White Ignorance and Hermeneutical Injustice: A Comment on Medina and Fricker, Charles W. Mills

Author Information: Charles W. Mills, Northwestern University, c-mills@northwestern.edu

Mills, Charles W. “White Ignorance and Hermeneutical Injustice: A Comment on Medina and Fricker.” Social Epistemology Review and Reply Collective 3, no. 1 (2013): 38-43.

The PDF of the article gives specific page numbers. Shortlink: http://wp.me/p1Bfg0-1ce

Please refer to:

In my “White Ignorance” (Mills 2007), I welcomed the development within formal epistemology of social epistemology, and the advent of journals like Social Epistemology, while complaining that the authors in this new branch of epistemology seemed in general to be working with a concept of the social that excluded social oppression. So I should begin by saying how delighted I am to find my essay being discussed years later in none other than Social Epistemology (and the Review Collective) itself — and by two philosophers, Miranda Fricker and José Medina, whose recent books (Fricker 2007; Medina 2013) I see as exemplary challenges to this unfortunate pattern of exclusion. I would claim that in the same way that the “ideal theory” famous from John Rawls’s A Theory of Justice (1999) has oriented — or, in my view, mis-oriented — philosophical discussions of social justice, so its epistemic analogue has arguably mis-oriented philosophical discussions of social epistemology. 

Norms and Ideology

According to Rawls, the devising of norms and principles appropriate for an ideally just society is a prerequisite for developing the non-ideal theory necessary for tackling corrective justice in non-ideal societies. But as both Fricker and Medina point out in their respective book introductions, once one recognizes the extent of the distance from ideality of actual societies — that injustice has historically been the norm and justice the deviation rather than justice being the norm and injustice the deviation — the rationale for such a strategy is radically put into question. The conceptual array, moral and epistemic, necessary for mapping and prescribing for a world profoundly non-ideal is going to have to be very different, and ideal theory, far from being necessary for accomplishing such a task, as Rawls thought, may actually be an obstacle to it.

From this perspective, real-life societies are more illuminatingly to be conceptualized as systems of group domination than Rawlsian “cooperative ventures for mutual advantage” among equal individuals. What Ann Cudd (2006) calls “non-voluntary social groups” (e.g., classes, genders, races) will be central to their formation, and these groups will be hierarchically arranged in relations of domination and subordination. So, the social ontology is fundamentally different from that presupposed by mainstream ideal theory, whether political or epistemological. Abstracting away for simplicity’s sake from issues of intersectionality and intra-group heterogeneity, then, for any particular pair, the G1s are structurally dominant over the subordinated G2s, socio-politically, economically, and, in particular time periods, juridically. But for our purposes, the crucial issue is that this domination is also manifest cognitively, in belief systems, conceptual frameworks, and normative assumptions.

G1 ideology is pervasive and justifies or obfuscates (in different ways at different times) G1 domination. The G2s are cognitively influenced by G1 ideology, so that even though there is resistance and attempts at counter-hegemonic ideation, many or most G2s are to a significant extent (again, historically varying) under its sway. Moreover, because of their social subordination, G2s are likely to have less access to the tools of intellectual development, and may internalize the G1 view of them as inferior cognizers. So they will be disadvantaged in developing oppositional views. If we distinguish material benefit (the chance at wealth, status, and opportunities) from epistemic benefit (the chance at getting it right, factually and morally), we could say that the G1s are generally materially advantaged while in crucial respects (at least re seeing the social truth) epistemically handicapped, while the G2s are generally materially handicapped while in crucial respects (at least re seeing the social truth) epistemically advantaged (though this, as standpoint theorists emphasize, is a matter of the potential insight afforded from their social location, given the obstacles just cited).

Theorizing Race and Privilege

This picture should be familiar from classic Marxism with respect to class, and as it has subsequently been taken up, developed, and modified over the years by “radical” or “oppositional” political theory of other kinds with respect to the theorist’s particular selected groups (feminism, critical race theory, postcolonial theory, etc.), though admittedly the rise of post-structuralism (for example in Foucault) complicates things both in terms of alternative analyses of power and different views of truth. But set these complications aside. The point is that in my “White Ignorance” essay I was presupposing such a background picture, and trying to bring these issues and concerns into the social epistemology conversation. In the heyday of academic Marxism, now long past, class ideology was widely discussed; in feminist circles today, androcentrism is a familiar topic. But race, white racism, and white privilege have been under-theorized in philosophy in general and certainly in epistemology in particular. So I was hoping to get a debate going on these matters via an attempted “translation” into the epistemology literature of such critical race theory claims about patterns of social cognition.

In keeping with the left tradition, and unsurprisingly considering what has already been said, I find most persuasive the analysis of racism as an ideology, an excellent account of which has been provided by Tommie Shelby (2003). Shelby argues that we should see racism as a set of misleading views that distort social realities in ways that serve to create, perpetuate, and justify racial domination and unfair racial advantage. So his account is doxastic and social-structural, as against the volitional and individualist account of Jorge Garcia (1996), which makes racism a racially-based “ill-will” that is the product of the individual’s vicious heart. What are the implications for social cognition? Well, it means that in a racially-structured social order, race will doxastically and conceptually affect one in multiple ways. Growing up in a white-dominated society historically characterized by widespread white racism, white privilege, and racially segregated and differentiated experiences will have a negative epistemic impact on white cognizers through a variety of mechanisms, both crude and subtle, “hot” and “cold.” My essay was an attempt to disentangle and itemize some of them, bearing in mind their frequent fusion in empirical reality.

So where, given this background theoretical story, do I stand on the debate between Medina and Fricker on “hermeneutical injustice” and “white ignorance”? To begin with, I think in part they might simply be talking past each other because of their possible different drawings of the boundary lines around the latter.

Regarding Fricker and Medina

Start with Fricker (2013). As she emphasizes from the start, she is not at all denying that interested group ignorance is an injustice, but insisting that it is not a hermeneutical one (49). Hermeneutical injustices for her are, while structurally unjust, epistemically non-culpable, unlike wrongful epistemic practices like “wishful thinking, denial, self-interested selectiveness as regards the evidence” etc. (50). In the latter, the dynamic of privileged group interest is crucial: “White ignorance names a certain kind of individual or collective motivated cognitive bias . . . a form of collective denial in the white community about some uncomfortable truths” (50). Citing me, she distinguishes two cases, the racist cognizer and the cognizer affected more impersonally by social-structural causation. The individual white racist cognizer is certainly epistemically culpable, but this is not hermeneutical injustice for two reasons (i) by her definition, “the person who suffers hermeneutical injustice is not [epistemically culpable],” and (ii) the “hermeneutical resources” for comprehension of the situation are available, but the white racist is refusing to apply them (51). In the social-structural case, “the individual may not be at any epistemic fault,” so this prerequisite may be met. But are there adequate hermeneutical resources? If there are, the problem is doxastic rather than conceptual. But even if there are not, so the problem is “at the level of conceptual repertoire,” it still cannot, for Fricker, count as hermeneutical injustice, because “it is not whites who are broadly disadvantaged by the hermeneutical lacuna” but nonwhites (52). She concludes that white ignorance should not be understood as hermeneutical injustice, since we “would find it odd to insist that a group X that broadly benefits from some local hermeneutical impoverishment nonetheless thereby suffers an epistemic injustice” (53).

It should be obvious, then, that Fricker is assuming throughout that white ignorance is limited to whites, which, apart from the motivational and doxastic/conceptual wrinkles, is why the characterization of hermeneutical injustice would be inappropriate (since whites are the beneficiaries rather than the victims of their ignorance). But though my primary focus in the essay was indeed on whites, I also wrote under item five on a ten-point clarificatory list that:

Fifth, the “white” in “white ignorance” does not mean that it has to be confined to white people. Indeed . . . it will often be shared by nonwhites to a greater or lesser extent because of the power relations and patterns of ideological hegemony involved. (This is a familiar point from the Marxist and feminist traditions—working-class conservatives, “male-identified” women, endorsing right-wing and sexist ideologies against their interests.) Providing that the causal route is appropriate, [nonwhites] can manifest white ignorance also (Mills 2007, 22).

Had I just spoken of “white racist ideology” rather than “white ignorance,” this implication would have been obvious, since as emphasized at the start, the whole point of the standard analysis given by radicals of ideology is that it is not confined to the dominant group, the G1s, but influences the G2s also, and in so doing contributes to the reproduction of the system. But the relatively unfamiliar phrase “white ignorance” would not have these straightforward connotations, especially since I do not attempt to provide a parallel causal account, except gesturally, of the explanation for nonwhite acceptance of these beliefs/frameworks/norms, etc.

Now if Medina is extending at least part of “white ignorance” (as white racial ideology) to people of color, while Fricker is restricting it just to whites, then one source of their disagreement is immediately clear. Since whites are not materially, but only epistemically, handicapped by embracing white ignorance, Fricker does not want to characterize hermeneutical gaps resulting from white ignorance as an injustice to whites. But she would surely not disagree with Medina that insofar as people of color are both epistemically handicapped in trying to understand their situation, and, of course, materially handicapped by the situation itself, such harms, if we want to categorize them as resulting from white ignorance, are indeed an injustice to people of color. And since nonwhites are not generally epistemically culpable for accepting these beliefs and/or problematic conceptual frameworks, “white ignorance” of this kind could count as hermeneutical injustice.

It might seem odd to still be speaking of “white ignorance” in this context of nonwhite belief and conceptual patterns. But if we have been reflexively reading the “white” as including both the source of generation and the scope of ascription, we do not have to do so; we can instead take the adjective as just indicating the former and leave the question of its scope open. Consider an enslaved black in the United States who has accepted the beliefs that (i) Africans are devil-worshiping savages, and Western enslavement had the virtue of bringing them to Christianity and civilization; (ii) the desire to run away from slavery is an indication of mental disease, drapetomania; and (iii) if they are emancipated, blacks as a race will quickly die out, since they will no longer be able to benefit from the care of their kindly slaveowners. (These are all—believe it or not—actual historical examples, not made up by me.) We would have no hesitation in saying that such sincerely-held but delusional white beliefs, clear-cut examples of “white ignorance,” would if endorsed by blacks be straightforwardly examples of black “white ignorance.” And as such, they would clearly be epistemically and materially disastrous for blacks, a great injustice in their obstruction of a veridical understanding of slavery as an institution and the appropriate normative and political attitude towards it. So if the scope of “white ignorance” is the source of the dispute between Medina and Fricker, then all we need is the appropriate disambiguation. We distinguish white ignorance as confined just to whites and white ignorance as shared, at least in some respects, by people of color. In other words, ideological socialization by another name: the acceptance of G1 ideology by the G2s.

But the other possibility is that Medina is referring to the injustice done to people of color by whites in the grip of white ignorance when nonwhites who have at least partially freed themselves from such ideology are trying to reason with them. Here it is not a matter of nonwhites accepting white ignorance themselves, but of having to overcome it in their attempt to get a fair hearing for their views. The objective evidence is on the side of people of color, but whites are unable to appreciate the force of this evidence, or perhaps even to recognize it as evidence, because of the distortion of their cognitive processes by white ignorance. Thus Medina (2012) writes that:

The hermeneutical disadvantages inscribed in white ignorance are not only harmful, but wrongful, although the harm is committed against someone else: interestingly and crucially, the hermeneutical harms are wrongful for others, not for those upon whom the epistemic harms are directly inflicted. . . . In fact, in white ignorance the primary [hermeneutical] and secondary [psychological, economic, political] harms diverge so radically that those who are unable to make sense of part of their identity and experience—the white subjects—at the same time enjoy practical benefits and ways to hold on to their privileges thanks to their hermeneutical disadvantages. . . . The privileged white subjects’ inability to understand . . .  is part of a pattern of injustice not against them, but against those who suffer the consequences of white privilege (214).

So here it is the barrier created by white ignorance to white comprehension of the true nature of the social order, or some crucial local section of it, that constitutes the injustice for people of color. Medina is not saying that white ignorance is a hermeneutical injustice for whites, but for nonwhites trying to advance alternative views. As Rebecca Mason (2011) has pointed out, Fricker’s notion of “collective hermeneutical resources” blurs the distinction between (in my formulation, not hers) the resources of G1 ideology and the collective resources that would be available were it not for G1 domination and suppression of G2 subordinate counter-hegemonic alternatives, in part because Fricker underestimates the extent to which such counter-hegemonic G2 ideation is possible. But once such “subversive” cognition is conceded, the “conflictual” (as against “cooperative”) assumptions of non-ideal theory undercut “symmetrical” framings. While the social-structural prerequisites of hermeneutical injustice stipulated by Fricker must be differentiated from the clearly interest-based conceptions of injustices resulting directly from individual denial, bad faith, etc., it is still the case that the “structure” functions overall to reproduce G1 domination, and that although individual members of G1 are not responsible for its reproduction, G1s as a group are complicit with it. So insofar as white ignorance functions to suppress emancipatory alternatives, injustice is being done to people of color.

Again, let me say how delighted I am that these issues are at last being discussed in such circles.


Cudd, Ann E. Analyzing Oppression. New York: Oxford University Press, 2006.

Fricker, Miranda. Epistemic Injustice: Power and the Ethics of Knowing. New York: Oxford University Press, 2007.

Fricker, Miranda. “How is hermeneutical injustice related to ‘white ignorance’?” Social Epistemology Review and Reply Collective 2, no. 8 (2013): 49-53.

Garcia, J. L. A. “The Heart of Racism.” Journal of Social Philosophy 27, no. 1 (1996): 5-45.

Mason, Rebecca. “Two Kinds of Unknowing.” Hypatia 26, no. 2 (2011): 294-307.

Medina, José. “Hermeneutical Injustice and Polyphonic Contextualism: Social Silences  and Shared Hermenutical Responsibilities. Social Epistemology 26, no 2 (2012): 201-220.

Medina, José. The Epistemology of Resistance: Gender and Racial Oppression, Epistemic Injustice, and Resistant Imaginations. New York: Oxford University Press, 2013.

Mills, Charles. “White Ignorance.” In Race and epistemologies of ignorance, edited by Shannon Sullivan and Nancy Tuana, 11-38. Albany, NY: SUNY Press, 2007.

Rawls, John. A Theory of Justice. Revised edition. (Original edition, 1971.) Cambridge, MA: Harvard  University Press, 1999.

Shelby, Tommie. “Ideology, Racism, and Critical Social Theory. The Philosophical Forum 34, no. 2 (2003): 153-188.

Categories: Critical Replies

Tags: , , , ,

1 reply


  1. The MMIWG Report, Genocide and Epistemic Injustice – The Decolonized Librarian

Leave a Reply