1. Introduction

Liberal social contract theories in Western analytic philosophy usually conceive of persons as abstract political subjects united by a common humanity and a common participation in a mutually beneficial social arrangement. Critics like Charles W. Mills, however, counter that these views ignore that the actually existing contract is far from being one of mutual advantage, but is instead designed to enforce economic and sociopolitical and relations that heavily favor whites over nonwhites; Mills calls this actually existing contract the “racial contract.” In his book The Racial Contracti he contends that as long as white philosophers defend theories of justice that are blind to the existence of the racial contract, they help cultivate an epistemology of systemic ignorance that threatens the recognition of nonwhites as full persons and propagates a neoliberal model for quiet racial domination.

I think that Mills’s argument is fundamentally correct, and the most prominent liberal contractarian theories, from Hobbes to Rawls, may indeed foster an epistemology of ignorance. But, of course, there is more to white political philosophy than social contract theories, and other approaches are more capable to acknowledge the racial contract as a serious problem. One such approach is Martha Nussbaum’s account of basic human capabilities, which she defends in collaboration with Amartya Sen. Here I contend that the capability approach to justice is not vulnerable to the theoretical failings that Mills finds with liberal contractarianism, specifically the propensity to foster an epistemology of ignorance by silencing the political voices of the oppressed. As I agree with most of Mills’s conclusions, this paper ought not to be construed as arguing against his claims. Rather, it contends that his critique of liberal contractarianism is one good reason to reject it in favor of a capability–based approach to justice.

Section I discusses Mills’s argument and makes explicit his account of racial, particularist, and communitarian social ontology. Section II shows why Mills’s argument thus strengthened constitutes an even stronger objection than Mills’s to John Rawls’s liberal contractarianism. Finally, section III shows why, in light of the above, we should prefer a capability approach over Rawls’s liberal contractarianism.

2. The racial contract

Once more, Mills argues that the racial contract is the “actually existing contract” that regulates the global interactions of people and peoples, as opposed to abstract ideal arrangements. For Mills, the actually existing racial contract has three main features. First, it is historical, for it is born from white domination and reinforced through the narrative propagation of white self–entitlement. Much white folklore still reinforces the European colonial ideals of cultural and technological superiority as the rightful catalysts of global supremacy.iv

Underlying these claims is a certain view of social ontology that Mills does not make as explicit as he could, but which is nonetheless crucial to his argument. To begin, his definition of race is anti–essentialist: terms like race, white, and nonwhite are to be understood as historically contingent and, therefore, “not really a color at all, but a set of power relations,” by which he means the colonial histories of white domination.v Mills is not anti–realist: race is a real social kind for him, but it is not determined by biology, ethnicity, tradition, or culture. Rather, it is born from power struggles and repeatedly redefined in those terms. Indeed, Mills concedes that membership into white or nonwhite “clubs” fluctuates depending on contingent local historical constructions of whiteness and nonwhiteness, and of course also on the “official” partitioning of these spaces dictated by the dominant hegemony, those who self–identify as “pure” white. Thus, personal identity is grounded in race, and race is grounded in the social and economic arrangements that have shaped the course of human history. Sensible as it is, this account of social ontology is also what makes it possible for an epistemology of systemic ignorance to subsist in the first place: locking people in racialized identity claims whose terms are defined by the dominant groups is an efficient method for the oppressors to broadcast their theories and silence the rest.

I think these claims are well grounded and receive ample evidential support beyond Mills’s own discussion. The epistemological claim is especially well–taken, as ignorance of the racial contract is propagated by many cultural staples of recent Western (and especially American) culture: segregation and miscegenation laws, misguided reparation efforts, racist textbooks, culturally blind literary canons, the insistence of mainstream Western media on “white guilt stories,” etc.vi Together, these tell a convincing story of how we actually form false beliefs that misrepresent our racial reality. But this is so only if we understand epistemology as a primarily social rather than personal activity.

The idea of a social epistemology, so central to critical theory, is also neither new nor especially controversial to the pragmatist and analytic traditions; indeed, most contemporary Western epistemologists do admit of some social dimension to knowledge. But an epistemology of ignorance demands the centrality of the social dimension. Individuals by themselves may have no means and little interest in producing or promulgating ignorance—but groups do. To borrow an example from feminist epistemology, Lorraine Code argues that since each gender identity has special epistemic access to certain facts and facets of the human experience, there are no such things as genderless, neutral, objective knowers.vii Human knowers are male and female, situated in space and time, and deriving their beliefs (or at least the inputs of their belief–forming practices) from their social milieus. In other words, knowers are specially situated in non–generalizable ways. Similar arguments apply to class, race, ethnicity, age, religion, sexual orientation, and other sources of situatedness.

Epistemologies of ignorance thrive especially well in these conditions, because a social epistemology provides the means and the motives for error to be created and propagated. Individual knowers are the sole obtainers of their own experiences and judgments. While vulnerable to Cartesian evil geniuses, they are mostly immune from the experiences and judgments of other knowers, save for using them as foils or terms of comparison. And even then, traditional “S knows that p” epistemologies—which are the formulaic and quasi–scientific standard of analytic philosophy—see ignorance only as a roadblock, or at most a latent skeptical pitfall to avoid along the path to knowledge. But if knowers are situated and knowledge is socially acquired and evaluated, the door opens to radical misinformation, because our epistemic practices derive in large part from sources we cannot control. To paraphrase Alcoff, epistemologies of ignorance construe error as central, both quantitatively and qualitatively.diktats from smoky rooms, but a simple and sensible story of how historically dominant groups (in this case whites of European descent) mistake their own socially construed epistemology for absolute knowledge, promulgate it as such, and ostracize the alternatives as “mistaken” or “illogical.” This is both easy and palatable to whites, as it operates by the same mechanism that led to white domination in the first place.

I can turn now to the primary target of Mills’s critique, the liberal contractarian tradition that is dear to many white Western philosophers. Mills discusses primarily Hobbes, Kant, and Rousseau, but I will focus on Rawls instead, given his popularity among 20th–century white political thinkers, showing how Mills’s argument is especially powerful against Rawls’s brand of liberal contractarianism.

3. Rawls’s liberal contractarianism

To many moral–political philosophers, the charge that white political philosophy seldom recognizes the existence of the racial contract, let alone engage it cogently, will just seem false. Many thinkers argue along the lines of Mills’s own historical claim. The better known may be Thomas Pogge, who echoes Mills in arguing that a theory of justice is incomplete unless it engages the systemic privilege that whites have created and maintained.xiii have made similar cases as well, though reaching different conclusions, from enforcing strong (indeed neocolonial) famine relief efforts to endorsing non–interventionism. In general, the fact that current sociopolitical and economic inequalities directly result from sustained European ethnocentrism is such an obvious historical truth that it is hardly a groundbreaking revelation in philosophical axiology.

But Mills is right that even when these inequalities are addressed, they are rarely said to be specifically racial, and the epistemic component of the racial contract is rarely taken seriously. Many of the aforementioned theorists, white and nonwhite alike, often couch their arguments in the supposedly race–neutral geopolitical language of “the West” and “the Third World.” Even if we could remap these semantics onto a race discourse—and it is not obvious that we can—they still betray an unwillingness to acknowledge that discourse frontally. And even when theorists do acknowledge the fact of the racial contract, they seldom elaborate on it at length or cast it as the conceptual core of their theories.

Mills’s argument goes deeper than accusing a few philosophers of being color–shy. He claims that the vast majority of work in the social contract tradition, always a stalwart of white sociopolitical philosophy, ought to be dismissed because all social contract theories are rooted in racism. The works of Grotius, Hobbes, Locke, Rousseau, Kant, and to some extent even Marx are tainted by their explicit endorsement of racist assumptions about non–European nonwhites. For the vast majority of its history in the intellectual tradition of the white West, the signatories of the social contract have been white males, and these theorists have provided little or no indication that their work was to mean anything else. Of course, theories of justice need not explicitly endorse the racial contract in order to support it. All they need to do is not acknowledge the existence of the racial contract and suggest social arrangements that either do nothing to abrogate it or that actively support its propagation.

Among liberal contractarian theories, Rawls’s Kant–inspired deontological constructivism fits this profile especially well. Rawls thought that the representatives of free and equal citizens, in an original position of initial fairness, will concur on a political conception of justice that everyone can and will endorse regardless of social standing. Even though persons privately endorse different reasonable comprehensive doctrines—religious, philosophical, traditional, historical, and moral—each doctrine yet comprises the theoretical elements needed to support a public and political conception of justice. Thus, when deciding the principles of justice to regulate their society, citizens must focus on this overlapping consensus and deemphasize their individual contingencies, stashing them away behind a “veil of ignorance,” lest they render the terms of the contract biased or unequal. Stated differently, people need not agree on anything except on what they find themselves agreeing upon after a process of fair bargaining. If it should be impossible for a person to support a political conception of justice from within his comprehensive doctrine, then that person’s comprehensive doctrine is unreasonable and he ought to be dismissed from the bargaining process. Rawls’s substantive claim is that those holding strong considered convictions about freedom, democracy, and equality will in fact also hold reasonable comprehensive doctrines and be able to support the political conception of justice.xiv

On the face of it, it seems surprising that this is a white man’s theory. One might think that whites would very much like to reject it, for why would those who have the upper hand in racialized historic relations agree to abandon them? Why would whites accept to disregard their historical privilege behind a fictitious veil of ignorance once they are made aware of it? But I think it is nonwhites—and everyone who is aware of the racial contract—who are most keen to reject Rawls’s theory, because his proposed equality is the wrong kind of equality. For one, to demand that the signatories of the contract disregard their individual socioeconomic contingencies is to propose a kind of censorship. Many philosophers have already criticized Rawls on these grounds, arguing that the ontological and moral demands imposed by the veil of ignorance are undesirable or even unattainable; Michael Sandel and Martha Nussbaum are notable examples. If the moral and social makeup of nonwhites is defined, even partly, by their nonwhiteness in a white–dominated world, then to demand that they ignore their nonwhiteness in the original position is to repress the key elements of their identity. Of course, this argument assumes (as the racial contract also does) a decidedly particularist social ontology that counters Rawls’s universalist essentialism.

The sort of nominal equality proposed by Rawls is a fantasy. Legal egalitarianism is a convenient escape route for whites to not be held accountable for their oppressive policies. Among the contingencies disregarded in the original position are not only individual ones, but also historical ones, including past and present social events. For instance, the facts of European colonialism and American slavery are deemed irrelevant to the present arrangement of justice, so again, insofar as those historical facts are ontologically relevant to the psychosocial makeup of nonwhites, crossing them off the rulebook is a form of censorship. Likewise, the desire for reparative measures harbored by certain nonwhite groups will never even be addressed in the original position. This has nothing to do with the moral permissibility of reparation and everything to do with its stifling: all reparative demands would be dismissed a priori on purely procedural grounds. And while we may charitably read Rawls and other contractarians as proposing a contract that would work in “ideal” conditions, the fact remains that the actually existing conditions are not ideal and that it is not up to them to wipe the slate clean to make them ideal.

If this argument is accepted, the proponents of the racial contract have a stronger reason than Mills’s to reject Rawlsian constructivism. The problem is not that Rawls’s argument is grounded in racist theories written centuries ago, nor that it is an instrument of oppression because the proposed equality is only pro forma. Instead, the problem is that the proposed equality is theoretically unachievable and runs the risk of systematically silencing nonwhite voices. Consider the phrase “We the People,” where “People” referred only to white male landowners. The problem is not that it could not be made to refer to all persons, but that the rules by which it was reached in fact did exclude nonwhites, non–males, and non–landowners from what counted as a “person” (the three–fifths compromise, etc). These were never given a voice in the establishment of the contract, just as nonwhites are not given a sufficient voice as nonwhites in the establishment of a Rawlsian society.

Far from rejecting an epistemology of ignorance, Rawls institutionalizes it by making the donning of a veil of ignorance the procedural centerpiece of his theory.xv Even if the resulting Rawlsian society eschewed the traditional media of ignorance propagation (say, by avoiding white guilt stories), the terms of the contract would still be agreed–upon by silencing certain voices and ignoring certain past events. Recall that the original position is not an actual point in time when freedom–loving people make decisions “from scratch,” but a thought experiment to guide our choices of justice from a historically specific and political mindset. Past events are likely to play a significant role in the culture and mindset of many groups and individuals—and yet, in a Rawlsian society, ignorance of them would be enforced at every turn, and when brought up they would be deemed irrelevant. Perhaps Rawls’s theory was never meant to uphold the status quo, but it does have the unpleasant consequence that the privileged enjoy their privilege thanks to the very rule by which they silence those who question it.

Thus, with specific regard to social epistemologies of ignorance, Rawls’s deontological constructivism contains a significant theoretical flaw that Mills does not fully appreciate but that would strengthen his argument for the pervasiveness of the racial contract. This theoretical flaw also distinguishes Rawls’s liberal contractarianism from other Western sociopolitical theories that are not vulnerable to the same objections and can instead openly discuss and work toward the elimination of the racial contract. Thus, we can use Mill’s argument against Rawls to show at least one significant way that liberal contractarianism is inferior to at least one of its main rivals. In the following section I will focus on the capability approach to justice championed by Martha Nussbaum and Amartya Sen.

4. The capability approach

Originally, the capability approach is rooted in Sen’s analysis of the concept of development in deeply impoverished economies. Many Westerners see development as an increase in wealth or living conditions, but Sen understands it as an increase in capability, defined as the realistic opportunity to achieve self–determined or socially determined life goals.xvii

Nussbaum then builds on Sen’s model to propose a full–fledged political theory of justice grounded in the concept of capability. Whereas for Rawls justice is procedural fairness, for Nussbaum, justice is measured in the actual availability of human capabilities. To specify which aspects of human life are essential and thus constitute “basic capabilities,” Nussbaum borrows Rawls’s idea of an overlapping consensus. She believes that we can reach a lasting agreement on which capabilities must be guaranteed for justice to exist, not from an abstract original position, but from concrete negotiation that values contingencies (past and present, collective and individual) instead of hiding them. Nussbaum thus proposes a list of ten basic capabilities that, she says, virtually everyone will endorse: being able to live a full biological life, the right to bodily integrity, being able to determine one’s emotional life, and so forth. She takes pride in pointing out that this consensus was itself reached through extensive dialogue and negotiation among privileged and underprivileged people of all classes, races, and ages, and that as such it is sufficiently representative of the disenfranchised and humanity as a whole.xviii

I will not attempt to analyze Nussbaum’s project in its entirety or discuss whether the ten capabilities are in fact sufficiently universal to be endorsed by all. Here I assume that such a project can be defended in its own right. If it can, then it can also welcome the postulations of the racial contract. First, and most importantly, it can address the epistemological claim. Whereas social contract theories propagate an epistemology of ignorance by silencing nonwhites, the capabilities approach does the opposite by promoting an epistemology of openness, which in turn exposes the historical and metaphysical falsehoods that allow a racial contract to persist. When procedurally unfettered dialogue is permitted and encouraged, the racial contract is likely to be the key subject. As Mills also argues, there is a deep Nietzschean divide between the sociopolitical philosophies of whites, which tend to focus on abstract justice; and those of nonwhites, which usually center on white oppression and variously racialized social arrangements (similar remarks also apply to other socially oppressed groups: refugees, women, children, persons with disabilities, etc). Nussbaum’s approach would help gap this divide by forcing whites to deemphasize abstract conceptions of justice and acknowledge the actually existing racial contract.

The capabilities approach is even more successful in addressing the social ontological claim of the racial contract. For one, far from being seen as nonwhite nonspaces, nonwhites occupy the same exact metaphysical place as whites: full people with full social identities and the full capacity to establish life paths and work toward their achievement, unhindered by oppression. The particularist social ontology dear to constructivists is thus vindicated, for the concepts of race determined by Eurocentric power relations are now brought to light and allowed to be part of the debate. Once the epistemic veil of ignorance has been lifted, all claims are again fair game, and what was censored under the pro forma egalitarianism of universalist essentialism now becomes viable.

Of course, this prospect does not dispel all worries. A most obvious obstacle is a strong resistance by whites themselves, philosophers and laypeople alike. After all, whites are the primary targets of the systemic ignorance propagated by their own dominant group; most nonwhites are well aware of the existence of a racial contract, but most whites are reluctant to acknowledge it. Even if they should acknowledge it, whites have no interest in actively working toward its suppression; and as under the capability approach they would lack the procedural epistemic shield of the veil of ignorance, they might find other ways to protect white privilege and enforce new versions of the racial contract. Some may be subtler; others may be overt and violent. Here one might protest that if this is what we think of whites—purely self–interested creatures who will always attempt to dominate regardless of circumstances—then one wonders why we should bother with constructive dialogue or theories of justice in the first place. Some may prefer a solution where nonwhites surge and revolt, violently or not, to take back equality after a half millennium of oppression.

A similar worry is that even if the capabilities approach were to yield positive results and genuinely address the terms of the racial contract, it may be seen as (and it may also be) just another instance of nonwhites getting their liberation at the hand of whites. After all, even those who are sensitive to the oppression of systemic ignorance still postulate that liberation is only possible only if whites agree that it is. It takes two to tango, so to speak, and the theory’s staunchest defenders are a white Jewish philosopher (Nussbaum) and an Indian–American economist who works in Western academia (Sen). The critic may thus complain that even this way out of a racial contract is just imperialism in a new guise. My pragmatic answer to that objection is to ask what the alternative is. If it is agreed that the capabilities approach properly addresses the negative claims of the racial contract—then to dismiss it on these grounds means (again) to say that nonwhites must liberate themselves and forcefully affirm their own voices and their own social epistemologies. Nothing in Mills’s writing indicates that he would favor this option, though it is possible. I have nothing to say about it here, except that it is strongly intuitively unappealing. Whether that is because I have been raised a privileged white and still don a veil of ignorance despite myself, or instead because revolt is revolting on objective moral grounds (if there are such things), I cannot say.

The Racial Contract (Ithaca, NY: Cornell UP, 1997).

ii Ibid., 19–27.

by white typeset. (I don’t blame Mills for using the term “nonwhites,” though: it’s semantically elegant and captures with irony the Eurocentric sentiment of singling out the Others).

iv Ibid., 17–19.

v Ibid., 127.


Feminist Epistemologies, ed. Linda Alcoff and Elizabeth Potter (New York: Routledge, 1993), 15–48.

Race and Epistemologies of Ignorance, ed. Shannon Sullivan and Nancy Tuana (New York: SUNY Press, 2007), 39–57.

The Ethics of Assistance: Morality and the Distant Needy, ed. Deen K. Chatterjee (Cambridge UP, 2004): 260–288.

Women, Culture and Development: A Study of Human Capabilities, ed. Martha C. Nussbaum and Jonathan Glover (Oxford UP, 1995): 61–104.

Philosophy and Public Affairs 1, no. 3 (1972): 229–243.

Diogenes 44, no. 173 (1996): 5–15.

Food Policy: The Responsibility of the United States in the Life and Death Choices, ed. Peter G. Brown and Henry Shue (Free Press, 1977): 54–62.

A Theory of Justice: Revised Edition (Cambridge, MA: Belknap, 1999).

in fact not socially constructed, in large part or at all, and that particularism is simply false.

Handbook of Development Economics, Vol. I, ed. Hollis Chenery and T.N. Srinivasan (Elsevier, 1988): 9–26.

xvii Ibid.

xviii Nussbaum, “Human Capabilities.”