What is a speaker owed?
2022; Wiley; Volume: 50; Issue: 3 Linguagem: Inglês
10.1111/papa.12219
ISSN1088-4963
Autores Tópico(s)Feminist Epistemology and Gender Studies
ResumoSuppose someone tells you something. Under what conditions do you owe it to her to accept what she's said? I will call this The Question. The Question raises an issue of philosophical significance. For one thing, the issue appears to be at the heart of #BelieveWomen, that part of the #MeToo movement focusing on the proper way to respond to allegations of sexual harassment. Do we owe it to women who tell us that they have been assaulted to believe them? How should we accommodate the fact that, rare as these examples are,11 According to Claire E. Ferguson and John M. Malouff, “Assessing Police Classifications of Sexual Assault Reports: A Meta-Analysis of False Reporting Rates,” Archives of Sexual Behavior 45, no. 5 (2016): 1185–93, of all reports of sexual harassment, only 5 percent are such that either they are false or there is not enough evidence to support a criminal charge. not everyone who reports having been assaulted does so truthfully? These are important questions; addressing them requires taking up The Question. But the importance of The Question transcends #MeToo and #BelieveWomen. Speakers feel a characteristic sting when their word is repudiated. Elizabeth Anscombe notes that it can be “an insult or an injury” not to be believed.22 Elizabeth Anscombe, “What is it to Believe Someone?” in Rationality and Religious Belief, ed. C. F. Delaney (South Bend: University of Notre Dame Press: 1979), 9. Miranda Fricker argues that certain ways of disbelieving another person—namely, those manifesting a credibility deficit driven by identity-prejudice—are tantamount to committing a “testimonial injustice” against her.33 Miranda Fricker, Epistemic Injustice: Power and the Ethics of Knowing (Oxford: Oxford University Press, 2007), 1. Other writers, too, recognize that, at least when it is undeserved, rejecting a speaker's say-so “slights” or “abuses”44 Both terms are used by Ted Hinchman, “Telling as Inviting to Trust,”Philosophy and Phenomenological Research 70, no. 3 (2005): 565 & 568. her or constitutes an “offense”55 Richard Moran, “Getting Told and Being Believed,” in The Epistemology of Testimony, eds. Jennifer Lackey and Ernest Sosa (Oxford: Oxford University Press: 2006), 301. or a “special insult”66 Allan Hazlett, “On the special insult of refusing testimony,” Philosophical Explorations 20, no. 1 (2017): S39. against her. Each of the forgoing authors would appear to be committed to the idea that an audience can owe it to the speaker to accept what she's said—on pain of committing such a slight or abuse or offense etc. against her. By addressing The Question we illuminate and characterize the scope of this phenomenon. Reflecting on The Question, however, can make us feel uneasy. Presumably the sense of “owing” that is in play, when we ask what an audience owes a speaker, is a matter of ethics or justice.77 There is also a line of argument on which relationships of intimacy, such as friendships, can generate obligations bearing on the acceptance of an another's say-so. The literature here is vast; see e.g. Simon Keller, “Belief for Someone Else's Sake,” Philosophical Topics 46, no. 1 (2018): 19–36; Simon Keller, “Friendship and Belief,” Philosophical Papers 33, no. 3 (2004): 329–51; Sarah Stroud, “Epistemic Partiality in Friendship,” Ethics 116, no. 3 (2006): 498–524; Allan Hazlett, A Luxury of the Understanding: On the Value of True Belief (Oxford: Oxford University Press, 2013); Sarah Paul and Jennifer Morton, “Believing in Others,” Philosophical Topics 46, no. 1 (2018): 75–96; and Berislav Marušiç and Steven White, “How Can Beliefs Wrong?—A Strawsonian Epistemology,” Philosophical Topics 46, no. 1 (2018): 97–114. For criticism see Jason Kawall, “Friendship and epistemic norms,” Philosophical Studies 165, no. 2 (2013): 349–70; Katherine Hawley, “Partiality and Prejudice in Trusting,” Synthese 191, no. 9 (2014): 2029–45; Nomy Arpaly and Anna Brinkerhoff, “Why Epistemic Partiality is Overrated,” Philosophical Topics 46, no. 1 (2018): 37–51; Sanford Goldberg, “Against Epistemic Partiality in Friendship: Value-Reflecting Reasons,” Philosophical Studies 176, no. 8 (2018): 2221–42; and Cathy Mason, “The epistemic demands of friendship: friendship as inherently knowledge-involving,” Synthese 199, no. 1 (2021): 2439–55. I will return, briefly, to this line of argument at the end of this paper. At the same time, The Question bears on a matter which at least traditionally has been thought to be the exclusive domain of epistemology: namely, when it is proper to accept the say-so of another. And this can make it seem as though The Question presupposes that there are ethical or justice-based constraints on the epistemic standards for proper acceptance. Insofar as the allegation of such constraints is highly controversial,88 To be sure, there are proponents of such constraints. The literature on doxastic wronging and on “moral encroachment” is full of proposals of this sort: see e.g., James Fritz, “Pragmatic Encroachment and Moral Encroachment,” Pacific Philosophical Quarterly 98, no. S1 (2017): 643–61; Rima Basu, “Can Beliefs Wrong?” Philosophical Topics 46, no. 1 (2018): 1–18; Rima Basu, “Radical Moral Encroachment: The Moral Stakes of Racist Beliefs,” Philosophical Issues 29, no. 1 (2019): 9–23; Sarah Moss, “Moral Encroachment,” Proceedings of the Aristotelian Society 118, no. 2 (2018): 177–205; Mark Schroeder, “When Beliefs Wrong,” Philosophical Topics 46, no. 1 (2018): 115–27; Rima Basu and Mark Schroeder, “Doxastic Wronging,” in Pragmatic Encroachment in Epistemology, eds. Brian Kim and Matthew McGrath (New York: Routledge, 2019); and Renée Jorgensen Bolinger, “Varieties of Moral Encroachment,” Philosophical Perspectives 34, no. 1 (2020): 5–26. But such proposals are highly controversial: see Endre Begby, “Doxastic Morality: A Moderately Skeptical Perspective,” Philosophical Topics 46, no. 1 (2018): 155–72; Georgi Gardiner, “Evidentialism and Moral Encroachment,” in Believing in Accordance with Evidence, ed. Kevin McCain (New York: Springer Publishing: 2018), 169–95; and Robert Carry Osborne, “What do we epistemically owe to each other? A reply to Basu,” Philosophical Studies 178, no. 3 (2021): 1005–22. I will return to this sort of argument at the end of this paper. to treat The Question as worthy of being addressed risks committing oneself at the outset to a highly controversial assumption. Given the importance of this topic, it is most curious that the perplexity raised by The Question has not received significant attention in the literature.99 But see Kimberly Ferzan, “#BelieveWomen and the Presumption of Innocence: Clarifying the Questions for Law and Life,” in Truth and Evidence: NOMOS LXIV, eds. Melissa Schwartzberg and Philip Kitcher (New York: NYU Press, 2021), 65–108; and Renée Bolinger, “#BelieveWomen and the Ethics of Belief,” in Truth and Evidence: NOMOS LXIV, eds. Melissa Schwartzberg and Philip Kitcher (New York: NYU Press, 2021), 109–44, for two excellent recent discussions. I speculate that this is because there is a simple strategy by which it appears that one can avoid such controversy. The strategy involves articulating what ethics or justice demands—what an audience owes a speaker—in distinctly epistemic terms. On such a view, we owe it to a speaker (morally speaking) to accept her say-so when and only when doing so is sanctioned by the standards of epistemology. Interestingly, insofar as she addresses the matter at all, Fricker herself appears to follow this sort of strategy. She represents the audience's “obligation” to the speaker as that of “match[ing] the level of credibility she [the audience] attributes to her interlocutor to the evidence that he [the speaker] is offering the truth.” 1010 Fricker, Epistemic Injustice, 19 (italics added). If such a proposal could be made to work, there would be no need to worry about ethics or justice impinging on epistemic standards; on the contrary, ethics or justice would then have a need to appeal to those very standards to articulate what it is that ethics or justice demands of audiences.1111 Geoff Pynn and Kimberly Ferzan also defend positions in which the demand can be articulated in exclusively epistemic terms. See Geoff Pynn, “Epistemic Degradation and Testimonial Injustice,” in Applied Epistemology, ed. Jennifer Lackey (Oxford: Oxford University Press: 2021), 151–70; and Ferzan, “#BelieveWomen and the Presumption of Innocence.” I will discuss these views below. There is hope, then, of being able to answer The Question in ways that do not invite controversy. This is where we will start, with candidate answers that I will call epistemic answers to The Question. These candidate answers are epistemic in that, while the demand in question—what an audience owes to a speaker—is a demand of ethics or justice, the content of what is owed is articulated in exclusively epistemic terms. One of the key points I wish to make in this paper is that no epistemic answer can be made to work. Before proceeding to our topic, several preliminary comments are in order. First, the language of The Question—Under what conditions does an audience owe it to a speaker to accept what she's said?—uses “accept” rather than “believe.” I do so in order to avoid a dispute with those who would insist that belief is not in our control, and so is not the sort of thing that can be “owed” to another. Acceptance is in our control,1212 Indeed, this was part of the basis on which the notion was originally introduced into the literature. See e.g., L. Jonathan Cohen, “Belief and Acceptance,” Mind 98, no. 391 (1989): 367–38. and so even if we think that there is a control requirement on what can be owed to another, we can make good sense of The Question.1313 Richard Holton, “Deciding to Trust, Coming to Believe,” Australasian Journal of Philosophy 72, no. 1 (1994): 63–76 is relevant to this issue. Second, in addressing when an audience owes it to a speaker to accept what she's said, I will be assuming that we can address this question without any concern for what the speaker said, that is, without any concern for the content of the saying. Not everyone will agree: perhaps we owe it to accept what a speaker said only when what she said (or our acceptance of it) is of some significance—to our relationship, to her well-being, or what-have-you.1414 Many of the papers on partiality in friendship make precisely this point; see the references in footnote 8. There is a world of difference, after all, between cases in which a speaker reports having been the victim of sexual assault and those in which she reports the morning's weather. Still, one of the deep insights of Fricker is that to reject another's say-so out of identity-prejudice is to wrong them “in their capacity as a knower,” and this is so for whatever it is that they said.1515 Fricker, Epistemic Injustice, 5. Following Fricker, I will assume that it makes sense to ask The Question at this level of generality, without attending to content—recognizing that there might be further obligations depending on what was said, or to whom.1616 Compare Bolinger, “#BelieveWomen and the Ethics of Belief.” She distinguishes the question, What sort of entitlement does a speaker have to being believed?, from the sorts of obligations that might arise “in the context of closer relationships.” Third, many authors make a big deal of the distinction between accepting what another says and believing them.1717 See Anscombe, “What is it to Believe Someone?”; Edward Craig, Knowledge and the State of Nature: An Essay in Conceptual Synthesis (Oxford: Oxford University Press: 1990); Hinchman, “Telling as Inviting to Trust”; Moran, “Getting Told and Being Believed”; Paul Faulkner, “On Telling and Trusting,” Mind 116, no. 464 (2007): 875–902; Benjamin McMyler, “Obedience and Believing a Person,” Philosophical Investigations 39, no. 1 (2016): 58–77; and Bolinger, “#BelieveWomen and the Ethics of Belief.” For an interesting critical discussion, see Finlay Malcolm, “How to Insult and Compliment a Testifier,” Episteme 15, no. 1 (2018): 50–64. Believing another person, on this view, is an action (or an event) that is imbued with interpersonal normativity, whereas accepting what they have said is a humdrum affair that can be understood in the antiseptic terms of epistemology. For my purposes here, I am happy to grant this distinction, and—modulo my first clarificatory point, above—I am happy to treat The Question as the question when we owe it to another to believe them. Still, I will continue to speak of “acceptance” rather than “belief,” for the reason indicated above. Fourth, I am taking no account of whether a speaker who tells us something is entitled to our attention in the first place.1818 This is a question that deserves much more attention than it has received. But see Sanford Goldberg, Conversational Pressure: Normativity in Speech Exchanges (Oxford: Oxford University Press, 2020), chap. 2. It may be that there are cases in which he is not: consider the unwelcome verbiage of the garrulous seat-mate on the international flight, the colleague who suffers from logorrhea, the neighbor who indulges in salacious gossip, or the acquaintance who drones on about inconsequential matters. Perhaps we do not owe anything to them because they were not entitled to our attention in the first place. Perhaps. I do not want to take a stand on this. But those who do should read The Question as further qualified: given a speaker who tells us something under conditions in which she was entitled to our attention, when do we owe it her to accept what she's said? Those who hope to answer The Question without making any controversial commitment to ethical or justice-based constraints on proper acceptance will attempt to provide an epistemic answer. Such an answer characterizes the ethical or justice-based demands on an audience in exclusively epistemological terms. In this section I argue that no such answer can be correct: whether their focus is on the epistemic condition of the speaker, the audience, or both, epistemic answers face insurmountable problems. (KSP) When a speaker tells us something, we owe it to her to accept what she's said just in case she spoke from knowledge. Of course (KSP) is far from the only speaker-based epistemic answer. We might arrive at other, weaker variants by replacing talk of knowledge with talk of some other epistemic property, such as justified or rational belief. Alternatively, a stronger version would talk of epistemic certainty rather than knowledge. Happily, the differences between these variants are irrelevant to what I say below. While I will focus my attention on (KSP), my case will naturally apply to any speaker-based epistemic answer. Mirroring the knowledge rule of assertion, we might suppose that there is a knowledge rule of acceptance, that enjoins an audience to accept any assertion that is known by the speaker.2121 Pynn, “Epistemic Degradation and Testimonial Injustice,” 161. An initial difficulty that arises for Pynn's suggestion is that, whereas the knowledge rule of assertion is a rule regarding when one is permitted to assert (namely, only when one knows), the knowledge rule of acceptance states an obligation. As Sanford Goldberg and Renée Bolinger have argued, it is not clear how to get from the former to the latter.2323 Sanford Goldberg, “Anti-Reductionism and Expected Trust,” Pacific Philosophical Quarterly 100, no. 4 (2019): 952–70; Bolinger, “#BelieveWomen and the Ethics of Belief.” My brief suggestions above go some distance in that direction, but they would need to be filled in. A more serious problem facing speaker-based epistemic answers is that there are circumstances in which (i) one has excellent (albeit misleading) evidence that a speaker is untrustworthy; and (ii) it would violate no obligation one has to the speaker to decline to accept her say-so on the basis of that evidence. Since these conditions can be satisfied even in cases in which the speaker speaks from knowledge, (KSP) is false. Consider the familiar children's tale involving the boy who cried wolf. The boy has a well-known history of lying to the townspeople, claiming to see a wolf when he does not, merely for the sport of it. By the time he actually sees a wolf and reports it, no one believes him. Nor is there any sense to the suggestion that they owe it to him to believe him on this occasion. The tale's lesson seems to be that one can lose any entitlement to be believed—to have one's say-so accepted—if one abuses others' trust. This lesson falsifies (KSP): it is not true that we owe it to a speaker to accept what she's said just in case she speaks from knowledge.2424 Fricker, Epistemic Injustice, 42 also brings up the used car salesperson: even when he speaks from knowledge, he is not wronged when we do not believe what he tells us about the used car on the lot. Pynn himself is aware of this sort of concern. But he thinks appearances are deceiving here: the audience does owe it to the boy to accept his report when he speaks from knowledge, his track record notwithstanding. Pynn thinks that the impression to the contrary can be explained away by appeal to a distinction, familiar in the literature on the norm of assertion, between violating a rule and being blameworthy. In particular, while we are enjoined to believe the boy who cries wolf on that occasion when he speaks from knowledge (as this is what the knowledge rule of acceptance requires), we are not to be blamed for our failure to do so, given (our knowledge of) his track record.2525 Pynn, “Epistemic Degradation and Testimonial Injustice,” 161–62. This reply is unpersuasive, and in any case it will not serve the aim of preserving a purely epistemic answer to The Question. The reply is unpersuasive. An audience who recognizes that she has been the victim of the boy's lies on previous occasions is (epistemically and morally) entitled to regard the boy as having violated her trust. But precisely because she is (epistemically and morally) entitled to regard the boy in this way, the claim that she is under a moral or justice-based obligation to believe him on the one occasion when he speaks from knowledge is, in a word, implausible. Consider a familiar theme in the literature on trust: trust itself is or involves an affective state in which one relies on another's good will.2626 The classics are Annette Baier, “Trust and Antitrust,” Ethics 96, no. 2 (1986): 231–60; Annette Baier, “Trust,” n Tanner Lectures on Human Values, Volume 13 (Salt Lake City: University of Utah Press, 1992); Karen Jones, “Second-Hand Moral Knowledge,” The Journal of Philosophy 96, no. 2 (1999): 55–78; and Karen Jones, “Trust as an Affective Attitude,” Ethics 107, no. 1 (1996): 4–25. Even though it is doubtful whether this is the proper way to characterize the nature of trust,2727 See Holton, “Deciding to Trust”; Onora O'Neill, Autonomy and Trust in Bioethics (Cambridge: Cambridge University Press, 2002); Karen Jones, “Trust and Terror,” in Moral Psychology: Feminist Ethics and Social Theory, eds. Peggy DesAutels and Margaret Urban Walker (Lanham, MD: Rowman & Littlefield, 2004), 3–18; and Katherine Hawley, “Trust, Distrust, and Commitment,” Noûs 48, no. 1 (2014): 1–20, among many others. it cannot be denied that such affective states are implicated in many relations of trust. For this very reason, it seems wrongheaded in principle to think that a speaker can violate another's trust, thereby generating strongly negative reactive attitudes in his audience, and yet still expect to be trusted by her when he speaks knowledgeably. It might seem that this reply can be met. After all, there are many examples in which we would be entitled to blame someone for their past wrongs—wrongs in which they might have violated our trust—without this undermining a present obligation we have to them.2828 I thank an anonymous referee for this example. Perhaps the boy is a case in point. But insisting on this is inconsistent with a purely epistemic answer to The Question. The boy's audience has excellent track record evidence to think that his present say-so is untrustworthy. To insist that, even so, they owe it to him to accept that say-so because he spoke from knowledge, is to cast the demand of ethics or justice as requiring the audience to accept/believe in a patently irrational fashion.2929 One response to this sort of argument might be to recast the demands of rationality so that they never conflict with the demands of ethics or justice. In effect this is the proposal of so-called “moral encroachers.” I will discuss this position below, in section V. Such a position is not in the spirit of an epistemic answer to The Question, since it acknowledges a conflict between the demands of ethics or justice and those of epistemology. (JBAU) When a speaker tells us something, we owe it to her accept what she's said just in case our total evidence supports the hypothesis that she spoke from knowledge. The last of these topics is of special importance, since one might think that an audience-based epistemic answer to The Question is rendered particularly plausible if we assume anti-reductionism about testimony. The basic idea behind an audience-based answer is this: one owes it to accept a speaker's say-so when (by one's own lights) that say-so constitutes an adequate reason to believe what she said. Combining this with anti-reductionism, we get the view that one owes it to a speaker to accept her say-so as long as one has no reasons for doubt. so long as [the audience's] credibility judgement was issued by a well-trained sensitivity to the epistemically salient features of the testimonial performance in the context…then it was justified.3232 Fricker, Epistemic Injustice, 77. #BelieveWomen can be understood through a prism that uses the respect that is owed as both a moral claim about what we owe women and an epistemic claim that given what we owe them we are also justified in believing them.3333 Ferzan, “#BelieveWomen and the Presumption of Innocence,” 8. as a call to trust and as a call to non-reductionism [= anti-reductionism]. That is, not only do we owe it to women to start with a baseline of trust but this trust will justify our believing them.3434 Ibid. Unfortunately, such a view faces what I regard as insuperable difficulties.3535 One objection against using anti-reductionism to articulate what speakers are owed is the existence of a lacuna noted in connection with the proposal in Pynn, “Epistemic Degradation and Testimonial Injustice”: whereas anti-reductionism appears to postulate an epistemic permission, claims about what is “owed” to a speaker state obligations. Though I will bring these difficulties out in connection with a version of (JBAU) which incorporates anti-reductionism about testimony—call this “(JBAU+AIT)”—what I have to say holds for every audience-based epistemic answer to The Question. (JBAU+AIT) misfires in two different ways: it yields false positives (cases in which (JBAU+AIT) predicts that the audience owes it to a speaker to accept, when this is not so), and perhaps more interestingly it yields false negatives as well (cases in which (JBAU+AIT) predicts that the audience does not owe it to a speaker to accept, where in fact she does). The false negatives are particularly interesting, in that the lesson they teach concerns the inadequacy of any epistemic answer to The Question—including those that focus on the epistemic condition of both the speaker and the audience.3636 It is perhaps worth noting that proponents of the so-called “assurance view” of testimony appear to hold this sort of epistemic answer to The Question—one that incorporates both a speaker-based condition and an audience-based condition. See e.g. Hinchman, “Telling as Inviting to Trust”; Moran, “Getting Told and Being Believed”; and McMyler, “Obedience and Believing a Person.” If they do, they are targeted by the argument to follow. (JBAU+AIT)'s false positives are easy to come by. If a speaker is an exceptionally talented liar, it can happen that there are occasions on which he lies but the total evidence possessed by an audience supports the hypothesis that he is speaking from knowledge. Whatever the epistemological situation is here, the audience does not owe it to such a speaker to accept what he's said. False negatives are harder to come by, but they are also much more interesting for the illumination they shed on The Question. I submit that we will get false negatives in any case that meets the following specification: (a) the speaker speaks from knowledge;3737 Here and throughout my description, you should feel free to substitute your favorite speaker-based epistemic condition here. (b) the audience has excellent (albeit misleading) evidence that a speaker is untrustworthy, so that the audience's total evidence does not support the hypothesis that the speaker speaks from knowledge; and (c) it would violate an obligation one has to the speaker to decline to accept her say-so on the basis of that evidence. The real task is in showing that condition (c) is sometimes met. Inspired by reflections on the nature of testimonial injustice, I believe we can show what needs to be shown by appeal to the norms of justice themselves. It can happen that, unbeknownst to one, the body of evidence one has was itself shaped by the distorting factors of racism or sexism (or some other pernicious -ism) prevalent in one's community, such that to disbelieve (or reject) a Black or female speaker's say-so on the basis of that evidence is to treat them unjustly. To be clear, the cases I have in mind are not cases in which the audience manifests prejudice in reaction to their own body of evidence. Such cases exist, of course, but they are uninteresting from the present perspective. Rather, the cases I have in mind are cases in which the audience reacts in an epistemically competent fashion to their total evidence, but where their total evidence itself reflects the prevailing prejudices of their community, and where their total evidence is such that they are not in a position to rationally discern this. If such cases are possible, the appeal to anti-reductionism about testimony cannot save audience-based answers to The Question: what is at issue are cases in which the audience's background evidence provides her with reasons to doubt the reliability of the speaker's testimony, and anti-reductionism does not sanction acceptance in such cases. Why think that such cases are possible?3838 Arguments for this possibility can be found in Kristie Dotson, “A Cautionary Tale: On Limiting Epistemic Oppression,” Frontiers 33, no. 1 (2021): 24–47; Endre Begby, The Epistemology of Prejudice (Oxford: Oxford University Press, 2021); Emmalon Davis, “Typecasts, Tokens, and Spokespersons: A Case for Credibility Excess as Testimonial Injustice,” Hypatia 31, no. 3 (2016): 485–501; and Rima Basu, “The Wrongs of Racist Beliefs,” Philosophical Studies 176, no, 9 (2018): 2497–515. In addition, empirical evidence that such cases are actual is presented in Eduardo Bonilla-Silva, Racism without Racists: Color-blind Racism and the Persistence of Racial Inequality in America, 3rd ed. (Lanham: Rowman & Littlefield, 2009). Well, to begin, consider that testimony is a large source of our evidence, and that one of the most insidious effects of the pernicious -isms is that they affect the testimonies one encounters in one's community. On this score, the appeal to anti-reductionism is decidedly unhelpful to the prospects of an audience-based answer, since anti-reductionism lowers the bar for justified testimonial belief, and so makes it easier for young children to acquire justified testimonial beliefs from the racist or sexist speakers in their communities. To be sure, when one has independent grounds for thinking that the testimonies one encounters are themselves shaped by racism or sexism or…, one has grounds to rejec
Referência(s)