Prejudice seems like a paradigmatic case of bad belief. But in what sense bad? Suppose we could agree that prejudiced beliefs tend to be false. That would be one way for them to bad: beliefs, we might think, should aim at the truth; false beliefs fall short in that important respect.
But prejudiced beliefs aren’t simply false, like the belief that the Great Wall of China is visible from space is false. They differ in that they also cause harm: they can be demeaning in their own right, but will also often cause discriminatory behavior further down the line. This would be another sense in which prejudiced beliefs could be bad, namely in terms of their moral consequences.
Nonetheless, when philosophers talk about beliefs in normative terms, they often have something else in mind, namely the question of how a particular belief stands with respect to its evidence. Philosophers say a belief is epistemically justified – i.e., “good” – if it is adequately supported by the evidence, otherwise “bad” (“unjustified,” “irrational,” etc.). Notice, right away, how this focus tends to push the question of truth to the margin. Maybe prejudiced beliefs tend to be false. But lots of false beliefs are nonetheless epistemically justified. The question of concern is whether these beliefs could be adequately supported by evidence, not whether they are true or false. For instance, most bygone scientific theories have turned out to be false. But this is no way settles the question of whether scientists of the era were justified in believing them on the basis of the evidence they had at the time.
It is common, in philosophical discourse no less than in everyday contexts, to encounter the supposition that prejudiced beliefs could never be epistemically justified in this sense. Prejudiced believers, then, are always doing something wrong, epistemically speaking. Perhaps they are generalizing on insufficient evidence or are disposed to retain their beliefs even in the face of strong evidence to the contrary.
In my recently published book Prejudice: A Study in Non-Ideal Epistemology (OUP 2021), I subject this assumption to critical scrutiny. To be sure, lots of prejudiced beliefs are probably unjustified. But this doesn’t make prejudice in any sense unique: much of our cognitive lives is only dubiously tethered in relevant evidence. Meanwhile, there’s nothing intrinsic to the concept of prejudice as such which entails that they couldn’t have relevant degrees of evidential support. In fact, this should be obvious when we take a closer look at the sorts of contexts in which people tend to form such beliefs. For instance, prejudice is often disseminated and reinforced through powerful information sharing networks and major social institutions (for instance, in the family, or via school curricula). Such “testimony” constitutes a form of evidence – and potentially a very strong form of evidence – that people can (and, in an important sense, must) draw on in forming their beliefs. Whether particular people holding particular prejudiced beliefs are epistemically justified is not something that we could settle on the basis of conceptual analysis alone. Instead, it requires a careful examination of the information environment they find themselves in. Some people are deeply unlucky to be born and socialized into information environments where prejudiced beliefs are peddled as baseline truths. We must ask ourselves, what else should they believe, given the information environment they happen to find themselves in?
Nonetheless, this analysis seems to leave the question of moral harm unaccounted for. I strongly suspect that one reason we are so drawn to simply classifying prejudiced beliefs as unjustified beliefs is that it would convenient. It would be convenient in the sense that it would allow us to find subjective fault all the way down whenever prejudice rears its ugly head: fault in forming the belief, fault in acting on the belief, fault for the harm that ensues.
This suspicion merits a closer look. It is often said that ignorance excuses. But ignorance excuses only if we are not at fault for our own ignorance. By arguing that some people might be perfectly justified in holding prejudiced beliefs, it would seem that I am opening the door for the claim that these people aren’t at fault for holding the beliefs that they do. In fact, these are, in some relevant sense, the beliefs they should hold given the evidential situation they are in. But if they are not at fault for holding these beliefs, it will be significantly harder to argue that they might be at fault for acting on these beliefs. And if so, it is hard to see why they should be at fault for the moral harm that their actions cause.
On the other hand, it’s no relief to victims of prejudice to learn that many of their victimizers might be perfectly justified in believing as they do, and therefore, presumably, blameless in acting on those beliefs. So there appears to be a clear tension between the perspective from which we might hope to vindicate victims’ rights and the perspective from which we are brought to recognize that their victimizers might not have been in a position to know any better. Allocating moral responsibility for harm is easy in cases where the wrongdoer’s action is based on an unjustified belief. But blameless ignorance – epistemically justified, false belief – complicates this picture significantly. In such a case, I may be in the wrong (in the sense that my belief is false). But I may have had no way of knowing that I was in the wrong: my evidence strongly supports the prejudiced belief. As such it seems unfair to blame me for any moral harm that my actions might cause: this would amount to blaming me for circumstances that are well outside my control.
Where does this leave the victims of prejudice? Are they just forced to accept their situation? No. The beginning of wisdom here is the realization that there is more to moral responsibility than blame. In many contexts, people can retain an important sense of moral responsibility for harms they have caused even though they were not at fault – not to blame – for causing them. These responsibilities may include offering restitution, compensation, or in any other way reversing the harms they have caused. I suspect that in many contexts, this is really the victim’s primary concern: to have things set right, to have opportunities unjustly denied restored.
Nonetheless, the impulse to blame prejudiced believers is strong. Sometimes, the urge to blame is entirely appropriate: they really should know better and their failure to believe as they should isn’t just an innocent mistake but an expression of their callous disregard for other people. But in other contexts, this impulse is misguided. It is misguided in two ways, at least. It is misguided, on the one hand, in failing to consider the possibility that prejudiced believers themselves might be subject to a certain kind of victimization, i.e., the victimization of being subject to systematic, socially and institutionally sanctioned misinformation, depriving them of the opportunity to know things that should matter to them.
But I believe it is misguided also in a second sense, one that speaks more directly to the victim’s perspective. The impulse to blame is misguided because it implicitly saddles victims of prejudice seeking redress with a greater burden that their case should require. The claim that should be in focus is that they have been unfairly targeted and discriminated against. There is nothing in this claim that should burden them with also making the case that their victimizers are generally bad people, who were in a position to know better. Differently put: whether one has been wronged by discrimination is an objective matter, a question of what rights one has. What rights one has is not, in general, dependent on the epistemic situation of others. Accordingly, making the claim that one has been discriminated against should in no way depend on the question of whether one’s victimizers were blameworthy for the discrimination. The notion that prejudiced belief is by its very nature epistemically unjustified invites the supposition that discriminatory behavior must be relevantly faulty all the way down to its cognitive causes. Severing that connection frees victims of prejudiced belief to make their case on objective grounds alone.
In sum, we should recognize that moral responsibility is a multi-dimensional concept. In one dimension, it tracks moral blame. To assign moral blame is to pass judgment on a person’s character: it is plausible that this is appropriate only if there is also epistemic blame at the root of the case, i.e., if we can make plausible that the agent was in a position to know better. But in another dimension, moral responsibility tracks an obligation to set things right. These dimensions of moral responsibility can come apart: there is nothing in the obligation to provide compensation which requires that one also be morally blameworthy for wronging someone. This insight allows us to harmonize the two perspectives that we described above. We can acknowledge that prejudiced believers are often not in a position to know better, and might in fact be perfectly epistemically justified in believing as they do. But we can also maintain that this has no bearing on the question of whether the targets of prejudice have suffered unjust discrimination, and so, whether they are entitled to seek redress.
Picture Alexas_Fotos https://pixabay.com/photos/violence-against-women-1169348/