Miranda Fricker’s Epistemic Injustice: Power and the Ethics of Knowing (2007) has inspired an incredible amount of work, both with philosophy and more broadly. Some of this work is more theoretical in nature: for example, trying to refine our understanding of the notion of epistemic injustice itself, and the two types Fricker identified, testimonial injustice and hermeneutical injustice; to identify further types; to point out oversights or problems stemming from Fricker’s characterisations and theories, and to explore alternative frameworks that might supplement or supersede Fricker’s; to figure out how to integrate theories of epistemic injustice with our broader theories of knowledge, testimony, understanding, communication, harm, injustice; and so on. Other work on epistemic injustice—including a lot of the research done by or with other disciplines—has focused more on how theories of epistemic injustice might be fruitfully applied to particular examples and social groups. However, this oversimplifies things, and discussions of epistemic injustice should—and frequently do—involve an interplay between more theoretical considerations and the details of real examples of the phenomena being theorised about…[please read below the rest of the article].

Image credit: Justin Chua via Flickr / Creative Commons

This is the spirit in which I engage with Dina Lupin Townsend and Leo Townsend’s paper (2021). It sits closer to the applied side of things, bringing to bear both of Fricker’s categories of epistemic injustice—testimonial injustice and hermeneutical injustice—in order to try to illuminate some of the epistemic challenges faced by Indigenous peoples and how those epistemic challenges contribute to the broader injustices and inequalities that Indigenous peoples face. However, I think that Townsend and Townsend may have misdiagnosed some of the features of the examples of this they discuss, and I’ll suggest an alternative account that draws on work by Quill Kukla (2014) rather than Fricker. Like Townsend and Townsend, I think that these issues about which theoretical framework to bring to bear matter primarily because they inform our consideration of possible remedies—about what appropriate, effective, and feasible steps towards epistemic justice might look like—and so I’ll turn to this question at the end.

I’ll focus on Townsend and Townsend’s discussion of testimonial injustice. I find their discussion of the ways in which framing Indigenous people’s relationships to land in terms of ownership can be a kind of hermeneutical injustice plausible and important. In particular, their suggestion that there’s a kind of injustice involved in having to use inapt or alien concepts and expressions in order to participate in certain institutional processes, particularly legal ones, seems not just like an insight about the particular examples they discuss, but a point that’s very likely to have much wider application. So in ignoring this aspect of Townsend and Townsend’s paper, I’m not being dismissive of it; on the contrary, it’s just not where I found myself with things to say.

‘Extreme’ Testimonial Injustice

Here I’ll examine Townsend and Townsend’s contention that indigenous people can face an ‘extreme’ or ‘radical’ form of testimonial injustice when testifying about the impact that industrial activity and resource extraction will have on them and their environment. Townsend and Townsend (2021) illustrate this claim with an example, the testimony of the Kichwa People of Sarayaku—henceforth the Sarayaku people[1]—to the Inter-American Court and Commission on Human Rights, where they were belatedly consulted on oil exploration activities on their traditional territory. To express and explain their opposition to drilling under the ground to look for oil, a spokesperson for the Sarayaku people said things like: “Beneath the ground, ucupacha, there are people living as they do here. There are beautiful towns down there, and there are trees, lakes, and mountains” (149). Townsend and Townsend argue that such claims were received as expressions of a ‘worldview’ or ‘cultural identity’, rather than straightforward claims to be taken as ‘bearing on the nature of the environment and the impacts of the drilling on that environment’ (149). Moreover, they contend that this is a form of testimonial injustice, where a speaker receives less credibility when testifying than they ought to due to prejudices held by their audience (Fricker 2007), and that in fact it’s an instance of a particular ‘radical’ or ‘extreme’ form of testimonial injustice in which the audience is so thoroughly prejudiced that the speaker isn’t even heard as testifying at all.

It’s this latter claim that I want to consider: whether Townsend and Townsend’s examples are ones in which the Sarayaku people were subject to a radical or extreme form of testimonial injustice. Here’s what they mean by describing a testimonial injustice as ‘extreme’:

In cases of ‘extreme’ testimonial injustice, the speaker is perceived as so utterly lacking in credibility that her testimony is given no epistemic consideration at all. So it is not that her testimony is heard, understood and dismissed because of ‘deflated’ credibility, but rather that there is no ‘genuine credibility judgment in respect of the speaker’s utterance . . . the [hearer] never really hears the [speaker] at all – her utterance simply fails to register with his testimonial sensibility’ (Fricker 2007, 140). This kind of extreme testimonial injustice deserves to be called ‘silencing’, according to Fricker, because although the speaker is not really prevented from speaking, she may as well be. As far as the hearer’s response goes, it is as though she (the speaker) has not spoken’ (150).

Elsewhere (McGlynn 2019), I’ve tried to illustrate what this involves by expanding on a parallel Fricker (2007, 71) draws to moral cases. To put the point in somewhat contentious terms, sometimes an ideology depicts a group in ways that are so dehumanising that members of that group aren’t seen as making any moral claims on those in its thrall. For example, people who are seen as ‘savages’ (e.g. Mills 2007, 27) or ‘super-predators’ (e.g. Curry 2017, chapter 3) may not merely be perceived as having making less weighty moral claims on others; rather, their pains, their plights, their suffering, the injustices done to them – those simply don’t morally register at all. But what does this mean? What Fricker has in mind is that registering on someone’s moral sensibility involves seeing them in ways which are inherently motivating and reason-giving; for example, seeing someone being treated in a morally wrong or unjust manner should motivate me and give me reasons to respond in certain ways (even if these are ultimately outweighed by counter-considerations, or if my courage isn’t up to it and so I fail to respond as I ought to). If a person or a group fails to register with my moral sensibility, my perception of them lacks any of this motivation and reason-giving moral force. As Fricker puts it, I don’t see those people in ‘moral colour’.

The epistemic case is to be understood along the same lines (or so Fricker proposes). If someone registers on my testimonial sensibility that should motivate me and give me reasons to accept what they say, or at least to weigh it in the balance when deciding what to believe, though those reasons are defeasible and I may fail to respond to them correctly. If a person or a group fails to register with my testimonial sensibility, all that is missing; their testimony is not perceived as motivating or reason-giving at all. Here’s how I previously put this:

[S]eeing a person in epistemic colour involves seeing them as more or less credible, where this involves a perceptual judgment which motivates me and gives me reason to respond doxastically in certain ways: believing what they say, forming a particular degree of credence or suspending judgment, and so on. When what someone says fails to register with their audience’s testimonial sensibility, their audience makes no judgment about the speaker’s credibility at all, not even an unfavourable one. In such cases, the motivational and reason-giving force of such credibility judgments is simply absent. The speaker might as well not have uttered anything; in this sense she has been silenced (McGlynn 2019, 408).

As in the moral case, Fricker suggests that this happens when certain people and groups are depicted in sufficiently dehumanising ways by an ideology held by their audience (2007, 140). This is the basis of Fricker’s proposed account of how best to make sense of Catharine MacKinnon’s much discussed claim that pornography silences women; the rough idea is that pornography depicts women in sexual contexts in ways that are so dehumanising that when women try to refuse sex, they are subject to such an extreme form of testimonial injustice that they are effectively silenced.[2]

Now, I don’t want to deny that how Indigenous peoples are perceived may sometimes be shaped by an extremely dehumanising ideology; that’s to say, I don’t want to pretend that the ideology mentioned above, which depicts Indigenous peoples as subhuman ‘savages’, is entirely a thing of the past or that it lacks powerful contemporary descendants.[3] My point is rather that there’s nothing in Townsend and Townsend’s account of how the Sarayaku people’s testimony was received by the Court and Commission that suggests that this was a context in which the Sarayaku people were dehumanised to anything like the degree involved in extreme cases of testimonial injustice. For this reason, I’m sceptical of Townsend and Townsend’s claim that they have presented examples of such an extreme form of testimonial injustice, involving speakers simply failing to register with their audiences’ testimonial sensibility and so being silenced; I suspect that this is the wrong diagnosis.

Moreover, the reasons that Townsend and Townsend offer for thinking otherwise cite features of the examples that are, plausibly at least, features of any case of testimonial injustice. One point they make is that the Court ‘fails to entertain questions about the truth or falsity of [the Sarayaku people’s] testimony’, which they suggest ‘stems from a perceived utter lack of credibility in this domain’ (150). However, it’s not clear in what contrasting sense the audience might be said to entertain questions about the truth or falsity of the relevant testimony in standard, more familiar cases of testimonial injustice. Consider Fricker’s two principal examples; in The Talented Mr Ripley Marge Sherwood’s well-founded suspicions about Tom Ripley are dismissed by her would-be father-in-law as the product mere ‘female intuition’, completely detached from the facts (2007, 9), while Tom Robinson fails to be seen as remotely trustworthy by the jury when defending himself from Mayella Ewell’s rape charge in To Kill a Mockingbird (2007, 23-27). In neither case is the speaker’s word given any weight on the topic at issue, but there’s no suggestion that these should be treated as extreme cases of testimonial injustice involving the speakers failing to register with their audiences’ testimonial sensibility and thereby being silenced; rather these are the two paradigm examples Fricker offers of standard cases of testimonial injustice.

Sources of Information

The other point Townsend and Townsend make is that the Sarayaku people are treated as mere ‘sources of information’ rather than informants, and are thereby in a sense objectified. We treat people as sources of information all the time, and often this is completely appropriate and benign. For example, Fricker (2007, 132) discusses an example in which I infer that it’s raining from a person’s wet appearance as they enter the building; in doing so, I treat them and their damp state as a state of affairs which I can infer things from, given other background information I have (in this example, about the most likely explanation of why someone would look wet when coming inside). In treating the person as a source of information, Fricker suggests, I treat them as an object: but in this case benignly so. Testimonial injustice, Fricker proposes, involves treating someone as a mere source of information; that is, treating them as no more than a source of information, and so being disposed not to treat them as an informant whose testimony on the topic at hand is valuable and to be taken seriously. As Townsend and Townsend (2021) observe, Fricker thinks that this more thoroughgoing and harmful kind of objectification is the primary harm of testimonial injustice:

[T]he words of the community are seen by the Court as having an altogether different significance: they are used as evidence of what their ‘worldview’ is. In our view, this is a good example of the sort of epistemic objectification that Fricker sees as the primary harm in testimonial injustice. Instead of being treated as an informant capable of sharing knowledge, the community is treated instead as a mere ‘source of information’ from which knowledge – in this case, knowledge of the community’s beliefs or ‘worldview’ – can be gleaned’ (150, note omitted).

However, the primary harm of testimonial injustice is by definition an inherent, intrinsic harm; the distinction between primary and secondary harms is precisely between those which are intrinsic to an epistemic injustice of that kind, and those which are merely contingent—though perhaps typical or frequent—accompaniments. To accept that the primary harm of testimonial injustice is that a speaker is epistemically objectified is to take it to be a feature of every case of testimonial injustice, not merely extreme ones. So from within this framework, even if we accept Townsend and Townsend’s claim that the Sarayaku community and their spokespeople are treated as mere sources of information and are thereby epistemically objectified, this isn’t any evidence that this is an extreme form of testimonial injustice; indeed, the framework commits us to defending the general claim that this kind of objectification is a feature of every case of testimonial injustice.[4]

I want to suggest, then, that if there’s testimonial injustice here, it’s of a non-extreme sort. I suspect that the impulse to think otherwise comes in part from failing to recognise just how extreme Fricker’s extreme form of testimonial injustice really is, in particular the high degree of dehumanisation by an audience that is a prerequisite for it to occur. But I also think that Townsend and Townsend’s discussion conflates extreme testimonial injustice with a somewhat different and more common phenomenon, which Quill Kukla (writing as Rebecca Kukla) has explored and labelled ‘discursive injustice’. Here’s how Kukla characterises this form of injustice:

When members of any disadvantaged group face a systematic inability to produce certain kinds of speech acts that they ought, but for their social identity, to be able to produce—and in particular when their attempts result in their actually producing a different kind of speech act that further weakens or problematizes their social position—they we can say they suffer a discursive injustice (by analogy with “epistemic injustice,” which is a phrase that has recently gotten a lot of play from Miranda Fricker and others [Fricker 2007] (Kukla 2014, 441).

In an attached note, Kukla explicitly cites ‘offering testimony’ as one kind of speech act that a group might be systematically prevented from producing, and so they suggest that ‘this kind of epistemic injustice would be a subspecies of discursive injustice’ (2014, 456, note 2). One kind of example that Kukla discusses seems particular relevant for thinking about Townsend and Townsend’s cases. Kukla argues that frequently when women try to assert, their utterances aren’t taken as assertions at all, but rather as expressives—speech acts which serve to express an individual’s personal feelings rather than to make a claim about the world that could be true or false—and Kukla’s description of how expressives are received by an audience is striking reminiscent of Townsend and Townsend’s account of how the Court treated the testimony of the Sarayaku people: ‘Expressives have no generalized epistemic consequences for others. They can be tolerated or suppressed, but assessing how they fit in with an overall story about how things are is not appropriate’ (2014, 451).

Kukla contention is that women’s assertions are often treated this way, as merely expressing some personal response rather than as a contribution to a shared overall picture of how the world is. I think this offers a useful analogue for thinking about how to understand the injustice faced by the Sarayaku people on Townsend and Townsend’s analysis, though in the latter case what’s taken to be expressed is a not an individual’s personal reaction, but a ‘worldview’ or ‘cultural identity’ shared by the community. This is, I think, what Townsend and Townsend should argue—that the case of the Court’s treatment of the Sarayaku people’s testimony is an example of discursive injustice, one that’s parallel to some of Kukla’s own examples in some respects, but with some interesting and important distinctive features. Indeed, although this proposal isn’t mentioned at all in Townsend and Townsend’s paper, Leo Townsend has favourably mooted it elsewhere (Townsend 2020, 35).[5]

Communicative Justice

Part of the significance that Townsend and Townsend attach to their conclusion that Indigenous peoples can be subject to an extreme of testimonial injustice is that they note that some of the remedies for testimonial injustice that have been proposed by Fricker and others seem completely inadequate as a response to radical testimonial injustices of this kind. For example, Fricker advocates the cultivation of ‘the virtue of testimonial injustice’, which involves hearers learning to consciously and unconsciously compensate for their initially prejudiced dismissal of what a speaker has said. Townsend and Townsend (2021) are, I think justly, sceptical that this will have any purchase on cases of extreme testimonial injustice:

[T]he virtue of testimonial injustice is meant to have an impact on the hearer’s credibility judgments, but radical testimonial injustice affects the hearer’s perception of the speaker prior to questions of credibility. In cases of radical testimonial injustice, the hearer does not even recognise the speaker’s speech for what is it – does not so much as recognise that the speaker is offering testimony – and so questions of the speaker’s credibility with respect to that testimony never arise (155).

If I’m right that the examples they discuss aren’t really extreme cases of testimonial injustice, then this point isn’t as relevant as it might first appear. Still, I share Townsend and Townsend’s sense that Fricker’s proposed virtue doesn’t seem to apply very cleanly to the kind of systematic misinterpretation that the Sarayaku people confront in their examples. Diagnosing this as an example of discursive injustice in Kukla’s sense, rather than as a standard case of testimonial injustice, helps make sense of this, and it also helps to motivate Townsend and Townsend’s proposal about the kind of virtue that they think we need to cultivate in order to avoid perpetrating the kind of injustice faced by the Sarayaku people, which they call the virtue of communicative justice:

Communicative justice means not allowing prejudicial views about a speaker, including views about the speaker’s authority in certain domains, to distort one’s perception of what a speaker is up to with her words – which communicative acts she means to perform. To have the virtue of communicative justice is to be open or attuned to what the speaker is doing (155).

This looks like a virtue that is close to tailor-made to address discursive injustice in Kukla’s sense; indeed, it would be fitting to call this the virtue of discursive justice.[6] So I suggest that framing the injustice done to the Sarayaku people in these terms rather than as an extreme form of testimonial injustice better motivates the kind of remedy that Townsend and Townsend themselves identify as what’s required.

References

Curry, Tommy J. 2017. The Man-Not: Race, Class, Genre, and the Dilemmas of Black Manhood. Philadelphia: Temple University Press.

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

Hornsby, Jennifer. 1993. “Speech Acts and Pornography.” Women’s Philosophical Review 10: 38-45.

Kukla, Quill Rebecca. 2014. “Performative Force, Convention, and Discursive Injustice.” Hypatia 29 (2): 440-457.

Langton, Rae. 1993. “Speech Acts and Unspeakable Acts.” Philosophy and Public Affairs 22 (4): 293-330.

Langton, Rae and Jennifer Hornsby. 1998. “Free Speech and Illocution.” Legal Theory 4: 21-37.

McGlynn, Aidan. Forthcoming. “Epistemic Objectification as the Primary Harm of Testimonial Injustice.” Episteme: 1-17.

McGlynn, Aidan. 2020. “Objects or Others? Epistemic Agency and the Primary Harm of Testimonial Injustice.” Ethical Theory and Moral Practice 23: 831-845.

McGlynn, Aidan. 2019. “Testimonial Injustice, Pornography, and Silencing.” Analytic Philosophy 60: 405-417.

Mills, Charles. 2007. “White Ignorance.” In Race and Epistemologies of Ignorance edited by Shannon Sullivan and Nancy Tuana,13-38. Buffalo: SUNY Press.

Townsend, Dina Lupin and Leo Townsend. 2021. “Epistemic Injustice and Indigenous Peoples in the Inter-American Human Rights System.” Social Epistemology 35 (2): 147-159.

Townsend, Leo. 2020. “Group Assertion and Group Silencing.” Language & Communication 70: 28-37.

Townsend, Leo and Dina Lupin Townsend. 2020. “Consultation, Consent, and the Silencing of Indigenous Communities.” Journal of Applied Philosophy 37 (5): 781-798.

Tsosie, Rebecca. 2017. “Indigenous Peoples, Anthropology, and the Legacy of Epistemic Injustice.’ In The Routledge Handbook of Epistemic Injustice edited by Ian James Kidd, José Medina, and Gaile Pohlhaus Jr., 356-369. Oxon: Routledge.

[1] I here follow Townsend and Townsend (2021, 149, note 5).

[2] I give a more careful presentation of Fricker’s proposal, and argue against it, in McGlynn 2019.

[3] For a brief overview of the history of this ideology, covering several of philosophy’s more inglorious moments, see Tsosie 2017, 357-359.

[4] For a start on this task, see McGlynn 2020 and McGlynn forthcoming.

[5] Kukla (2014) notes that discursive injustice, as they understand it, is similar to but distinct from the notion of illocutionary silencing found in Rae Langton’s and Jennifer Hornsby’s work (e.g. Langton 1993, Hornsby 1993, Langton and Hornsby 1998); in particular, illocutionary silencing is a matter of a speaker’s illocutionary intentions not being recognised by their audience, whereas discursive injustice is a matter of how a speech act ‘gets recognized and responded to in practice’ (453). Given this, classifying Townsend and Townsend’s examples as involving discursive injustice may undercut their motivation for treating them as involving illocutionary silencing (see Townsend and Townsend 2021, 156, notes 4 and 12 and Townsend and Townsend 2020), though it does so in a way that explains the appeal of that proposal. As noted in the text, Leo Townsend seems sympathetic to this (2020, 35), but in their joint work Townsend and Townsend seem to express a preference for instead employing the kind of Austinian framework that Langton and Hornsby appeal to (Townsend and Townsend 2020, 796, note 8).

[6] I add the qualifier ‘close to’ here because, as noted in the previous footnote, Kukla’s notion of discursive injustice is less focused on questions about the speaker’s intentions—’which communicative acts she means to perform’—than this characterisation of the relevant virtue is. This superficial mismatch could be easy addressed, though.

Categories: Critical Replies

Tags: Aidan McGlynn, Dina Lupin Townsend, epistemic injustice, extreme testimonial injustice, indigenous peoples, Kichwa People of Sarayaku, Leo Townsend, Miranda Fricker, Quill Rebecca Kukla, Sarayaku