Why You Ought to Defer: Moral Deference and Marginalized Experience Feminist Philosophy Quarterly Volume 8 | Issue 2 Article 2 Recommended Citation Pearlman, Savannah, and Elizabeth Williams. 2022. “Why You Ought to Defer: Moral Deference and Marginalized Experience.” Feminist Philosophy Quarterly 8 (2). Article 2. 2022 Why You Ought to Defer: Moral Deference and Marginalized Experience Savannah Pearlman Indiana University–Bloomington savannah.pearlman@gmail.com Elizabeth Williams Indiana University–Bloomington ecw3@iu.edu Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 1 Why You Ought to Defer: Moral Deference and Marginalized Experience Savannah Pearlman and Elizabeth Williams Abstract In this paper we argue that moral deference is prima facie obligatory in cases in which the testifier is a member of a marginalized social group that the receiver is not and testifies about their marginalized experience. We distinguish between two types of deference: epistemic deference, which refers to believing p in virtue of trusting the testifier, and actional deference, which involves acting appropriately in response to the testimony given. The prima facie duty we propose applies to both epistemic and actional deference, though defeaters may quash either or both obligations. Even if one fails to epistemically defer to the marginalized testifier, we argue that they may still be ethically obligated to act in accordance with their testimony. Keywords: moral deference, marginalized testifiers, moral testimony, epistemic injustice, marginalized identity Imagine that a nonbinary person tells you, a cisgendered person, that you should not post an all-gender bathroom sign with whimsical or mythical creatures in your restaurant, no matter how well-intended the message.1 Representing the bathroom as “all gender” by denoting it with a mermaid, alien, superhero, or a person with two heads is harmful—they say—and if you have one, you should take it down. Should you defer to their testimony? In this paper, we argue that not only is it justified for you to defer, but that you prima facie ought to defer.2 1 For examples of such signs, see: https://cdn.customsigns.com/media/catalog/pro duct/cache/3/image/600x600/9df78eab33525d08d6e5fb8d27136e95/1/2/12285- custom-engraved-unisex-stick-figure-funny-bathroom-sign-8-x-8.png and https://i .pinimg.com/originals/19/bd/fb/19bdfbdb3cb796cd957c44be1334f4a9.jpg 2 By “prima facie” we do not mean to convey that this merely appears to be a duty, which we will later disprove. Rather, we mean this is, indeed, a duty, but that such a duty is defeasible—other considerations may override it. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 2 How should we go about forming our moral judgments? On the one hand, it seems that we ought to use our own reason to understand the moral issue, as an act of autonomy. On the other hand, we are limited moral agents who only have access to certain experiences. We frequently lack the knowledge necessary to understand a wide variety of moral issues. In cases such as these, deferring to others is often our only practical option for obtaining moral information in areas outside of our expertise. In this paper, we set aside the traditional puzzle of moral deference, which focuses on explaining (or denying) an asymmetry between the straightforward acceptability of nonmoral deference and the controversial nature of moral deference.3 Instead, we focus on a subset of moral deference, which we take to be not only permissible but prima facie obligatory in cases in which the testifier is a member of a marginalized social group that the receiver is not and testifies about their marginalized experience.4 Deference is justified in these cases in part because marginalized people have more relevant nonmoral information with regard to moral claims about their own experience as a member of that marginalized group as compared to their nonmarginalized counterparts.5 We argue that there are two distinct varieties of deference: epistemic and actional.6 At its broadest level, epistemic deference refers to believing p in virtue of 3 For discussion of this puzzle, see McGrath (2009, 2011), Hills (2009), Sliwa (2012), Enoch (2014), Howell (2014), Davia and Palmira (2015), Callahan (2018), and Lewis (2020). Much of the contemporary moral deference literature questions the permissibility of pure moral deference—i.e., cases in which the receiver believes some moral claim, p, on the basis of another’s testimony that p, though both testifier and receiver share the same nonmoral information. Our discussion will be limited to impure deference—i.e., cases in which the testifier has more relevant nonmoral information that the receiver lacks. The pure/impure distinction is Sarah McGrath’s (2009, 322). 4 In this paper we will use the terms “testifier” and “receiver” instead of “speaker” and “hearer.” We prefer the language of “testifier” and “receiver,” as it is more inclusive of those with disabilities. 5 This is not to say that individuals are wholly marginalized or wholly privileged. Recent trends in feminist literature have highlighted the complex, interwoven threads of various identities. It will be a frequent occurrence where someone will belong to a marginalized group but also a privileged group—an individual can be simultaneously oppressor and oppressed. 6 Robert Howell (2014) comes the closest to providing the distinction we have in mind, with his doxastic and active deference. However, whereas Howell’s active deference is limited to acting as though the judgment is true, our actional deference will contain this possibility as well as varieties of partial deference, as we explain in section 3. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 3 trusting the testifier, whereas actional deference refers to acting in accordance with the testimony given. The two can come apart, such that even if epistemic deference is inappropriate, the receiver may still be required to actionally defer to the testimony of marginalized individuals. The prima facie duty we propose applies to both epistemic and actional deference. For the purpose of this paper, we take it that someone has membership in a marginalized group if they belong to a social group that is subject to a combination of interpersonal, institutional, and structural oppression.7 Interpersonal oppression refers to the intentional or unintentional transmission of beliefs and acts that uphold and perpetuate systems of oppression between individuals. Institutional oppression is found at the level of institutions such as health-care, education, and legal systems. Institutions are oppressive when certain policies have the ramifications that marginalized individuals cannot easily navigate the institutional system, lack fair access to resources, or are otherwise treated unequally. Structural oppression, on the other hand, is not caused by specific policies or individuals; rather, it cuts across institutional and individual boundaries.8 In section 1, we clarify the distinction between epistemic and actional deference. In section 2, we argue that there is a prima facie duty to epistemically defer in cases in which the testifier is a member of a marginalized community that the receiver is not a part of. In section 3, we argue that there is also a prima facie duty to actionally defer in the kinds of cases with which we are concerned. Finally, in section 4, we address common worries, including the understanding objection—which 7 Implicit in the division of marginalized membership is a basic question of social ontology: How do we go about defining and distinguishing these groups? We follow Iris Marion Young (2014) in her suggestion that social groups exist in relation to other groups and constitute the identity of their members (11). While social relations produce groups, there is not one feature that all group members necessarily share (12). Still, members of the same group have “a specific affinity with one another because of their similar experience or way of life” (7). See Frye (2000) for more on oppression. 8 According to Young (2014, 39), structural oppression harms members of a marginalized identity via the “often unconscious assumptions and reactions of well- meaning people in ordinary interactions, media and cultural stereotypes, and structural features of bureaucratic hierarchies and market mechanisms—in short the normal processes of everyday life.” Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 4 suggests that there is something suspicious about deferring to moral testimony without sufficiently understanding the moral reasoning behind it.9 1. Epistemic and Actional Deference We take “epistemic deference” to be as follows: The receiver epistemically defers to the testifier iff given the testifier’s testimony that p and their assurances that p is the case, the receiver accepts these assurances and, in virtue of trusting the testifier, believes p.10 Our account of epistemic deference is within the tradition of Richard Moran’s assurance view.11 On the assurance view, the testifier—in providing their assurance— takes responsibility for the truth of their testimony. In doing so, the testifier gives the receiver their personal guarantee that the receiver has reason to believe what the testifier has to say.12 Moran (2018) argues that the testifier’s assurance that p offers the receiver an epistemic reason to believe p. Then, certain background conditions on the part of the testifier (such as their reliability or trustworthiness) determine whether or not the assurance provides a good reason to believe p. The relation between testifiers and other contextual features help to identify the type of speech act, which in turn will determine if we should attend to certain background conditions. For instance, if the 9 A number of philosophers (Hills [2009] most prominently) have argued that the fishiness of moral deference is rooted within the deferrer’s lack of moral understanding about the moral claim on which they defer. 10 In this, we follow Karen Jones (1999, 67), who contends that “someone accepts the testimony of another if she forms the belief that p on the basis of the other’s saying that p is the case.” 11 Jennifer Lackey (2008, 2011) and Frederick F. Schmitt (2010) have raised notable criticisms of the assurance view. Lackey (2008, 2011) has argued that the view faces a two-horned dilemma: “Either the view of testimony in question is genuinely interpersonal but epistemologically impotent, or it is not epistemologically impotent but neither is it genuinely interpersonal” (Lackey 2008, 222). As we will show in section 2, the three conditions we take to be necessary to have good reason to defer are both epistemologically potent (truth-conducive and evidentially based) and genuinely interpersonal (sensitive to assurance and trust). For a full-fledged defense of the assurance view against these charges, see Moran (2018). 12 Note that the assurance can be implicit, otherwise deference would be limited to very few cases indeed. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 5 testifier is being sarcastic, the receiver need not consider whether they are presumed to be reliable about that topic, since the purpose of the act is meant as a joke rather than to convey information. In this way, interpreting the speech act is a gatekeeping feature. If the act indicates sincerity and the receiver trusts the testifier, the testifier’s assurance provides a positive epistemic reason to believe that p is the case. When the receiver trusts the testifier’s assurance, they are doing more than trusting in a merely evidential sense. That is, trusting the testifier’s assurance does not simply reduce to “the fact that this person is committed to p is evidence that p is true.” We embrace two varieties of trust, one positive and one negative. The positive version is identified by its interpersonal features, which Berislav Marušić (2017) highlights in his account of trust. This kind of trust is a “belief held from the participant stance,” which licenses us to react with various participant attitudes to the person giving us the assurance (5).13 Giving an assurance is like making a promise; epistemically deferring is, analogously, like believing the promisor. The interpersonal burden is on the person giving the assurance to ensure that it is true— if the assurance is false, the person who received the assurance can rightly feel betrayed, disappointed, or indignant, to give some examples. Marušić (2017, 4) writes that positive trust (our term) has two “functional roles: It stops enquiry and precludes doubt.” When the receiver epistemically defers, they trust the testifier in this positive way, accepting their assurance—they do not doubt that the testimony is true. This is not to say that they cannot change their mind if presented with new information: epistemic deference is defeasible. However, when the receiver epistemically defers, they do not continue to seek out others’ opinions or purposefully inquire on the topic in order to determine the validity of the claim given by the assurance. This differs from the evidential account, which sanctions the receiver to continue actively attending to new evidence. In contrast, when the receiver trusts in this positive sense, they take the matter to be settled (though their belief might change if new information comes to light). In contrast, the negative version of trust is warranted in the absence of good reasons to distrust. In this, we follow Edward Hinchman (2005, 578–79): 13 Drawn from P. F. Strawson’s notion of the “participant point of view,” the participant stance is meant to capture the way in which we interact with other people as persons, capable of holding reactive attitudes towards one another, rather than as mere things, “posing problems simply of intellectual understanding, management, treatment, and control” (Strawson 2008, 18). For accounts of trust rooted in the participant stance, see Marušić (2017), Helm (2015), Hawley (2014), Hieronymi (2008) and Holton (1994). Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 6 Trust is a source of epistemic warrant just when it is epistemically reasonable. Trust is epistemically reasonable when the thing trusted is worthy of the trust—as long as there is no evidence available that it is untrustworthy. Assuming satisfaction of this negative evidential condition . . . , when an epistemic faculty is trustworthy by serving as a reliable guide to the truth, it makes available an entitlement to believe what it tells you whose basis lies simply in the fact that you trust it. Negative trust makes available a minimal justification to believe the testimony in a truth-connected way—but positive trust gives you thicker interpersonal reasons to believe, not simply because the information is available but because someone has assured you that it is the case. Taken together, the positive and negative kinds of trust capture the interpersonal features of trust while ensuring that it is sufficiently truth- tracking.14 Epistemic deference requires that the receiver trusts the testifier in both of these ways. From here on, when we speak generally of “trust,” we mean this as an umbrella term, which encompasses both varieties. While we do not dismiss the importance of evidence, we take evidence to be distinct from assurance, which is an appeal to trusting the testifier in virtue of their interpersonal relationship rather than merely by what evidence they have on offer.15 14 We think this combination is sufficient to avoid Lackey’s (2008, 2011) concerns. While Lackey (2011) takes Hinchman’s view to successfully place “the Trust View on the epistemological map,” she also contends that on it, “trust itself turns out to be epistemically superfluous” (18). Lackey’s primary worry is that someone could overhear the information and, without having any of the interpersonal relations, succeed in gaining the epistemic information. To this, we are convinced of Moran’s (2005) reply to such cases: That the asymmetry between cases where you are the intended receiver of testimony versus an unintended overhearer lies in the risk that such testimony is deceptive. In the case where you are the intended receiver, the testifier acts as a guarantor—they accept the risk on themselves that what they are saying is truthful (and not an inside joke or a sarcastic quip). In the case where you are not the intended recipient of testimony, the burden of risk shifts to you. 15 The importance of the interpersonal relationship also helps to block various unjust beliefs that would seem warranted on the basis of evidence alone. For example, Basu (2019) highlights the “rationally racist” belief that Black customers leave worse tips. Moss (2018) describes a case where, statistically, one could infer that a female employee is an administrative assistant. Franklin (2005) describes a case where he was mistaken as a worker at a social club that he attended, since the club had very few Black members and many Black workers. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 7 The assurance—which is sometimes implicit and sometimes explicit—will provide the epistemic justification, rather than the background conditions themselves. To illustrate how epistemic deference works within the assurance view, let us begin with a nonmoral scenario. Take the case of a plumber who, upon inspecting my sink, diagnoses the broken garbage disposal as a result of my having carelessly disposed of artichoke leaves too fibrous for my disposal to manage. The plumber’s assurance as to the cause of my broken garbage disposal acts as a type of guarantee— that according to the evidence he has as the relevant expert, this is the proper diagnosis. When I epistemically defer, I accept the reason given by the assurance as a good reason and form the belief that the artichoke leaves were the cause of my clogged sink. Whereas epistemic deference involves trusting the testifier in virtue of believing the assurance, actional deference does not require trust but merely requires that the receiver relies upon the testimony.16 Our definition of “actional deference” is as follows: The receiver actionally defers to the testifier when they act in accordance with and in response to the testimony that they have received. Acting in accordance with the testimony does not require belief or trust—it only requires that the receiver act in a way that is appropriate given the practical reason supplied by the testimony. What does “acting in accordance with the testimony” mean? With some kinds of moral testimony such as “Never do x,” “Do x if y,” or “Always do x,” the answer seems clear. If the testimony is “Never do x,” and the receiver never does x in response to the testimony, the receiver has clearly acted in accordance with the testimony.17 With less straightforward testimony, such as “Generally do x” or “Sometimes do x,” the receiver may have to investigate further to discover the right 16 In this we roughly follow Marušić’s (2017, 3) account, on which “trust and reliance are fundamentally different—as different as judgment and action.” 17 Testimony may be more or less clear with regard to what actions the receiver should take. For example, an organizer might testify that you should sign the petition in front of you, show up to Saturday’s protest, etc. The receiver might also receive advice that is targeted to specific situations, such as “Don’t comment on your daughter’s weight when she comes home for Thanksgiving.” But frequently, moral advice will be difficult to translate into action. Take, for instance, the task of determining which maxims are licensed by Kant’s categorical imperatives or the task of determining what the most virtuous person would do in a given situation. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 8 conditions for actional deference and do the best they can in the meantime.18 Often, depending on the content of the testimony and the context in which it is given, there will be a number of appropriate ways to act. For instance, if my child’s pediatrician tells me that it’s important to vaccinate children and warns me against anti-vax propaganda, I may vaccinate my children, encourage my friends and family to vaccinate their children, or say something if I see a friend post anti-vax misinformation on social media. If I do any of these things, I act in accordance with the testimony given and, in doing so, rely on the doctor’s testimony. When the receiver actionally defers in response to testimony, they take the testimony to provide a reason for action. Like any moral reason, it may be overridden by stronger reasons for action. For example, if I am told by a woman that I shouldn’t touch women without their explicit consent, I do not fail to actionally defer to that testimony by grabbing a woman who is about to fall off a cliff, even if she does not explicitly consent. There is a stronger reason to preserve her life that reveals an exception to the general moral claim. Even if the reason for action presented by the testimony is overridden in certain cases, I actionally defer if I act in accordance with the strength of the reason given by the testimony. While ideally we would epistemically defer in cases in which we have good reason to trust the testifier and hence believe the testimony, actional deference does not require those things. Importantly, it is governed by practical reasons for action. For example, suppose I am playing a game in which my teammates are attempting to give me information from a manual they possess to disarm a digital 18 One worry is that, when the demands of actional deference are unclear, actional deference will collapse into the epistemic. For instance, if we must try acting on the advice or do our own research to determine its applicability in a particular instance, it seems that we will eventually either form the belief and epistemically defer or determine that the testimony was inaccurate. Though this might occur, we do not agree that actional deference necessarily reduces to the order of the epistemic. You might continue to act in accordance with what someone tells you to do but remain morally unconvinced that it is the right thing to do, even if you never run across any potential counterexamples in experience. See, for instance, the male colleague who is given the advice, “Don’t comment on female colleague’s appearance.” The male colleague may be unconvinced that this is wrong in all cases but may adhere to the advice out of self-preservation and never encounter any good reason to act differently. Likewise, we could imagine someone who is constitutionally agnostic, such that they are unable to fully trust and believe any testimony or evidence provided to them. Despite their doxastic agnosticism, this person might act on the testimony given to them for practical reasons that do not reduce to belief in the testimony. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 9 bomb, where my teammates only have access to the manual and I only have access to the bomb. Suppose that in past instances of this game I have found that my teammates have often lead me astray because they have missed important details in the complex manual. In this particular game, I now have only two seconds left to disarm the bomb and my teammates tell me to press the red button. Even though I might reasonably fail to trust my teammates and believe that this is the right thing to do, I must rely on their testimony to have a chance of winning the game and defusing the bomb. When I press the red button, the bomb might explode, and I might lose the game, despite my practical reason being to win the game. Thus, I actionally defer to my teammates without epistemically deferring to them. Epistemic deference and actional deference can also come apart in the other direction. I may epistemically defer to the testimony of a local community activist when they say that this holiday season, it is important to hold your relatives responsible when they say racist things at the dinner table. And yet, though I may truly believe this is the morally right thing to do, I may also fail to actionally defer. Say, for example, that at Christmas my ninety-year-old grandmother makes a veiled racist comment in passing before continuing onto another tangent. Perhaps I am unwilling to cause my grandmother and other family members discomfort by backpeddling the conversation to the inappropriate comment and focusing in for a teaching moment. I may make excuses for myself—“It happened so fast!” or “She’s ninety years old. Is this really the moment to talk to her about racism?” Such a case shows that it is possible to split the distinction in the opposite direction—to sincerely epistemically defer and yet fail to actionally defer. In this section we have defined two distinct varieties of deference. We now turn to motivating a prima facie duty to epistemically defer. 2. Why You Ought to Epistemically Defer Imagine again that you are a cisgendered person who has been told by a nonbinary person that your well-meaning, whimsical bathroom sign is actually rather offensive. Assume no explanation is given. Should you epistemically defer? We argue that there is good epistemic reason to epistemically defer when (1) the receiver justifiably believes that the testifier has more relevant expertise about p (and as such the receiver believes the testifier to have greater epistemic authority), (2) the receiver lacks good overriding reasons against p, meaning that negative trust is warranted, and (3) the receiver should positively trust the testifier on the issue about which they are testifying.19 19 Condition 2, coupled with a general epistemic norm that receivers should be generally aware of the basic conditions for trustworthiness and reliability, fortifies our view against gullible receivers. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 10 The first condition provides us with an epistemic reason to defer.20 While full- fledged expertise provides the strongest basis for epistemic authority, it is not necessary. A much weaker condition is sufficient for epistemic deference, depending on the context of the situation: the testifier need only to be better informed about the relevant information than the receiver in order for the receiver to be justified in deferring to the testifier.21 The second condition ensures that we are unjustified in deferring to a testifier about a proposition that we have good reason to believe is untrue, misleading, or distorted.22 Depending on the circumstances of the case, the third condition can represent either an epistemic “should” or both an epistemic and a moral “should.” The epistemic “should” captures the requirement that epistemic deference meets the conditions of positive, interpersonal trust. The moral “should” excludes those who would withhold trust for the wrong reasons, such as prejudice or discrimination. Conditions 1 and 3 positively justify deference, and condition 2 is a negative condition: the receiver must lack any good overriding reason that would defeat the testimony. As such, our account both values the interpersonal features of trust and assurance and is epistemically truth-tracking. Taken together, these three conditions provide us with a justified foundation for epistemic deference. That is, they provide us with sufficiently good reason to epistemically defer, though we may be obligated when there is moral reason to defer. Return to our central case: do you have good reason to epistemically defer when the testifier is testifying about their marginalized experience and they are a member of a marginalized group that you are not? Take our condition 1, that the testifier has more relevant expertise about p than the receiver and as such has greater epistemic authority than the receiver. We argue that members of marginalized groups have greater relevant expertise regarding claims about their experiences qua member of that marginalized group than nonmembers. For example, a person with a disability that requires them to use a scooter will have (at the very least) greater nonmoral 20 John Hardwig (1985) argues that rationality sometimes requires a layperson to defer to an expert. Or, as he puts it “rationality sometimes consists in refusing to think for oneself” (336). 21 Take the following example: I am a flight attendant on an airplane, and there is a medical emergency. I have very little medical training. Concerned, I call for a doctor over the loudspeaker. Though no doctors are available, there is a veterinarian ready and willing to help. Certainly, she is not an expert on the human body, but I am justified in deferring to her medical orders in handling this sick passenger. 22 Our second condition is similar to an amendment offered by Hinchman (2005), which we referenced in section 1. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 11 knowledge than an abled person with regard to the accessibility of the various places they go. Take our condition 2, that the receiver lacks good, overriding reasons against p. If the receiver has no good reason to distrust the testifier, then our conditions for negative trust are fulfilled. Receivers should be careful not to be misled by various unjustified rationales that cast doubt on the truth of the testimony given by marginalized individuals—for example, “You’re just emotional,” “You’re just an angry Black woman,” “You can’t be objective about this,” and so on. Take our condition 3. Should you trust the marginalized testifier on the basis of the assurance? If our first two conditions are fulfilled, then you have good epistemic reason to positively trust the testifier, where that reason is provided by the assurance itself. In sum, we have good reason to epistemically defer when we have reason to believe that a testifier is in a better position to know p, there is no reason to distrust the testifier, and positive trust is warranted by the testifier’s assurance. The preceding discussion has focused primarily on epistemic reasons to defer; however, there are also moral reasons to defer in the kinds of cases with which we are concerned. When our three conditions are met and you fail to defer to marginalized individuals testifying about their own experience, you fail to treat them with proper respect. This failure of respect might occur in two different ways.23 First, it might be a matter of failing to give proper appraisal respect, or respect for a person in light of their merits or excellences. If you do not defer to a marginalized person because you do not appreciate their expertise and authoritative standing as an epistemic agent, you fail to give proper appraisal respect. This is straightforwardly a case of epistemic injustice, especially if you do not give proper appraisal respect due to some prejudiced view of the members of a particular marginalized group.24 Second, withholding deference may represent a failure to give proper recognition respect, by which you fail to treat the testifier as a moral person and give them the status value they deserve. In order to give members of marginalized groups the respect they deserve, we must recognize them as experts regarding testimony about their experience qua members of that group.25 Historically, members of these groups have been ignored, 23 The appraisal respect/recognition respect distinction was introduced by Darwall (1977). 24 For accounts of epistemic injustice, see Miranda Fricker (2007), Davis (2016), Anderson (2012), and Pohlhaus (2017). 25 We recognize the harm of assuming marginalized persons to be spokespersons for their identity. We will address this and other similar worries in section 4. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 12 and we should be careful not to repeat those injustices.26 The injustice of a single instance of disrespect is compounded by a history of dominant groups failing to give proper appraisal and recognition respect to marginalized persons. As such, there is good moral reason to positively trust marginalized testifiers and epistemically defer. We have argued that there is both good epistemic reason and good moral reason to epistemically defer in the general kinds of cases with which we are concerned. Each taken alone might be insufficient to justify a prima facie obligation to defer; however, taken together, they show that we ought to defer. If there is good epistemic reason to defer, then the receiver is at least justified in deferring. If, in addition, we consider the potential moral costs of failing to epistemically defer (informed by a history of oppression), then it becomes clear that the receiver ought to defer in such cases. As such, we have a prima facie obligation to epistemically defer in cases in which the testifier is a member of a marginalized group that we ourselves are not and they testify about their marginalized experience. Returning to the case of the nonbinary acquaintance, does it meet our three epistemic conditions? First, given that our nonbinary acquaintance is a member of the relevant marginalized group and that the receiver, a cisgendered person, is not, the testifier has more relevant expertise as a member of that marginalized group. This informs their testimony about the moral status of whether it is harmful to use a sign with mythical creatures to represent an all-gendered bathroom. Second, the receiver has no reason to withhold negative trust from the testifier. Third, the assurance provides the receiver with an interpersonal reason to defer. As such, the receiver has good epistemic reason to epistemically defer. Does the receiver have good moral reason to epistemically defer? If the receiver fails to defer in this case, then they either (a) do not show proper appraisal respect by failing to treat the acquaintance as the relative expert that they are on this matter or (b) do not show proper recognition respect by failing to treat the acquaintance as a person who can make moral claims upon them. Failing to believe the acquaintance may risk committing another epistemic injustice in a history of injustice. As such, the receiver has good moral reason to defer. Hence, as there is both good epistemic reason and good moral reason to defer, there is a prima facie obligation to epistemically defer in this case. 26 Our view does not reduce to the claim that deference is sufficient to counteract epistemic injustice. We recognize that epistemic injustice is a complex interpersonal, institutional, and structural phenomenon, and our view only addresses a small part of this larger problem. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 13 3. Why You Ought to Actionally Defer Unlike epistemic deference, actional deference can be appropriate even when belief is unwarranted. Whether we should actionally defer depends on a variety of practical considerations that cannot easily be reduced to what is epistemically justified—we must rely on our practical judgment to determine how to act in light of the testimony that we have received. In this section, we gesture towards two primary considerations that should guide our practical judgments. First, we have general moral reason to do what is right, to act in ways that are appropriate, to do what is good, and so on. When we have good reason to believe that the moral testimony given to us has been provided by someone with greater expertise and that it presents the strongest reason for how we ought to act in a particular situation, we have good reason to follow the advice the testimony gives in that situation.27 For example, if a friend (who I know to be kind and socially adept) gently points out that I should let others talk a bit more in group conversations, I should follow that friend’s advice and allow more space for others in conversation. We all have moral blind spots and sometimes must rely on the testimony of others to determine what we ought to do. Second, there is an interpersonal dimension to actional deference that itself provides reason to act on the testimony. Though positive trust is not required for actional deference, the interpersonal relationship that the receiver has with the testifier will influence how the receiver ought to act, as the receiver’s actions reflect upon that relationship. If I fail to actionally defer to the testimony of a trustworthy friend, I communicate to them that I did not take their assurance seriously. Depending on the particular relationship between the testifier and the receiver, the receiver’s actions may have a different import. However, all that is needed to get interpersonal reason to defer off the ground is the moral relationship between testifier and receiver. Failing to actionally defer can be disrespectful to the testifier, either by violating expectations of trust or respect within an interpersonal relationship or by acting in a way that would cause harm regardless of any breach of trust or respect. Members of dominant groups should be careful to avoid either kind of disrespect. For someone who is outside of a marginalized group, it is easy to be ignorant (sometimes willfully so) of the moral experiences of those within that marginalized group and of the harms that members of the dominant group inflict upon the marginalized. As such, those in dominant groups have special reason to pay heed to the testimony of marginalized individuals in particular so as to avoid 27 This, of course, depends on whether the testifier is sufficiently trustworthy and reliable. The receiver might have good reason to do the opposite of what a constitutional liar says. Depending on the stakes involved, a slight degree of untrustworthiness might be sufficient to make actional deference inappropriate. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 14 perpetuating epistemic injustice and the other day-to-day harms that marginalized individuals experience. Even if a member of the dominant group does not understand or believe the claim that the member of the marginalized group is putting forward, they still ought to actionally defer and rely on the testimony in action. To do otherwise is to fail to appreciate the interpersonal claim being made on the receiver and to risk doing further harm to members of the marginalized group. Take the example of the nonbinary acquaintance again. If the receiver, a cisgendered person, has good epistemic and moral reason to trust the acquaintance, they also have good reason to act in accordance with the testimony and refrain from hanging a sign with whimsical creatures to indicate the “all gender” status of the bathroom. While they may lose some pleasure by using a different sign that they find less humorous, there does not appear to be any good reason that trumps the acquaintance’s testimony that posting such a sign is a harmful thing to do. Hence, they shouldn’t post such a sign, and if they have one, they should replace it. Importantly, our distinction between epistemic and actional deference aids in the explanation of certain cases with which some traditional explanations of moral deference have difficulty—namely, cases in which epistemic deference is unwarranted. Take, for instance, an example of a friend who confides in me that she was molested in the past by a high school teacher. Let us stipulate that she has once lied to me in the past about a related matter. Many accounts of moral deference (McGrath [2009], Hills [2009], and Lewis [2020], to name a few) would recommend that I do not defer in this scenario—full stop. And in some ways, it is obvious that we ought not defer. Knowing that my friend has lied about a similar matter in the past undermines her trustworthiness. As a result, her testimony fails to be credible enough for me to have good epistemic reason to believe that she has been harmed in this way. And yet, though I may fail to believe her, it seems patently the wrong thing to do to act fully in line with this disbelief. I should, perhaps, act in certain ways as though her testimony is true, even if I do not have full-fledged belief that it is.28 It would be appropriate for me to comfort her, to offer my support, and to warn her if a movie 28 Some might suggest that a credence notion of belief would better describe those cases in which beliefs seem to be partial. This could amount, potentially, to partial epistemic deference. However, the credence view has its own difficulties; namely, a penchant for vagueness and mathematical complication. For this reason, our account uses a more traditional epistemic framework. Still, while we take belief to be all-or- nothing, the way we determine how to act will depend on our degree of confidence that the testimony is correct (or incorrect). Thus, determining how we act in response to testimony parallels the credence view about belief. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 15 I’ve picked out for us to watch contains sexual violence.29 Despite my reasons to doubt her testimony and not accept it as fully credible, I still have reason to act in certain ways as if the testimony is true. Because of the nature of the testimony, I have reason to be sensitive and caring around what is very possibly a real and heavy trauma for my friend. In the literature on moral deference, philosophers have gestured at the importance of actional deference even in the absence of belief. David Enoch (2014, 6) writes: “Even if . . . deference-based moral judgments are epistemically suspicious, still we should act on them.” Similarly, Matheson, McElreath, and Nobis (2018, 14) argue that for any proposition “we have three options (believe, disbelieve, or suspend judgment)” whereas with actions “we have two options: do it, or don’t do it.” Of course, there are an array of ways in which we can act (or fail to act). Thus, in cases where we lack belief (and therefore do not epistemically defer), we must decide how to act (or to refrain from acting) and to what extent our action should be responsive to the testimony at hand. Return to our example: It is irresponsible to epistemically defer to my friend, as a result of her history of being untrustworthy regarding such matters. We might, as Matheson, McElreath, and Nobis suggest, suspend judgment. However, we must determine when it is appropriate to act in accordance with the testimony and when it is not. These determinations should take into account the possibility that my friend is insincere. Thus, this case appears to be one in which only partial actional deference is warranted. While epistemic deference is all-or-nothing (belief or failure to believe), actional deference admits to degrees. Actional deference is a strain of deference distinct from its epistemic counterpart, subject to its own appropriateness conditions. We can be morally required to actionally defer even when it is epistemically inappropriate to believe the testifier. 4. Objections We have argued for a prima facie obligation to defer both epistemically and actionally in cases in which the testifier is a member of a marginalized community that the receiver is not. We now turn to potential concerns for our account. 29 Depending on the relationship, it might also be appropriate for me to attempt to have a deeper conversation with her about whether this particular claim is true, knowing that she has lied in the past. Finding the right way to have that conversation might be incredibly fraught. But doing so might be necessary to discover whether this claim is truthful, so I can either get behind my friend fully or help her to grow with regard to her tendency to lie about these things. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 16 Perhaps the most common objection to epistemic deference on moral matters is the understanding objection. Several philosophers have sought to pinpoint the fishiness of moral deference within the deferrer’s lack of understanding why they should act in accordance with a particular moral claim.30 There are two important features of this set of cases that allow our central claim to avoid the understanding objection: (1) Since certain phenomenological experiences of being a member of a marginalized group are not accessible to the receiver, the receiver cannot have a full and complete understanding of the moral reasons that support the testimony. (2) The receiver gains a good bit of moral information about the testimony by virtue of recognizing that the person giving the testimony is a member of a marginalized group and that it is salient to their experiences as a member of that marginalized group. The first feature highlights that the direct, full-fledged grasping of reasons is too strict a threshold for nonmembers of a marginalized group to meet, since they will be unable to have the relevant phenomenological experiences. One might object that, given enough personal narratives of these experiences or by extrapolation, one could come to know what those experiences are like at the phenomenological level. However, as Laurence Thomas (1998, 362) argues, there are serious limits to reflective imagination, such that “no amount of imagination in the world can make it the case that one has the subjective imprimatur of the experiences and memories of another.” The second feature, however, seems to provide the cognitive grasping of at least some moral reasons—namely that there is some way in which the moral claim is tied to the experience of marginalization. It may be fairly clear from context that the reason some action is wrong is because it is racist, sexist, homophobic, and so on. It is unclear what further explanation beyond “x is wrong because x is racist” might be needed in order to justifiably hold the belief that x is wrong—if more were required, it is unclear how many of our moral beliefs would survive scrutiny. If it is plausible that certain features of the testimonial exchange are sufficient for the kind of understanding required by the understanding objection, then our central claim about epistemic deference is not overturned. Even if our view survives the understanding objection, one might worry that the belief produced from deference is still somehow epistemically inferior to a belief formed from a more complete moral understanding. Yet, if the receiver already has substantial understanding, why would they need to defer in the first place? Second, while we agree that in some circumstances it is preferable to have in-depth moral understanding, we are too limited as agents to be able to do this for all, or even most, of our beliefs. Given these limitations, beliefs formed from deference have a distinct 30 See Hills (2009), Callahan (2018), McGrath (2011), and Nickel (2001). “Fishiness” is Enoch’s word (2014). Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 17 value of their own, both epistemically, as they allow us to expand our moral knowledge, and morally, as they deepen our interpersonal relationships. Deference, then, should be seen as a practice of community building and shared epistemic labor, in the same way that the university is an institution of community building and shared epistemic labor. The experts in the philosophy department do not have the time or energy to fully understand everything in every other department, yet they can trust the experts in other parts of the university. In our view, communal epistemic trust based on differential expertise is a distinctly valuable enterprise, and it is unclear that it is inferior to a more individual view of epistemic achievement. In cases where marginalized people testify about the identity-harms that they have experienced, these community-building relations are important both morally and epistemically. Morally, trusting marginalized people communicates respect. Epistemically, the fact that the nonmarginalized receiver cannot understand (or understand fully) makes deference the proper response. However, one related issue is that a failure to internalize understanding makes it difficult to transmit beliefs. Return to our mermaid sign. Imagine that another employee at the café sees the owner taking down the sign. They might ask, “Hey, why are you doing that?” The owner might say, “It’s offensive.” The employee asks, “Oh really? Why is that?” The owner could reply, “I had a chat with a nonbinary customer who assured me it is. They didn’t have time to explain, but I believe them.”31 This might be insufficient for the employee to come to believe that the sign is, indeed, offensive—especially since he did not participate in the interpersonal exchange, himself.32 However, the owner’s explanation of the relevant features of the testimonial exchange will provide at least some understanding for why she is taking down the sign. Both the owner and the employee could take it upon themselves to gain a better understanding by deciding to research the issue online, reading testimony from other LGBTQ+ people on the same topic. In this way, deference can open the door to learning—it needn’t be construed merely as an epistemic backstop. Pursuing your curiosity is consistent with epistemic deference, so long as you believe the testifier and do not seek out information with the intent to undermine their testimony. Yet another way of motivating the understanding objection is that epistemic deference suffers a double bind: Either the testifier can provide us reasons to support 31 Thank you to the anonymous reviewer for imagining this scenario. 32 We have argued in a previous footnote that although knowledge can usually transmit via testimony (or overheard testimony), this fact does not indicate that assurance is epistemically superfluous, as Lackey claims. Moran (2005) highlights that in such cases, the fact that the receiver is not the target of the assurance puts them at greater risk to misconstrue the testimony’s content or intended speech act. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 18 the belief in question, or they cannot. If they can, why do we need deference at all? And if they cannot, why should we be motivated to believe them in the first place?33 To the first horn we have two replies. First, we take it that marginalized people can, generally speaking, provide reasons in support of their testimony.34 A disabled friend could provide the reasons why a building is inaccessible, or the nonbinary acquaintance could outline why the sign is offensive. However, sometimes these reasons aren’t easily understood by those who do not have the proper experience to conceptualize them. Perhaps, in an ideal world (unconstrained by the limitations of time, labor, and conceptual experience) there would be something preferable about learning information for oneself over deferring to others.35 Importantly, we do live in a world constrained by these social and temporal features. Just because a marginalized testifier can provide us reasons for their belief does not mean they have the time, desire, or will to do so. A nonmarginalized person might ask for those reasons if they are interested in understanding rather than deferring. But they ought to be prepared to receive a “no” in reply. At the end of the day, there is a fine line between asking and insisting that a marginalized person provide reasons for the testimony about their own identity harms. Too often, the task of providing a privileged person understanding is done at the expense of marginalized people’s time and energy, resulting in the harms of spokespersoning and epistemic exploitation (both topics we will return to towards the end of this section). Turn to the second horn—if they can provide reasons, why do we need deference at all? To answer this, we return to the participant stance and the relationship between persons. In the cases of the sort with which we are concerned, where limited evidential reasons are provided, it is the interpersonal exchange rather than the evidence that does the work to produce the receiver’s belief or action. The testifier appeals to their own experience qua a marginalized person, which is first- hand experience from the relevant social identity. This lends weight to the credibility and persuasiveness of the testifier, framing them as a person worthy of trust. When the receiver is (or would be) unconvinced by the evidential reasons alone, but trusts and believes the testifier in virtue of their assurance that p is the case, they 33 Thank you to an anonymous reviewer for this formulation of the worry. 34 We say “generally speaking” because there will be a limited number of cases where the reasons may be inaccessible or difficult to parse. Marginalized people may also have a good understanding of the moral features of their experience but lack the ability to effectively communicate it to others. 35 We will bracket this conversation, but others have discussed the virtue or vice of epistemic autonomy at length. Elizabeth Fricker (2006) has argued that epistemic autonomy is an ideal we should seek to achieve. Zagzebski (2012) and Wiland (2021) deny that epistemic autonomy has any intrinsic value. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 19 epistemically defer to them. Deference, rather than mere learning, captures the interpersonal heft of this exchange. Even if the understanding objection or some other worry were to succeed against our argument for a prima facie duty to epistemically defer, we argue that actional deference may still be required. Imagine a slightly edited case where the café owner who has posted the whimsical “all gender” bathroom sign is a cis person who is new to the notion of being trans or nonbinary. Perhaps this person is self-aware enough that they should be attentive to the nonbinary speaker, but they do not fully grasp the relevant concepts, and therefore need time to understand what they mean. In this case, the owner still has the obligation to actionally and epistemically defer, though they may not be aware of the latter. Even if the lack of conceptual background may prevent them from realizing they should believe what the customer has to say, the conversation alone should be enough for the owner to realize they should take down the sign—that is, to act in line with the testimony given. The previous objections have targeted epistemic deference, but the following objections target both strains. One of the most pressing worries is that, were deference of this kind to become a more widespread social practice, there would be several negative outcomes that would undermine the benefits of the practice. First, it seems that such deference norms might encourage superficial deference, by which individuals might purport to believe marginalized people and act in accordance with their testimony for the wrong reasons. For instance, it may become a norm to “defer” to marginalized testifiers as a means of virtue signaling rather than with the primary intent of harm reduction or treating others with respect. Of course, it is often difficult to separate instances of virtue signaling from deference done for the right reasons. The descriptive possibility of virtue signaling, however, does not count as a mark against our account because the norms of deference we propose can explain why virtue signaling is not done in the proper spirit of deference. The point of deferring is to properly listen to marginalized persons and respect them, and virtue signaling takes this basic aim and corrupts it to the ends of the privileged who seek to maintain their social power. Second, it seems that the expansion of deference norms within our social practices may allow people with internalized oppression to be “used,” not as a source for deference but instead merely as support for the receiver’s own foregone conclusions. Again, this kind of move undermines the core spirit of deference and makes marginalized people into pawns to be used for winning argumentative battles. It is not acceptable to “defer” to the one marginalized individual who shares your view simply because it reinforces your own perspective. Finally, it seems that structural features of who makes it into the conversation might encourage deference to a select group of unrepresentative marginalized agents Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 20 who have made it into the conversation through luck or relative privilege, even if the receiver defers for the right reasons. As Olúfẹḿi Táíwò (2020) argues, the good intentions of deference may be led astray by the limitations of who is in the room. If we only pay attention to the people in the room, we will fail to see how the room is constructed and who has been left outside of it. Due to its limited scope, our account does not touch on the construction of the room itself; however, it can distinguish between different experiences of oppression that marginalized people have, often due to intersecting oppressions. While you should listen to a class-privileged Black colleague when she testifies about the experiences she has had of racial discrimination within the halls of power, you should not take her experiences to be representative of all Black people. As our account emphasizes respect for all marginalized people, especially those who are frequently disrespected, we are convinced that receivers have a duty to seek out the testimony of different members within different intersecting marginalized groups. What should you do, however, if there is a consensus among members of that identity group and this consensus conflicts with the testimony given? In our view, this acts as a good reason for you to at the very least suspend belief (or defer to the consensus opinion at most).36 However, there may be no consensus at all. To whom do we defer in cases in which there is not a clear consensus among members of the marginalized community? If multiple testifiers meet the conditions we have suggested (that the testifier has more relevant expertise about p and the receiver lacks overriding reasons against p), and these testifiers are providing conflicting moral testimony, we argue that the receiver is given a sufficient reason against the prima facie duty to defer to any one of them. If there is no reason to treat one testifier as more trustworthy than the other, then it may be that the correct attitude is, as Matheson, McElreath, and Nobis (2018) suggest, to suspend belief. The receiver will have to make a considered judgment based upon the reasons they have available or do further investigating. In some cases, there is not a straightforward answer as to whether the receiver should actionally defer, and in those cases the receiver will have to rely upon their own judgment. There will also be instances in which we ought to defer to one marginalized individual over another, even if the two disagree.37 As Enoch (2009) writes, there are 36 While Alvin Goldman (2001) argues against “going by the numbers” when it comes to deferring to experts, we are convinced by David Coady’s (2006) reply. In it, Coady contends that Goldman’s formal argument wrongly assumes that the probabilities involved will remain constant, when they are usually in flux (70). 37 It is generally agreed upon within the epistemology of disagreement that, when a testifier has better evidence or a better ability to correctly respond to that evidence, we ought to defer to that person in the case of disagreement. See: Hardwig (1985), Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 21 a variety of possible explanations for moral disagreement: moral matters are complex, people are not as rational as we often think, and we can be self-deceived about how our interests affect our beliefs. In the case of marginalized individuals, there may be other reasons for disagreement, including that many people might have internalized oppression that makes it more difficult to see the harms done to themselves, some might not be reflective about how various actions make them feel, and others might have different histories that make them less susceptible to various harms. But if it is important to seek out consensus views, then why do we need deference at all? Wouldn’t it be preferable to simply do our homework and avoid relying on individual testimony that could be wrong?38 First, we might worry that jumping over individuals to the group view treats individuals within that marginalized group as a monolith. Marginalized people have different views, and this knowledge should both lead us to seek out whether there is a consensus view and to listen to individual marginalized people who may have different views. More importantly, there is moral value to participating in testimonial exchanges with marginalized people so that those individuals are actually listened to and recognized as full-fledged moral and epistemic agents. By going straight to the group view, we bypass valuable interpersonal social goods, such as directed care, empathy, and relationship-building between people with salient identity differences. Another point of concern is that internalized oppression might interfere or distort the marginalized person’s own beliefs such that they are not a reliable or good source of moral advice about aspects of their experience. For instance, a gay man might testify that his same-sex attraction is morally reprehensible and that LGBTQ Foley (2001), Kelly (2005), Christensen (2007), Simpson (2013) and Matheson (2015). See Goldman (2001) for a full-fledged account of who to trust when experts disagree. 38 In order to know that you should do your homework in the first place, it is likely that you will have to encounter at least some testimony from marginalized individuals. This is in some way a chicken-and-egg problem, by which, in order to get to a consensus view and know you should seek out such a view, you will likely have to defer to others. What this suggests is that there isn’t really a path to understanding marginalized issues without deferring to or at least provisionally relying on the testimony of marginalized individuals. As we have argued, the former is preferable to the latter because failing to defer can often communicate a lack of trust and subsequently disrespect, especially if the receiver only holds marginalized testimony in this provisional light. Perhaps the receiver could be generally skeptical and withhold belief in any domain without full understanding, but that seems to us to be an epistemically and socially impoverished way to move through the world, as that receiver will never positively trust anyone else. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 22 people should resist their desires and live heterosexual lives. In a less drastic example, marginalized individuals might treat certain arguments or statements as reasonable even when they are not, because they are used to hearing them. Internalized oppression complicates the picture, but there will be cases in which internalized oppression is not a strongly distorting factor, such as in cases where there is a consensus view among members of the marginalized group about an aspect of their experience. In these cases, (assuming our other conditions are met) you ought to defer. In cases in which the testimony contradicts the consensus, suspension of belief and more investigation may be in order. Relatedly, marginalized experiences are often traumatic, and trauma often distorts not only the initial experience(s) of harm but subsequent experiences that appear to be similar to the original experience(s) of harm.39 If trauma is distorting, then why should we defer in cases in which the testimony is shaped by it?40 We agree that trauma is often distorting, but, as a general rule, it is not sufficient to block deference. Part of respecting others is taking a default stance of trust towards them when they speak about their own experiences unless we have a special reason to disbelieve them. The mere existence of trauma in someone’s background does not provide a sufficient special reason. If it did, it would undermine a great amount of human testimony, which occurs in important contexts (e.g., refugees from war, the experience of the incarcerated, etc.) There must be something specific about how the trauma manifests (e.g., if someone who is diagnosed with PTSD has lost their memory of the traumatic incident and tries to testify about the specifics of that incident) in order to create a sufficiently strong reason to set aside the receiver’s obligation to defer. Individual people deal with trauma very differently and have very different reactions, some of which are epistemically illuminating rather than clouding. Ideally, for epistemic reasons, we would like to find testifiers who have processed their trauma and learned from it in constructive ways, but, for moral reasons, it is important to listen to those who are still processing through their trauma. Our account emphasizes these norms of interpersonal respect, and we should be wary of dismissing those with trauma, especially as that trauma is often informed by a failure of respect. We should also be wary of assuming that the testifier’s perspective is warped due to trauma, especially since those who are not in marginalized groups are 39 For instance, a woman who has experienced sexual harassment in the workplace might interpret an otherwise benign comment about her wardrobe to imply sexual content. That’s to say, she may be primed to interpret comments in such a way as a result of her experience with past trauma. 40 For an objection of this kind, see Táíwò (2020). Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 23 often blind to instances of marginalization, even after they have been pointed out by the people who have experienced them.41 A similar worry is that deference might be patronizing or fail to treat marginalized persons as full-fledged persons capable of withstanding and repelling disagreement or criticism. This worry assumes that to do justice to a marginalized person, the nonmarginalized person with whom they speak ought to treat them as epistemic equals. We have argued, however, that such testifiers and receivers are not epistemic peers. While thinking of another as capable of “holding their own” in response to criticism may be a show of respect among epistemic peers, the relation manifests as disrespect between a relative novice and a domain expert. We reject the notion that there is something patronizing about deferring to someone to whom deference is owed. Domain-experts are owed deference when the receiver is a comparative novice about the topic at hand. While we do not have a right to regulate other’s emotional responses, including their response to deference, the testifiers in our cases are providing an invitation to trust—an invitation to defer to them. It would be odd for a person to invite us to defer to them on an issue about which they are a relative expert and for our subsequent deference to be patronizing. Imagine a pulmonologist recommending an inhaler and then feeling patronized when I fulfill the prescription. There’s something incongruent about providing this invitation as a relative expert and being patronized upon its acceptance. There is also the problem that some receivers may be unaware of the testifier’s marginalized status. A testifier may have an invisible disability or may perhaps present as straight, cis, or gender-conforming when they are not.42 The result is that the receiver may accidentally treat the testifier as an epistemic peer, when they are not. In cases such as these, there exists a prima facie duty to defer, but the testifier is unaware that it applies. While these circumstances do not affect the duty itself, they will determine whether the receiver is blameworthy if they are unaware that the testifier’s marginalized identity is epistemically relevant to their testimony. In the innocent 41 As Sarah Schulman (2016) argues, supremacy mindsets (held by the privileged) can distort in very similar ways to trauma. Even if a person’s individual trauma presents a sufficient reason not to defer, there will be similar cases in which privileged persons will be equally epistemically corrupted due to supremacist mindsets. Think about, for instance, the white person that cannot accept any suggestion that their community might be racist. Additionally, trauma is not unique to the marginalized. As such, we should be especially wary of any suggestion that marginalized people as a class are unique in being subject to epistemically contaminating experiences. 42 Those with invisible marginalized status will not be subject to all of the same experiences as someone with a visible marginalized status. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 24 case—where the receiver could not have known, and it is not the case that they should have known that the testifier is a member of a marginalized group—the receiver is not blameworthy for treating the testifier as an epistemic peer. However, when the receiver is willfully ignorant—say the receiver is being purposefully insensitive to relevant identity conditions—the receiver is blameworthy for minimizing the role of identity-experience as it relates to their testimony. A final issue here is the problem of tokenizing a marginalized person as a spokesperson for their identity group. As Emmalon Davis (2016) notes, the harms of tokenization and other forms of identity-prejudicial credibility excess include (A) that the testifier is only valued for the information they can give and not valued as an individual, (B) that the testifier is presumed to have the same experiences as any other member of that group, (C) that the testifier is expected to only contribute knowledge and experiences that are recognized as “distinct” to their marginalized identity by dominant others, and (D) that the testifier is subject to pressure to deliver on behalf of the whole marginalized community. Our view avoids A, as we emphasize valuing and respecting marginalized individuals as individuals, not merely as dispensers of moral knowledge. It avoids B, as our view only pertains to experiences that marginalized individuals themselves have had qua members of that group, and we recognize that marginalized experience is heterogenous within a marginalized group. It avoids C, as our view does not entail that marginalized individuals should only be deferred to when giving testimony about their experience as a marginalized person. Our framework emphasizes the epistemic reason to defer in the case of expertise, which entails that we have good reason to defer to marginalized individuals who are testifying on other subjects (outside of marginalization) about which they are experts. The last harm, D, is more difficult to avoid. One way to read D is that it picks out the harm of epistemic exploitation. As Nora Berenstain (2016, 570) writes: “Epistemic exploitation occurs when privileged persons compel marginalized persons to produce an education or explanation about the nature of the oppression they face.” The epistemic labor of producing an explanation is often “unrecognized, uncompensated, emotionally taxing, [and] coerced” (570). To avoid epistemic exploitation, we warn nonmembers against making demands or insisting that a marginalized person explain their experiences. It is for this reason that we emphasize that testimony must be offered voluntarily and that nonmembers are responsible for educating themselves. However, even if the testifier is speaking voluntarily about their experiences as a member of a marginalized group, there may still be pressure to deliver for the entire marginalized community. There is often disagreement within different communities, and speaking for the entire group can sometimes risk harming other individuals within that group. At the same time, it is still important to listen to the Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 25 voices of those who are speaking authoritatively in their representative role to others outside of the group. Sometimes it may be difficult to determine when a marginalized person testifies from this stance, and we should be careful not to put pressure on a marginalized person to be speaking authoritatively when they are, for instance, trying to work out for themselves a part of their experience. But even in cases in which a marginalized person is working out a part of their experience, their testimony can often provide information about a common kind of experience, even if it is not universal. Even in this kind of case, deference may still be warranted, so long as the testimony is not taken to be representative of all who have this identity feature. Marginalized experience and systems of oppression are complex, and as such, a number of issues may arise in testimonial exchanges that are not covered in this paper. These complexities warrant further philosophical investigation. In conclusion, we have argued that we have a prima facie duty to defer—both epistemically and actionally—to marginalized people in their moral testimony about harms that relate to their identity when the receiver is not a member of that identity group. We have shown that, in the cases with which we are concerned, we ought to epistemically defer for both moral and epistemic reasons. Even if the receiver rejects the duty to epistemically defer, they should accept a prima facie duty to actionally defer. Receivers are not off the hook simply because belief is not warranted. Distinguishing epistemic deference from actional deference allows us to explain the appropriateness of a divergent response—epistemic deference without actional deference, or actional deference without epistemic deference—to cases that traditional accounts of moral deference must take as all-or- nothing. Our account might be extended to a broader variety of cases, and there is more work to be done in understanding the conditions under which the everyday varieties of moral deference are appropriate and even morally required. References Anderson, Elizabeth. 2012. “Epistemic Justice as a Virtue of Social Institutions.” Social Epistemology 26 (2): 163–73. Basu, Rima. 2019. “The Wrongs of Racist Beliefs.” Philosophical Studies 176 (9): 2497– 515. Berenstain, Nora. 2016. “Epistemic Exploitation.” Ergo 3 (22): 569–90. https://doi.org/10.3998/ergo.12405314.0003.022. Callahan, Laura. 2018. “Moral Testimony: A Re-conceived Understanding Explanation.” Philosophical Quarterly 68 (272): 437–59. Christensen, David. 2007. “Epistemology of Disagreement: The Good News.” Philosophical Review 116 (2): 187–217. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 26 Coady, David. 2006. “When Experts Disagree.” Episteme: A Journal of Social Epistemology 3 (1–2): 68–79. Darwall, Stephen. 1977. “Two Kinds of Respect.” Ethics 88 (1): 36–49. Davia, Cory, and Michele Palmira. 2015. “Moral Deference and Deference to an Epistemic Peer.” Philosophical Quarterly 65 (261): 605–25. Davis, Emmalon. 2016. “Typecasts, Tokens, and Spokespersons: A Case for Credibility Excess as Testimonial Injustice.” Hypatia 31 (3): 485–501. Enoch, David. 2009. “How is Moral Disagreement a Problem for Realism?” Journal of Ethics 13 (1): 15–50. ———. 2014. “A Defense of Moral Deference.” Journal of Philosophy 111 (5): 229–58. Foley, Richard. 2001. Intellectual Trust in Oneself and Others. New York: Cambridge University Press. Franklin, John Hope. 2005. Mirror to America: The Autobiography of John Hope Franklin. New York: Farrar, Straus, and Giroux. Fricker, Elizabeth. 2006. “Testimony and Epistemic Autonomy.” In The Epistemology of Testimony, edited by Jennifer Lackey and Ernest Sosa, 225–50. New York: Oxford University Press. Fricker, Miranda. 2007. Epistemic Injustice: Power and the Ethics of Knowing. New York: Oxford University Press. Frye, Marilyn. 2000. “Oppression.” In Gender Basics: Feminist Perspectives on Women and Men, 2nd ed., edited by Anne Minas, 10–16. Belmont, CA: Wadsworth. Goldman, Alvin I. 2001. “Experts: Which Ones Should You Trust?” Philosophy and Phenomenological Research 63 (1): 85–110. Hardwig, John. 1985. “Epistemic Dependence.” Journal of Philosophy 82 (7): 335–49. Hawley, Katherine. 2014. “Trust, Distrust and Commitment.” Noûs 48 (1): 1–20. Helm, Bennett W. 2015. “Trust as a Reactive Attitude.” In Oxford Studies in Agency and Responsibility, Vol. 2: “Freedom and Resentment” at 50, edited by David Shoemaker and Neal A. Tognazzini, 187–215. Oxford: Oxford University Press. Hieronymi, Pamela. 2008. “The Reasons of Trust.” Australasian Journal of Philosophy 86 (2): 213–36. Hills, Alison. 2009. “Moral Testimony and Moral Epistemology.” Ethics 120 (1): 94– 127. Hinchman, Edward S. 2005. “Telling as Inviting to Trust.” Philosophy and Phenomenological Research 70 (3): 562–87. Holton, Richard. 1994. “Deciding to Trust, Coming to Believe.” Australasian Journal of Philosophy 72 (1): 63–76. Howell, Robert J. 2014. “Google Morals, Virtue, and the Asymmetry of Deference.” Noûs 48 (3): 389–415. Jones, Karen. 1999. “Second-Hand Moral Knowledge.” Journal of Philosophy 96 (2): 55–78. Pearlman and Williams – Why You Ought to Defer Published by Scholarship@Western, 2022 27 Kelly, Thomas. 2005. “The Epistemic Significance of Disagreement.” In Oxford Studies in Epistemology, vol. 1, edited by Tamar Szabó Gendler and John Hawthorne, 167–96. New York: Oxford University Press. Lackey, Jennifer. 2008. Learning from Words: Testimony as a Source of Knowledge. New York: Oxford University Press. ———. 2011. “Testimony: Acquiring Knowledge from Others.” In Social Epistemology: Essential Readings, edited by Alvin I. Goldman and Dennis Whitcomb, 71–91. New York: Oxford University Press. Lewis, Max. 2020. “A Defense of the Very Idea of Moral Deference Pessimism.” Philosophical Studies 177 (8): 2323–40. Marušić, Berislav. 2017. “Trust, Reliance and The Participant Stance.” Philosopher’s Imprint 17 (17): 1–10. Matheson, Jonathan. 2015. “Disagreement and Epistemic Peers.” In Oxford Handbooks Online, 1-24. Oxford University Press. https://doi.org/10.1093/ox fordhb/9780199935314.013.13. Matheson, Jonathan, Scott McElreath, and Nathan Nobis. 2018. “Moral Experts, Deference & Disagreement.” In Moral Expertise: New Essays from Theoretical and Clinical Bioethics, edited by Jamie Carlin Watson and Laura K. Guidry- Grimes, 87–105. Cham: Springer. McGrath, Sarah. 2009. “The Puzzle of Pure Moral Deference.” Philosophical Perspectives 23:321–44. ———. 2011. “Skepticism about Moral Expertise as a Puzzle for Moral Realism.” Journal of Philosophy 108 (3): 111–37. Moran, Richard. 2005. “Getting Told and Being Believed.” Philosopher’s Imprint 5 (5): 1–29. ———. 2018. The Exchange of Words: Speech, Testimony, and Intersubjectivity. New York: Oxford University Press. Moss, Sarah. 2018. Probabilistic Knowledge. New York: Oxford University Press. Nickel, Philip. 2001. “Moral Testimony and Its Authority.” Ethical Theory and Moral Practice 4 (3): 253–66. Pohlhaus, Gaile, Jr. 2017. “Varieties of Epistemic Injustice.” In The Routledge Handbook of Epistemic Injustice, edited by Ian James Kidd, José Medina, and Gaile Pohlhaus Jr., 13–26. New York: Routledge. Schmitt, Frederick F. 2010. “The Assurance View of Testimony.” In Social Epistemology, edited by Adrian Haddock, Alan Millar, and Duncan Pritchard, 216–42. New York: Oxford University Press. Schulman, Sarah. 2016. Conflict Is Not Abuse: Overstating Harm, Community Responsibility, and the Duty of Repair. Vancouver: Arsenal Pulp Press. Simpson, Robert Mark. 2013. “Epistemic Peerhood and the Epistemology of Disagreement.” Philosophical Studies 164 (2): 561–77. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 2 Published by Scholarship@Western, 2022 28 Sliwa, Paulina. 2012. “In Defense of Moral Testimony.” Philosophical Studies 158 (2): 175–95. Strawson, P. F. 2008. “Freedom and Resentment.” In Freedom and Resentment and Other Essays, 1–28. New York: Routledge. Táíwò, Olúfẹ́mi. 2020. “Being-in-the-Room Privilege: Elite Capture and Epistemic Deference.” The Philosopher 108 (4). Thomas, Laurence M. 1998. “Moral Deference.” In Theorizing Multiculturalism: A Guide to the Current Debate, edited by Cynthia Willet, 359–81. Malden, MA: Blackwell. Wiland, Eric. 2021. Guided by Voices: Moral Testimony, Advice, and Forging a ‘We’. New York: Oxford University Press. Young, Iris Marion. 2014. “Five Faces of Oppression.” In Diversity, Social Justice, and Inclusive Excellence: Transdisciplinary and Global Perspectives, edited by Seth N. Asumah and Mechthild Nagel, 3–32. Albany: State University of New York Press. Zagzebski, Linda Trinkaus. 2012. Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief. New York: Oxford University Press. SAVANNAH PEARLMAN is a PhD candidate and an associate instructor at Indiana University–Bloomington. Her dissertation argues that non-group members have a moral and epistemic duty to defer to marginalized testifiers when they testify about the harms that they, themselves, have experienced. In addition to her research on normative epistemology, Savannah is interested in a philosophical treatment of mutual aid, solidarity, and community care. ELIZABETH WILLIAMS is a PhD candidate at Indiana University–Bloomington. Their dissertation argues that we are morally responsible for both our choices and our judgments, which do not reduce to each other, and that the patterns in our choices and our judgments reveal our character. In addition to their research on what it is to be morally responsible, they are interested in the treatment of accountability, attributability, and other forms of responsibility. 14386 Pearlman and Williams title page 14386 Pearlman Williams final format