PicoBlog

People often do what they do because of prejudices they have. They make false stereotypical assumptions about a group because that assumption is prevalent in their society, and their behaviour concerning members of that group is based on those assumptions. One sort of thing they do on the basis of prejudice is to adjust how confident they are in a proposition in the light of what someone from the group in question has told them about that proposition. Here’s a variation on an example due to Mona Simion:

Directions. Cal is standing outside Glasgow Queen Street train station poring over a map and looking confused. Elspeth approaches him and asks where he’s going. He tells her he needs to get to Glasgow Central train station, and she points him in the right direction. Before she tells him this, he is 50% confident it’s to his right and 50% confident it’s to his left. After she tells him it’s to his right, he becomes 70% confident it’s to his right and 30% it’s to his left. However, given his background evidence, he should be about 95% confident it’s to his right on the basis of her testimony. His background evidence tells him that, regardless of gender, a person who approaches you outside a train station offering to give directions is right about 95% of the time. He is less confident than he should be because of a prejudice that women are, on average, not as competent at giving directions as they in fact are, on average; and that they are, on average, less competent than men. If a man had approached him and offered the same directions, he’d have become 95% confident the station was to his right on the basis of that, which matches his evidence.

Hearing this story, we might agree with Miranda Fricker and others that Cal has wronged Elspeth in some way, and indeed that this wrong amounts to an injustice. Fricker calls the specific form of injustice she thinks Cal has committed a testimonial injustice. In this post, I’m interested in what sort of wrong Cal has done and how it might amount to an injustice.

There are some pretty clear ways in which what Cal does might wrong Elspeth because it foreseeably harms her or others. Seeing that Cal doesn’t take up her testimony as strongly as he should might diminish Elspeth’s self-confidence; if other women see Cal act like this, it might diminish their self-confidence, or further entrench their feeling that society doesn’t value them; and if men see it and they are already somewhat prejudiced in favour of deferring to a man’s judgment over a woman’s, this might lead them to strengthen that prejudice; and so on. But these are possible contingent consequences of Cal’s epistemic act. Fricker and others think that, over and above these, and whether or not they result from Cal’s actions, the act itself is intrinsically wrong; that is, it is wrong regardless of its consequences.

I want to think through three accounts of this intrinsic wrong: (i) Fricker’s own, developed further by Aidan McGlynn, which is based on Martha Nussbaum’s account of sexual objectification; (ii) an alternative proposed by Gaile Pohlhaus, Emmalon Davis, and Matthew Congdon, which is based on Simone de Beauvoir’s and Frantz Fanon’s conception of recognition and the subject/other distinction; and (iii) an analogy with Rachel Fraser’s proposal, made in the context of what she calls aesthetic injustice, which draws on Elizabeth Anderson’s theory of relational egalitarianism.

My worry about the first two is that, in some sense, they’re too heavy duty to explain the wrongs we find in some of the most everyday and banal instances of testimonial injustice, such as Cal’s injustice towards Elspeth. My worries about the third are less clearly formulated, largely because I think something in the vicinity must be right, but I’ll try to pin down what concerns me. In a number of places, I’ll draw on the sort of case I sketched in last week’s post: as Aidan McGlynn brought to my attention, it turns out that Katherine Hawley explores pretty much the exact same thought in her trust-based treatment of testimonial injustice, and Aidan himself has explained how the basic idea works without appeal to trust, which comes very close to the version I gave.

So let’s start with epistemic objectification. Fricker begins with a distinction that Edward Craig draws between a source of information and an informant:

Broadly speaking, informants are epistemic agents who convey information, whereas sources of information are states of affairs from which the inquirer may be in a position to glean information. Thus, while objects can only be sources of information, people can be either informants (as when someone tells one something one wants to know) or sources of information (as when the fact that one’s guest arrives bedraggled and shaking her umbrella may allow one to infer it has been raining). (Epistemic Injustice, p. 132)

According to Fricker, you objectify someone epistemically if you only ever treat them as a source of information, and never as an informant, and are disposed to so treat them in any interaction. And doing this is the source of the injustice in a testimonial injustice.

The problem with this is that, while it might account for some of the cases with which Fricker introduces the notion of testimonial injustice, such as when Herbert Greenspan dismisses Marge Sherwood’s suspicions about Tom Ripley as ‘female intuition’ in The Talented Mr Ripley, it just doesn’t account for the case of Cal and Elspeth in Glasgow. Cal does not deny Elspeth any epistemic agency. Indeed, he treats her as someone who is a reasonably competent epistemic agent, able to investigate the world and come up with reasonably reliable judgments. Being 70% reliable about something is pretty good—if Elspeth were genuinely that reliable, rather than Cal just thinking she is, we wouldn’t say she lacked epistemic agency, just that she was an epistemic agent who could improve a bit. But nonetheless Cal commits a testimonial injustice, for he treats her as less reliable than his evidence suggests, less reliable than he’d take a man to be in exactly the same situation, and he does so because of a prejudice against women as testifiers about this topic. Sure, it’s less of an testimonial injustice than if he’d ignored her evidence completely and remained 50-50 between the two directions, which is close to what Greenspan does to Sherwood, but it’s a testimonial injustice all the same. The point is that prejudices sometimes lead us to completely dismiss a person’s judgment and deny them any epistemic agency, as in Fricker’s example from The Talented Mr Ripley, but very often they lead us only to set our confidence in the truth of what the person tells us lower than our evidence warrants, and doing that denies no-one any epistemic agency.

To drive the point home, consider the sort of example I gave in my last post, and that Katherine Hawley and Aidan McGlynn have previously discussed. Given a body of evidence and a proposition, sometimes the evidence picks out a unique credence you should have in that proposition, but sometimes it doesn’t—or so say epistemic permissivists, and I’ll assume they’re right here. So suppose that, in fact, Cal’s evidence gives him a bit of leeway when he hears Elspeth’s testimony. It warrants a credence between 90% and 98%, but any credence in the range is permissible. And now suppose that, when Elspeth points him towards Glasgow Central, he becomes 91% confident she’s right whereas, if it had been a man pointing him that way, he’d have been 97% confident. Then, in that case, I think we would say that Cal has committed a testimonial injustice, but interestingly there is no credibility deficit and no credibility excess. Everyone receives a degree of credibility that they in some way deserve: they deserve something that falls in the range, and they receive that. But prejudice leads Elspeth to receive less than a man would and in a way that we can suppose would be systematic in Cal’s epistemic behaviour. In this case, it’s even less plausible to say that Elspeth has been denied any of her epistemic agency: she’s been treated fully in line with what Cal’s evidence permits.

Now, as Aidan McGlynn notes, Fricker picks up on two of the many varieties of objectification that Nussbaum catalogues (instrumentality and inertness), but there are five others (denial of autonomy, fungibility, violability, ownership, and denial of subjectivity), and perhaps they provide us with the resources to understand the original case of Cal and Elspeth, and the variation, as instances of epistemic objectification—after all, McGlynn thinks that fungibility might cover Emmalon Davis’s cases of credibility excesses given to members of a group that is falsely stereotyped as having greater than average knowledge of an area. But I don’t think these will help. Cal doesn’t deny Elspeth any epistemic autonomy; he might treat her as epistemically fungible, in the sense that he’d have treated any other woman who approached him outside the station in the same way, but everything he does is compatible with him granting her epistemic individuality in many other interactions, and he’d also treat any man who approached him as fungible in that sense, because he’d treat that particular man as he’d treat any other man (barring racial, class, and other prejudices); he doesn’t treat her as violable or apt to be owned, and nor does he treat her as someone whose subjective experiences don’t matter and can be ignored. In the original case, he simply treats her as less competent than he should, though still quite competent; and in the variation in which there’s a range of permissible credences, he simply gives less credence to her testimony than to a man’s, though in both cases he gives a permissible level of credence. So, whatever Cal does wrong to Elspeth, he doesn’t epistemically objectify her.

Does he, perhaps, treat her as an epistemic other, in the sense to which Pohlhaus appeals to understand the wrong of testimonial injustice? Here is how she explains the idea:

the sole purpose of the epistemic other is to provide epistemic support for navigating the experienced world of those deemed subjects. In this relation, those persons treated as “other” serve to recognize and maintain epistemic practices that make sense of the world as experienced from dominant subjectivities, but do not receive the same epistemic support with regard to their distinct lived experiences in the world. In the subject/other relation, recognition is monodirectional as opposed to intersubjective. This type of relation is not one in which objects are capable of participating; it is also a kind of relation that Beauvoir judges to be fundamentally unethical, since it denies a person’s full status as a free subject capable of experiencing and giving significance to the world uniquely. (‘Discerning the Primary Epistemic Harm in Cases of Testimonial Injustice,’ p. 106)

But again, I don’t think this can be right. Cal just doesn’t treat Elspeth in anything like the way described. He doesn’t deny Elspeth “full status as a free subject capable of experiencing and giving significance to the world uniquely”. In the original example, he just thinks of her as less competent than he should based on his evidence, but he still thinks of her as pretty competent and indeed competent to a degree that, were she truly just that competent, we’d have no hestitation in counting her as an epistemic subject, and so Cal believing it of her doesn’t amount to treating her as other. And in the permissivist’s variation of the case, he doesn’t even treat her as less competent than his evidence suggests.

Given the emphasis that Fricker places on credibility deficits—cases in which you treat someone as less credible than you should—and the fact that attempts to broaden the concept have largely involved drawing attention to credibility excesses, as Emmalon Davis and Jennifer Lackey do, what is striking about the permissive version of the example of Cal and Elspeth is that it involves neither deficits nor excesses. What it involves instead is inequality. For that reason, a natural place to look for an account of the source of the injustice in that case is Rachel Fraser’s appeal, in her paper on aesthetic injustice, to Elizabeth Anderson’s relational egalitarianism. Indeed, in a comment on my previous post, Rachel suggests exactly this. So let’s turn to that next. I think I’ll only have the space to consider testimonial injustice here, but I hope to return to aesthetic injustice in the future, and I wholeheartedly recommend the paper.

Here’s the crucial idea, which I quote from Rachel’s comment on last week’s post:

One agent does another a relational injustice when she treats her as an inferior. […] A treats B as an inferior when A’s conduct is guided by norms which play a role in marking or constituting members of A’s social group as the social superiors of B’s social group.

So Cal does Elspeth a relational injustice by treating her as an inferior, and he treats her as an inferior when he is guided by a prejudiced and false stereotype about women’s competence at giving directions that plays a role in marking or constituting a hierarchy in which women are socially inferior to men.

This proposal seems very plausible to me. For one thing, it helps us to understand what’s going on in the permissive version of the example with Cal and Elspeth. It doesn’t matter than Cal would give permissible credences to men and to women; what matters is that he routinely gives more credence to men than to women and does so because of a prejudice that plays the sort of role in marking or constituting social hierarchies that Fraser describes. What’s more, it allows us to separate cases in which the stereotype from which the unequal treatment derives is widespread or pernicious from cases in which it is not. If Elspeth tells Cal not the direction to Glasgow Central, but which day of the week it is, and if Cal has somehow got hold of the idiosyncratic prejudice that women are less reliable about that sort of information and downgrades his credence on that basis, no injustice occurs, since that stereotype plays no role in marking or constituting women as inferiors, not least because it’s just not widespread enough to do so, perhaps limited to Cal alone. The prejudice that women are poor at giving directions, while perhaps not so weighty on its own as to mark as women inferior to men, is part of a whole network of prejudices about women’s competence in certain sorts of domain that do have the weight to do that.

There’s a natural worry, which Fraser mentions briefly in the paper. Suppose that, in our example, Cal were a woman, not a man. Then no testimonial injustice would be committed, because in that case Cal does not belong to a group that sits higher in the social hierarchy than Elspeth—they sit in the same group. In a footnote, Fraser says:

In such a case, A does not treat B as her inferior. But she does treat B as someone’s inferior, because her conduct – given that it is guided by hierarchical norms and assumptions – is expressive of a hierarchy. In my view, this also makes for a violation of egalitarian justice. (‘Aesthetic Injustice’)

So what matters is that there is a social hierarchy and the injustice arises because someone, wherever they sit within this hierarchy, does something that is guided by its norms, and this constitutes an injustice because doing something guided by its norms expresses the hierarchy.

I want to make what seems initially to be a nitpicky point. Maybe it also seems merely a nitpicky point on closer inspection as well! But it helps me to think about the question. Think about another variation on the example of Cal and Elspeth. In this case, as Elspeth approaches, Cal notices she’s wearing a badge that signals that she’s en route to a Scottish mountaineering event. Given this evidence, which strongly suggests that she has a sense of direction far better than the average, when she points him in a particular direction, Cal should become 99% confident that it is the correct direction, but he becomes only 90% confident that it is, as before because of prejudice about women’s average competence—he believes that there are women with higher than average competence and in particular that mountaineering women have higher average competence; but, when we restrict attention to mountaineers, among whose number Cal now counts Elspeth, he thinks women have lower competence on average than men. Again, he commits a testimonial injustice. But in the colloquial sense, it doesn’t seem right to say that Cal treats Elspeth as an inferior. Let’s say that Cal has a pretty average sense of direction and is well aware of this. Even though he downgrades Elspeth’s sense of direction because she’s a woman, he still considers it much better than his own because she’s a mountaineering woman. In the relevant domain, he treats her not as an inferior but as a superior. And yet he nonetheless commits a testimonial injustice against her.

In some ways, this is irrelevant to Fraser’s point, since she is not appealing to the colloquial sense of treating someone as inferior. On her more technical account, A treats B as inferior if A is guided by a norm that marks or constitutes a social hierarchy that ranks the group to which A belongs above the group to which B belongs. So the treatment is indirect in two ways: first, the action is bad because it is guided by a norm that marks or constitutes a hierarchy that is bad, and thereby expresses that hierarchy; and second, the bad hierarchy places the whole group to which A belongs higher than the whole group to which B belongs, though not necessarily ranking A higher than B. And in this technical sense, Cal does treat Elspeth as an inferior in the example just described.

So we’ve got a case in which a word might be used in two sense. This doesn’t seem like a big deal, particularly since Fraser is very clear about her usage at all times in the paper. But I think it does make a difference when we evaluate the claim that one person does another an injustice by treating her as their inferior. That might sound plausible because we hear it in the colloquial sense, and it might seem less obviously true if we hear it in Fraser’s technical sense. And it might sound explanatory when we hear it in the colloquial sense, and it might seem less so if we hear it in the technical sense.

In fact, I’m inclined to agree with Fraser that an injustice nearly always occurs when A’s treatment of B is guided by a norm that marks or constitutes a social hierarchy in which the group to which B belongs is treated as socially inferior (I’ll sketch a case below in which it seems either A doesn’t commit an injustice or doesn’t commit one against B, but I’m not certain of it). My worry is rather this doesn’t give us all we want from an account of the wrong of testimonial injustice. To see this, note that, in the epistemic case, to say that A’s treatment of B is guided by a norm that marks or constitutes a social hiearchy in which the group to which B belongs is treated as inferior is often just to say that A forms their beliefs on the basis of B’s testimony in a way that is guided by prejudice against B’s competence in the given area of knowledge and inquiry. That is, the norms that constitute the social hierarchy that are relevant here are just these prejudices concerning epistemic competence. Now it does help to see such cases as instances of a more general sort of injustice, and to connect it with social hierarchies—Fraser’s account does substantial explanatory work, and gives principled ways of drawing distinctions. But when we ask for the wrong that testimonial injustice does, I think we ask for something more than this. For instance, we want to know why doing something that merely expresses a hierarchy counts as an injustice, even if that expression has no downstream harmful causal effects on anyone; even if it doesn’t perpetuate the hierarchy, or endorse it, or bolster it. And we want to know why something that expresses a hierarchy between whole groups of people counts as an injustice against one of the members, even if, in the specific case, the perpetrator of the injustice actually treats the victim as superior to them in the relevant competence, as we see in the second variation on the case of Cal and Elspeth the mountaineer described above. So, when we’re looking for the source of the injustice that Cal commits in these various cases, I think we need to know more than that his beliefs were guided by prejudices that are intimately related to the oppression or social standing of women. This is close to Fricker’s definition of testimonial injustice, and what we want is an explanation of why forming beliefs in that way counts as an injustice.

What’s clear is that the existence of the hierarchy that Fraser describes is unjust. So it is clear that there is an injustice collectively perpetrated by those who express, endorse, bolster, or perpetuate that hierarchy. The question is whether there is always an injustice individually perpetrated by someone who doesn’t bolster or perpetuate it, but does express it by acting in ways that are guided by the norms that constitute it. And I’m not sure whether the relational egalitarian has quite yet answered that.

This is already a long post, but let me finish by describing a sort of case, similar to that of Elspeth the mountaineer, that pushes me towards thinking that these relational injustices are really done to the whole group, rather than the person directly affected by the behaviour. Prejudiced people nearly always allow themselves an exception to their prejudice. Some of the prejudices against gay men are that they are irresponsible, shallow, and flippant (alongside the much darker assumptions that seem to be returning to public discourse with alarming force). But if you dress a particular way and live your private life in a particular way, you’ll often be treated as an exception to these stereotypes (though not always the darker ones). Now suppose that B is a gay man, and A, knowing this but observing the way he dresses and knowing the way he lives his private life, behaves towards him in a way that treats him as an exception to the standard prejudices that A has: A treats B as responsible, thoughtful, and morally serious, whereas they’d treat a gay man who didn’t dress this way as irresponsible, shallow, and flippant. Indeed, A treats B as they’d treat most straight people. However, their actions are guided by a norm that marks or constitutes a social hierarchy in which gay men are treated as socially inferior. The norm that guides them encodes the prejudices as well as the specified exemptions. What are we to say about such a case? I’m not sure, but I find it harder to think that B himself suffers an injustice in it, while also remaining convinced by Fraser’s claim that an injustice occurred. And I wonder whether this is the general case: it is the whole social group ranked as inferior in the hierarchy who suffer the injustice, not the individual.

ncG1vNJzZmiqmZi1or7DqZytrJmcv6bDjaysm6uklrCsesKopGioX6m1pnnWq6ann12ks27AxKyroqWfo7aiuIyipaOto6m2pLHS