1 Introduction

Consider the following cases, each featuring an epistemic norm violator (V) and a blamer (B) who has also violated a similar epistemic norm:

  • Unreliable Stereotypes: Steve (V) sees a number of news stories featuring terrorists who are Muslim and forms a stereotype that Muslims are dangerous, failing to consider that an extremely large Muslim population means there is an extremely low base rate of terrorists, which Stacey (B) blames him for. However, Stacey herself forms a stereotype that police are dangerous as a result of seeing a number of news stories featuring police killing civilians on the news, failing to consider base rates.

  • Uncritical Deference: Dimitri (B) criticizes his uncle Dominik (V) for believing claims from the state-owned RN News that other news networks are unreliable, without critically examining RN’s reliability. But Dimitri has formed his views that RN News is unreliable simply from watching CCN, without questioning CCN’s accuracy.

  • Echo Chamber: Ellie (V) is aware that news networks can be unreliable, so she tries to find multiple sources from which to verify what she hears. However, those independent sources are all known to be owned by the same company, Newspaper Corp. Eamon (B) blames her for this, but Eamon is also in an echo chamber as he accesses all of his news from one social media site, which filters out news sources that don’t share its political outlook.

  • Inconsistent Standards: Isaac (B) believes we should strongly defer to scientists even when it strongly seems to us that the scientists are wrong, so he blames Isla (V) for not trusting scientists that vaccines are safe and effective. However, Isaac himself doesn’t believe scientists when they say that blaming and shaming people who are vaccine-hesitant is counter-productive, making them even less likely to become vaccinated. He instead insists on blaming people like Isla, taking it to be ‘common sense’ this will improve vaccine rates.

  • Conceptual Creep: Cory (V) hears and accepts an argument that people living in poverty are victims of state-imposed violence. Coco (B) criticizes him for unknowingly inflating the concept of ‘violence’ and not noticing that the kinds of responses we have towards paradigmatic violent offenders are not the kind we should have towards a state that has failed to eradicate poverty. However, Coco herself believes that schools teaching socialism harms students, and doesn’t reflect on whether ‘harm’ is the most accurate way of capturing her objections to it being taught.

  • Motivated Reasoning: After many years of friendship, Molly (B) knows Maurice (V) well enough to know that he falls prey to motivated reasoning whenever the topic of immigration comes up. He appeals to otherwise unreliable sources, accepts new evidence favouring his position quickly while strongly scrutinizing disconfirming evidence for flaws, and places more weight on personal anecdotes rather than representative statistics. Molly blames Maurice for this. However, Molly herself falls prey to motivated reasoning on this topic too, without realizing. Maurice is unsure whether he exhibits motivated reasoning, but he knows Molly does, and so he objects to her blaming him for exhibiting motivated reasoning.

Let me acknowledge that it is possible to fill out the details of each case so that B or V may not have violated an epistemic norm, or so that the epistemic activities (e.g. believing, inferring, trusting) of the agents are not comparable. Norms admit of exceptions, and context matters greatly in determining when they should be followed, how much weight they should be given, and how reasonable it is to expect people to adhere to them. A host of background considerations might make one party’s (dis)believing epistemically justified despite the apparent surface similarity to the other agent’s criticisable beliefs.

Though it is possible to fill in the details of each case so that some parties are epistemically innocent, it is a familiar fact of life that even people who agree with us on many topics and share our conclusions do so for bad reasons, or while violating many epistemic norms. Kahan (2015a), for instance, famously found that people who believe in anthropogenic climate change do not, on the whole, understand more about climate science and scientists than people who are skeptical. This seems to make their belief in climate change somewhat precarious, epistemically speaking, and may undermine any entitlement to criticize climate sceptics’ understanding of the topic.Footnote 1

Let’s take it to be a stipulated feature then, that in each of these cases V has criticisably violated some norm, and B has criticisably violated the same norm or a relevantly similar norm in a relevant similar manner (so e.g. it’s not the case that one party has a good excuse). My question in this paper regards the status of B’s blame. I take it to be intuitive that in each of these cases, V can object to B’s blame. It seems to be a part of our practices to consider B to be doing something objectionable which someone who has not committed that norm could not be accused of. Blaming someone for a norm violation, while having committed a relevantly similar violation, seems to manifest a distinctive kind of fault. And the fault exhibited by the agents in these cases seems very similar to the fault of moral hypocrisy, which has recently received a lot of attention in the ethics of moral blame literature (Fritz & Miller, 2018; Herstein, 2017; Isserow & Klein, 2017; Lippert-Rasmussen, 2021; Piovarchy, 2020, 2023a, c, forthcoming; Riedener, 2019; Roadevin, 2018; Rossi, 2018; Todd, 2019; Wallace, 2010). If I commit adultery, I lack an entitlement to blame other people who are also guilty of adultery, for instance. If I do, the target of my blame is within their rights to respond, ‘Who are you to blame me?’ or ‘Look who’s talking!’ I am thought to lack what is known as the ‘standing’ to blame. The importance we place on standing to blame is reflected in our many cultural, literary and religious admonitions against hypocrisy (e.g. Alighieri, 2003; Marvin, 1922).

Could something similar be happening in the epistemic domain? Is each B in these thought experiments guilty of what we might call ‘epistemic hypocrisy’? Each B seems to be doing something criticisable, and each V seems entitled to object to B’s blame. There is a clear structural similarity between these cases and instances of moral hypocrisy, and moral hypocrisy is widely taken to undermine standing to blame (cf. Bell, 2013; Dover, 2019). Given these considerations, we have a strong prima facie case for thinking there is such a thing as epistemic hypocrisy, and that it is objectionable for reasons similar to those that make moral hypocrisy objectionable. This proposal would be stronger, however, if we had some idea of what counts as epistemic hypocrisy, the circumstances in which one lacks the standing to blame others for epistemic norm violations, a clear idea of what kinds of norms our blame is concerned with, and an account of epistemic standing to blame that was consistent with our account of moral standing to morally blame.

This paper fills this lacuna, having two main goals. One is to provide a careful examination of the kinds of phenomena at issue in the above thought experiments, the various things we might be referring to when discussing blame directed at beliefs, and to identify some choice points for philosophers interested in these interactions. To do this, I first home in on the target of V’s objection, identify a few related faults worth disentangling, and clarify what is at issue when discussing one’s lack of standing to blame. As we shall see, because there are multiple type of norms that beliefs can violate, which in turn can be appropriate targets of multiple types of blame, this means there are potentially a few different ways that we can (mis)understand standing to blame in the epistemic domain. It will thus be valuable to disambiguate a few things we are (and are not) referring to when talking of epistemic standing in the context of blame directed at beliefs, and to situate this topic among other related areas of inquiry.

The second goal is to provide a substantive account of the nature of epistemic hypocrisy. I argue that a commitment account of standing to blame, once suitably developed, can make sense of our responses to epistemic hypocrisy. Agents lack the epistemic standing to blame when they are not sufficiently committed to the epistemic norm they are blaming the other agent for violating. The notion of ‘sufficient commitment’, however, requires some elucidation, and some questions arise when it comes to considering our commitments to epistemic norms. Once these questions are answered, we have an account that not only gives us what we want from a theory of epistemic hypocrisy, it is one with some very particular strengths regarding our broader theories of blame and standing.

2 Clarifying the Investigation

Let me grant that in the above cases, each B might be manifesting a number of faults. One is inconsistency. Each B seems to have a belief about the importance of a particular epistemic norm and when it should apply, but also holds a belief that is inconsistent with that very norm. Another related fault is that they are failing to believe in accordance with their reasons. Given what B takes to be the case, or given their evidence of what is the case, they ought to be forming certain different beliefs to the ones they do.

But these faults don’t get at the nature of V’s objection. Suppose James sincerely asserts P in some circumstances, but then sincerely asserts not-P in other relevantly similar circumstances without noticing this inconsistency. To be sure, he manifests a fault and is open to blame. But suppose he now blames Stacey for holding inconsistent beliefs and making inconsistent assertions. Stacey now gains an additional, new complaint against James, that isn’t equivalent to what she would gain if she’d instead discovered James holds some other set of inconsistent attitudes. Inconsistency alone doesn’t capture the phenomenon we’re interested in.

Some readers might think we should object here not to B, but to V’s claim to have any legitimate objection at all. V’s objecting smacks of a tu quoque fallacy, or an attempt to deflect criticism from behavior that is, in fact, criticisable. After all, that B has committed a similar norm violation does nothing to show that V hasn’t committed a norm violation, or that V is not blameworthy. Given B’s blame and criticism are, in fact, fitting (i.e. they correctly represent their target), one might be tempted to think we should not give any legitimacy to V’s objection.

But this way of putting things misunderstands the interaction. When B objects, they are not saying this somehow changes the facts about their faults, or that there is nothing about their own performance that could warrant blame and criticism. Rather, they are objecting to B in particular doing the blaming. V could intelligibly say to some other norm complying blamer ‘What you say might be true, but they [B] are in no position to blame me’.

We can clarify our inquiry by examining different senses in which someone can be criticisable. To see what I mean, consider a case of a very young child believing badly. Suppose they see one swan and conclude that all swans are white. Their inference is unjustified, unsupported, or too hasty, and is criticisable in the sense that it does not conform to good reasons. But there is a sense in which it would be unreasonable to blame and criticise the child. Though one can say that the child is bad at making inferences (compared to adults, or relative to our standards of good reasoning), there is a relevant difference between a child who makes such an inference and a typical adult. Because the child lacks the kinds of capacities that make one a certain kind of agent, their believing badly is excused. In contrast, the adult is an epistemically responsible agent who is open to blame, praise and criticism.

The kind of criticism at issue is not a mere grading, or dry assessment of how well one reasons. To be epistemically irresponsible is to be eligible for certain kinds of negative responses on the basis of one’s poor performance, not just for it to be true that one instantiates certain less-than-perfectly rational properties (e.g. having inconsistent beliefs). When one violates epistemic norms without a good excuse, this reflects badly on them as an epistemic agent—as the kind of being who is capable of recognizing and responding to reasons—and warrants certain responses from other members of the epistemic community.

It is in this domain—regarding how to respond to norm violations—that questions of standing arise. Whereas questions of the appropriateness of blame and criticism depend primarily upon facts about our target (e.g. how strongly should they be expected to adhere to the norm, how egregious their violation was, whether there were mitigating factors), questions of standing to blame concern facts about the would-be blamer, and these have received scant attention in the literature on blame for epistemic norm violations (cf. Boult, 2021a). For example, some philosophers think that before one can appropriately (morally) blame a norm violator, one must have a sufficient stake in the matter, or be suitably related to the target (e.g. Herstein, 2017; Smith, 2007). If not, one should keep their nose out of other people’s business. More importantly for our purposes is that if one has manifested a fault that is relevantly similar to that which they are criticizing or blaming others for, they are not entitled to blame.Footnote 2

We can now clarify some responses B is (not) entitled to make. I believe that B can ‘criticize’ in the sense of simply identifying faults in V’s argument. That B is themselves guilty of a similar fault does not make V’s fault any less of a criticisable, and it seems that B retains the right to accurately identify features of the world. They do not have to remain silent altogether. But I take it that the kinds of responses in the above thought experiments involve something stronger. They involve expressing disapproval of the agent’s conduct, and this kind of response is what B is not entitled to make if they have committed a similar norm violation. I understand blame as an expression of disapproval, with manifestations of that disapproval coming in degrees, from (disapproving) criticism, to reproach, to rebuke, to ‘full-blown’ outrage. As is standard, I will focus on expressions of blame, and not take a stand on whether standing to blame also affects private (i.e. unexpressed) blame.Footnote 3

One last clarification concerns there being a number of ways moral norms and epistemic norms intersect or contrast with one another which might be relevant to our inquiry. The literature on doxastic responsibility and doxastic blameworthiness mostly focuses on moral blameworthiness for violating epistemic norms. The literature on moral encroachment is concerned with whether moral considerations can determine the relevant epistemic standards in any given situation e.g. whether high stakes of getting it wrong determine what counts as justification. The literature on pragmatism and evidentialism concerns whether extra-evidential moral or prudential considerations can make it so that one is required to believe against relevant epistemic standards (or, at least, that doing so is not blameworthy). Finally, a number of philosophers have recently argued there is a distinctively epistemic form of blame, one which targets violations of epistemic norms qua epistemic norms, i.e. where our blame is not ultimately grounded in any moral duties the agent had to adhere to said norms (Boult, 2021b; Brown, 2020a, b; Piovarchy, 2021).

Following Osborne (2021), I take ‘doxastic norms’ to refer to norms on belief, and ‘epistemic norms’ to concern norms on believing that have knowledge, warrant, and true belief as their aim. Given our interest in blaming practices, what sorts of things should we consider to be instances of ‘epistemic hypocrisy’?Footnote 4 Here’s two proposals we might consider. The first is to say that someone manifests epistemic hypocrisy when they direct blame (of any kind) towards someone who has violated an epistemic norm they themselves have violated. Someone who has violated an epistemic norm will lack the epistemic standing to blame others who violate relevantly similar norms. A difficulty with this approach is that in some contexts, we might require elaboration of whether moral blame and blameworthiness or epistemic blame and blameworthiness are at issue. The second proposal is to say there are ‘doxastic hypocrites’, with two sub-types of hypocritical blame: hypocritical doxastic moral blame, and hypocritical epistemic blame (given epistemic blame concerns violations of epistemic norms qua epistemic norms). On this proposal, only agents who engage in the latter are epistemic hypocrites. And rather than talking of the epistemic standing to blame (which may refer to one’s standing to engage in moral blame), we would instead be concerned with the standing to engage in epistemic blame.

While this is perhaps more precise, there are a few costs here. When we morally blame someone for violating an epistemic norm, we often experience our blame as being for violating that norm, even if that (epistemic) norm’s importance is ultimately grounded in something else (like moral considerations).Footnote 5 This is particularly relevant to standing because our ‘Who are you to blame me?’ response seems much more apt when it is the epistemic norm that the blamer has themselves violated. ‘Doxastic hypocrite’ risks confusion because, since multiple considerations might bear on what one ought to believe, two agents might both fail to believe what they ought in relevantly similar circumstances, but do so via violations of very different norms.Footnote 6 Additionally, whether hypocritical doxastic moral blameworthiness even exists will be disputed by strict evidentialists, who think that epistemic reasons are the only kinds of reasons for belief. The first proposal, in contrast, allows us to set that debate aside. One can intelligibly accuse others epistemic hypocrisy, leaving it as an open question whether we care about the epistemic norm violation qua epistemic norm violation, or due to some other consideration. As a result, I propose we go with the first option. Someone who blames others for violating an epistemic norm while having criticisably violated a relevantly similar (in some sense to be specified) norm themselves manifests epistemic hypocrisy.Footnote 7 Whether we are concerned with identifying epistemic hypocrisy vis-à-vis standing to engage in moral blame, or their standing to engage in epistemic blame (or both) will simply depend upon our interests. As we shall see, in this paper we don’t need to restrict our attention to either option, or specify what ultimately grounds our expectations that others comply with epistemic norms. A fortunate feature of the account I propose is that it is relevant for both kinds of blame, because my account of epistemic standing to blame is continuous with a popular account of moral standing to blame.

This rounds out the first goal of this paper: we have clarified the nature of V’s objection, the ways in which B and V are each criticisable or blameworthy, and disambiguated how our topic intersects with various other related debates and terminology concerning blame and beliefs. I shall now move onto my second goal: providing a substantive account of when agents lack epistemic standing to blame.

3 Commitment and Standing to Blame

I believe we can make sense of V’s objection and how agents can lack the epistemic standing to blame by considering what I’ll call a commitment-based account of standing (Friedman, 2013; Isserow, 2022; Piovarchy, 2023c, forthcoming; Rossi, 2018; Todd, 2019). On this account, agents lack the standing to blame other agents for a moral norm violation when they are not sufficiently committed to that kind of norm.Footnote 8 It is not the inconsistency between one’s epistemic conduct and blaming behavior, or professed beliefs, that is itself objectionable. The agent’s failure to comply with the norm is evidence that they are not, in fact, committed to the norm, but it is the lack of commitment that undermines entitlement to blame. Interestingly, this has the consequence that one can lack standing to blame even if one has never violated the norm, because one might nevertheless lack the relevant commitment e.g. one would freely violate the norm if given the chance, but simply not yet had the opportunity to do so (cf. Isserow, 2022). One can also retain standing to blame even if one has violated the norm, so long as one’s violation does not demonstrate a present lack of sufficient commitment. This might happen when e.g. the violation was extremely out of the ordinary, or the violation was a long time ago and the violator has since reformed their ways, now taking the norm seriously. I propose that, with some development, a commitment account of standing to engage in moral blame can help us make sense of what counts as epistemic hypocrisy.

One question which has not received sufficient attention by proponents is this: how should we understand the idea of commitment? Given we are interested in epistemic standing to blame, one tempting proposal is to understand ‘commitment’ as a rational commitment. Perhaps one has standing if one is rationally committed to the norm, meaning they recognize there are good reasons to comply with that norm, and appropriately weigh those reasons in their deliberations about what to believe. The problem with this is that someone might be rationally committed to a norm while having little motivation to follow it, resulting in them doing things which indirectly cause them to not comply with the norm. One might simply not care about the norm, intentionally performing actions which undermine their tendency to comply with the norm. They might become motivated by other factors to violate the norm regularly e.g. jealousy might cause them to dismiss information from a reliable source who happens to be a rival. Or perhaps they single-mindedly pursue self-interest, adopting any misinformation that helps them advance their career. If such a person then blames someone else for not being ‘rationally committed’ to a norm (e.g. not recognizing the importance of the norm), they seem guilty of a kind of false-pretense, and their blame doesn’t seem to be motivated by any concern for the norm itself. In short, such a person seems hypocritical.

An alternative proposal is to think of commitment in motivational and affective terms. Perhaps one has standing to blame if one cares about the norm, wants to comply with it, and experiences relevant affective responses, e.g. regret upon realizing they violated the norm. The problem with this proposal is that one can sincerely profess to care a lot about a norm and yet do terribly at conforming to it, and such agents seem like apt targets of charges of hypocrisy. Perhaps one asserts and believes that they should trust experts, but always finds themselves with an excuse for why this expert doesn’t need to be deferred to, or generating exceptions to the principle that aren’t well-supported. A proposal that leaves us with an account in which epistemic standing to blame is lost entirely as a result of one’s motivations and affective responses, with little regard for one’s actual epistemic performance, seems somewhat odd. Whereas B’s blame targets V’s epistemic activities and epistemic norm violations, B’s loss of standing which warrants the ‘Who are you to blame me?’ response from V results from B’s motivations, which seems somewhat disconnected from B’s own epistemic activities and norm violations.Footnote 9

I think each of these proposals gets something right. Commitment requires some kind of rational or cognitive component, in that agents need to be able to recognize and understand the norm itself, its importance, and the way in which someone who violates it is blameworthy. It is hard for me to blame someone for a fault that I do not understand or am not capable of recognizing. But an agent’s commitment also needs some kind of affective or motivational component, as if one lacked this, it is hard to see how one could muster up the emotional energy to truly blame someone for that norm violation.

I propose that commitment to a norm can be understood as internalization and endorsement of a norm.Footnote 10 Internalisation consists in experiencing the norm as inherently reason-giving (so one does not feel they should abide by the norm for norm-independent reasons e.g. to avoid punishment). Endorsement of a norm consists in being such that, upon reflection, one would agree the norm is valuable when one was cool, calm and collected and able to successfully reflect upon their beliefs and values.Footnote 11 These components can themselves be understood in terms of a certain set of cognitive, affective, and behavioural dispositions. Regarding themselves, people who have successfully internalised and endorse a norm will believe that they should abide by that norm, they will feel motivated to abide by that norm, and they will successfully abide by that norm in circumstances in which it would be reasonable to expect them to do so. Regarding others, agents who have internalised and endorse a norm will believe that others ought to abide by that norm, communicate that they find violations unacceptable, and abstain from behaviours which would encourage others to violate the norm.

Another thing we need to clarify is the notion of sufficient commitment. Whether someone is sufficiently committed to a norm is evidenced not only by how they respond to norm-relevant considerations, but also how they respond to those considerations relative to other considerations. When we are looking for evidence of commitment, the circumstances in which agents comply with the norm matter; if one fails to comply, it matters greatly why they failed to comply. Commitments, I take it, are not fragile. Attributing them to an agent requires the agent exhibit a certain sort of robustness in their conduct with regard to the norm. If one’s disposition to comply is too easily masked by, say, one’s other, more criticisable commitments, then this might give us reason to think the agent wasn’t sufficiently committed after all.Footnote 12

Having outlined how to think of commitments vis-à-vis epistemic norms, we now have a better grasp on when agents lack the epistemic standing to blame. This account makes sense of the thought experiments considered at the beginning of this paper: V’s objection seems legitimate because B’s own norm violation seems like good evidence that they are not sufficiently committed to the norm they are blaming V for violating.

4 Commitments and Compliance

Readers may have some questions regarding the relationship between commitment to a norm and compliance with a norm. Though there are many similarities between how we think of commitments to moral norms and commitments to epistemic norms, one potential difference is that epistemic norm violations are much less sensitive to an agent’s motivations than moral norm violations are. And this might generate a few worries.

Call the kind of control that one exhibits over most ordinary decisions and actions, or the control one manifests when one’s actions stem from one’s intentions in the right way, voluntary control.Footnote 13 Voluntary control seems to enable one to comply with a great many moral norms. Although there are some moral dilemmas where it is difficult to discern what the right option is, for the most part we are adept at identifying what (not) to do morally in most of our lives when we have the opportunity to commit wrongdoing. We know and succeed in not stealing from the shopkeeper, punching our co-workers, or assigning grades to students randomly to save time. We are adept at both translating our existing commitments into norm compliance through voluntary control, as well as reflecting on our commitments and exercising our voluntary control to instead make different choices if we think other considerations or commitments are more worthwhile. This kind of sustained action can, over time, even make it the case that one has changed what commitments they have.

But this doesn’t seem to be the case in the epistemic domain. As noted in the many analyses of doxastic involuntarism—the thesis that we lack direct voluntary control over our beliefsFootnote 14 (Alston, 1988; Bennett, 1990; Chuard & Southwood, 2009; cf. Frankish, 2007; Ginet, 2001; Weatherson, 2008)—one cannot comply with epistemic norms simply by dint of sheer will in the way that one can comply with moral norms. Admittedly, one can exert voluntary control and choose to do things which make it more likely one will come to comply with epistemic norms (e.g. enroll in a particular college course). But such indirect choices can only influence our beliefs so much; a course is going to be of little help if one is simply not intelligent enough to understand the course material, or conscientious enough to stick to an effective studying schedule. One might, despite their best efforts to improve their reasoning and epistemic habits, routinely get things wrong.

It thus seems that, while many moral norm violations can be grounded in an agent’s choices, relatively few epistemic norm violations can.Footnote 15 And so, one might be suspicious of modelling our understanding of epistemic commitments on our understanding of moral commitments. When someone has the opportunity to comply with moral norms (e.g. they were aware of the norm, and nothing was preventing them from complying) their failure to comply (or a pattern of failures) can be quite good evidence they were not sufficiently committed to the norm. But it is often not obvious to one when they are about to violate an epistemic norm like it is when one is about to violate many kinds of moral norms. Without voluntary control, the link between commitment to epistemic norms and compliance with epistemic norms seems much less reliable, and this might worry us for a few reasons. One is too many false positives; we’ll erroneously take many failures to be evidence of insufficient commitment when they are not. While this won’t necessarily show epistemic hypocrisy doesn’t exist, it might favour being overly cautious about calling out epistemic hypocrisy because it is inherently more risky than calling out moral hypocrisy (cf. O’Brien & Whelan, 2022; Piovarchy, 2023a; Piovarchy & Siskind, forthcoming). Another worry is that, while most agents can regain their moral standing to blame by simply trying to make better choices, many who lack the epistemic standing to blame might continue lacking it despite their (honest, but failed) attempts to improve their beliefs. This might strike some readers as unfair. Finally, there’s a risk that whatever the explanation is for why insufficient moral commitment undermines moral standing to blame, if it involves something like voluntary control, then this story won’t be able to be extended to epistemic standing to blame.

Whichever of these worries readers might have, they can be assuaged by realizing that compliance with moral norms and compliance with epistemic norms are much more similar than seems at first glance, in both directions. Start with how our moral behavior is much more like our epistemic behaviour than we assume. Although our moral actions (or omissions, or choices) are more subject to our will than our epistemic behavior is, meaning we have a kind of control over our actions that we might lack over our beliefs, a number of philosophers have noted that explaining our moral actions eventually ‘bottoms out’ in features of agents that said agents have no voluntary control over (Hieronymi, 2007; Jaster, 2020; McGeer, 2018; Smith, 2003). Suppose Stacey decides to shoot James today. To be sure, she exerts a kind of control over her decision. She has a sense that the choice is up to her, she experiences the choice as her own, with her wrongdoing stemming from her forming and acting upon a certain intention. But how did she decide to decide to shoot him? How did she form the intention in the first place? If that itself required a decision with an intention, an infinite regress looms. Moreover, how did she ensure that her intention, once formed, would manifest in action? Sometimes one can form an intention and yet not act on it. But choosing to act on the intention itself seems like another action, risking another regress.

When we reflect on how it is that Stacy responded to this reason rather than that reason, whether that response be recognizing, understanding, deciding, intending, or acting, we cannot point to some further kind of control or motivation that she exerted. But this means aspects of our agency that result in voluntary control are, themselves, not under our voluntary control. Rather taking this to show control is immaterial to thinking about moral responsibility, many philosophers instead propose that the relevant sense of ‘control’ required for moral responsibility and blameworthiness consists in exhibiting a certain kind of sensitivity to reasons. One does not need to also have voluntary control over that sensitivity. If we like, we can flesh out the notion of control in terms of capacities (Murray & Vargas, 2020; Piovarchy, 2022; Rudy-Hiller, 2017), or facts about nearby possible worlds (Fischer & Ravizza, 1998; cf. Jaster, 2020), or conditional analyses, but all that matters for us is that the way in which someone forms an intention to act cannot itself be accounted for in terms of some additional, prior form of control. And importantly, thinking of our agency and responsibility in this way looks a lot like the way our beliefs respond to epistemic reasons (Hieronymi, 2008, 2019). Even if we don’t exhibit voluntary control over our beliefs, we still manifest a certain kind of sensitivity to reasons which can reflect well or badly upon us, making us apt targets of certain responses from others (Vargas, 2020).Footnote 16

The second way that compliance with epistemic norms is more similar to compliance with moral norms than one might think comes from the realization that our epistemic life is much more strongly affected by our motivations than philosophers typically make out in their arguments. That we cannot exhibit voluntary control over our beliefs does not mean they are unresponsive to our motivations, desires and commitments. Readers are no doubt familiar with the very large body of psychological evidence suggesting that much of our epistemic activity is affected motivated reasoning (Kahan, 2015b; Lodge & Taber, 2013; Molden & Higgins, 2012). And this applies especially to our reasoning on the kinds of topics in which violations generate the strongest forms of blame, namely, topics with political and moral dimensions.

As an example of how this manifests, Levy (2020) notes there is a widespread ‘suspiciously convenient’ correlation between many people’s moral beliefs on what actions are called for, and their beliefs about what the consequences of those actions will be, even when their moral beliefs are (supposedly) justified without reference to what the consequences will be. For example, people who make a rights-based argument for why guns should be allowed tend to also believe that guns are much safer and deter crime more than people who believe we do not have a right to guns, even when neither side takes said consequences to be what justifies their position. People who think we have a moral duty to reduce greenhouse gas emissions to prevent harm to citizens of other nations also tend to think that this would have positive economic effects for their own nation, whereas people who think climate change is not that harmful to others think proposed economic policies would be disastrous domestically. Although this might not be suspicious if our beliefs about dispute-independent consequences was driving our beliefs about what actions are called for, Levy demonstrates that much of the time the effect moves in the other direction: we take the world to co-operate with our moral beliefs even when we have not formed those beliefs with any reference to the implicated facts.

One notable piece of evidence for how large an impact such tendencies have on our beliefs is that degree of political polarization increases as one’s level of education increases (Kahan et al., 2012; Pew Research Center, 2016). One would expect that there would be more convergence of politically relevant beliefs as level of education increases, since people with more education tend to have higher intelligence, training, and more access to higher quality epistemic resources, which are factors that should enable one to get closer to the truth on most matters. However, the opposite is observed. A leading explanation for this observation is that people’s higher intelligence makes it easier for them to discount or explain away any evidence or arguments that don’t fit their political identity. For example, Kahan et al., (2017) found that although numerical ability predicted ability to choose the right answer when assessing the efficacy of a skin cream, greater numerical ability resulted in more polarization when it came to assessing efficacy of a ban on carrying concealed weapons, despite the task using identical numbers to the skin cream task.

Indeed, Mercier and Sperber (2017) even famously argue that our ability to reason evolved primarily because of the fitness-enhancing benefits of devising arguments that are effective at persuasion. While reasoning is also fitness-enhancing because it helps us to collectively get at truth, the evolution of our reasoning ability was driven primarily by the need to know who to trust to avoid the high costs of misinformation, and the benefits conferred by enhancing one’s reputation.

Though one’s commitments don’t necessarily determine what beliefs one ends up forming, this body of research suggests that for many of our norm violations—particularly those on the kinds of topics that provoke heated emotions and blaming—our commitments, values, identity, desires, and incentives have a strong influence.Footnote 17 And since one’s commitment to an epistemic norm is not only measured by looking at one’s beliefs, attitudes and behavior directly related to that norm itself, but also comparatively (i.e. how much they prioritise that norm compared to other things that matter to them), many epistemic norm failures will be attributable to the agent not being sufficiently committed.

To bring this all home: epistemic standing to blame requires that one be sufficiently committed to the kind of epistemic norm one is blaming a target for violating. ‘Commitment’ has cognitive, affective and motivational components. One must not only believe the norm ought to be given a certain weight in relevant circumstances, and be capable of understanding how the norm applies in practice, one must have some level of motivation to comply with it. Whether any particular norm violation is evidence of insufficient commitment depends on a number of factors, including the agent’s prior behavior and success or failure rates, what was at stake, and what explains the agent’s failure to comply with the norm. Though it can seem we have a kind of control over our actions that we lack over our beliefs, this is no barrier to holding agents responsible, blameworthy, attributing to them certain commitments, or assessing the strength of those commitments. What matters is that we are agents who are sensitive to reasons, this sensitivity is what grounds assessments of our performance and eligibility for blame, and our sensitivity can reflect our commitments which, in turn, grant or undermine our standing to blame. To be committed to a norm is to give that norm priority in one’s conduct. It is to see that there are reasons favouring that thing and to experience those reasons as weighty. When one blames others for violating an epistemic norm, while themselves not being committed to that norm, one manifests epistemic hypocrisy, and their blame is objectionable.

One particular strength of this model is it gives us an account of epistemic standing to blame that is unified with a popular account of moral standing to blame.Footnote 18 The literature on epistemic blame has made good use of the literature on moral blame, attempting to show that said accounts can also account for our practices of blaming agents for their epistemic norm violations (Brown, 2020b; Piovarchy, 2021). However, at time of writing there is only one other account of epistemic standing to blame. Boult adopts a Scanlonian (2008) account of blame qua relationship modification to make sense of epistemic blame (Boult, 2021b), its significance (Boult, 2023) and the ‘business condition’ on standing to blame Boult (2021a). It is yet to be shown this account can make sense of epistemic hypocrisy, and Scanlon’s account of moral standing to blame faces some objections which any attempt would be likely to inherit (Wallace, 2010). Moreover, a Scanlonian account of moral standing has not received as much support in the moral blame literature as other theories. It thus seems that a commitment-based account of epistemic standing to blame currently has no competitors, and its coherence with our account of moral standing provides a source of support beyond its ability to account for our practices of blaming epistemic norm violators.Footnote 19

5 Implications

Having shown that a commitment account of standing can make sense of the phenomenon of epistemic hypocrisy, it is reasonable to wonder how much impact this should have on our actual practices going forward. Let me close by offering some reflections on what implications this may have for how and when we hold each other epistemically accountable.

Accusations of hypocritical blame are particularly effective at silencing criticism, drawing attention to the blamer’s own faults, and spurring dialogue about what our collective norms should be, when they should apply, and what is the best means of ensuring they are upheld going forward. The costs of being thought a hypocritical blamer provide a strong incentive to ensure that if we are going to preach it, we best also practice it, but this incentive only works if the risk of incurring said costs is credible: that is, if people are in fact disposed to call out hypocrisy and see ‘silencing’ as legitimate.

Norms against hypocritical blame in the moral domain are thought to ensure that we treat each other fairly (we don’t apply a standard to others that we don’t hold ourselves to; Fritz & Miller, 2018), that we relate to each other as equals (Piovarchy, 2020), that we don’t illegitimately present ourselves as being morally superior (Isserow & Klein, 2017) and that we all have a shared sense of what can reasonably be expected of one another. I believe similar things are true when it comes to epistemic norms. Pointing out that we are being epistemically hypocritical is likely to act as a strong source of pressure to either cease preaching, or start practicing. It ensures we treat others fairly, relate as equals, don’t present ourselves as epistemically better than we in fact are, and helps us coordinate on a set of shared epistemic expectations.

The last of these is particularly valuable. A number of epistemologists have argued that epistemic norms and evaluations serve important functions in society, namely, assisting with the coordination of the epistemic division of labour (Brennan, 2010; Goldberg, 2010; Graham, 2015). Given we are so reliant on one another’s epistemic expertise, shared standards are very important (Grimm, 2009; Piovarchy, 2023d). If we don’t agree on what it means to be ‘reliable’, then when you tell me your incompetent friend is reliable, I’m likely to be exposed to significant costs by trusting you (Dogramaci, 2012; Henderson & Greco, 2015).

Of course, there is often lots of disagreement about what norms we should endorse, when they should apply, and what counts as a violation. But that someone is being epistemically hypocritical manages to cut through much of this potential dialectical baggage: given the blamer is blaming, they already recognise this kind of behaviour is not acceptable (at least in the abstract). However, given they themselves have engaged in similar behaviour, this shows they are not (in practice) as committed to that norm as they take themselves to be. It’s fine if some people are unreliable so long as they are honest about their limitations, but by blaming the blamer presents themselves (by signalling; Shoemaker & Vargas, 2021) as someone who is committed to this kind of norm and can be counted on to uphold it. Their blame is thus a form of dishonest signalling, which we have an interest in calling out and minimising (Piovarchy, forthcoming).

Norms against epistemic hypocrisy are also likely to provide an important source of pressure towards something like humility. While humility is widely considered valuble, injunctions to be humble are often made in more supererogatory terms: it would be nice or praiseworthy to be humble, but most people aren’t particularly criticisable if they don’t actively cultivate it (so long as they don’t slide into being arrogant). In contrast, reasons to avoid epistemic hypocrisy are interpreted in more categorical terms, and we are likely to feel uneasy at the prospect of having violated them.

The thought experiments at the beginning of this paper help invoke this sense of unease. Fake news, stereotypes, vaccines, echo chambers, and immigration are hot-button, polarising issues, in which it is common to think that while we are responding to evidence and fulfilling our procedural episteimc duties, they are being epistemically vicious, selective, and lack evidence for their claims (cf. Piovarchy & Siskind, 2023e, §6). But realising that actually, there are similarities in the ways that we and they both reason about such topics (Levy, 2021) reduces some of the perceived distance between us and them. It also helps highlight how reasoning about these topics and having consistent principles can be hard, and requires effort. We know from psychology that we often suffer from things like overconfidence and an illusion of explanatory depth (where we overestimate our understanding of a topic and ability to explain it; Rozenblit & Keil, 2002) as well as mistaken beliefs about what our political opponents actually believe (Levendusky & Malhotra, 2016; Pettigrew, 2020). Having our attention drawn to the fact that we are doing something similar to our opponents, and just ending up at a different conclusion, can be an important corrective. It makes salient how we are failing to meet epistemic standards more often than we think, and that we might have misunderstood how exactly our opponents arrived at their beliefs (cf. Piovarchy, 2023b). It is much easier to engage constructively with someone when we think they are reasonably mistaken, rather than inexcusably wrong.

This, I believe, is also likely to take some of the heat out of discussions about each other’s epistemic conduct. Though blame can be an important source of motivation to comply with norms, it doesn’t follow from this that more blame is always better: too much can backfire, leading to spite and digging one’s heels in, particularly when it is seen as unfair (cf. Nikiforakis et al., 2012). While broad injunctions against blaming too much may have some merit, an alternative approach is to argue that inapt or standingless blame is, itself, blameworthy. (If you can’t beat ‘em, join ‘em). If we care about a norm, and want others to comply with it, but we recognise that blaming will be unproductive (because the reasons we’d cite as justifying blame weren’t enough to get us to comply with the norm either), we are likely to instead try to start reasoning with our target, convincing them that their conduct was subpar. This dialogic exchange is, in turn, makes us open to recognising the ways in which our conduct may need changing (Dover, 2019), and helps diagnose why we failed, why our existing habits or environment were insufficient to ensure compliance despite our recognition of the importance of the norm (in the abstract), and motivate us to consider alternative remedies (Piovarchy, 2023a).

6 Conclusion

People often seem blameworthy for their beliefs. In heated discussions featuring multiple parties blaming one another, it can be particularly objectionable when they blame us for committing the very kind of fault that they themselves are committing. A commitment account of epistemic standing to blame allows us to make sense of the conditions in which our objection is legitimate, and provides theory that is commensurate with our theory of the moral standing to blame. It is sometimes hard to know, in practice, when a blamer lacks standing. But such an account helps us make sense of and respond to the more obvious instances of epistemic hypocrisy, and can also give us tools for recognizing when we might be guilty of it ourselves.