2. Social and environmental epistemic virtues

I begin by explaining how, in general, virtue epistemology can function as a suitable middle ground between excessively individualistic and excessively socio-environmental explanations of epistemic successes and failures. This might seem like a dead-end. At least on the face of things, virtue epistemology is a paradigmatically individualistic approach to accounting for epistemic successes and failures (cf. Greco Reference Greco2020: 5.2). To see this, consider the common distinction between two forms of virtue epistemology: virtue reliabilism and virtue responsibilism (Axtell Reference Axtell1997; Montmarquet Reference Montmarquet and Battaly2018). While some epistemologists doubt that there is a neat distinction between virtue reliabilism and virtue responsibilism (Baehr Reference Baehr2011; Fleisher Reference Fleisher2017), or that this distinction exhausts the space of viable virtue epistemological approaches (Battaly Reference Battaly and Battaly2018a; Slote Reference Slote and Battaly2018), a useful distinction can at least be drawn between distinct types of epistemic virtues. In what follows, I will assume that a distinction between virtue reliabilism and virtue responsibilism can be drawn based on the approach each view takes toward epistemic virtues. However, this assumption serves principally to structure the discussion to follow, and no finding of substance here depends upon it.

Virtue reliabilism is a successor to process reliabilist approaches within epistemology. Process reliabilists take epistemically valuable properties—often including justification and knowledge—to be rooted in the reliability of the processes by which beliefs are produced (Armstrong Reference Armstrong1973; Goldman Reference Goldman1976, Reference Goldman and Pappas1979). In a similar vein, virtue reliabilists regard epistemic virtues as cognitive abilities that reliably contribute to the production of true belief and take epistemically valuable properties to be connected to the manifestation of such abilities (Greco Reference Greco1999, Reference Greco2009, Reference Greco2010; Sosa Reference Sosa2007). Thus, candidate reliabilist virtues include well-functioning color vision and inferential capacities. On the face of things, virtue reliabilism is a highly individualistic approach within epistemology. Virtue reliabilists emphasize epistemic successes that arise through the exercise of one’s epistemic abilities. In particular, virtue reliabilists tend to hold that only such epistemic successes—namely true beliefs that manifest one’s epistemic abilities—constitute knowledge. In this way, virtue reliabilists take appeal to epistemic virtues to be a way of advancing long-standing epistemological pursuits, including the analysis of knowledge.

In contrast, the ambitions of virtue responsibilists are typically more distanced from the traditional problems of epistemology.Footnote ¹ Whereas virtue reliabilism has much in common with earlier process reliabilist approaches to epistemic normativity, virtue responsibilism is more closely connected to virtue theoretic approaches in ethics (Baehr Reference Baehr and Battaly2018). Like virtue ethicists, virtue responsibilists construe virtues as excellent traits of character. In contrast to virtue reliabilists, who tend to define epistemic virtues in terms of their effects—namely the reliable production of true belief—virtue responsibilists treat the motivation toward truth as central to epistemic virtue (Battaly Reference Battaly and Battaly2018a; Montmarquet Reference Montmarquet and Battaly2018). Virtue responsibilists thus tend to count traits like intellectual humility and intellectual courage among the epistemic virtues. While one might in principle appeal to such traits within an analysis of knowledge, or in furtherance of some other traditional epistemological question, virtue responsibilists tend to articulate the epistemic virtues and vices as a way of accounting for epistemic successes and failures, without necessarily defining knowledge or other epistemic properties in terms of the virtues. Thus, for example, Heather Battaly (Reference Battaly2018b) has considered whether, and in what contexts, closed-mindedness might constitute an epistemic virtue. Similarly, Quassim Cassam (Reference Cassam2018, Reference Cassam2019) has articulated the vice of epistemic insouciance and described the role of this vice in causing various kinds of epistemic problems.

The discussion thus far suggests that both virtue reliabilism and virtue responsibilism are highly individualistic views. However, recent research in social and virtue epistemology suggests that these views need not be individualistic. Let us begin with virtue reliabilism. This view, as we have seen, focuses on epistemic abilities for true belief formation. Such a view appears individualistic insofar as both the epistemic abilities and the true beliefs in question belong to the same individual. But this description of an individualistic version of virtue reliabilism immediately suggests non-individualistic versions. First, if one countenances the existence of group beliefs, one might consider non-individualistic belief-forming processes and hence non-individualist reliabilist virtues (Kallestrup Reference Kallestrup2020; Palermos Reference Palermos2015, Reference Palermos2022). Second, even if there are no group beliefs, virtue reliabilism might move beyond individualism by allowing that the epistemic abilities that contribute to true belief formation need not be possessed solely by the same individuals possessed of the resultant true beliefs. For example, John Greco (Reference Greco2020, Chapter 5) has argued that the transmission of testimonial knowledge involves exercises of competences by both sources and recipients of testimony.Footnote ² Similarly, in a recent paper, I (Reference Harris2022) have argued for the existence of at least one “outward-facing” reliabilist virtue: testimonial effectiveness. Roughly, testimonial effectiveness is the ability to reliably transmit one’s knowledge to others through one’s testimony. The exercise of such a virtue would make one responsible, to some degree, for the epistemic states of others toward whom one directs it. While much work toward articulating a non-individualistic virtue reliabilism remains to be done, I focus in this paper on a non-individualistic approach within virtue responsibilism.

Like virtue reliabilism, virtue responsibilism seems on its face to be an individualistic approach. When we consider epistemic virtues like open-mindedness and intellectual courage, it is natural to think of these as traits related to an individual’s absorption and pursuit of evidence. However, also like virtue reliabilism, virtue responsibilism admits of various extensions beyond the individual. First, as Will Fleisher (Reference Fleisher2023) has recently argued, whether an inquiry manifests intellectual courage or a corresponding intellectual vice may depend on features of the social context, and in particular on how intellectual labor is distributed. Second, some collectives arguably possess responsibilist virtues and vices (Baehr Reference Baehr and Battaly2018; Baird & Calvard Reference Baird and Calvard2019; Fricker Reference Fricker2010; Lahroodi Reference Lahroodi2007, Reference Lahroodi and Battaly2018).Footnote ³ Third, some epistemic virtues and vices are plausibly directed at persons other than their possessors. Jason Kawall (Reference Kawall2002), for example, offers a pioneering exploration of other-regarding responsibilist epistemic virtues. His examples include honesty, sincerity, integrity, and creativity. The exercise of such virtues often yields improvements to others’ epistemic positions. Turning to vices, Jason Baehr (Reference Baehr2010) discusses epistemic malevolence, and I (Reference Harris2022) discuss the vices of dishonesty, communicative recklessness, communicative impatience and non-creativity, and epistemic insensitivity. Whereas Kawall’s virtues are oriented toward epistemically benefiting others, these latter traits are oriented toward epistemically harming others or, at least, are characterized by a lack of appropriate motivation toward epistemically benefiting others.

The just-mentioned virtues and vices are non-individualistic in that they are traits of a given person that are oriented toward the epistemic states of others. However, there is plausibly another non-individualistic role for responsibilist epistemic virtue, beyond the promotion of positive epistemic states in others. In short, some responsibilist epistemic virtues are plausibly oriented toward structuring the wider epistemic environment to facilitate epistemic success. Neil Levy, writes, for example, that “it is largely by contributing to a knowledge-conducive epistemic environment that the virtues lead to a better life” (Reference Levy, Alfano, Klein and Ridder2022b, p. 113). Notably, in light of the present context, Levy’s contention occurs within an argument to the effect that virtue epistemology is, by and large, too individualistic to provide helpful epistemic guidance. In particular, Levy argues that the cultivation of virtues like open-mindedness would often lead us astray, given our limitations and the pollution that exists in our epistemic environments. Given these complications, the most helpful epistemic virtues will be those that facilitate truth-conducive reliance on others. In a reply to Levy, Steven Bland expands on this basic point. Bland suggests that we ought to recognize a category of embedding virtues, “whose value consists in their tendency to promote benign cognitive environments” (Reference Bland, Alfano, Klein and de Ridder2022, p. 132). Just as one may shape one’s social epistemic environment in a way regulated by virtue, one may also shape one’s social epistemic environment in a way that reflects a vicious lack of regard for the truth. Consider, as an example, the corporate CEO who surrounds himself with yes-men, rather than those that can be counted on to supply truth. Or consider the conspiracist who tolerates and even encourages the spread of outrageous falsehoods in order to maintain social standing within the conspiracist community.

Levy and Bland are mainly focused on the role of intellectual virtue in structuring our social epistemic environments and how we engage with such environments. It is also worth noting, however, that responsibilist epistemic virtues have a place in a non-individualistic virtue epistemology, even beyond the regulation of our epistemic interactions with others. Consider cognitive offloading, that is, the exploitation of extra-neural structures to facilitate the completion of cognitive tasks. While cognitive offloading might take a social form—as when we rely on others to retrieve or store information—cognitive offloading need not be interpersonal. Consider two parallel examples. In the first, a server at a restaurant internally memorizes the orders from everyone at a given table. In the second, a server simply jots down the orders on a notepad. The latter utilizes a form of cognitive offloading to simplify the task of recording orders. Servers often display remarkable memorization abilities that are in many instances bolstered by tricks for enhancing memorization. However, a server lacking such abilities, who nonetheless insisted on attempting to remember orders internally, would thereby display a sort of hubris. In contrast, the appropriate reliance on notepads or other external tools demonstrates a form of intellectual humility. In addition to one-off or otherwise short-lived reliance on features of the environment, one might more systematically structure one’s environment to facilitate sustained cognitive successes. Appropriate reliance on such external aids often reflects the sort of concern for the truth that is central to responsibilist epistemic virtue.Footnote ⁴ Notably, while some (Pritchard Reference Pritchard and Baehr2015) have proposed to use such examples to motivate an extended virtue epistemology based on the hypothesis of extended cognition (Clark & Chalmers Reference Clark and Chalmers1998), no such controversial commitments are required to recognize that reliance on external features of the environment to facilitate completion of cognitive tasks may manifest epistemic virtue.

Let us take stock. In this section, I have introduced virtue reliabilism and virtue responsibilism. While both such views appear individualistic at first glance, we have seen that there are various ways in which such views are consistent with a non-individualistic account of responsibility for epistemic successes and failures. To these, we might add that—to the extent that development of various intellectual traits and abilities are traceable to the instruction of others, or to one’s social position more generally—purely individualistic explanations of successes arising from such virtues are inadequate. I am mainly interested in non-individualistic virtue responsibilism here so, to conclude this section, let us summarize the ways in which such a view can incorporate non-individualistic elements. First, social factors can influence the development—or not—of responsibilist epistemic virtues. Second, collectives can arguably possess responsibilist epistemic virtues and vices. Third, some responsibilist epistemic virtues and vices are oriented toward the epistemic states of others. Fourth, some epistemic virtues and vices guide one’s reliance on the surrounding (social) epistemic environment. Arguably, these include familiar virtues and vices like humility and arrogance, alongside more specifically environmentally oriented traits like Bland’s embedding virtues. In the next section, I draw attention to an overlooked trait that belongs to this latter category, but that differs in noteworthy respects from the traits considered thus far.

3. Epistemic exactingness

As we have seen, some responsibilist epistemic virtues shape their possessors’ interactions with features of their social epistemic environments. For example, Levy Reference Levy2024) stresses the value of intellectual interdependence, understood to involve a disposition to appropriately rely on, and defer to, others. Similarly, those that possess intellectual humility will refrain from placing undue weight on their own judgments, when those judgments conflict with the claims of others better positioned to know. These virtues need not simply shape how one reacts to the social evidence available at a time, but may lead individuals to develop better networks of social feedback and better routines for accessing this feedback. For example, the intellectually virtuous person might strive to befriend individuals with different academic or professional histories, or life experiences more generally, partly to develop a better understanding of how the world works.

The preceding paragraph suggests two basic roles for social epistemic virtues: to shape one’s social epistemic networks and to regulate one’s treatment of the evidence received from those in one’s network. But there is, I now suggest, a third role for social epistemic virtue. To introduce this role, consider an example:

That Mark always initially believes the mechanic is a sign of his gullibility, but gullibility is not the only epistemic vice he exhibits. Not only does Mark believe the mechanic when the mechanic lies to him, Mark fails to confront the mechanic concerning the past lies of which Mark is aware through the advice of his friends. In effect, Mark fails to stand up for himself, instead allowing his mechanic’s past lies to pass without comment. Additionally, by returning to the same mechanic, Mark effectively rewards the mechanic’s lies. In doing so, he makes himself an easy mark, increasing his chances of being exposed to additional misleading testimony in the future. Moreover, Mark fails to punish, and indeed incentivizes the mechanic’s lies, thereby increasing the chances that others will be lied to in the future.

Mark is lacking in a virtue I will label epistemic exactingness. The epistemically exacting agent is, among other things, appropriately concerned with the quality of the evidence which others provide and is disposed to insist on quality in this regard. A more epistemically exacting agent in Mark’s position would thus act so as to discourage the mechanic from being deceptive. Epistemic exactingness is thus not manifested in the establishment of a social epistemic network or in how one responds to evidence proffered by those within the network. Rather, the exacting agent is motivated to act so as to encourage quality evidence from those within the network. In this way, the exacting agent is disposed to regulate the epistemic conduct of others, rather than merely deciding which others to interact with.

In Easy Mark, Mark is exposed to low-quality testimony that is traceable to the insincerity of the mechanic. But the quality of testimony does not depend solely on the sincerity of the testifier. Sometimes, testimony is lacking in quality not because the testifier is insincere, but because the testifier has failed to exercise competence in the lead-up to the testimony. For example, we might imagine a corporate executive who regularly makes bad marketing decisions because the company’s marketing researchers consistently base their recommendations on weak evidence. Let us suppose that, despite having the power to do so, this executive fails to pressure the researchers to produce better-evidenced recommendations. Such an executive, like Mark, fails to manifest proper concern for truth and thus tolerates low-quality testimony from members of their epistemic network.

It is commonplace in the epistemological literature on testimony to recognize that liars and the incompetent typically do not produce testimony that yields knowledge (Graham Reference Graham2000; Greco Reference Greco2020, p. 190), unless perhaps the quality of the testimony is bolstered by some external factor (Goldberg Reference Goldberg2007, p. 226; Greco Reference Greco2020, pp. 139–140). Thus, for example, Regina Rini (Reference Rini2020) suggests that one reason for which we can generally trust public figures not to brazenly lie is that such persons are aware that their words are likely to be recorded, and lies are likely to be punished. Similarly, punishments for perjury bolster the epistemic value of testimony within contexts in which such punishments are in effect. The present suggestion is that one’s own epistemic virtues or vices are sometimes important factors in influencing the quality of others’ testimony. Even an individual disposed toward dishonesty might suppress this tendency in certain company. Likewise, one who often speaks with confidence not warranted by the available evidence might moderate this tendency—either by refraining from certain claims or by seeking more evidence—when in conversation with epistemically exacting persons.

We have already seen some illustrations of a vicious counterpart of epistemic exactingness. We might say that Mark and the executive manifest excessive epistemic leniency. On the other hand, it is possible to be overly demanding when it comes to the testimony of others. Consider one who regularly interrupts others’ stories to insist on small details, thereby failing to appreciate the main points of the stories themselves. Such a person has the vice of epistemic pedantry. The epistemic pedant risks missing out not only on the social function of others’ testimony, but also certain epistemic functions that can be fulfilled by imperfect testimony.

The virtue of epistemic exactingness can be usefully described in terms of norms. Epistemologists have long suggested that assertion is subject to certain norms. The epistemically exacting agent, we might say, is disposed to encourage others to adhere to certain epistemic norms of assertion, if they are able to do so. What epistemic norms would the epistemically exacting person enforce? At a minimum, I suggest that the epistemically exacting person would be disposed to enforce a belief norm and a justification norm of assertion. The failure to do so is what suggests a lack of exactingness on the part of Mark and the corporate executive, respectively. Whether this suggests that the epistemically exacting person would enforce a knowledge norm of assertion (Williamson Reference Williamson2000), a truth norm of assertion that speakers can attempt to follow by asserting only what they justifiedly believe (Weiner Reference Weiner2005), or some further possibility, is a question I will not attempt to address here. Rather, it is worth noting that, while epistemic exactingness can be usefully illuminated in terms of a disposition to enforce the norms of assertion, assertion is not the only activity plausibly regulated by exactingness. Rather, the epistemically exacting person will be inclined to regulate certain other epistemically significant activities, plausibly including the sharing of news, photos, and other content on social media. The norms of such activities have received less attention from epistemologists than assertionsFootnote ⁵ but, given how much (mis)information is spread in this way, social media is an important arena for the exercise of exactingness.

Thus far, I have focused on the connection between epistemic exactingness and the enforcement of norms related to information sharing, including assertion. There are, however, closely related traits, which may or may not be understood as constitutive of epistemic exactingness, that are worth mentioning in this connection.Footnote ⁶ For example, epistemologists have devoted considerable attention to the norms of belief (Chan Reference Chan2013; Gibbons Reference Gibbons2013; Simion et al. Reference Simion, Kelp and Ghijsen2016; Williamson Reference Williamson2000) and inquiry (Flores & Woodard Reference Flores and Woodard2023; Friedman Reference Friedman2020; Haziza Reference Haziza2023). Notably, it has been argued that some such norms are social norms, and are reinforced in part by criticism and sanctions on those that violate them (Graham Reference Graham, Henderson and Greco2015). It is thus tempting to understand epistemic exactingness somewhat more broadly than I have thus far, such that an epistemically exacting person is one who is motivated to enforce a wide range of epistemically relevantFootnote ⁷ norms on information sharing, belief, action, and inquiry.

While I have no decisive argument against this more inclusive understanding of epistemic exactingness, I maintain a relatively narrow understanding of epistemic exactingness here. There are a few reasons for this. The first is pragmatic. One of my main aims here is to enrich social virtue epistemology by highlighting a class of virtues that involve the motivation to promote epistemic goods by regulating the conduct of others. This task is best pursued, I think, by maintaining a relatively narrow focus. Secondly, when it comes to the enforcement of epistemic norms of inquiry and belief, tricky questions arise concerning the boundary between a virtuous concern for epistemic goods and a vicious meddling in the mental lives of others. It is commonly thought that individuals are entitled to their beliefs,Footnote ⁸ but not to just any actions that might be predicated on these beliefs. Similarly, one might think, the epistemically exacting person is motivated to intervene only when deficiencies in beliefs or inquiries issue in damage to the epistemic conditions of others or to the epistemic environment. I do not take this consideration to be decisive, but I think it at least complicates matters sufficiently that it is best to adopt a relatively narrow construal of epistemic exactingness here. This narrow construal of epistemic exactingness is, it should be noted, consistent with the existence of closely related virtues that involve the motivation to enforce further epistemic norms.

To conclude this section, it is worth emphasizing that, as the discussion in Section 2 anticipated, expressions of exactingness need not be purely self-interested or individual-centric. First, although our initial examples involved agents whose lack of exactingness was chiefly epistemically detrimental to themselves, an exacting agent might well be motivated to safeguard or improve the epistemic states of others. In general, responsibilist virtues are tied to appropriate concern for epistemically valuable states like knowledge, and not necessarily to the knowledge of any particular individual. Consequently, responsibilist epistemic virtues can manifest in the promotion or defense of others’ valuable epistemic states (Harris Reference Harris2022). Epistemic exactingness is well-suited to playing this role. Thus, for example, the epistemically exacting agent may be motivated to discourage others from sharing misinformation or gossip, even if this agent is certain that she will be unmoved by such misleading testimony. Even if an exacting agent in Mark’s situation was not themselves gullible, and thus at no risk of being deceived by the mechanic, that agent might be motivated to confront the mechanic as a means of protecting future customers. Similarly, Sandy Golberg (Goldberg Reference Goldberg2007, Chapter 8, Reference Goldberg2008) offers an example of such an agent in the form of a mother who safeguards her child’s epistemic states by restricting the access of would-be tellers of falsehoods to the child.

Second, although I have thus far focused on individual-centric cases of epistemic exactingness and its lack, these traits may—like those described in Section 2—also manifest at the collective level. I return to this point in Sections 5 and 6.

4. Epistemic exactingness and self-respect

As I noted in Section 2, the responsibilist virtues resemble in some respects those elucidated within virtue ethics. One consequence of this point is that there are many close parallels between responsibilist epistemic virtues and moral virtues. For example, there are both intellectual and non-intellectual forms of courage. In the case of some virtues—especially honesty—it is unclear whether there is a single moral-epistemic virtue, or two distinct virtues for two distinct domains of normativity. Given the close connection between responsibilism and virtue ethics, it is worth considering, if only briefly, the relation between epistemic exactingness and the more familiar virtues of virtue ethics.

At first glance, epistemic exactingness might seem distant from the sorts of traits explored in virtue ethics. Epistemic exactingness serves to regulate the quality of others’ testimony in a way that is, as the examples given in the previous section suggest, often practically beneficial to the exacting agent. For this reason, epistemic exactingness might seem to lack the other-regarding nature of many moral virtues. However, it would be a mistake to conclude on this basis that epistemic exactingness is radically divorced from the familiar virtues of virtue ethics.

First, as I noted near the end of the previous section, epistemic exactingness need not involve merely the motivation to improve or safeguard one’s own epistemic condition. Rather, the exacting agent might be principally motivated by concern for the epistemic conditions of others. Second, while many moral virtues—generosity and compassion, for some examples—plausibly involve regard for others as a central component, some moral virtues are best understood as principally self-regarding. Prudence, for example, is a paradigmatic self-regarding virtue. Some moral virtues, while self-regarding, are also other-involving. Consider an example. Self-respect—which I take to include having proper regard for one’s own status as a person, for one’s moral standing, and for one’s interestsFootnote ⁹—is not only a virtue. It is arguably an especially central virtue, the possession of which is required for the possession of other virtues. If one does not value oneself, then the sorts of altruistic acts typically understood as manifesting great generosity or compassion do no such thing. If one does not respect oneself, then sacrifices made in the interests of others are hardly sacrifices at all (Hill Reference Hill1973, p. 104).

Those with self-respect not only regulate themselves—for instance in terms of how they think about their own interests—but are also disposed to regulate the conduct of others insofar as it bears upon themselves. There are certain kinds of treatment that, given the choice, no self-respecting person will abide (Boxill Reference Boxill and Dillon1994; Taylor Reference Taylor and Dillon1994). Drawing on Kant, Robin S. Dillon offers the following helpful discussion of what self-respecting persons demand of others:

In short, self-respect is a virtue that does not merely regulate its possessor’s own conduct. Rather, the self-respecting person will, given the choice, refuse to tolerate certain types of behaviors from others. In this way, self-respect is both self-regarding and other-involving. Epistemic exactingness has a similar structure. The epistemically exacting person has appropriate concern for her own epistemic condition, but this concern manifests—at least under suitable conditions—in the regulation of others’ testimony and other epistemically significant conduct.

Some readers will have noticed a degree of hedging in the above remarks concerning the sort of treatment that those possessed of self-respect and epistemic exactingness tolerate. Self-respect and epistemic exactingness regulate the conduct of others other things being equal, given the possessor’s choice, under suitable conditions, and so on. In the next section, I articulate an important condition on the regulatory effectiveness of epistemic exactingness.

5. Epistemic exactingness and power

In Section 3, I gave examples of people who—due to a lack of exactingness—tolerated low quality testimony from others. Yet the lack of epistemic exactingness is not the only reason for which one might fail to properly regulate the testimony and other evidence furnished by others. Even those who have an appropriate concern for their own epistemic conditions might fail to regulate others because they lack the power to do so. Let us turn to some examples.

Above, I briefly presented a case involving a corporate executive whose lack of epistemic exactingness allowed for market researchers to make poorly evidenced recommendations. This is a case of an individual holding power over others who nonetheless fails to enforce epistemic norms over the quality of offered evidence as a consequence of individual traits. But, in other cases, individuals possessed of epistemic exactingness—who would for example insist on truth if positioned to—simply lack the power to do so effectively. For example, consider an employee who suspects that a colleague is routinely dishonest in professional contexts, resulting in that employee being regularly fed bad information. Even if this employee is exacting, he may recognize that he lacks the standing to effectively pressure the colleague toward honesty. More generally, epistemic exactingness does not guarantee that members of one’s network offer high-quality testimony or other social evidence. Epistemic exactingness involves the motivation to do so, the effectiveness of which will depend on the agent’s ability to influence others. This is typical of the responsibilist virtues, which—at least according to some prominent virtue epistemologists (Baehr Reference Baehr2011; Montmarquet Reference Montmarquet1993)—are more closely connected to motivations than to effects.

Let us turn to a more complex example. Ordinary citizens often bemoan the lack of honesty in politics. We may notice even our preferred candidates distorting or concealing the truth. Some of these instances can be written off as necessary simplifications. In other cases, political speech serves more of a social than epistemic function. It would in some cases be pedantic to challenge the details of a politician’s anecdote, where that anecdote serves principally to signal commitment to certain values. In other cases, however, politicians—even those we prefer—straightforwardly lie. Suppose one notices that one’s preferred candidate for national office has lied about a significant matter within a debate. A common reaction one might have is to think to oneself that, while one’s preferred candidate has lied, the other candidate is worse. Does such a reaction betray a failure of epistemic exactingness? It need not. Most ordinary citizens have no power to regulate the behavior of high-profile politicians. This might be due in part to a lack of access—in particular the inability to reach politicians with one’s complaints. It might also be due to the nature of politics. No single individual’s attempts to regulate politicians’ behavior—whether by withholding donations or votes, or by protesting—is likely to succeed.

Notably, even if no individual is capable of regulating politicians’ behavior, collectives are capable of this. Even if one vote or one donation is not a difference-maker, many votes or many donations do, collectively, make a difference. An electorate that values the truth sufficiently highly can collectively punish failures of honesty by, for example, withholding votes or donations from dishonest politicians. Consequently, collectives might effectively manifest epistemic exactingness even if no individual can do so. For this reason, a culture of epistemic exactingness may be practically significant in constraining political developments.

To conclude this section, it is worth highlighting that the dependency of the effectiveness of epistemic exactingness on the agent’s power underscores the complexity of the virtue epistemological answer to the responsibility question with which this paper began. Consideration of the virtue of epistemic exactingness might tempt one toward a sort of victim blaming. Those that are deceived are often blameworthy for their resultant false beliefs, not because these are due to gullibility, but rather because the deceived fail to insist on the truth. Or so one might think. But, as this section emphasizes, the effectiveness of one’s insistence upon the truth may depend on features that go far beyond one’s virtues, or lack thereof, and that involve imbalances in power for which one cannot reasonably be held responsible.

6. Epistemic exactingness and strange beliefs

To what extent, then, does the possession or lack of epistemic exactingness account for epistemic success and failure in the real world? This is partly an empirical question, a rigorous answer to which would require study of the prevalence of epistemic exactingness and its role in shaping social epistemic environments. Still, there is good reason to expect that the possession and lack of epistemic exactingness plays an important role in shaping the epistemic successes and failures of real-world individuals and collectives.

Consider, first, the individual level. Most of us will be aware of individuals who attempt to regulate the epistemic offerings of others in their networks and individuals who do not do this. Those in the former class may be prone to issuing commands like “don’t bullshit me,” “give it to us straight,” and so on. Such expressions make clear a desire to hear the truth and, especially when delivered in a certain tone of voice, may imply that failures to satisfy this desire will be met with some form of retaliation. At least in some contexts, there is no reason to doubt that such expressions effectively regulate others’ epistemic offerings. So long as the targets of these expressions care about the desires of their utterers or are otherwise incentivized to follow such commands, they will present evidence accordingly.

In addition to encouraging or deterring specific acts, epistemically exacting individuals are likely to have policies that are aimed, in some small way, at regulating the epistemically significant conduct of others. For example, those that aim to promote good epistemic practice and to deter bad practices will avoid supporting excessively partisan or otherwise biased media outlets, either through subscriptions or online engagement. There is some reason to think that a lack of epistemic exactingness has allowed for such outlets to flourish, as false and otherwise sensational online content tends to be widely shared by ordinary social media users (Vosoughi et al. Reference Vosoughi, Roy and Aral2018). In this way, a lack of epistemic exactingness has plausibly contributed to current patterns of epistemic dysfunction.

We thus have good reason to expect that epistemic exactingness plays an important role in epistemic success and failure within both small-scale interpersonal contexts and on a larger scale. But, to return to the question with which this essay began, does epistemic exactingness help to account for phenomena like the outsized influence of QAnon and other bizarre conspiracy theories? At least on the face of things, it seems that the epistemically exacting could function to deter the spread of such theories by discouraging others from sharing them. Similarly, the absence of epistemic exactingness in some communities is a plausible culprit in the success of QAnon and similar theories.

Yet there are at least two reasons why one might think otherwise. First, as in the political examples discussed in Section 5, ordinary individuals have limited ability to regulate the spread of conspiracy theories. For example, false claims associated with QAnon, as well as misleading evidence for those claims, were spread in large part by political figures and online personalities with large enough followings that individual followers could exhibit little control over the distribution of misleading evidence. Thus, one might think, epistemic exactingness would be ineffective in regulating the spread of conspiracy theories and so failures of epistemic exactingness cannot account for the influence of such theories. According to a second dismissive line of response to this question, the influence of such theories is not attributable to absences of epistemic exactingness, or to failures of epistemic virtue more generally, because the influence of conspiracy theories has little to do with beliefs at all. Instead, as some scholars have argued, the appearance of widespread belief in conspiracy theories and other outrageous falsehoods is due to individuals’ performative signaling or expressive endorsement of claims known to be false, as ways of indicating political loyalties and values (Levy Reference Levy2022a, Chapter 1; Mercier Reference Mercier2020). Let us consider these objections in turn.

Consider, first, the objection that epistemic exactingness is inert with respect to the influence of conspiracy theorizing because putative evidence for conspiracy theories is largely spread by influential figures over whose offerings individual audience members have little control. The QAnon conspiracy theory offers an especially good test of this suggestion as, unlike narrower conspiracy theories whose content concerns specific events, QAnon offers a sprawling worldview that promises to weave together disparate events and existing conspiracy theories. QAnon is thus, in Michael Barkun’s sense, a superconspiracy theory (Barkun Reference Barkun2013; Greer & Beene Reference Greer and Beene2024). The expansiveness of QAnon offers a good test case for the importance of failures of epistemic exactingness, because the wide range of false claims and predictions associated with that theory offer many opportunities for followers to demand better evidence.

Notably, while many commentators have tended to focus on the “Q drops”—cryptic messages from a supposed high-ranking government insider—as drivers of QAnon, the reach of the theory is due in large part to the efforts of intermediary QAnon promoters on mainstream and fringe platforms. These figures offer interpretations of Q drops, along with QAnon-centric predictions and analyses of real and imagined events. Strikingly, the online followings of many such figures continued to grow in the face of consistently false analyses and failed predictions, including predictions concerning the ultimate triumph of Donald Trump in the 2020 US Presidential election—perhaps on the historical inauguration date of the 4^th of March (Argentino et al. Reference Argentino, Crawford, Keen and Rose2021). While some followers might have become disillusioned, and there is some evidence that such influencers cracked down on dissent through the removal of content and users (Greer & Beene Reference Greer and Beene2024), many QAnon promoters maintained dedicated followings despite their record of failures. Confident conclusions about the role of epistemic leniency in preserving the influence of QAnon would require a more careful empirical study of the traits of individual followers. Still, the ability of QAnon influencers to retain large and uncritically engaged followings despite consistent false assertions is some evidence that epistemic leniency, at least at the collective level, has contributed to the influence and longevity of the QAnon conspiracy theory. What is more, there is reason to expect that such uncritical engagement incentivized the production of more misinformation, thereby doing harm to the broader epistemic environment.

But what of epistemic exactingness at the individual level? Here it is important to note that, although the claims constitutive of the QAnon conspiracy theory are traceable in large part to the person or persons referred to as “Q” and to influencers with large followings, small-scale exchanges of conspiratorial claims between ordinary persons played an important role in the spread of QAnon. QAnon followers, “Anons,” conceive of themselves as “digital soldiers” (Hannah Reference Hannah2021), whose mission is, in part, to “wake up” the members of their social circles. Thus, for example, a given Anon might attempt to influence family members and friends by posting QAnon-centered misinformation on social media or sharing it in face-to-face conversations. Such attempts are occasions for the virtue of epistemic exactingness to push back on the spread of the theory. We might, for example, imagine two communities that are alike except in the prevalence of epistemic exactingness. In a community in which such exactingness is sparse, a given Anon might attempt to spread the QAnon theory and, being met with little pushback, manage to seed the theory for at least the more gullible members of the community. In a second community, in which gullibility is equally common but exactingness is more prevalent, the exacting members of the community may exert social pressure to deter the Anon from spreading wild theories, thus protecting the more gullible members of the community.

Notably, it is implicit in this optimistic scenario that those possessed of epistemic exactingness also take the right attitude toward the (im)plausibility of QAnon. This is not guaranteed by the nature of epistemic exactingness itself, although exactingness—as we have seen in the discussion surrounding Easy Mark—can promote individual reliability in some cases. It is worth noting that, in the real world, epistemic exactingness paired with individual misconceptions about the truth plausibly has negative consequences. Consider, for example, those individuals who misapply the term “fake news” and chastise others for sharing what they mistakenly regard as fake news. Such cases help to illustrate that, as is commonly the case with responsibilist virtues, the effects of epistemic exactingness will depend in part on the reliability of its possessor. Epistemic exactingness is thus by itself no guarantor of epistemic successes but, when paired with true beliefs on the part of the exacting, can be expected to make a positive epistemic difference.

Let us turn then to the second basis for thinking that the lack of epistemic exactingness might have limited application in accounting for the prominence of conspiracy theories and other outlandish falsehoods. According to this concern, the lack of epistemic exactingness and other virtues is of limited significance in this context because most individuals who claim to do so do not really believe outrageous falsehoods. Instead, their endorsement of such claims is merely expressive. It should first be acknowledged that, to the extent that apparent support for conspiracy theories and other outlandish falsehoods reflects expressive gestures, rather than sincere beliefs, certain epistemic vices likely play less of a role in the prominence of these falsehoods than might otherwise be thought. For example, if a person endorses QAnon only to express distaste for Democrats or Hollywood celebrities, then this person’s endorsement of the theory is not well-explained—at least directlyFootnote ¹⁰—in terms of gullibility.

However, unlike gullibility, epistemic exactingness and its absence are not about how one incorporates existing evidence into one’s belief system. Instead, epistemic exactingness is about one’s motivation to regulate the epistemically significant conduct of others. In fact, toleration of the expressive endorsement of falsehoods is well-explained by the absence of epistemic exactingness. One who tolerates such falsehoods allows for other desires—perhaps for the denigration of political opponents or for glorification of co-partisans—to override the desire to maintain a healthy epistemic environment. A culture lacking in epistemic exactingness is one that will be especially prone to the spread—whether credulous or otherwise—of falsehoods and the development of channels for spreading them.

7. Concluding remarks

One of the most important developments in contemporary epistemology has been an increased attention to our epistemic dependence on others (Hardwig Reference Hardwig1985). The insistence upon epistemic autonomy—at least if this is understood as non-reliance upon othersFootnote ¹¹—would leave us with radically diminished epistemic prospects. However, it does not follow from our extensive epistemic dependence upon others that our epistemic conditions are simply thrust upon on. A socio-environmentally oriented virtue epistemology helps to illuminate how it can be true both that what individuals believe is powerfully influenced by features of their social-epistemic environments and that individuals are often responsible for what they believe. Here, I have sought to advance the development of such a virtue epistemology by introducing the virtue of epistemic exactingness, its role in regulating the epistemically significant conduct of others, and some limitations on its effectiveness.