What’s Wrong with Ethnocentrism? (DRAFT: Please do not circulate or cite without permission) ABSTRACT Despite its prominence as a pejorative term in moral and political philosophy, the phenomenon of ethnocentrism has largely escaped the focused attention of moral and political philosophers. This article attempts to address that gap in the philosophical literature. The article begins by drawing a core distinction between ethnocentrism as a moral phenomenon (i.e., a form of moral partiality), on the one hand, and as an epistemological phenomenon (i.e., a cognitive bias), on the other. After singling out the epistemological aspect of ethnocentrism as its focus (in Section II), the article takes on three main tasks. First, it offers a detailed account of the nature of ethnocentric bias (Sections III & IV). Second, it develops an account of the most natural and common concerns that are raised by ethnocentrism so defined (Section V). And third, the article examines how these concerns might be justified (Sections VI, VII, VIII, and IX). The main conclusions of the article are twofold: (i) that ethnocentrism is a problem chiefly in light of its deleterious effects on the reliability of our judgment; and (ii) that despite the longstanding tradition of associating concerns about ethnocentrism with the acceptance of some form of relativism or skepticism, only realism (broadly conceived) can make sense of the full range of concerns normally and naturally associated with ethnocentrism. I. INTRODUCTION Ethnocentrism is a pejorative concept in moral and political philosophy. When understood as something akin to nationalism – e.g., as a way of showing special concern towards the members of one’s cultural community – then, it is true, it becomes less obviously negative. And ethnocentrism is sometimes discussed in value-neutral terms: for instance, when it is cited as evidence in support of some brand of moral skepticism or relativism.1 But in general, moral and political philosophers tend to think of ethnocentrism as something worth avoiding, if at all possible. And, as such, when philosophers discuss it, their discussions are usually about how to avoid ethnocentrism, or about how it has been avoided by them and/or hasn’t been avoided by others. Despite this, philosophers have all too rarely specified just what sort of problem ethnocentrism is, and what makes it so. My goal in this paper is to do just that. Crucial to this task is developing an adequate working definition of ethnocentrism itself, one that can form the focus of any account of its problematic nature. That will be my aim in the following three Sections (II, III, and IV), where I argue that ethnocentrism is best understood not as a kind of belief (e.g., in one’s cultural superiority) but as a cognitive bias that affects the character of one’s belief-forming process. Section V provides an overview of the key (and most commonly cited) concerns that are raised by such a bias. And Sections VI, VII, VIII, and IX attempt to examine both how those 1

See, e.g., John Mackie, Ethics: Inventing Right and Wrong (London: Penguin Books, 1977), pp. 3638; Melville J. Herskovits, Man and His Works: The Science of Cultural Anthropology (New York: Alfred A. Knopf, 1948), pp. 63-64.


concerns might be justified and how they vary depending on one’s theory of truth and justification. Ultimately, I hope to dispel the widely held notion that worries about ethnocentrism are logically tied to the acceptance of some form of relativism or skepticism. On the contrary, I argue that only realists, broadly conceived, can make sense of the full range of concerns that are ostensibly raised by ethnocentric bias. II. MORAL AND EPISTEMOLOGICAL ELEMENTS Ethnocentrism can mean many things, and often means different things to different people. Part of what I want to do in this article, then, is to disentangle some of these meanings, as well as pick out what I believe to be the most prominent and interesting among them. As an analytical starting point, we can begin by tracing the term back to its origins in the work of the 19th Century sociologist, William Graham Sumner. As Sumner writes in his most popular work, Folkways: a Study of Mores, Manners, Customs, and Morals, Ethnocentrism is the technical name for this view of things according to which one’s own group is the center of everything, and all others are scaled with reference to it… Each group nourishes its own pride and vanity, boasts itself superior, exalts its own divinities, and looks with contempt on outsiders. Each group thinks its own folkways the only right ones, and if it observes that other groups have other folkways, these excite its scorn.2

According to Sumner, ethnocentrism is a way of thinking, or a ‘view of things,’ whereby centrality is attributed to one’s own cultural group or ethnos.3 Moreover, implicit in this attribution of centrality, as Sumner understands it, is a companion ascription of superiority.4 This superiority seems to have two major dimensions. One dimension is moral. Insiders are morally privileged; they are seen as more deserving of respect and moral consideration, and less deserving of ‘contempt’ than outsiders are. On the other hand, there is also an epistemological dimension to ethnocentrism as Sumner defines it. Insiders are not only regarded as morally privileged vis-à-vis outsiders, they are also seen as cognitively privileged, and uniquely so. Unlike outsiders, insiders understand the truth about which ‘folkways’ – i.e., longstanding and deeply-held customs of kinship, authority, 2

William Graham Sumner, Folkways: a Study of Mores, Manners, Customs, and Morals (Mineola: Dover Publications, 1907/2002), p. 13. Herodotus made similar observations about the Persians in particular but also more generally: “For if anyone, no matter who, were given the opportunity of choosing from amongst all the nations in the world the beliefs which he thought best, he would inevitably, after careful consideration of their relative merits, choose those of his own country.” Herodotus, The Histories, 3.38, (London: Penguin, 1972), Translated by A. de Sélincourt, p. 169. Thus, we can attribute the term, ‘ethnocentrism,’ to Sumner, but not the concept that it picks out; the latter has a much longer history. 3 By “culture”, or “cultural group”, or “ethnos” I don’t mean anything as narrow as ethnicity. Rather, I define culture or ethnos both classically and in the broadest of terms, as “that complex whole which includes knowledge, belief, art, morals, law, custom, and any other capabilities and habits acquired by man as a member of society.” Sir Edward Brunett Tylor, Primitive Culture (London: Bradbury, Evans, and Co., 1871), Vol. 1, p. 1. Thus, I take cultures to be individuated, however vaguely, not by race or ties of history, origin, and kinship per se (although these may also play a part) but primarily by certain necessarily loose conformities of belief and social practice. 4 See: Robert K. Merton, On Social Structure and Science (Chicago: University of Chicago Press, 1996), ed. Piotr Sztompka, p. 248.


conflict resolution, worship, medicine, education, etc. – are best, or ‘right.’ And if one looks at the concrete examples of ethnocentrism that Sumner provides in his text, it is clear that this sort of epistemic hubris is a general one, and not confined only to moral or practical matters.5 Ethnocentric groups see themselves as having a monopoly over the correct understanding of what is truly good and evil, beneficial and harmful, real and illusory, divine and mundane, beautiful and ugly, etc. Both of these connotations of ethnocentrism, moral and epistemological, still resonate today. However, the epistemological connotation seems to overshadow its moral counterpart, at least in contemporary debates. This is suggested, for instance, by the fact that so many thinkers have found it perfectly coherent to ask whether the Western commitment to universal moral equality is itself ethnocentric.6 If “ethnocentrism” were standardly understood as a moral notion – i.e., as a way of seeing outsiders as morally unequal to members of one’s cultural group – then this would not be nearly as coherent a question as it is.7 However, even if the moral interpretation of ethnocentrism remains highly relevant, as I suspect it does, the justifiability (or lack thereof) of moral partiality towards one’s ethnos or conationals is a well-developed topic in moral and political philosophy, and I don’t intend to say much about it here. By contrast, besides being frequently invoked in arguments for moral skepticism and relativism or volleyed as a critical barb in normative theoretical debates, comparatively far less has been written about its epistemological counterpart, i.e., our tendency to assume that the beliefs and practices of our own culture or society are superior to all others. That is the phenomenon that I intend to address here. III. BIAS AND ENCULTURATION With the present object of analysis now in clearer sight, a puzzle quickly emerges. This is because, on closer inspection, this kind of ethnocentrism is by no means an inappropriate attitude for an individual or society to adopt towards its beliefs and practices. Far from it: it seems only normal for a group that endorses some belief or custom x to consider x to be in some way demonstrably superior to its alternatives. That is just the regular course of events for any ostensibly reasoned commitment to x.8 And so, a group’s belief in the comparative superiority of 5

See: Sumner 1907/2002, pp. 13-14. See, e.g., Seyla Benhabib, The Claims of Culture: Equality and Diversity in the Global Era (Princeton: Princeton University Press, 2002), pp. 24-49, esp. 27; Martin Hollis, “Is Universalism Ethnocentric?” in Multicultural Questions (Oxford: Oxford University Press, 1999), eds. Christian Joppke & Stephen Lukes, pp. 27-44; Daniel Zolo, “Against Universalism” online: www.theglobalsite.ac.uk 7 I don’t mean to suggest here that there is no way of making sense of this question if we understand ethnocentrism in moral terms. There is a vast literature, for instance, that argues, in various ways, that the Western moral and political commitment to moral equality and human rights is intimately connected to the violation of that very commitment, i.e., the systematic moral oppression and political abuse of nonWesterners by Westerners. (See, e.g., Makau Mutua, Human Rights: A Political and Cultural Critique (Philadelphia: University of Pennsylvania Press, 2002), pp. 67-70; Costas Douzinas, Human Rights and Empire: The Political Philosophy of Cosmopolitanism (Abingdon: Routledge-Cavendish, 2007), pp. 180181; Daniel Zolo, Cosmopolis: Prospects for World Government (Cambridge: Polity Press, 1997), Translated by D. McKie.) Still, when questions are raised about the ethnocentricity of human rights specifically, my impression is that the question being raised is most often about their being espoused as a result of the epistemic hubris described above, and not about their egalitarian credentials. 8 All that I’m assuming here is that a reasoned commitment is one that we adopt on the basis of some 6


its customs and beliefs cannot on its own count as an instance of ethnocentrism, so long as we take that notion to indicate that something (whatever it might turn out to be) has gone awry. Nor would it seem to matter much if that comparative belief turns out to be false. For we can imagine a cultural group deliberating extensively, self-critically, and constructively on the question of whether, say, creationism or evolutionary theory (or both, or neither) should be taught in secondary schools. The group might, for instance, converge on the wrong answer without realizing it. But while the group would, in that case, falsely believe that it has adopted the best solution to the matter at hand (i.e., a solution it takes to be superior to the alternatives adopted by other cultures), it’s not at all clear that this would amount to an instance of ethnocentrism. There must be some cases of error that are not cases of ethnocentrism, and nothing in this example seems to clearly settle the matter one way or the other. This last set of reflections already points the way towards a more satisfactory definition. For, what seems to save the aforementioned group from condemnation as ethnocentric, despite its error, is the thought that its deliberative process was, or at least seems to have been, sound, i.e., open-minded, self-critical, rigorous, evidence-based, etc. But if we remove that thought and imagine that the relevant deliberative process was in fact skewed in favor of pre-existing beliefs (e.g., in evolution, creationism, etc.) prevalent among the community at hand, then we have indeed arrived at something genuinely troubling as well as intuitively more worthy of being counted as an instance of ethnocentrism. What all this points to is that Sumner’s definition becomes more powerful and illuminating if we take exegetical liberties and read him as describing a kind of bias in favor of local folkways and opinions, rather than simply a belief in their superiority – although that kind of belief will be a natural product of such a bias. That is, on this reading, ethnocentrism (or at least its epistemological aspect) affects the quality and character of the belief-forming process, i.e., the process by which one comes to endorse some belief and/or practice over others.9 The term “bias” is rather vague and unhelpful itself, although it should already be clear that the kind of bias at issue here is not moral but cognitive. Ethnocentrism, or rather the epistemological aspect of it that I am examining here, is not to be understood principally as a way of showing special moral concern towards the members of one’s cultural group. It is not a bias towards certain persons. It is rather a bias towards certain judgments, predisposing us to accept certain descriptive and/or normative claims over others.10

reason or consideration that we take to speak in favor of it over its alternatives. Presumably, not just any consideration will do, so there are further criteria to be filled in, but I take no positive position on what they are here. 9 One of the implications of this is that the ethnocentricity of a belief is not determined by its content or place of origin per se, but by the character of the process through which it came to be endorsed. Thus, the mere fact that ‘rights’ were originally conceived by ‘Westerners’, even if true, is not enough to establish the ethnocentricity of a belief in rights, since a belief in rights may be the result of what I above called a sound deliberative process, rather than a culturally biased one. In this way, this definition is sensitive to the concerns of those who dismiss such content or origin-based allegations of ethnocentrism as committing the “genetic fallacy”. See e.g., Jack Donnelly, Universal Human Rights in Theory and Practice: Second Edition (Ithaca and London: Cornell University Press, 2003), pp. 69-70. 10 In what follows, I use the terms “belief”, “judgment”, and “claim” interchangeably. However, sometimes I use the term “judgment” in a different but no less familiar sense, i.e., to refer to a higherorder capacity for forming individual beliefs, judgments, or claims. Context will (I hope) make clear in which sense I mean to use this term below.


Moral and cognitive biases do intersect. For instance, a moral bias can sometimes manifest itself in cognitive form, since it can affect the set of cognitive judgments that one is willing to accept: as when we are, say, unwilling to accept any evidence that corroborates the criminal guilt of a loved one, or unwilling to consider the plausibility of an argument raised by a despised political foe.11 But the sort of cognitive bias that I have in mind here is more general, since ethnocentrism is typically and plausibly understood to have sources beyond that of mere loyalty or partiality towards certain individuals. Historically, the most often cited source of ethnocentrism is “enculturation,” a learning process understood to begin in infancy and to extend well into adulthood (if not indefinitely), through which one comes to internalize the beliefs and conventions (linguistic, moral, social, political, scientific, religious, economic, etc.) of one’s culture.12 While enculturation can certainly involve the development of special loyalties towards one’s cultural associates, it is normally thought to involve more than that. Particularly important for our purposes is the way in which it is understood to affect normative judgment, inculcating in its subjects an instinctive and stubborn attachment to the moral values, social and institutional practices, and broader intellectual traditions that are endorsed by local culture.13 Moral psychologists have disagreed about the strength of this socially inculcated attachment as well as the exact nature and workings of its influence on moral reasoning, but few would deny its pervasiveness.14 And moral and political philosophers have highlighted different aspects of it. For instance, in his well-known “argument from relativity”, John Mackie notes how participation in a way of life can engender positive valuation of its distinctive social practices: “people approve of monogamy because they participate in a monogamous way of life”.15 And in his discussion of ethnocentrism (or “parochialism”) as a problem facing the theory and practice of human rights, Allen Buchanan has noted how enculturation can produce a strong attachment to certain moral and political values – e.g., autonomy and individual liberty – a fact often noted in debates about the ethnocentricity of such rights.16 11

Miranda Fricker has done very interesting work on the harms and injustices that can result from this sort of incredulity. See: Miranda Fricker, Epistemic Injustice: Power & the Ethics of Knowing (Oxford: Oxford University Press, 2007). 12 The term, I believe, originates in the work of Melville J. Herskovitz, and is to be distinguished from the term “acculturation” (which standardly describes a process of cultural change or transformation). Herskovitz writes: “Judgments are based on experience, and experience is interpreted by each individual in terms of his own enculturation… When we reflect that such intangibles as right and wrong, normal and abnormal, beautiful and plain are absorbed from infancy, as a person learns the ways of the group into which he is born, we see that we are dealing here with a process of first importance. Even the facts of the physical world are discerned through the enculturative screen, so that the perception of time, distance, weight, size, and other ‘realities’ is mediated by the conventions of any given group.” (Herskovitz 1948, pp. 63-64). 13 See e.g., Melville Herskovitz, “Some Further Comments on Cultural Relativism” in American Anthropologist, Vol. 60, April 1958, p. 270. 14 I do not intend to explore the ‘mechanics’ of enculturation here. For a useful exchange on this topic, see: Jonathan Haidt, The Righteous Mind: Why Good People are Divided by Politics and Religion (New York: Pantheon Books, 2012) & S. Matthew Liao, “Bias and Reasoning: Haidt’s Theory of Moral Judgment” in New Waves in Ethics (London: Palgrave MacMillan, 2011), pp. 108-128. 15 Mackie 1977, p. 36. 16 Allen Buchanan, “Human Rights and the Legitimacy of the International Order” in Legal Theory, 2008, Vol. 14, No. 1, pp. 44-55. See also, American Anthropological Association (AAA), “Statement on


Perhaps because contemporary complaints about ethnocentrism often take inspiration from 20th Century anthropological critiques of earlier evolutionary theories of culture – i.e., theories that ranked societies according to a set of (purportedly) universal developmental criteria – there is a strong temptation to understand ethnocentrism as involving normative judgments of universal scope. For instance, some theorists have built universalism into the very notion of ethnocentrism itself. Martin Hollis understands the charge of ethnocentrism to be “that the accused did unwarrantedly presume the truth of a universal proposition and/or its applicability to persons of contrary opinion, such presumption being of cultural origin.”17 But there is reason to resist this conflation. Biased judgments about the merits of a (native) set of beliefs and/or practices vis-à-vis (foreign) alternatives can be made even if the judgment in question is non-universal in form, or only about what we ought to do around here. For instance, an American considering which conjugal practice is best for Americans (and no one else) may nevertheless biasedly rule in favor of a locally entrenched practice such as monogamy. And if such a judgment can be as influenced by enculturation as any other, it’s hard to see why it should not be able to count as an instance of ethnocentrism despite its limited scope of application. Thus, there is reason to hesitate before assuming, as Hollis does, that ethnocentric judgment must be universalist in form. IV. NORMATIVE AND DESCRIPTIVE ELEMENTS My primary interest here is in the nature of ethnocentrism as it applies to normative judgment, both moral and practical. Accordingly, I’ve emphasized how our cultural upbringing can influence our judgments about values and virtues, on the one hand, and about social and institutional practices, on the other. But what I’ve failed to note thus far is how quickly this attempt to focus exclusively on normative (as opposed to descriptive) judgment breaks down. Once we take it upon ourselves to evaluate concrete social practices, or to make judgments about what we actually have reason to do and what obligations we owe towards others, our assessment of non-normative facts becomes crucial. For instance, in order to comprehensively appraise a political practice like representative democracy, we presumably need to understand, among other things, what its real world (e.g., social, institutional, psychological, economic) consequences are and how these differ from its alternatives, if at all. And if, furthermore, we want to make a specific prescription about the fittingness of representative democracy for a specific group, then we will need to know special information about that group’s size and nature, its present and past political practices, as well as about what the likely consequences of any attempted move away from those practices would be. These matters, important as they are for normative judgment, are all highly empirical. And so any complete account of the influence of ethnocentrism on normative judgment must also be alive to the possibility of its influence on normatively relevant descriptive judgments as well. Which descriptive judgments ultimately are normatively relevant will to a great extent depend on what normative assumptions we take to be true, so it’s difficult to offer a comprehensive account of this in advance. For instance, if we subscribe to the classical principle Human Rights” in American Anthropologist, Vol. 49, No. 4, 1947, pp. 539-543; Jack Donnelly, “Human Rights and Asian Values: A Defense of ‘Western’ Universalism” in The East Asian Challenge for Human Rights (Cambridge: Cambridge University Press, 1999), eds. Joanne R. Bauer & Daniel A. Bell, esp. pp. 79-87. 17 Hollis 1999, p. 31.


of utility, then the only relevant consideration in any normative judgment will be that of how much pain and/or pleasure is generated by the act, practice, or policy in question. And although it may be hard to see how enculturation might affect our answer to this sort of question at first glance, the possibilities are in fact various. For instance, it has been widely noted that both ‘pleasure’ and ‘happiness’ are differently described, and their enabling conditions differently conceived, by different cultures.18 This raises a host of questions about what pleasure and happiness really are, their biological roots, and their cultural variability. But it also raises the possibility that, say, a ‘Westerner’ may fail to recognize legitimate instances (or enabling conditions) of pleasure or happiness – and so fail to correctly discern the right action on a classical utilitarian account – simply because they fail to conform to Western preconceptions thereof. This opens up to a broader observation, because it is illustrative of how ethnocentrism can affect our understanding of a wide range of culturally alien phenomena, not just those relevant to classical utilitarian reasoning. Like other cultural misunderstandings, the previously imagined one is generated by our enculturated tendency to understand the beliefs, attitudes, behaviors, and practices of foreign cultures in light of experiences, categories, assumptions, and conventions particular to our own. Concern about ethnocentrism of this kind has long been a mainstay of twentieth-century sociocultural anthropology, and much has been written about it.19 In order not to get bogged down in the elaborate nuances of that literature, however, I will simply outline two ways in which this bias can interfere with our judgment of other cultures. Illustrating these two forms of interference turns out to be a useful way of getting an initial grasp on the problematic nature of ethnocentrism, since the most natural way to describe them is via their production of two different kinds of error, respectively. The first kind of ethnocentric blunder is relatively straightforward, and occurs when the cultural baggage (e.g., experiences, categories, assumptions, and conventions) that one brings to the table when trying to understand some foreign cultural phenomenon leads them to overlook important interpretive cues. Just above, I imagined this sort of misinterpretation occurring in the context of recognizing pleasure or happiness across cultures, i.e., sensations or emotions. But it can also mar our attempts to appreciate the intentions behind culturally foreign actions and practices. For instance, in his ethnographic work, the Danish explorer and anthropologist, Peter Freuchen, notes how Western conventions regarding the treatment of children led him to see the routine abandonment of orphans among the Eskimos as a cruel and pernicious act, only to later realize, after raising the topic with members of the Eskimo community itself, that such neglect is part of a sincere and concerted attempt on the part of the community to prepare the youth for the formidable hardships of adult life.20 Certainly, this sort of information about background 18

See e.g., Happiness Across Cultures: Views of Happiness and Quality of Life in Non-Western Cultures (Springer, 2012), eds. Helaine Selin & Gareth Davey; Richard Nisbett, The Geography of Thought: How Asians and Westerners Think Differently… and Why (New York: Free Press, 2003), pp. 56. 19 See e.g., Franz Boas, “The Mind of Primitive Man” in Journal of American Folklore, Vol. 14, 1901; Herskovitz 1948; Peter Winch, “Understanding a Primitive Society” in American Philosophical Quarterly, Vol. 1, No. 4, 1964; Clifford Geertz, The Interpretation of Cultures (New York: Basic Books, 1973), Ch. 1; Also: Charles Taylor, “Understanding and Ethnocentricity” in Philosophy and the Human Sciences: Philosophical Papers 2 (Cambridge: Cambridge University Press, 1985), pp. 116-134. 20 For the reference, see: Peter Freuchen, Book of the Eskimos (New York: Fawcett, 1961), pp. 41-42. In his insightful book, Morality and Cultural Differences (Oxford: Oxford University Press, 1999), John


intentions can bear on our evaluation of the practice in question, as well as many others.21 So too can closely related information about a practice’s meaning or categorization, which is also something that the bias under discussion can lead us to misdiagnose across cultures. For instance, in a famous essay Peter Winch argues that, due to their profound attachment to the achievements and methods of the sciences, Western interpreters tend to understand Zande witchcraft – which involves, among other things, magical rituals that are meant to promote the flourishing of agricultural crops – as a primitive form of science, i.e., a clumsy attempt to discover and manipulate relations of cause and effect. In fact, Winch argues, Zande witchcraft is something subtly different: a culturally alien way of coming to terms with the world, and submitting oneself to its contingencies, rather than attempting to discover and control them.22 I am not in a position to elaborate or verify Winch’s argument here. The point is simply to offer particular examples of one general way in which ethnocentrism can distort our descriptive judgments, and to illustrate the resultant influence of such distortions on normative judgment. In the cases described just above, that influence was derogatory, leading us, for instance, to see the Eskimos as more cruel than they perhaps are, or to see the Zande as more scientifically naive than they may be. But the evaluative fallout from these factual mistakes can also, at least in principle, be approbatory. The second kind of error is only subtly different from the first, since it too is a result of the application of homegrown experiences, categories, assumptions, and conventions to the study of a foreign culture. What makes it distinct, however, is that it is generated by an ulterior fixation (conscious or unconscious) on the otherness of the foreign culture in question. Thus, while it is possible for someone making the first kind of error to misunderstand, say, the unique mentality of a foreign society because she naively assumes that it is no different from that of her own, this possibility and others like it fade out of view in the second case. Here, foreign cultural phenomena of all sorts (including foreign individuals) are seen through the lens of a set of facile oppositions that distinguish “us” from “them”, e.g., familiar vs. exotic, normal vs. abnormal, civilized vs. barbarian, human vs. inhuman, rational vs. irrational, clean vs. dirty, peaceful vs. violent, self-governing vs. dependent, etc.23 Examples of this sort of error are abundant, and particularly evident in the way that cultural, religious, and political identities (Western vs. Eastern, Muslim vs. Christian, Israeli vs. Palestinian, Democrat vs. Republican, etc.) are often used to evoke grossly over-exaggerated thoughts about the differences between groups, overshadowing their common interests, capacities, beliefs, practices, history, hopes, and intentions.24 The drive to understand another culture in this way, as different or other, can have any number of sources: fear, pride, a history of doing so, conflict, past injustice, power disparities, the demands of modern identity politics, etc. W. Cook discusses this case and generally names misinterpretations of this sort “projection errors”, providing a number of useful examples (esp. pp. 93-101). Also see: Bernard Williams, Morality: And Introduction to Ethics (Cambridge: Cambridge University Press, 1972), pp. 23-24. 21 For a similar illustration of how background considerations about the meaning of and motivations behind some foreign cultural practice can affect our moral evaluation of it, see: David Wiggins, Ethics: Twelve Lectures on the Philosophy of Morality (London: Penguin Books, 2006), pp. 337-356. 22 Winch 1964, esp. p. 321. 23 I am thinking here chiefly of Edward W. Said’s portrayal of “orientalist” discourse in, Orientalism (New York: Vintage Books, 1979). 24 For a much-discussed example of this, see: Samuel P. Huntington, “The Clash of Civilizations?” in Foreign Affairs, Vol. 72, No. 3, 1993, pp. 22-49.


And while it usually has a derogatory effect on our normative evaluation of other cultures, it can also do the opposite; sometimes the otherness of a foreign people can be exaggerated by adopting an idealized vision thereof (e.g., the myth of the noble savage), one in which the problems and ills that we recognize in our own society are conspicuously absent. This completes my account of the range of cognitive biases that fall under the heading of ethnocentrism. It is not meant to be an exhaustive account, but it is meant to provide a solid working overview.25 According to this account, ethnocentrism can affect normative judgment at two major levels: first, at the level of so-to-speak pure judgments of value or principle, as when, for example, we biasedly affirm the value of autonomy over social harmony, or vice-versa; and second, it can affect normatively relevant descriptive judgments in various ways, including in the ways described just above. Any given normative judgment may be affected by ethnocentrism at the first level, the second, or both. This definition of ethnocentrism is both wider and narrower than Sumner’s. It is narrower because it addresses only the epistemological aspect of ethnocentrism. And it is wider because it understands ethnocentrism to be first and foremost a kind of cognitive bias, one that can result in a variety of (derivatively) ethnocentric judgments: false, true, cross-cultural, universal, local, particular, normative, descriptive, derogatory, approbatory, etc. Thus, I count as ethnocentric any judgment that is skewed, in an unchecked manner, by the judger’s predisposition towards the (normative and/or descriptive) beliefs, categories, assumptions, practices and conventions that are current in his or her culture. I have cited the (largely passive) learning process of enculturation as the chief source of ethnocentrism understood in this sense. But I’ve also indicated that other sources can come into the mix, including an ulterior (intentional or unintentional) fixation on the ‘otherness’ of foreign cultures that will affect cross-cultural understanding in particular, and which is likely to have complex sources of its own.26 I understand that this is an extremely broad definition. But neither its breadth nor its departure from Sumner’s account should, I think, be seen as a defect. Rather, these are simply reflections of the fact that, in the last one hundred years, ethnocentrism has been understood in ways that stretch the concept beyond the limits that Sumner originally envisaged for it. That the present definition does greater justice to those developments should be seen as a benefit, if anything. V. THE PROBLEM OF ETHNOCENTRISM 25

One very important form of ethnocentrism that I have not discussed here, for instance, isn’t defined by deference to local cultural beliefs and practices per se, but rather involves a tendency to apply (potentially true and justified) critical normative standards more strictly to foreign persons and cultural practices than to their domestic counterparts. In this regard, see e.g., Yael Tamir, “Hands off Clitoridectomy: What our Revulsion Reveals about Ourselves” in, The Boston Review, Summer 1996. 26 One point worth noting here is that ethnocentric bias can, in principle, have intentional sources. For instance, many will intentionally denigrate or ignore opinions, beliefs, points of view, social practices and forms of life that they consciously recognize to be in some way foreign to their own. We see this very often in inter-societal debates about certain polarizing social practices, such as homosexuality. In many African countries, for instance, arguments in favor of accepting or at least tolerating homosexuals are routinely stigmatized as “Western” or “Anti-Christian,” and are therefore rejected outright. See e.g., Rahul Rao, Third World Protest: Between Home and the World (Oxford: Oxford University Press, 2010), pp. 179-189.


It is plain that, very often, intentionally or unintentionally adopting the standpoint of some culturally local set of categories, conventions, and beliefs can be tremendously useful or even necessary for understanding. Consider, for instance, how important this is when we want to understand some communicative act. To use a famous example, only if we are well aware of local conventions – according to which intentionally closing one’s eyelid constitutes a conspiratorial sign – will we be able to tell a twitch from a wink.27 Part of what makes enculturation so useful and important, then, is surely that it provides us with the tools and background information – conceptual, linguistic, behavioral, psychological, social, etc. – to be able to navigate and understand our own social environment effectively. I have no interest in denying that ethnocentrism can be useful in this sense. But the limits of its epistemic usefulness need to be highlighted, and they quickly become apparent once we broaden the scope of desired understanding. Knowledge (or even unthinking acceptance) of local conventions and beliefs will generally serve us well when we are trying to understand communicative acts (including art, music, literature, etc.) and other forms of human behavior around us.28 It may also help us understand how best to apply some general set of normative principles or propositions to our own local circumstances: as when, for instance, we try to determine how best to satisfy a specific population’s right to adequate food and housing. But such enculturated knowledge is not likely to help us understand actions and circumstances beyond our local cultural horizon (The examples of error offered in Section IV vividly illustrate this). Moreover, the process of enculturation does not merely give us knowledge of local beliefs and practices; it biases or predisposes our judgment in their favor. And it’s not at all clear that such a bias will help us identify, say, true higher-order normative principles, i.e., those which normative reasoning ought to apply to concrete circumstances.29 The same could certainly be said about its tendency to help us understand a wide range of empirical matters beyond human action, e.g., the truths of physics, mathematics, biology, economics, etc. For this broader set of objects, ethnocentrism seems more like an impediment than an aid to true or accurate understanding. And it is this more than anything else, I shall argue, that makes ethnocentrism a ‘problem’. What I want to do in this section is provide a basic overview of this problem and highlight its various elements. I shall list five in total, almost all of which have been (at least) implicitly illustrated in the preliminary analysis above. This list is not meant to be definitive. Its point is simply to outline five natural and widespread concerns about ethnocentrism as I’ve described it. In the following Sections (VI, VII, VIII, and IX), I will examine how these concerns might be motivated, as well as examine how their motivation alters (and sometimes disappears) depending on one’s background theory of truth and justification. (1) Unreliable. Ethnocentrism is often derided for its tendency to lead our judgment into errors of various forms. As I mentioned in Section IV, social scientists and cultural anthropologists have frequently understood ethnocentrism to be a danger of just this sort, one


See: Geertz 1977, Ch. 1. There is a temporal dimension to this as well; since they change over time, present social conventions and beliefs may be a poor guide to understanding the historical behavior of one’s cultural group. 29 Of course, one may be a ‘particularist’ and not believe that there are such higher principles. Even so, particularists may agree that biasedly endorsing culturally local beliefs and practices is of no use in discerning normative truths. 28


that leads observers to misinterpret the meaning and intent of actions in foreign cultures.30 And the same is true of other kinds of scientists and theoreticians. If we want knowledge of broader empirical phenomena (e.g., biological, physical, astronomical, economic, etc.), the normal understanding is that we need to directly and carefully observe the things themselves, rather than stubbornly heed the judgment of persons with whom we randomly happen to be associated with by birth and/or enculturation. If anyone’s opinion merits consultation in this respect, it is the opinion of scientists who are competent evaluators of the latest observational (and/or theoretical) data. And since there is little reason to think that our cultural (as opposed to scientific) associates are going to be (a) those people, or (b) reliably deferential towards those people’s judgment, there is correspondingly little reason to think – as a matter of general principle – that ethnocentric deference to culturally established opinion will tend to land us with true scientific and theoretical beliefs.31 Similar concerns arise in the context of normative inquiry. Even if there is far less general consensus about how best to access normative truths, upon reflection most will agree that being stubbornly partial to the normative attitudes of the culture(s) into which one randomly happens to be enculturated (whether by birth or some other happenstance) is a very poor method indeed. Worries of this sort have occupied a prominent place in longstanding philosophical debates about toleration. For instance, Saint Augustine saw our natural tendency to internalize the beliefs and customs of those around us as a dangerous obstacle to true belief (i.e., Catholic moral, political, and theological doctrine) in general: one that, he reasoned, should be counteracted by the threat of punishment whenever necessary.32 And John Stuart Mill saw the same general danger in ethnocentrism – or what he called the “magical influence of custom” – but argued that it should be averted not by punishment but by free, regular, and vigorous intellectual debate.33 Amongst contemporary moral and political philosophers, these concerns about ethnocentrically induced error tend to be framed in at least two main ways: first, as a worry about our propensity to mistakenly assume that culturally local normative principles and practices are ‘the only acceptable ones’, when in fact there is an array of acceptable (or ‘reasonable’) alternatives;34 and second, in a more Augustinian-Millian vein, as a worry about the ‘conservative’ or uncritical nature of ethnocentric normative judgment, and of any prescribed method of normative justification that strongly defers to our current beliefs.35 (2) Unjustified. This last observation turns us towards a second set of concerns that follows directly from the first. In Section II, I suggested that there is nothing inherently ethnocentric about understanding one’s cultural beliefs and practices to be superior to foreign alternatives, even if that understanding is false. What seems to matter most is the deliberative route by which one arrives at such a belief. If that route is sound or justified – i.e., open-minded, 30

See: References listed in fn. 19 above. To substantiate this point, one only has to note how often the popular opinion of whole cultures, or large sub-cultures, can oppose scientifically-established opinion, e.g., about the likelihood of global warning, the origins of life and the universe, or the consequences of same-sex marriage. 32 Saint Augustine, “Letter 93” in Letters (New York: New City Press, 2001), pp. 376-409. 33 John Stuart Mill, On Liberty and Other Writings (Cambridge: Cambridge University Press, 1989), ed. S. Collini, p. 9. 34 I shall discuss this in greater detail in Section IX. 35 For a good contemporary example, see: Thomas Scanlon, “Rawls on Justification” in The Cambridge Companion to Rawls (Cambridge: Cambridge University Press, 2003), Ed. S. Freeman, pp. 150-151. 31


self-critical, rigorous, fair, grounded in good evidence, etc. – then allegations of ethnocentrism seem out of place. But if it were biased in favor of prior beliefs and practices prevalent among the epistemic agent’s cultural group, then the resulting belief would seem to be both ethnocentric and unjustified, regardless of its veracity. What this suggests is that part of what is wrong with an ethnocentric judgment, if anything is, is its justification. After all, the alternate deliberative routes I outline here represent different modes of justification, i.e., different ways of grounding one’s belief in evidence. And since we don’t think that the belief patterns of one’s cultural associates constitute appropriate or reliable evidence as to what is true in all but the narrowest of circumstances (i.e., those outlined at the very beginning of this section), ethnocentrically deferring to those beliefs seems generally foolish and unjustified from an epistemic point of view.36 This conclusion holds independently of whether we think of justification as a deontological requirement (i.e., a matter of conscientiously fulfilling one’s intellectual duties or responsibilities) or in more reliabilist terms as a matter of tapping into a reliable belief forming process, consciously or not. If one takes the former view, justification will typically involve duties both to make sure one’s beliefs are grounded in appropriate evidentiary considerations and to follow that evidence wherever it may lead, even if it leads one towards beliefs that contradict previously held views, popular opinions, and desired realities.37 But the fulfillment of these duties is precisely what ethnocentric bias – i.e., one’s predisposed (and characteristically stubborn) commitment to the beliefs and practices of one’s culture – militates against. If, on the other hand, one is a reliabilist about justification, and thinks that all that is required for epistemic justification is that one’s belief is formed, caused, or maintained by a cognitive process (e.g., of memory, perception, learning, or reliance on testimony) that has a propensity to produce true beliefs, then for the very reasons discussed just above ethnocentrism will also seem inadequate.38 On either account, judgments unmitigatedly influenced by ethnocentric bias appear unjustified. (3) Inferior. One way of disarming the preceding concerns is to admit that ethnocentric judgment is both unreliable and unjustified (the noted exceptions notwithstanding), but to claim that there is no better alternative. While this wouldn’t entirely dispel these concerns, it certainly would change their character. As things stand, the natural reaction to the above would be to try to avoid ethnocentrism, or at least mitigate its influence on our intellectual pursuits. But if there were no better alternative, then the more natural response to the preceding concerns would seem to be reconciliation rather than avoidance. Some such line of thought stands behind Richard Rorty’s resigned attitude towards ethnocentrism. Because Rorty rejects the notion that we can transcend our cultural biases, he recommends that we give up on the prospect of avoidance altogether, and become more “frankly” or avowedly ethnocentric in our epistemic endeavors.39 But ethnocentric judgment is commonly understood to be flanked by an array of better alternatives. Indeed, it is this very thought that gives allegations of ethnocentrism their 36

Richard L. Kirkham provides an excellent account of the “epistemological enterprise” and its relationship to theories of truth in, Theories of Truth: A Critical Introduction (Cambridge: MIT Press, 1992), pp. 41-49. 37 See, e.g.: James A. Montmarquet, Epistemic Virtue and Doxastic Responsibility (Lanham: Rowman and Littlefield, 1993). 38 For the classic statement of reliabilism, see: Alvin Goldman, “What is Justified Belief?” in Liaisons: Philosophy Meets the Cognitive and Social Sciences (Cambridge: MIT Press, 1979/1992), pp. 105-127. 39 See: Richard Rorty, “Justice as a Larger Loyalty” in Cosmopolitics: Thinking and Feeling beyond the Nation (Minneapolis: University of Minneapolis Press, 1998), eds. P. Cheah & B. Robbins, pp. 45-59.


characteristic normative bite. Because we think that there are other available forms of judgment (whether non-ethnocentric, or merely less ethnocentric) that are more reliable and justified than it is, ethnocentrism seems like a reprehensible (rather than inevitable) pitfall for a mature epistemic agent. This line of thought is typical of sociocultural anthropologists and social scientists, for one, who have devoted considerable energy to explaining how ethnocentrism can be avoided or at least mitigated in ethnographic fieldwork, whether by engaging in “thick description”,40 “the interpretive view”,41 or some other methodology. It is familiar to those involved in the ‘hard’ sciences, too, who will typically point to the revelatory power of direct empirical observation as an alternative to ethnocentric deference to popular cultural belief. And many moral and political philosophers affirm it as well.42 Even if the prospects for entirely avoiding ethnocentrism were slim in any domain of inquiry, there still seems to be an array of common-sense, available, and generally applicable ways of at least mitigating its influence and (one might hope) thereby improving our epistemic condition: e.g., adopting an attitude of humility towards one’s subject matter, being aware of the cultural sources of (at least some of) one’s commitments, embracing the possibility of error, sincerely engaging (to the best of one’s ability) views, arguments, and perspectives different from one’s own, etc. And although our ability to avoid or mitigate ethnocentrism in any domain of inquiry is ultimately an empirical question, the possibility of at least some degree of mitigation would seem to be a safe bet.43 (4) Generally so. In Section III I argued that ethnocentrism can be at play regardless of the scope of one’s judgment, i.e., regardless of whether one is thinking about what is true for everyone, or only for their cultural associates. Connected to this observation is a further one. For just as ethnocentrism can be generally operative in this way, so can its ills. Providing one is not, as already noted, trying to interpret contemporary social phenomena within their own society, a stubborn attachment to local patterns of belief will seem problematic regardless of the scope of one’s judgment. For instance, consider two epistemic agents: one interested in examining the consequences of global warming for the entire human population, and the other interested in examining its consequences only for their own cultural group, e.g., Tahitian islanders. Such differences in focus do not change the fact that both persons should want their beliefs on this matter to be based as much as possible on the best available evidence, and/or expert advice, and not on opinions (e.g., about the gravity and likelihood of global warming) that merely happen to be in local cultural vogue. The same would be true if one’s inquiry were normative in character. Anyone who formulates conclusions about what members of her own culture have reason to do simply by consulting what those persons tend to think they have reason to do, would seem to be too facile and deferential in her judgment. Regardless of its (local, cross-cultural, or universalistic) scope, 40

Geertz 1977, Ch. 1. Taylor 1985, pp. 116-133. 42 See, e.g.: Ronald Dworkin, “Objectivity and Truth: You'd Better Believe it”, in Philosophy and Public Affairs (1996), Vol. 25, No. 2, pp. 87-139; John Rawls, The Law of Peoples: with, The Idea of Public Reason Revisited. (Cambridge: Harvard University press, 1999), pp. 121-122; Charles Taylor, “Explanation and Practical Reason” in, The Quality of Life (Oxford: Oxford University Press, 1993), eds. M. Nussbaum & A. Sen, pp. 208-232; 43 Even Haidt, who in recent years has been the foremost exponent of the influence of ethnocentrism on moral and political judgment, still admits that there are feasible and significant ways in which its influence can be mitigated (Haidt 2012, pp. 68-70). 41


such inquiry seems to require at least some measure of critical distance from beliefs current in one’s culture. The historical prospect of being raised under the influence of Nazi ideology serves to prove the point, but so can less extreme examples. If, for example, the modern practice of human rights were discovered to be (what many suspect it is) an unmitigated product of ‘Western’ ethnocentric biases (e.g., in favor of norms of autonomy and individual entitlement), this wouldn’t become any less of a problem if we suddenly restricted the application of such rights only to the ‘West’. What we normally want, and intuitively should want, for any schedule of rights that applies to us, is for it to be based on sound normative reasoning, and not merely on reasoning that some culture (even our own) takes to be so.44 (5) Morally and Practically Hazardous. So far I’ve been outlining the epistemic hazards of ethnocentrism, i.e., its unreliability and unjustifiability, and the manner in which it is so. But ethnocentrism is also morally and practically hazardous, or at least is normally and naturally thought to be. These normative hazards come in at least three main varieties. The first is a straightforward result of ethnocentrism’s unreliability, or its tendency to produce error. And it stems from the fact that reaching the right conclusion about some moral and/or practical matter is not just an epistemic concern; it is a moral and/or practical concern. For instance, if my culture draws me into the belief that it is permissible for me to slaughter animals, or that buying expensive things is the best route to happiness, and these beliefs are in fact false, then I will be drawn into immoral and impractical action. That is a moral and practical hazard if anything is. A second kind of normative hazard is slightly less straightforward, and follows from enculturation’s tendency to lure its subjects into an overinflated appreciation of their culture’s beliefs and practices. For instance, as noted above (in relation to ethnocentrism’s unreliability), enculturation can lead one to believe that culturally local normative beliefs and practices are the only acceptable ones, when in fact there is a variety acceptable alternatives, including some endorsed by other cultures. Mistakes of this sort are dangerous because they can promote intolerant attitudes towards dissenting views and practices (both at home and abroad) when tolerance, or even full acceptance, is what is really required.45 Finally, ethnocentrism is a moral and practical hazard in so far as it can lead to distorted understandings of foreign cultures more generally. This is true with respect to both kinds of error discussed towards the end of Section IV. For instance, Mr. Freuchen’s originally unsympathetic interpretation of the Eskimo treatment of orphans might have lead him to intervene when, in fact, a fuller understanding of that practice shows it to be sensitive to the interests of orphans in a way that makes intervention less obviously appropriate. And an exaggeratedly othered picture of foreigners (e.g., as weak, childlike, ignorant, inhuman, dirty, transgressive, malevolent, etc.) has played a pivotal role in the wrongful justification of imperialistic misadventures throughout the course of human history.46 In all these ways and surely more, the problem of ethnocentrism can have extremely grave moral and practical consequences. This five-part account provides us with a reasonably complete overview of the most natural and persistent concerns about ethnocentrism, and clarifies their nature and scope. It outlines the key senses in which ethnocentrism (at least in its epistemic aspect) can be 44

See e.g., Jeremy Waldron, “How to Argue for a Universal Claim” in Columbia Human Rights Law Review, Vol. 30, No. 2, esp. pp. 310-314, for a convincing version of this observation. 45 This is a classic complaint of 20th Century anthropologists. See e.g., AAA 1947; Herskovits 1948, p. 76. 46 See again, Said 1979, but there are innumerable other examples.


understood to be a problem. What I want to do in the following sections is take a look at how these concerns might be justified at a deeper metaphysical and epistemological level, as well as examine how such concerns change, or even disappear, depending on our background theory of truth and justification. Beyond its clarificatory value, one of the advantages of doing this is that it will help dispel the common misconception that the problem of ethnocentrism is motivated by skeptical and/or relativistic conceptions of truth and justification.47 Although allegations of (and worries about) ethnocentrism have often been advanced on skeptical and relativistic grounds, what I argue below is that, in fact, only realists (broadly conceived) can make sense of the full range of concerns listed above. In their allegations of ethnocentrism, those who endorse skeptical and relativistic premises often accuse others of failing to meet concerns that, properly understood, such premises cancel out. VI. REALISM One very natural way of making sense of the preceding concerns about ethnocentrism is via a realist theory of truth and justification. Because I am interested in ethnocentrism’s influence on both normative and descriptive judgment, the sort of realist theory I have in mind here is general, i.e., it is not confined to normative matters, nor just to descriptive ones, but to both. For the purposes of this discussion, I take realism to involve the following five claims: (a) Cognitivism: normative and descriptive judgments purport to represent facts about the world, and are not simply expressions of attitudes or emotions. (b) Ontological Objectivity: the facts that normative and description judgments purport to represent hold regardless of what we take them to be, and how we feel about them.48 (c) Epistemological Objectivity: the existence of a subjective belief about these facts does not in itself provide anyone with a reason to believe that the facts are actually so.49 (d) Optimism: there are indeed some such facts in light of which normative and descriptive judgments are either true or false. (e) Accessibility: there are methods (e.g., modes of perception, reasoning, memory, etc.) available to us through which we can reliably form true beliefs about these facts.50 47

See the references listed in Sections VII, VIII, and IX below. This condition obviously excludes cases in which what we are trying to truly or correctly ascertain is, in fact, what we believe about some matter, or cases in which references to what we believe are somehow contained within the judgment in question, e.g., “If Suzie believes that John needs to seek out help, then she should tell him so.” 49 Epistemological objectivity does seem like a natural entailment of ontological objectivity, after all: if some object of understanding is not subjective in nature, nor subjectively determined, then, in the absence of some longer story, there seems to be no grounds for assuming that my (or anyone else’s) thinking something about x (e.g., that x is good) is any reason to think that x is actually so. 50 This dimension is important since the claim that judgments purport to be, and can be, true in light of objective facts [i.e., (b) & (c)] is no great revelation unless we also have some idea of how we – as in, you and I – can come to ascertain such facts. In other words, it is no good to know that judgments can be objectively true or false unless we also have a reasonably clear idea of what feasible circumstances, modes of judgment, or epistemic conditions can justify our feeling confident, or at least more confident, that we actually have a true judgment in hand. 48


If one accepts these five premises, then the preceding concerns begin to make good sense. This is first and foremost because realism provides us with a clear explanation of why ethnocentrism undermines the reliability of our judgment. Given that ethnocentrism involves an indiscriminate bias towards culturally prevalent beliefs, it is bound to appear unreliable from a standard realist point of view, which understands facts to be objective, i.e., to hold independently of what we de facto, here and now, take them to be. Indeed, in light of the objective nature of the facts that one is after on such an account, what one should want from the outset, at the very least, is an open mind – e.g., towards conclusions based on appropriate evidence, towards expert opinion, and towards a diversity of views on the matter at hand, etc. – which is exactly what an enculturated (and typically stubborn) attachment to local cultural perspectives deprives us of. With few exceptions, then, ethnocentrism’s alignment of our judgment with that of the particular culture(s) into which one (more or less) randomly happens to be enculturated will look disastrously arbitrary from a realist point of view. Form here it’s not difficult to understand how the other concerns about ethnocentrism (i.e., concerns two through five) fit in as well. Because concerns about justification closely track concerns about reliability and/or epistemic responsibility that I rationalized just above, their activation is already guaranteed. But crucially, realists can also readily vindicate our sense that ethnocentrism is an inferior mode of judgment. This is because they optimistically believe not only that there are some objective facts but also that there are some feasible ways of accessing these facts that, unlike ethnocentric judgment, are reliable, or are at least more reliable than it is. This allows them to frame the problem of ethnocentrism in the way I did above, i.e., as a problem that merits efforts towards avoidance rather than resignation, and that one can be faulted for not at least trying to avoid. Realism can also make good sense of the generality of ethnocentrism’s failings. This is because the objectivity of the facts one is after on a realist account is, again with few exceptions, not going to change depending on the scope of one’s inquiry. For instance, on a realist account, the objective authority of basic normative principles – like the fact that the pain caused by torture is a reason not to torture others – will typically hold regardless of whether one is using such principles to evaluate one’s own actions, those of one’s cultural associates, or those of everyone in the world. This means that the same objectivist epistemic demands (of evidence-based judgment, deference towards plausible epistemic authorities, and openness to a diversity of viewpoints, etc.) will be in effect regardless of the scope of one’s inquiry, be it individual, local, cross-cultural, or universal. And so long as such epistemic requirements are in effect, ethnocentrism will continue to threaten the epistemic quality of one’s judgment. Finally, there is no difficulty in understanding how realists can accommodate concerns about the moral and practical hazards of ethnocentrism as I portrayed them above. Since realists can make fine sense of ethnocentrism’s propensity to generate mistakes (e.g., about the intentions and otherness of foreign cultures), on the one hand, and the existence of normative demands (e.g., of tolerance, acceptance, forbearance, etc.) that such mistakes can lead us to violate, on the other, the moral and practical dangers of ethnocentrism will have a natural salience from their point of view. At this point, it’s worth noting that not all five of the realist premises listed above are necessarily crucial to motivating the problem of ethnocentrism as I’ve understood it. In particular, premise (b) can seemingly be rejected without great consequence. This is because one can still affirm the unjustifiability of unmitigatedly ethnocentric judgment on the basis of the epistemic objectivity of some fact alone. Indeed, if the existence of a subjective belief about x


(e.g., that x is good) does not in itself provide anyone with a reason for believing that x is actually so, then indiscriminately assimilating the beliefs that those around you hold about x would be a fruitless and distracting epistemic strategy. One’s resultant judgment about x, in that case, would not be grounded in appropriate evidentiary considerations, i.e., appropriate reasons for belief. This means that some forms of anti-realism, such as quasi-realism and constructivism – which deny the ontological objectivity of moral facts (and so deny moral realism) while nevertheless affirming their epistemic objectivity – can motivate the problem of ethnocentrism just as naturally as realism can. So, for instance, if we are quasi-realists and, despite claiming that ethical commitments are at bottom merely expressions of our dispositional attitudes, still maintain that such attitudes (whether our own or those of others) cannot justify those commitments, I see no reason why the problem of ethnocentrism would change shape at all.51 Similarly, if we are metaethical constructivists, and so understand the truth or correctness of normative judgments to depend on whether they withstand some specified procedure of scrutiny rather than on whether they correspond to ontologically independent moral facts, the potential objectivity of the requisite procedure (i.e., its invariance with respect to de facto beliefs) can motivate the same concerns as well.52 Thus, realism, at least in its full-blown form, is not necessarily a preordained recourse for anyone wishing to vindicate the various concerns at issue. VII. SKEPTICISM But things do begin to change if we knock out more of the realist premises listed above. This is true if we are non-cognitivists that reject premises (a) through (e), and it is also true if we embrace some form of error theory and so reject premises (d) and (e). For instance, if one is a thoroughgoing non-cognitivist, then the problem of ethnocentrism could not be understood in anything like the form I’ve understood it here. This is because, on such an account, our judgment would not be in the business of trying to discern or represent realities about which it can be mistaken, and therefore unreliable. It instead be in the business of simply expressing the dispositional attitudes and emotions of its authors, e.g., as in, “Hooray for autonomy!” and “Global warming: boo!” etc. And although there might be some non-cognitivist analog to the problem of ethnocentrism on this account, one rooted in the fact that enculturation will tend to bring our attitudes and emotions into conformity with those of our cultural associates, it’s not clear to me what shape it would take. At the very least, what is clear is that such a version of the problem could not also be rooted in concerns about error and misrepresentation.


For instance, as Simon Blackburn writes: “our actual responses are inappropriate anchors for the wrongness of cruelty. What makes cruelty abhorrent is not that it offends us, but all those hideous things that make it do so.” Simon Blackburn, Essays in Quasi-Realism (Oxford: Oxford University Press, 1993), p. 172. Thanks to Jeremy Waldron for raising this point in discussion. 52 For instance, according to Christine Korsgaard’s version of metaethical constructivism, there are certain normative judgments – in particular those that value “humanity” – to which any agent who accepts any normative judgment at all is committed. Christine M. Korsgaard, The Sources of Normativity (Cambridge: Cambridge University Press, 1996). On such a view, ethnocentrism can perhaps lead one astray from realizing this latent implication of their de facto normative judgments – say, if one is born into a society where ‘humanity’ is discursively taken to have no value at all.


If, by contrast, one is an error theorist, and rejects only premises (d) and (e),53 then some aspects of the present understanding of the problem of ethnocentrism can be motivated and some cannot. For one, it is perhaps unsurprising that an error theorist can easily make sense of ethnocentrism’s unreliability, i.e., its tendency to lead its subjects into error. After all, so long as any judgment, normative or descriptive, is cognitive in nature (i.e., makes a claim to truth), it will turn out to be false on their account: there are simply no such facts to be correctly represented. Ethnocentrism would therefore be eminently unreliable. It would also be necessarily unjustified, since, properly understood, there are no good reasons for believing anything to be true on such an account, nor any (objectively) existent epistemic responsibilities to speak of. Moreover, these conclusions will hold with perfect generality, since, ex hypothesi, our judgment aspires to represent facts that do not exist in any domain of inquiry (individual, local, crosscultural, universal, etc.) and so will be unreliable and unjustified across the board. Despite all this, error theorists cannot do justice to the commonplace notion that ethnocentrism is a uniquely regrettable pitfall that renders our judgment less reliable and justified than it otherwise could be, and that one ought to try to avoid because there are ways of at least mitigating its influence and thereby improving one’s epistemic condition. For instance, if we think back to the interpretive worries raised about ethnocentrism and its tendency to produce two kinds of cross-cultural misunderstanding in Section IV, it is clear that those worries presume that there is a truth about (or at least more and less accurate interpretations of) the intentions behind some action, or the meaning of some social practice, that ethnocentrism leads us to miss. While cognitivist skepticism can certainly validate the notion that ethnocentrism will lead us into error, since all claims to truth will turn out to be wrong on its account, it cannot do justice to this basic thought about its leading us away from true belief or better understanding. Nor can it therefore do justice to the consecutive thought that individuals and communities can be faulted for not at least trying, in good faith, to avoid ethnocentrism. Accusations of ethnocentrism therefore lose their characteristic normative bite on this account. Finally, error theorists will also have difficulty vindicating the moral and practical concerns about ethnocentrism that I highlighted above. This is due to the fact that, while such theorists can, again, make sense of ethnocentrism’s epistemic failings, they cannot straightforwardly affirm the existence of normative demands that such failings (or any of ethnocentrism’s aspects, for that matter) may lead us to violate. It’s not at all clear then, that ethnocentrism would constitute any sort moral and/or practical hazard from the point of view of error theory. Despite all this, it is not uncommon to find critics of ethnocentrism who embrace some thoroughgoing form of skepticism. This is particularly true of outspoken critics of the ethnocentricity of the human rights movement, for example. Often, such critics will appeal to skeptical premises (e.g., rejecting the possibility of objective moral facts) in order to unmask the purported objective and universal authority of such rights as a fallacious product of ‘Western’ ethnocentric bias.54 Moreover, they will typically lament its facilitation of intolerant, undignified, and unjust behavior towards ‘non-Western’ peoples, as well as suggest that those who fall victim 53

Mackie (1977), is an example of one such theorist. See e.g., Zolo 1997, pp. 118-119; Alan Boyle and Christine Chinkin, The Making of International Law (Oxford: Oxford University Press, 2007), p. 13. These are also quoted in John Tasioulas’ valuable discussion of skeptical interpretations of the problem of ethnocentrism in his, “Parochialism and the Legitimacy of International Law” in Parochialism, Cosmopolitanism, and the Foundations of International Law (Cambridge: Cambridge University Press, 2012) ed. M. N. S. Sellars, pp. 26-33. 54


to it could have avoided doing so. Criticisms like this have contributed, it seems, to a general sense that skepticism and worries about ethnocentrism are somehow importantly linked at a conceptual level.55 But given that these kinds of concerns cannot intelligibly or straightforwardly be grounded in a skepticism that itself denies the objective veracity of all claims, we ought to remain cautiously aware of the inherent limitations of any skeptically grounded articulation of the problem of ethnocentrism, as well as any broader proposition of a conceptual link. VIII. RELATIVISM Historically, relativists (and not realists, nor skeptics) have been the most salient critics of ethnocentrism. Several of the early 20th Century anthropologists who were most responsible for bringing the problem of ethnocentrism to the forefront of contemporary academic (and popular) discourse were themselves cultural relativists. Their understanding of the problem was heavily undergirded by the notion that the truth about things (whether moral, physical, spiritual, etc.) could not be understood independently of one’s ‘cultural framework’ – a collection of assumptions, categories, beliefs, and experiences that is impressed upon us through enculturation – and that, as a result, different cultures produced equally valid or acceptable interpretations of the world.56 As a bias in favor of one’s own culture’s standard interpretations of things, ethnocentrism was understood to be a problem precisely because it promotes ignorance of this basic fact about equal acceptability. And in so far as it served to remind people of this fact and so reduce such ignorance, the doctrine of cultural relativism was often seen as a kind of antidote to ethnocentrism.57 In line with this sort of thinking, many theorists, including many contemporary moral and political philosophers, still understand the problem of ethnocentrism against the background of some form of relativism. For instance, Steven Lukes argues that commonplace concerns about ethnocentrism are a direct result of equally commonplace relativistic “intuitions.”58 James Griffin suggests that worries about ethnocentrism are easily understood as worries about the truth of relativism itself.59 And John W. Cook has argued that the very meaning of the term “ethnocentrism” cannot even be grasped without having recourse to the notion of relativism.60 I have no interest in evaluating the rationale behind relativism here. This has been strongly debated elsewhere. All I intend to do is examine the implications of relativism for our understanding of problem of ethnocentrism. These implications are, I think complex in a manner that has not yet been adequately appreciated. And this complexity starts with the fact that relativism is readily capable of annulling all of the concerns about ethnocentrism that have preoccupied us so far. Unlike realism, cultural relativism (which is the kind of relativism that those relativists who worry about ethnocentrism tend to have in mind) denies both premises (b) and (c): 55

For this reason, Tasioulas (2012, pp. 26-33) astutely dedicates a whole section to skeptical interpretations of the problem of ethnocentrism. 56 See e.g., Boas 1901; the Herskovitz quote above in fn. 12. 57 See e.g., Cook 1999, pp. 24-31, for an array of references. 58 Steven Lukes, Moral Relativism (London: Profile Books, 2008), pp. ix-xi. 59 See: James Griffin, “Are Human Rights Parochial?” in Parochialism, Cosmopolitanism, and the Foundations of International Law (Cambridge: Cambridge University Press, 2012), ed. M.N.S. Sellars, pp. 158-162. 60 Cook 1999, pp. 80-81.


epistemic and ontological objectivity. On its account, not only is the truth about some matter x determined by the content of culturally established judgments about x (so that, for example, infanticide will be permissible in some culture if enough of its members believe it to be so); it also logically follows that the cultural prevalence of a certain belief about x constitutes a reason for believing that x is actually so. All this makes ethnocentrism likely to be an effective way of forming true and/or justified beliefs from the point of view of cultural relativism. Since enculturation aligns our judgment with culturally pervasive beliefs that, on this account, dictate the truth, ethnocentrism suddenly looks more like an epistemic virtue than a vice. The problematic nature of ethnocentrism, from a relativistic point of view, emerges only when we refine the doctrine in specific ways. First and foremost, it is important that the version of cultural relativism we appeal to be “agent-based” rather than “assessor-based”. To illustrate why this is so, consider the following assessor-based version of cultural relativism. In order to simplify matters, I shall focus only on normative relativism in what follows, i.e., cultural relativism about normative facts rather than about all facts or descriptive facts.61 Assessor-based: Whether x is a normative reason for agent m to y depends on whether the conclusion that x is a normative reason for agent m to y follows from the global set of normative judgments that are endorsed by the assessor’s cultural group, in combination with the non-normative facts.62

On this sort of view, the problematic nature of ethnocentrism disappears, at least in the context of normative inquiry. Since, in order to correctly evaluate what agents have reason to do, an assessor only has to consider what follows from the global set of normative judgments endorsed by their own (i.e., the assessor’s) culture – which is precisely what ethnocentrism impels them to do anyways – nothing in the nature of ethnocentrism will steer an assessor wrong on such an account.63 If, for example, my culture understands the infliction of pain to be morally and practically desirable, then on this account I will be right to assume what the process of enculturation will naturally lead me to assume: that agent a (whoever they may be) has at least a prima facie reason to inflict pain on others. Of course, you would also be right to assume just the opposite – that agent a has at least a prima facie reason not to inflict pain on others – if that is indeed what follows from the normative judgments current in your culture. Either way, the preceding concerns about the unreliable and unjustified nature of ethnocentrism seem to vanish on this sort of account: and not because the very idea of pursuing the truth is deemed bankrupt, as in the case of skepticism, but because in this relativistic context ethnocentrism becomes an excellent way of accessing the truth. There is thus no foothold for anything like the present understanding of the problem of ethnocentrism available here. The kind of relativism that is usually appealed to by cultural relativists who condemn ethnocentrism is slightly different from this, however. The usual idea is not that the facts that apply to (or “hold” for) some agent are determined by the culture of a third-party assessor, but that they are determined by the agent’s own culture. That is, the relevant doctrine is an “agentbased” form of relativism, according to which a culture’s established judgments (e.g., its “moral 61

I am borrowing heavily here from Sharon Street’s insightful article, “How to be a Relativist about Normativity” (DRAFT). Available online. 62 Note how nothing special hangs on the notion of a “reason” here. I could just as well have talked about x making it “good” for agent m to do y, without any loss or serious change in meaning. 63 Let’s assume henceforth that, for illustrative purposes, the relevant non-normative facts are both known and their normative implications easily understood.


code” and/or its broader system of belief) have local authority, since they determine the facts that hold for members of the culture in question and no one else.64 To illustrate, consider the implications of this alternative version of normative cultural relativism: Agent-based: Whether x is a normative reason for agent m to y depends on whether the conclusion that x is a normative reason for agent m to y follows from the global set of normative judgments that are endorsed by m’s cultural group, in combination with the non-normative facts.

On this account, enculturation is not similarly guaranteed to steer our judgment right, at least not in all instances of normative inquiry. If one’s aim is to determine the reasons that apply to members of their own culture, then ethnocentrism will again be unproblematic, since it will naturally align their thought with the global set of normative judgments that dictates the matter in question. But if, on the other hand, one turns to the question of what reasons apply to individuals who belong to other cultures, different from their own, then enculturation will have aligned their judgment with the wrong global set (i.e., the wrong determinant of truth), and so ethnocentrism becomes undesirable. For instance, the fact that my culture takes the infliction of pain to be morally and practically desirable is entirely irrelevant to the question of what some culturally foreign agent m has reason to do, on this account. In order to decide the matter appropriately, I need to adopt the standpoint of the global set of normative judgments endorsed by m’s culture, which is of course not what enculturation will naturally lead me to do (unless our cultures coincidentally happen to endorse the same set). Thus, by contrast with the previous form of relativism, the epistemic risks of ethnocentrism reappear on this account, but their presence depends on the scope of normative inquiry. If our aim is only to determine the reasons that apply to members of our own culture (including ourselves), then ethnocentrism will be benign, even useful; but if our aim is to determine the reasons that apply to members of alien cultures (like m) then it will naturally lead us into error and so become a problem. In this way, relativists are not self-deluded if and when they speak out against ethnocentrism and emphasize the risks it poses for judgment across cultures, provided they adopt something like this agent-based view. But the limited scope of their condemnation is significant. For, while such relativists can certainly complain about the inferior reliability and justification of ethnocentrism in the context of cross-cultural judgment, they can do no such thing in the context of intra-cultural judgment. On their view, it will always be appropriate to assimilate the established judgments of one’s culture – no matter how apparently backwards or outlandish – providing one’s aim is to determine the facts that hold for oneself and/or one’s cultural associates. This means that cultural relativists cannot make sense of the generality of the problem of ethnocentrism as I’ve understood it here and as it is commonly understood. Nor can they make ready sense of the moral and practical hazards of ethnocentrism as these are understood even, in at least some instances, by relativists themselves. For instance, the 20th Century cultural anthropologists that I mentioned just above, along with their contemporary followers, understand the attitude of cultural superiority engendered by ethnocentrism to be not only unmerited but also morally hazardous. They frequently warn against the propensities towards interventionism and intolerance that can accompany such an attitude.65 But it’s not at all 64

See e.g., AAA 1947. For more contemporary examples of this kind of relativism, see: David Velleman, Foundations for Moral Relativism (Open Book Publishers, 2013), Ch. 4; Street DRAFT. 65 See e.g., AAA 1947; Herskovits 1948, p. 76; Alison Dundes Renteln, “Relativism and the Search for


clear how their commitment to cultural relativism is compatible with such normative pleas for universal tolerance and non-intervention. After all, the bindingness of the norm of toleration for any given agent will depend on the specific content of their culture’s global set of normative judgments, or its ‘moral code’, which leaves the universal bindingness of any such demand entirely vulnerable to the existence of intolerant and interventionist cultures, of which many examples can be provided.66 Because the content of the reasons that apply to us will be subject to deep variation in this way, cultural relativism’s ability to do justice to the straightforward moral and practical hazards of ethnocentrism that I illustrated above is unstable and ambiguous at best. IX. THE ‘EQUAL ACCEPTABILITY’ OF CULTURES Before concluding, it is worth mentioning that one of the lingering advantages of relativism with respect to motivating the problem of ethnocentrism, it might be thought, is its ability to undergird the familiar notion that cultures are in some sense equal arbiters of the truth. Such an idea has figured prominently in discussions of ethnocentrism in just the way described above, i.e., as something that ethnocentrism is thought to lead us to overlook.67 And while (agent-based) cultural relativists can affirm the equality of cultures so far as their local epistemic authority goes – since the cultures of alien peoples determine what is true for them just as authoritatively as our own culture determines the truth for us and our cultural associates – realists and skeptics seem far less well-positioned in this respect. This is because realists, for their part, understand truth to be objective in a way that makes it entirely possible for some cultures to have failed to grasp it and therefore to be, in this sense at least, unequal to other cultures. And skeptics, for their part, can only affirm a peculiar kind of equality – the equality of all cultures’ complete ignorance or disengagement from truth seeking – that is too different from the notion at hand to be relevant. There are two main points that I’d like to make here. First, as noted early on (in Sections III and IV), ethnocentrism is not only maligned for its tendency to promote ignorance of the supposed equal acceptability, validity, or reasonableness of different cultural practices and points of view. It is just as often maligned for its tendency to thwart our understanding of the intentions behind alien cultural behavior and practices, regardless of how malicious or benevolent these may be. And Augustinian-Millian style complaints about the conservatism of ethnocentrism, also noted above, and which remain no less prevalent, often make no mention of the ‘equal acceptability’ of cultures. Indeed, that important genre of complaint often assumes that there is only one right answer to objective normative and descriptive questions (e.g., about the permissibility of slavery, or the existence of black holes, etc.) and that, by stubbornly aligning one’s judgment with that of the culture that one (more or less) randomly happens to be a member of, one significantly lowers one’s chances of arriving at the right answer. In light of all this, it’s important not to overstate the significance of the idea of the ‘equal acceptability’ of cultures for understanding the problem of ethnocentrism. One of the upshots of defining ethnocentrism as a kind of bias rather than as a specific kind of belief, as I have, is that it can be associated with various types of error, not just that of overlooking the (supposed) equal validity, acceptability, or reasonableness of different cultural beliefs and practices. Human Rights” in American Anthropologist, Vol. 90, No. 1, 1988, pp. 56-72. 66 This is a widely noted logical inconsistency, and defined as the “anthropologist’s heresy” in Williams 1972, pp. 20-22. 67 See: fn. 57 above.


Second, relativists are not at all unique in being able to make sense of this notion of equal acceptability. Realists can do so as well, and in a number of interesting ways. I shall mention four here: two general, and two that are particular to normative affairs. First, given that realism typically requires us to form beliefs about the world on the basis of good or appropriate evidence, but provides us with no guarantee that such evidence will in all cases (or in any case!) be clear, abundant, and decisively point us towards one conclusion, it is entirely within the scope of realism to affirm the justifiability, acceptability, validity, or reasonableness of various different beliefs about some subject matter. Consider, for example, how it was hardly unreasonable for some ancient Greeks to assume that earthquakes were caused by the release of pressurized air in underground caves, and for others to assume that Poseidon was to blame, given the evidence and instruments available to them at the time. Realists are also perfectly capable of recognizing that some questions can be arbitrarily decided by convention. For instance, our conventional ways of measuring time, mass, and space – e.g., ‘days’, ‘seconds’, ‘meters’, ‘miles’, ‘grams’, ‘pounds’, etc. – are entirely arbitrary. We could equally well measure time on the basis of the lunar rather than the solar calendar, or even abandon the seven-day ‘week’ in favor of an eight-day alternative, and we wouldn’t be making any sort of objective mistake if we did. There are innumerable junctures at which the world allows us to carve it up (more or less) as we see fit. And realists can readily acknowledge the equal acceptability or validity of various culturally distinct ways of doing so via the adoption of some coherent but essentially arbitrary set of conventions: whether linguistic, scientific, temporal, spatial, social, conjugal, etc. Third – and this brings us on to the normative domain in particular – it is also open to the realist to recognize that objective normative principles can be realized in a multiplicity of acceptable ways. For instance, if we want to ensure the political participation of citizens in their government, there are a number of ways of achieving this: at an institutional level, we can adopt various forms of democratic government (e.g., direct, representative, majoritarian, consensual, etc.), and at an individual level, we can engage in various participatory activities (e.g., petitioning, protesting, running for office, disobeying, etc.). Surely, different societies can achieve this same end in different ways. And sometimes the achievement of a single end will require different means in different cultural circumstances. For example, if we were Classical Utilitarians, and discovered that the members of one society found the greatest happiness in obedience to authority, and that the members of another were happiest when able to decide for themselves how to live, then presumably we would recommend different principles of political morality for those respective societies: the one more authoritarian, and the other more liberal. In these ways, cultures with importantly different normative beliefs and practices can be equally acceptable or valid from a realist point of view. Finally, realists can affirm the equal acceptability of distinct normative beliefs and practices by endorsing a pluralistic understanding of normativity, i.e., one that asserts the equal importance of different fundamental values and the equal acceptability of different rankings thereof.68 For example, a pluralist can (in principle) affirm the equal acceptability of a social order that prioritizes the values of equality and harmony over that of individual autonomy, on the one hand, and one that does just the opposite, on the other. Thus, on such a view, someone raised under the first kind of social order and convinced of its superiority would risk judging alien societies of the second kind by (locally practiced and acceptable) standards to which they 68

See e.g., Isaiah Berlin, “The Pursuit of the Ideal” in The Crooked Timber of Humanity (Princeton: Princeton University Press, 1998), ed. H. Hardy, for a classic statement of this view.


nevertheless need not conform: just the sort of error the agent-based cultural relativist warns against.69 But since the range of acceptable systems of value, on such an account, is neither infinite (e.g., a regime that overwhelmingly prioritized equality and harmony over individual liberty would presumably be unacceptable) nor subjectively determined (because the equal acceptability of different normative codes is an objective feature of normative reality itself) this does not amount to relativism as I have defined it. And so recourse to relativism is neither necessary to justify the amorphous idea of the equal acceptability of cultures, nor is it a necessary antidote to ethnocentric failures to recognize such equality in general. X. CONCLUSION In this paper, my aim has been to evaluate the dangers of ethnocentrism in much the same way as we might evaluate the dangers of any cognitive bias: be it one rooted in laziness, overeagerness, loyalty, self-esteem, or self-interest. Because my aim was to offer an account of the various ways in which such a bias is normally and naturally thought to be problematic, the preceding discussion has been rather one-sided. That is, I haven’t bothered to spend any time thinking about the possible virtues of ethnocentrism, or the ulterior benefits of such a bias. Surely there are some such benefits. Allan Bloom, for example, was probably right to point out that social stability is one: “Only if they think their own things are good can they rest content with them.”70 It is also true that there are circumstances in which there is nothing wrong with being biasedly partial to beliefs that one grows up with. Ronald Dworkin, for instance, has plausibly argued that, in cases of rationally irresolvable normative disagreement – including cases where the disagreement in question is a result of enculturation – each interlocutor is entitled to assume the superiority of their own view, i.e., to assume that they are, in effect, epistemically luckier than their counterpart.71 And he is likely right about that. We do need to start from somewhere, so to speak, and it seems unobjectionable for us to hold a normative belief to which we are already intuitively committed but cannot rationally defend (nor rationally defeat) as a kind of default view: that is, true until we are presented with evidence to the contrary. The fact that the problem of ethnocentrism, as I’ve argued, can only be motivated in all its intuitive dimensions from a broadly realist point of view, is significant. For one, it lays ruin to the common perception that worries about ethnocentrism are somehow inherently linked to the acceptance of some form of relativism or skepticism. And along with this, it also debunks an equally common sense that the problem of ethnocentrism is one that will disappear as soon as we discover convincing arguments against relativism and skepticism and in favor of realism or something very much like it.72 In fact, quite the opposite is the case. If anything, the problem of ethnocentrism is graver and more pressing for realists than it is for anyone else.


See: Susan Wolf, “Two Levels of Pluralism” in, Ethics (1992), Vol. 102, No. 4, pp. 785-798, for a useful demonstration of this. 70 Allan Bloom, The Closing of the American Mind: How Higher Education has Failed Democracy and Impoverished the Souls of our Students (New York: Simon & Schuster, 1987), p. 37. 71 See: Dworkin 1996, pp. 121-122. 72 This is a suggestion that seems implicit in Tasioulas’ version of the problem in, “From Utopia to Kazanistan: John Rawls and the Law of Peoples” in Oxford Journal of Legal Studies, Vol. 22, No. 2, 2002, pp. 390-395.


This is true not just because realists can make sense of the full range of concerns discussed above, but also because realists will inevitably have a harder time addressing those concerns than skeptics and relativists will. Skeptics, for their part, have no interest in addressing the problem of ethnocentrism, since for them it either does not exist or is incapable of remedy. And relativists, for their part, can easily resolve the problem whenever it arises: all we need to do to discover the truth about some matter is, in some sense, to identify what the relevant culture already thinks about it. But for realists, matters are more difficult, since they affirm the possibility of avoiding or at least mitigating ethnocentrism but lack an obvious epistemological theory that tells us exactly how this is supposed to be done. This is particularly true in the normative domain, where the question of how truth is attained remains alive but plagued by intense controversy. This gives the problem of ethnocentrism a further dimension still, on a realist account: the burden of developing a theory of avoidance.73 That is of course a task for another occasion, but one that the present analysis has contributed to by at least giving us a clearer sense both of what needs avoiding and why.


Buchanan’s recent paper (2008) is a useful first step in this respect.