I. Relevant-alternatives Contextualism

Consider the following pair of cases:

Assuming that Hannah is in the same epistemic position in both cases and that the bank is open, contextualists claim that both Hannah's knowledge ascription and Sarah's knowledge-denial are true. Contextualists explain this by claiming that the truth-conditions of knowledge claims vary with the context of the ascriber. According to relevant-alternatives contextualists, for “S knows that p” to be true, S's evidence must rule out all of the relevant alternatives to p, where different alternatives are relevant in different contexts. Truth also requires that p be true, that S believe that p, and that S's belief be properly based.Footnote ³ Prominent defenders of relevant-alternatives contextualism include David Lewis, Michael Blome-Tillmann, and Jonathan Jenkins Ichikawa (Lewis Reference Lewis1996; Blome-Tillmann Reference Blome-Tillmann2009; Reference Blome-Tillmann2014; Ichikawa Reference Ichikawa2017).Footnote ⁴ The set of alternatives to the proposition that the bank is open on Saturdays that's relevant in NARROW is narrower—it doesn't include the possibility that the bank has changed its opening hours—than the set of alternatives that are relevant in WIDE—which does include the possibility that the bank has changed its opening hours. Hannah's evidence rules out the set of relevant alternatives in NARROW, but not those that are relevant in WIDE. Thus, it's true to say that Hannah “knows” in NARROW but not in WIDE.

There are a few important points to keep in mind going forward: First, contextualism is a view about knowledge-ascribing sentences—sentences of the form “S knows that p”—and knowledge-denying sentences—sentences of the form “S does not know that p,” not about the metaphysical property of knowledge itself. The injustice that I'll be describing involves knowledge ascriptions and denials, rather than knowledge itself. Second, we can make mistakes in our judgments about the truth or falsity of knowledge-ascribing and knowledge-denying sentences. “S knows that p” might be true, yet someone might mistakenly judge it to be false. “S knows that p” might be false, yet someone might mistakenly judge it to be true. As we'll see, irrelevance-injustice involves these kinds of mistaken judgment about the truth and falsity of knowledge ascriptions and denials. Third, one cause of mistaken judgments about the truth-values of knowledge claims can be mistaken judgments about whether alternatives are relevant or irrelevant. When a speaker raises an alternative by saying, “what about this not-p alternative?,” it might be judged to be relevant when it is irrelevant, leading to mistaken knowledge-denials, or the alternative might be judged to be irrelevant when it is relevant, leading to mistaken knowledge ascriptions. As we'll see, irrelevance-injustice involves these kinds of mistaken judgments about the relevance and irrelevance of raised alternatives, which in turn leads to mistaken judgments about the truth-values of knowledge ascriptions and denials.

What makes an alternative relevant? This is a difficult and much debated question (see, for example, Austin Reference Austin1946; Dretske Reference Dretske1970; Goldman Reference Goldman1976; Stine Reference Stine1976; Lewis Reference Lewis1996; Blome-Tillmann Reference Blome-Tillmann2009 for discussion). Lewis's attempt to specify a list of rules that tell us when an alternative is relevant is the most comprehensive attempt to answer this question. Lewis's rules include the rule of actuality, which says that if some not-p alternative is actual, then it is a relevant alternative (Lewis Reference Lewis1996, 554–55), the rule of belief, which says that alternatives that the subject of the knowledge-claim believes, or should believe, are relevant (555), and the rule of resemblance, which says that alternatives that saliently resemble other relevant alternatives are themselves relevant (559).Footnote ⁵ What's important for our purposes is that speakers can change which alternatives are relevant by making certain conversational moves. Which conversational moves affect which alternatives are relevant is a matter of dispute. Lewis proposed the rule of attention, which says an alternative becomes relevant when a conversational participant brings it to the attention of other conversational participants (559). For example, when Laura raises the possibility of the bank changing its opening hours, it becomes relevant. The truth-conditions for the ascription change such that the possibility of changed hours must be ruled out if the knowledge-claim is to be true. This rule is now widely rejected because it makes “knows” too difficult to satisfy. It allows that I can always undermine the truth of knowledge ascriptions by raising a far-out, skeptical hypothesis (see, for example, Williams Reference Williams2001, 15; Blome-Tillmann Reference Blome-Tillmann2009, 246; Ichikawa Reference Ichikawa2017, 25 for discussion of this criticism). In the second part of section II, I'll explain that another reason to reject the rule of attention is that it leads to misdiagnoses of cases of irrelevance-injustice. Blome-Tillmann proposes a different view on which, in order for an alternative to be relevant, not only must attention be drawn to it, but it must also be taken seriously by the participants in the conversation (Blome-Tillmann Reference Blome-Tillmann2009, 249–56). Ichikawa has argued that this rule is subject to counterexamples (Ichikawa Reference Ichikawa2015). In the next section I’ll make some of my own suggestions for what makes an alternative relevant. In particular, I'll suggest that if an alternative is raised in an exercise of intellectual virtue, then it is relevant. I'll also suggest that if an alternative is taken seriously by someone who is competent in a domain, then it is relevant. These suggestions are motivated and supported by their ability to help us to diagnose cases of irrelevance-injustice and develop a theory of irrelevance-injustice.

Let me reiterate the important points from this section to take into the rest of the article: relevant-alternatives contextualism is the view that the truth of sentences of the form “S knows that p” requires that S rule out different alternatives in different contexts. Which alternatives are relevant can be a matter of conversational participants making certain conversational moves. People can make mistakes about the relevance and irrelevance of alternatives. These mistakes can cause further mistakes about the truth or falsity of knowledge ascriptions and denials. The project of the rest of the article is to show that these mistakes can be a site of injustice.

II. Irrelevance Injustice

Running parallel to conversations among contextualists about how speakers can affect which alternatives are relevant have been conversations about how speakers’ social identities and the power that attends to them affects which conversational moves a speaker can pull off. The literature on epistemic injustice, and testimonial injustice in particular, has been one important site of discussion. Canonically, Miranda Fricker argues that identity power operates in testimonial exchanges to produce testimonial injustice. Fricker defines social power as a capacity to control others’ actions (Fricker Reference Fricker2007, 13). Identity power is social power that depends on shared imaginative conceptions of some identity (14–15). In cases of testimonial injustice, a hearer has an identity prejudice that operates as identity power to prevent a speaker from conveying knowledge (28). Among the many others to discuss this theme are Karen Jones, Kristie Dotson, Gaile Pohlhaus Jr., José Medina, and Derek Egan Anderson (Jones Reference Jones, Antony and Witt2002; Dotson Reference Dotson2011; Pohlhaus Reference Pohlhaus2012; Medina Reference Medina2013; Anderson Reference Anderson2017). Another important site of discussion is the silencing literature. The core idea is that speakers can be silenced by being made unable to pull off certain speech acts. For example, Rae Langton has famously argued that women are silenced in their attempts to refuse sex as they are unable to secure uptake from men for their locutions because of the effects of pornography on men's ability to recognize women's intention to refuse (Langton Reference Langton1993). Among the many others to discuss this theme are Langton and Jennifer Hornsby, Langton and Caroline West, Ishani Maitra, Mary Kate McGowan, Alex Davies, and Samia Hesni (Langton and Hornsby Reference Langton and Hornsby1998; Langton and West Reference Langton and West1999; Maitra Reference Maitra2009; McGowan Reference McGowan2009; Davies Reference Davies2016; Hesni Reference Hesni2018).

Largely missing from these discussions is an investigation into how the two conversations interact (Ichikawa Reference Ichikawa2020 provides the one very recent exception that I'll discuss at length in section III). In this section I investigate how social identity and power affect whether speakers can successfully raise alternatives to relevance, how this affects the practice of ascribing and denying knowledge, and whether injustice lies therein. A distinctive form of epistemic injustice is revealed: irrelevance-injustice.

I'll proceed by examining three cases of irrelevance-injustice before arriving at a definition. Each case demonstrates a different form that irrelevance-injustice can take. Along the way I highlight points of difference from already recognized forms of epistemic injustice and some implications for how to understand epistemic injustice. Here's the first case:

BOARDROOM is a case of irrelevance-injustice. Miss Triggs raises a relevant alternative that is not taken seriously by her more powerful audience because of a prejudice they harbor against women that results in direct epistemic harm to her, and in distortions in the pattern of knowledge ascriptions. The alternative is relevant because Miss Triggs raises it in an exercise of her competence as a business analyst. This is in line with Georgi Gardiner's claim that it is dogmatically stubborn to ignore error possibilities that other competent people take seriously (Gardiner Reference Gardiner2020, 12). A further reason why Miss Triggs's alternative is relevant is that she raises it in an exercise of the intellectual virtue of open-mindedness. As the disposition to consider a variety of viewpoints in order to further one's understanding or to get at the truth, open-mindedness is the intellectual virtue that is closely associated with raising relevant alternatives (see Riggs Reference Riggs and Battaly2019 for an overview of open-mindedness). Miss Triggs plausibly displays open-mindedness when she raises the alternative that Punch's decline in profits is at least partly explained by structural changes in the economy. These facts mean that the alternative is relevant even if we suppose that structural changes in the economy are no part of the explanation for Punch's recent misfortunes, and so are not ruled in by virtue of being actual.

Miss Triggs is not taken seriously by her more powerful audience. The relevant form of power here is epistemic power. Dotson defines epistemic power as “differing ranges of privilege and underprivilege that's co-extensive with one's epistemic status” (Dotson Reference Dotson2018, 130). She defines epistemic status as “positive or negative assessments of one's epistemic position” (131). Epistemic status is domain-specific and can be legitimate, such as when positive epistemic status is conferred on a climate scientist with respect to climate change, or illegitimate, such as when positive epistemic status is conferred on a man with respect to business simply because he's a man. The privileges conferred by having high epistemic status include the ability to have one's claims taken seriously and being able to dismiss the claims of others. In Dotson's terms, having epistemic power enables one to seize the epistemological high ground, which is a “contrastive privilege that generates defense of one's claims and the ability and/or authority to challenge competing claims” (139).

Applied to BOARDROOM, the men have epistemic power, in that they enjoy positive epistemic status of being judged competent in the world of business. This gives them the privilege of being able to dismiss Miss Triggs's (relevant!) alternative and subsequently continue being (falsely!) ascribed knowledge.

A helpful way to see that it's not the mere failure to take an alternative seriously that makes for irrelevance-injustice, but rather failure to take an alternative seriously by those who possess power in the context, is by considering what happens when those who lack power fail to take an alternative seriously. Suppose that it's Mr. X. who raises an alternative, and that it's not taken seriously by Miss Triggs. It's hard to imagine, given her social identity and attendant lack of social and epistemic power in the context, that Miss Triggs not taking Mr. X. seriously prevents his alternative from being judged to be relevant and in need of ruling out before “knowledge” is ascribed.

Now to the harms and consequences. In being taken to lack the epistemic competence required to contribute to the investigation at hand, Miss Triggs suffers an affront to her epistemic agency that is the hallmark of epistemic injustice. This affront plausibly has further harmful consequences. Epistemically, Miss Triggs may suffer from diminished intellectual courage—the intellectual virtue of not ceasing to contribute to an epistemic activity too easily—due to having her confidence shaken by being so swiftly dismissed. Practically, Miss Triggs might miss out on professional benefits that come with being seen as a meaningful contributor to a workplace meeting.

The harms to Miss Triggs highlight one of several important differences between irrelevance-injustice and testimonial injustice, the injustice in which a speaker is not given the credibility they are due when testifying because of a hearer's identity prejudice. According to Fricker, the victim of testimonial injustice is harmed in their capacity “as a giver of knowledge” (Fricker Reference Fricker2007, 45). But whether Miss Triggs knows is never at issue. In raising an alternative, she is not attempting to give knowledge; rather, she's attempting to take part in an inquiry into a particular question. Irrelevance-injustice thus supports Mikkel Gerken's proposed broadening of epistemic injustice from involving a harm done to someone in their capacity as a knower (Gerken Reference Gerken2019, 2), as Fricker proposes (Fricker Reference Fricker2007, 1), to harm done to someone in their capacity as an epistemic subject.

The harms suffered by Miss Triggs are significant, but the downstream effects are perhaps even more significant. Dotson points out that epistemic power can be accumulated, which can in turn serve to uphold and exacerbate unjust power relations that set up the conditions for the perpetration of further injustice (Dotson Reference Dotson2018). BOARDROOM provides a stark illustration of this process of the accumulation of epistemic power. In irrelevance-injustice, alternatives that are in fact relevant and alternatives that are taken to be relevant come apart in a way that leads to distortions in the patterns of knowledge ascriptions that facilitate the accumulation of epistemic power and the stabilization of injustice. Because Miss Triggs's alternative is relevant, Mr. X.'s knowledge ascription is literally false. But because those with epistemic power don't take Miss Triggs seriously, they mistakenly judge that Mr. X.'s knowledge ascription is true. This is the distortion: Mr. X. is ascribed knowledge when he should not be, and the merely apparent truth of Mr. X.'s knowledge ascriptions is sustained by the judgments of those with epistemic power.

The significance of this distortion is tied up with the socially valuable functions of knowledge ascriptions. It has been suggested that knowledge ascriptions function to identify reliable informants (Craig Reference Craig1990), signal the appropriate end of inquiry (Kelp Reference Kelp2011), and track the epistemic norms governing assertion (Williamson Reference Williamson2000), belief, and practical reasoning (Hawthorne and Stanley Reference Hawthorne and Stanley2008). In being ascribed knowledge when he should not be, Mr. X. accumulates greater epistemic power by maintaining statuses and permissions that he is not entitled to: he's seen as a reliable informant, as someone who has reached the end of inquiry, and as someone whose epistemic position permits him to assert and act on the proposition. In these ways, Mr. X. benefits from the injustice Miss Triggs faces. These benefits serve to stabilize unjust power relations that set up the conditions for further injustice. Being judged to bear the positive epistemic status of a reliable informant, and as a place where inquiry can legitimately stop, means that the other conversational participants will be led to misplace their trust in Mr. X. Such misplaced trust causes the other conversational participants to be misled both about who “knows” and about what is known. This consolidates power in the hands of Mr. X., as in the eyes of his audience he is a reliable informant who is well placed to dismiss challenges to his knowledge claim. When the further relevant conditions obtain, such dismissals constitute further injustices.

How much power must an audience have if their reception of an alternative is to have the harmful consequences characteristic of irrelevance-injustice? It's hard to provide a precise answer, but we can at least say that the harms of irrelevance-injustice will be most stark when there are serious power differences between the audience of the alternative and the victim of irrelevance-injustice. Things will be more complicated when there is a less dramatic power differential. Some but not all of the harms might take hold, or the harms may be less severe.

In BOARDROOM, the men do not take Miss Triggs seriously because of a prejudice that they hold against women. A variation on BOARDROOM shows that prejudice is not a necessary condition of irrelevance-injustice. Imagine a version of the case in which everything is the same apart from the reason that the men do not take Miss Triggs seriously. Imagine that instead of not taking Miss Triggs seriously because of prejudice that women are not well placed to make claims about business and economics, they instead fail to take her seriously because they feel threatened by her intelligence. In this version of the case, the men use their epistemic power to work together to make it appear as if Miss Triggs's alternative is irrelevant and thereby undermine her ability to take part in the discussion on an equal footing.Footnote ⁷ The reason that this is an injustice, despite the absence of prejudice, is that the judgment of irrelevance and the harmful consequences that flow from it conform to a general pattern of oppression that unfairly harms some, and privileges others, in the same ways as those judgments that are caused by prejudice. Judgments of relevance that conform to the general pattern of norms that work to harm the oppressed by making it difficult for them to be ascribed knowledge, and that privilege the dominant by making it easier for them to be ascribed knowledge, are of concern even when conformity does not arise from prejudice. Following the Fricker paradigm, in which negative identity-prejudice is a necessary condition for epistemic injustice, will cause us to miss out on cases like this.Footnote ⁸

Consider a second case.

CLASSROOM is also a case of irrelevance-injustice, but in a different way from BOARDROOM. Whereas BOARDROOM involves the dismissal of a relevant alternative because of a prejudice against the speaker, in CLASSROOM the principal raises an irrelevant alternative that is taken seriously, by those with power in the context, because of a prejudice they harbor in the principal's favor.Footnote ¹⁰ The principal's alternative is irrelevant, and Rashaan is competent to judge it to be. Above I suggested that an alternative is relevant if it is raised by someone who is competent in the relevant domain. We can take this idea further by saying that we can also figure out if an alternative is irrelevant by deferring to those who are competent in some domain. Sometimes it might be unclear to some whether or not an alternative is relevant, yet extremely clear to competent others because they are experts on the matter at hand or because they occupy the relevant standpoint. It's been convincingly argued that we often have good reason to accept the testimony of those who are experts or who occupy particular standpoints (see, for example, Wylie Reference Wylie and Harding2004; Alcoff Reference Alcoff2005). The same reasons carry over from how we should respond to testimony to what are appropriate judgments of relevance. We can see this more clearly by adding to CLASSROOM that, in addition to the observational evidence possessed by Rashaan, after several years of being repeatedly targeted by racial stereotyping and being sent to the principal's office for minor offenses, he's become attuned to patterns of racial discrimination in his school life. Taking seriously the thought that particular social locations often come with epistemic advantages, when it comes to accessing knowledge about the racism faced by Black people, Black people are well positioned to know when and how it manifests because they experience it frequently and in a patterned way. If this is correct, then this gives us additional reason to take Rashaan seriously when he resists the principal's suggestion that perhaps he's seeing racism where there is none. Generalizing, we can look to those who occupy epistemically privileged standpoints to determine relevance or irrelevance.

The irrelevant alternative is judged to be relevant by those with epistemic power in the context because of a prejudice they have in favor of the principal. Both the principal and the assistant principal's epistemic power derives from their institutional as well their social positions. The principal has a positive prejudice toward himself that leads him to overestimate his own ability to detect racism based on his recent training. The assistant principal also has a positive prejudice in the principal's favor that leads him to overestimate the principal's racism detection capabilities, and that leads him to take the principal's alternative seriously. The assistant principal thereby confers positive epistemic status on the principal, enabling the principal to accumulate even greater epistemic power, and likely boosting the principal's own self-trust further. The role of the assistant principal mirrors the mechanism described by Jones by which dominant agents can scaffold one another's self-trust to excessive degrees and shore up one another's epistemic power beyond what is warranted, thereby creating the conditions in which epistemic injustice can flourish (Jones Reference Jones2012, 245–47). The principal is judged not only to be well placed to detect racism, but is judged to be better placed than Rashaan. The overestimation claim is plausible. Evidence suggests that the measures employed in antiracism courses are largely ineffective.Footnote ¹¹

CLASSROOM involves different harms and consequences from BOARDROOM. Unlike in BOARDROOM, the “knowledge” of the target of the injustice in CLASSROOM is at issue. This is again connected to the distorting effect of the reception given to the raised alternative on the economy of knowledge ascriptions. Because the principal's alternative is irrelevant, Rashaan's knowledge ascription remains literally true after the alternative is raised. But because those with epistemic power take the principal's alternative to be relevant, they mistakenly judge that Rashaan's knowledge ascription is false. This is the distortion: Rashaan is denied “knowledge,” when he shouldn't be, because an irrelevant alternative is taken seriously by those with epistemic power, when it shouldn't be. Consequently, Rashaan loses out on the statuses and permissions that come with being ascribed knowledge. He suffers an affront to his epistemic agency in being seen as an unreliable informant. He is perceived as lacking permission to assert and act on the proposition that he was a victim of racism. Being perceived to lack the permission to act, any actions that he takes will be open to criticism, potentially shaking his confidence in the righteousness of his protest and his determination to take action. Thus, alternatives being mistakenly taken to be relevant by those with epistemic power can undermine actions and place barriers in the way to achieving justice.

Like testimonial injustice, this form of irrelevance-injustice involves a testifier being unfairly judged not to “know,” but the mechanism that brings about this judgment is quite different. In testimonial injustice, a speaker is judged not to know because the audience harbors a prejudice against the speaker. In irrelevance-injustice, a speaker raises an alternative that is taken seriously when it should not be because of a prejudice in the speaker's favor, which in turn leads an audience to mistakenly judge that an alternative needs to be ruled out and to a mistaken knowledge-denial.

Looking at BOARDROOM and CLASSROOM side by side reveals an asymmetric ability to influence what's perceived to be a relevant alternative, which in turn affects the ease or difficulty with which the differentially socially situated agents are ascribed knowledge. Those who have their alternatives taken seriously can increase the size of the set of alternatives that others must rule out prior to being ascribed knowledge, but the reverse is not true. Consequently, there is an asymmetry in the amount of epistemic labor that differentially situated agents must expend, both in gathering evidence and in demonstrating their possession of such evidence, prior to being ascribed knowledge. Miss Triggs not being taken seriously reduces the amount of epistemic labor that Mr. X. must expend in order to be ascribed knowledge. The principal being taken seriously increases the amount of epistemic labor that Rashaan must expend in order to be ascribed knowledge. Irrelevance-injustice shows how the powerful can make epistemic life easier for themselves and harder for others.

Now for a third and final case.

DINING CLUB presents a third case of irrelevance-injustice which is importantly different from the other two we've looked at. In DINING CLUB, an alternative is taken seriously when it should not be,Footnote ¹³ but not because of any prejudice toward the speaker. Rather, the alternative is taken seriously because the audience harbors a prejudice pertaining to the alternative's content. As a first pass we can say that the audience harbors a content-prejudice in that they are prejudiced against a group that the alternative is about. I'll develop a more thoroughgoing definition below in the third part of this section (Condition (iii)). The diners hold the prejudicial belief that women are unreliable in their judgments about whether they have been sexually harassed.Footnote ¹⁴ They would've taken the alternative seriously whatever the identity of the speaker. This shows that irrelevance-injustice can be a matter of an alternative's content in the sense that it involves an alternative being problematically received because of its content, rather than because of who raises it. This marks another important difference between irrelevance-injustice and testimonial injustice. Testimonial injustice, as Fricker develops it, always involves an identity prejudice toward the speaker, DINING CLUB shows that this is not required for irrelevance-injustice.

Unlike in BOARDROOM and CLASSROOM, those who are harmed by the injustice that takes place in DINING CLUB are not themselves participants in the conversation. Rather, the reception given to the raised alternative causes distortions in the economy of knowledge ascriptions that harm people outside of the conversation. Particular alternatives being seen as irrelevant in one context can reinforce what alternatives are seen as relevant in other contexts. Imagine that having taken Charles's skeptical possibility seriously, those inside the conversation less readily ascribe knowledge to women they later encounter, who claim to have experienced sexual harassment, because they apply skeptical standards. The women are seen as lacking the statuses and permissions that come with being ascribed knowledge. The judgment that the women are unreliable informants reinforces the stereotype that women are unreliable testifiers about sexual harassment. They will be judged as not being positioned to act on the proposition that they were sexually harassed. Such judgments may feed into criticisms of actions, and such criticisms may stop the women in their tracks, undermining the pursuit of justice.

With these three examples in mind, we can say that irrelevance-injustice occurs when:

1. (i) S raises p as an alternative to some proposition already under discussion, q; and

2. (ii) p is not taken seriously when it is relevant, or p is taken seriously when it is not relevant, by those members of S's audience who possess epistemic power in the context;

3. (iii) for the reason that either S's audience is prejudiced toward S, or S's audience is prejudiced toward the content of what S says, or taking/not taking p seriously conforms to a general pattern of oppression.

I will now discuss each of the conditions listed in greater detail. In section III I will spend some more time comparing irrelevance-injustice with two other forms of epistemic injustice: testimonial injustice (Fricker Reference Fricker2007) and contextual injustice (Ichikawa Reference Ichikawa2020).

III. Irrelevance-Injustice and Other Forms of Epistemic Injustice

I'll finish up by noting some important differences and connections between irrelevance-injustice and other forms of epistemic injustice.

IV. A Previously Unrecognized Form of Epistemic Injustice

Our investigation into the relationship between social context, conversational context, and the mechanisms of raising alternatives to relevance emphasized by epistemic contextualists has revealed an unrecognized form of epistemic injustice: irrelevance-injustice. Irrelevance-injustice occurs either when a speaker advances an alternative that is not taken seriously when it should be, or when a speaker advances an alternative that is taken seriously when it should not be. Irrelevance-injustice causes distortions in the economy of knowledge ascriptions, affecting how much epistemic labor members of different social groups must undertake in ruling out alternatives prior to being ascribed knowledge, and thus whether they end up being ascribed knowledge. Given that knowledge ascriptions have socially valuable functions, irrelevance-injustice generates asymmetries in who gets to take advantage of these socially valuable functions.