environmental and genetic factors produced ‘human variation’.⁵⁴

In 1950 UNESCO put out its first statement on race.⁵⁵ The statement begins boldly and unequivocally: ‘Scientists have reached general agreement that mankind is one: that all men belong to the same species, Homo sapiens.’⁵⁶ Rejecting the idea that races were distinct species, the statement affirms that from ‘the biological standpoint’, race may refer to several different populations that constitute humanity, but whatever the physical differences, they are ‘variations’ upon ‘a common theme’.⁵⁷ Genetic concentrations in these populations exist, but they change and disappear over time.⁵⁸ As such, for many, save the biologists studying genetics, the related object of ‘ethnic group’ is more suited to orientate discussions around national, religious, cultural and other forms of social difference. Indeed, for ‘all practical social purposes “race” is not so much a biological phenomenon as a social myth’.⁵⁹

Cultural ‘achievements’ are, therefore, not attributable to genetics, and the statement highlights that the ‘one trait which above all others has been at a premium in the evolution of men's mental characters has been educability, plasticity’, something that ‘all human beings possess’.⁶⁰ The statement concludes with an ethical commitment to human equality that ‘in no way depends upon the assertion that human beings are in fact equal in endowment’. Individual and group differences exist, but the latter has been ‘exaggerated’ to ‘[question] the validity of equality in the ethical sense’.⁶¹ As Sonali Thakkar argues, the 1950 statement, alongside others, is less about a ‘paradigmatic shift’ when it comes to natural scientific thinking on race than an ‘ethicopolitical aspiration’ and pedagogical tool premising anti-racism on the educability of all.⁶² Rather than a fundamental break from the logic of colonial orders, it reproduced a familiar liberal justification of empire in that all were educable.⁶³

As discussed in prefatory notes,⁶⁴ the 1950 statement did not represent the view of physical anthropologists and geneticists, resulting in a further statement in 1951. There was deep contestation over the nature of race as viewed from different disciplines.⁶⁵ Though Ashley Montagu, an anthropologist who studied under Ruth Benedict and who found the concept of race scientifically useless, was a prominent contributor to the first statement and is credited for reducing the importance of race as a scientific concept in the statement, his views were not shared by all.⁶⁶ Indeed, the 1950 statement was contentious among the transnational scientific community, with many European physical anthropologists and geneticists rejecting the idea that race was more ‘social myth’ than ‘biological fact’.⁶⁷

Perrin Selcer uncovers the complexity of this contest, discussing how physical anthropologists were at a particular ‘vulnerable moment’ wherein dismissals of taxonomic classifications of races threatened their disciplinary ‘turf’ and the rise of population genetics promised more sophisticated approaches to understanding the dynamics of human diversity.⁶⁸ Thus, in the 1951 statement representing the natural sciences, the sweeping claim to sameness across humanity is qualified by raising the question of how ‘different human groups diverged from this common stock’⁶⁹ and affirming that race ‘is unanimously regarded by anthropologists’ as a productive tool to classify humanity and study its evolution.⁷⁰ Humanity, the statement maintains, can be classed into three major groups despite the complexity of differences.⁷¹

Robbie Shilliam reads the 1950 and 1951 statements as evidencing the underlying episteme of postwar order-building, one that separated ‘race from politics’ and ‘allowed no place for the ongoing story of the sufferers [of colonial subjugation] and their epistemic and practical strategies for the meaningful re-humanization and reclamation of personhoods’.⁷² Attending to the link between epistemic questions and the creation of political order, Shilliam argues that these statements enabled western states at once to ‘sweep away’ their own sins and ignore the ongoing racial orders that continued to shape domestic and imperial relations.⁷³

UNESCO published four statements on race, in 1950, 1951, 1964 and 1967, each one implicitly or explicitly representing either a sociological or biological emphasis. While Shilliam and Thakkar rightly identify the ways in which imperial relations were reproduced, reading all four UNESCO statements complicates a linear narrative of uncontested reproduction of colonial logics and attunes us to the epistemic divisions underpinning these statements. The 1964 proposal, premised on outlining the ‘biological aspects’ of race, reduces the concept of race as ‘purely biological’, concluding that the biological data do not support racism.⁷⁴ The 1967 statement on ‘race and race prejudice’, however, focuses on racism as a product of imperialism and slavery, sharpening a particular sociological critique of the West.⁷⁵ The problem of racism is a false discourse claiming a scientific or biological basis for human difference—one that has emerged from ‘the conditions of conquest, out of the justification of Negro slavery and its aftermath of racial inequality in the West, and out of the colonial relationship’.⁷⁶

Rather than an emerging consensus, multiple statements evidence the split among social and natural scientists and, therefore, the production of an ambiguous epistemic object. Foregrounding the knowledgeable practices that constitute objects, I argue, demonstrates how ambiguity is generated by the unclear distinction between related objects such as ‘ethnicity’ and interdisciplinary contests. A split was emerging between the social and natural sciences (and indeed between physical anthropologists and geneticists) over the epistemic quality of race and its analytical utility: whereas social scientists sought to abandon the concept, preferring related ones like ‘ethnic group’,⁷⁷ natural scientists affirmed its analytical utility. The epistemically ambiguous nature of race was not just differing ideas about humanity but was embedded in material practices and supported by institutional investments that made scientific divergence possible. UNESCO itself was administratively ‘divided between departments of natural and social sciences’.⁷⁸ Moreover, different practices around the study of racial differences continued, as was the case with the State Institute for Human Genetics and Race Biology in Sweden, led from 1935 to 1956 by Gunnar Dahlberg, an author of the 1951 statement. The institute oversaw the continuation of race science even as it claimed an orientation to population genetics.⁷⁹

If these statements help us understand the production of ambiguous objects, what effects result? Getting at the complex and contradictory effects of the statements entails contending with the work that epistemic ambiguity performs. As theorized above, epistemic ambiguity permits elasticity in how actors wage politico-normative contests, allowing for both the reproduction of domination and strategies of resistance. In the next section I turn to these effects, arguing that we cannot account for the relationship between racism and the LIO through the changing balance of ideological power brought on by newly independent African and Asian states. Exploring the counterfactuals, I show how the making of the ICERD was shaped by ambiguity in the meaning of race.

Legislating anti-racism: ICERD, race, and overlapping objects

The ambiguities of race and its political effects appear in 1965 with ICERD, an instrument that continues to serve as the core international institution for governing racism.⁸⁰ Widely ratified by states (though with reservations from several western states about regulating speech),⁸¹ the Convention commits states to the project of anti-racism.⁸² A handful of African states put this problem on the international agenda, arguing that dealing with racial hierarchy was indeed intimately related to maintaining peace and cooperation in the world.⁸³ Drawing on principles laid out in the UN's Declaration on Human Rights (UNDHR, 1948), as well as previous declarations against racism in employment and education and against colonialism, the Convention makes clear the fundamental equality of all and puts the onus on states to eliminate all forms of racial discrimination within their borders. That is, alongside the UNDHR, the Convention is a legal instrument that takes the individual rather than the state as the rights-bearing subject of international law.⁸⁴

Immediately, though, ambiguities in terms of what race signified enveloped early discussions. Precursors to the Convention, for instance, included a rise in anti-Semitism in 1959–1960⁸⁵ leading to the pairing of racial and religious intolerance in UN spaces, but a decision was taken to separate religious discrimination from racial discrimination, with the latter taking precedence.⁸⁶ But parsing race from religion was only one of the epistemic issues that shaped governance of racial discrimination. Those involved in drafting the Convention found it difficult to disentangle race from other related objects, especially as they involved different material meanings in different places. ICERD, for instance, begins by noting that ‘racial discrimination’ refers to ‘any distinction, exclusion, restriction or preference based on race, colour, descent, or national or ethnic origin’.⁸⁷ As such, racial discrimination was not only about ‘race’ per se, but skin colour, ‘descent’, nationality and ethnicity.

Such ambiguous overlap with other epistemic objects prompted sustained debate. National origin, for instance, generated confusion over the idea of nationality as a ‘politico-legal’ term referring to membership of a state, common in international law, or a ‘historico-biological’ one, common in the social sciences.⁸⁸ Legal opinion at the time noted inconsistencies and ambiguities in the final text, arguing that both senses are apparent.⁸⁹ But the inclusion of national origin may be a link to other similar conventions on human rights, civil and political rights, and employment and occupation.⁹⁰ Other terms that are connected to ‘race’ include ‘descent’—a category that is not found in other similar conventions and does not specify its meaning against other categories⁹¹—and ‘ethnic origin’.

Indeed, slipperiness between race and ethnicity occurs in article 1, paragraph 4 of the Convention, where ‘special measures’ are permitted to promote the ‘advancement of certain racial or ethnic groups or individuals’ without it being classified as racial discrimination, so long as it does not ‘lead to the maintenance of separate rights for different racial groups’.⁹² Why both racial and ethnic—but not national or other terms—are used here is not clear. Similarly, article 7 references the necessity for states to advance education and information that combat racial discrimination and promote ‘understanding, tolerance and friendship among nations and racial or [ethnic] groups’ in accordance with similar legal instruments. Here, the collapsing of race and ethnicity in relation to national origin is apparent though, again, the politico-epistemic rationale is obscure.⁹³

Not only was race scientifically ambiguous, but linguistic differences also came to make the fundamental object of race a difficult one to define, further evidencing the material complexities of diplomatic practice in addressing governance problems. In discussing article 1, the representative of Poland at the UN General Assembly cautioned that different cultural communities epistemically viewed the same concept differently, noting that in ‘English and French, “race” represented a sociological concept and “colour” an anthropological one, while in other languages, such as Polish, “race” and “colour” were concepts of physical anthropology exclusively’.⁹⁴ In other places, distinctions between ‘sociological’ and ‘legal’ concepts were discussed.⁹⁵

Crucially, it was not only the ambiguous overlap of multiple objects that generated governance constraints in the making of ICERD. There was also the question of whether racial differentiation itself constituted a form of discrimination. As Patrick Thornberry shows, drafters of the Convention grappled with whether ‘doctrines’ of racial differentiation and racial superiority were to be rejected as scientifically false or whether the root cause of discrimination depended only on the latter.⁹⁶ Both sides of the debate drew from the UNESCO statements to stake their claims, with some pointing to the clear existence of racial differentiation and others pointing to the idea that ‘race’ itself was scientifically false. The preamble of the Convention ultimately only identifies racial superiority based on racial difference as a problem, leaving open whether race could be a meaningful object to categorize human diversity.

While ICERD marks the emergence of racism as an object of global governance, the epistemic object of race complicated its legal institutionalization. Race had different epistemic and legal meanings depending on geography, making it difficult to arrive at some common minimal understanding. That is, what race meant as opposed to other objects was unclear: it seemed to overlap with religious categories (some of which were posed in quasi-biological and physiological terms), nationalities (understood in legal and ‘historico-biological’ terms), and ethnic groups (defined culturally and geographically). Moreover, anti-racists who held on to a social notion of race found themselves in a difficult position: racial emancipation implies uplift of particular racial groups that have faced discrimination, thereby reinforcing the idea that race is a meaningful way to organize human diversity. In other words, those parties playing an active role in drafting the Convention could not reasonably define race in a legible and universal way, such that one could know when discrimination targeted a ‘racial’ group as opposed to some other kind of social group, as this would imply that there are meaningful and relatively fixed racial groups rather than a singular or common humanity.

While we cannot discount how the shifting balance of ideological power at the UN helps to explain the fraught integration of anti-racism into international law and institutions of the LIO, I maintain that more is needed to understand the complex unfolding of this relationship. Epistemic ambiguity provided opportunities and constraints for postcolonial actors to make democratic demands in this historical formation of LIO.

The significance of epistemic ambiguity can reasonably be evidenced by exploring counterfactuals.⁹⁷ Had the object of race been clearly bordered—hegemonically defined such that there existed some shared criterion of what race meant—governance of racial discrimination would look different. For instance, had it been bounded as a biological thing, the Convention may have focused on promoting educational and welfare policies that ensured equal opportunity to genetically distinct, minority populations. Were it clearly defined as a social fiction in which theories of racial differences were rendered as false, policies may have been orientated towards denying the use of race as a classification of people within state policies, scientific research and educational content. Yet, we do not see such specificity in how states should tackle racism.

Similarly, were deliberate ‘silencing’ of race significant as might be expected in the literature on epistemological ignorance, we would expect a different outcome in relation to ICERD. To consider this counterfactual, we might look to the example of post-revolutionary Cuba, where the state claimed that socialism had dissolved racial antagonisms.⁹⁸ Analogously, we might expect strong resistance to formulating international law on racial discrimination or claims that other liberal instruments such as the Universal Declaration of Human Rights were already sufficient. At the very least, we might expect limited mechanisms for collecting information on race and racial discrimination, but here the Convention is relatively robust, requiring biannual reporting on actions taken by party states to address racism.

Thus, epistemic ambiguity does not mean there are no links between knowledge and politics or that in such situations actors revert to other political calculations, but it allows for both novel intensifications of political domination and new opportunities for contesting domination. The ambiguity of race, and its permeable boundaries with respect to other categories, both allowed colonial powers to reproduce imperial relations in the postwar order and provided the means by which postcolonial actors could contest this order. Such a meeting of ‘traditional’ and ‘transformational’ coalitions, however, cannot be fully mapped without attending to the epistemic contestations over race that underpinned this meeting.

Conclusion: race and racism in the future of international order

This article theoretically and empirically contributes to the literature on race, racism and the LIO. Theoretically, the article contributes to the growing literature on object-orientated IR by theorizing the drivers of epistemic ambiguity and its political effects. I show how ambiguous objects—particularly when an object already structures international order—are produced when multiple related objects also exist and when disciplinary fragmentation is supported by institutional infrastructures (like disciplinary silos). In such situations, we expect to see wider flexibility in how actors engage in politico-normative struggles.

Empirically, this article provides the story of how scientific struggles around race shaped the institutionalization of anti-racist norms, bringing to the fore a divide in how different scientists and political actors thought about race. Though historical in approach, this argument suggests areas of further research and entails normative implications for today. First, while this article does not offer an answer to the question: ‘What is race?’, it does ask IR scholars to take this question seriously in contemporary analysis of race, racism and international politics. While many social scientists such as Debra Thompson hold that race is a social construct, ‘a powerful set of ideas or norms about the identity, difference and organization of a society and its constituents’,⁹⁹ others continue to think about race or related objects in biological terms. Fields such as epigenetics and public health implicitly or explicitly make use of race. As Stefanie Marie Fullerton argues, population genetics, which dismisses a ‘typological ontology of race’ nonetheless finds that genetic variation does correlate with geography and, while ‘such differences are typically matters of degree rather than kind … such differences are widely recognized and the continuing object of scientific and clinical attention’.¹⁰⁰

According to Fullerton, ‘race’ in some form remains an object of analysis for several scientific fields and the reduction among social scientists and humanists to locating the category as purely ‘social’ ignores the socio-material making of racial distinctions, reinscribes an unhelpful nature/culture divide, and ‘leaves the ongoing scientific construction of racial biology immune from evaluation, critique, and correction’.¹⁰¹ Moreover, as Zakiyyah Iman Jackson argues, the recent rise of epigenetics, which investigates how social behaviours and environments are imbricated with gene expression, is sold as posing the end of ‘biocentric conceptions of race’.¹⁰² But in so far as it is bound to genetics, it is necessary to see how it might be entangled in biopolitical and necropolitical projects.¹⁰³ The question of what race is in some ways still persists. IR, then, needs to attend to how different scientific communities use race and their connection to international politics.

Second, there are normative implications for policy-makers. Today, there is robust discussion about the ‘challenges’ and ‘threats’ to the LIO. Such challenges are largely—though certainly not exclusively—orientated around the rise of powers such as China, India and a resurgent Russia, and what such multipolarity might entail for liberal order. Frequently, academic policy discussions involve questions about whether the LIO can withstand such challenges, or whether its constituent institutions are resilient enough to accommodate shifting material power dynamics as well as alternative ways of organizing international life.

Western anxieties of pluriversality echo mid-century fears of decolonization that promised challenges to international hierarchies and bipolar competition. Today, as before, these anxieties express themselves as cultural and material threats to the seemingly neutral, open and rules-based order, a sentiment that ignores 1) the entanglements between liberal political order and colonial/neocolonial relations and 2), as demonstrated in this piece, the ways in which the LIO has always incompletely dealt with the problem of race and racism. Policy-makers and those concerned about the institutions, norms and principles that make up the LIO must take more seriously how these institutions are not separate from racial orders, that issues around negotiating racial difference and addressing racial hierarchies were integral to an historical formation of the LIO. Calls for ‘decolonization’ of the histories of international order¹⁰⁴ or demands for epistemic justice¹⁰⁵ are necessary practices to negotiate and cultivate a world of difference.

Footnotes

Author notes