Outward-facing epistemic vice

The epistemic virtues and vices are typically defined in terms of effects or motivations related to the epistemic states of their possessors. However, philosophers have recently begun to consider other-regarding epistemic virtues, traits oriented toward the epistemic flourishing of others. In a similar vein, this paper discusses outward-facing epistemic vices, properties oriented toward the epistemic languishing of others. I argue for the existence of both reliabilist and responsibilist outward-facing vices, and illustrate how such vices negatively bear on the epistemic prospects of others. I pay special attention to how outward-facing epistemic vices may manifest in online activities that promote the epistemic languishing of others by negatively influencing the online epistemic environment.


Introduction
Virtue and vice epistemology have tended toward individualism in several respects. First, epistemic virtues and vices are most commonly ascribed to individuals, rather than collectives.
1 Second, and relatedly, appeals to epistemic virtues and vices provide individualist explanations of epistemic outcomes. For example, some epistemologists have discussed how individuals' epistemic vices may contribute to belief in outlandish conspiracy theories (Cassam, 2016;Harris, 2018;Meyer & Alfano, 2022). Third, the possessors of epistemic virtues are typically thought to be the proximate beneficiaries of these virtues, and the possessors of epistemic vices are typically thought to suffer most proximately from such vices. 2 Thus, while a scientist's intellectual tenacity may benefit her community in the long run, it will typically 3 do so by first effecting improvements in her own epistemic condition. Likewise, the intellectually arrogant conspiracy theorist might deceive or confuse members of his online community, but will typically do so only after himself succumbing to misconception or confusion. Some epistemologists have taken forays away from this third form of individualism, especially in the case of the intellectual virtues. For example, Jason Kawall (2002) discusses several other-regarding epistemic virtues, which primarily epistemically benefit individuals other than their possessors. In this paper, I discuss the vicious counterparts of such virtues. In contrast to the inward-facing epistemic vices that are principally oriented toward the subject's own epistemic condition, I label these the outward-facing epistemic vices. In Sect. 2, I make the case for a conception of epistemic vice broad enough to include the outward-facing epistemic vices discussed in this paper. I argue that both reliabilist and responsibilist conceptions of epistemic virtue and vice allow for outward-facing epistemic vices. In Sect. 3, I discuss epistemic vices that consist in lack of competence for transmitting beliefs through testimony. Then, in Sect. 4, I discuss epistemic vices that consist in a lack of the motivation to promote others' epistemic flourishing through effective testimony. In Sect. 5, I argue that focusing solely on outward-facing epistemic vices related to testimony threatens to underappreciate various other ways in which agents shape the epistemic environments encountered by others. I discuss how outward-facing epistemic vice can contribute to online activities that promote the epistemic languishing of others. Section 6 concludes with some brief reflections on the importance of outward-facing epistemic vices to social epistemology.

Epistemic vice
One way to understand epistemic vice is by contrast to epistemic virtue. This strategy can only get one so far, as the nature of epistemic virtue is itself a matter of contention. There is a well-known distinction in virtue epistemology between reliabilist and responsibilist epistemic virtues. 4 Whereas the former are, roughly, reliable and cognitively-integrated faculties for belief-formation, the latter are traits of character that contribute to the epistemic excellence of their possessors. According to a description of epistemic virtue suitably broad to capture both reliabilist and responsibilist varieties, epistemic virtues are properties of individuals (or perhaps collectives) that promote or aim to promote intellectual flourishing (Turri et al., 2021). The disjunctive nature of this description of intellectual virtue serves to recognize that reliabilist and 2 Jason Baehr's (2010) discussion of epistemic malevolence is a rare exception. Baehr defines epistemic malevolence as opposition to the epistemic good. While epistemic malevolence need not be oriented toward the epistemic states of other parties, several of Baehr's core examples emphasize the role of epistemic malevolence in preventing the achievement of epistemically valuable states by others. 3 Arguably, there are exceptions to this general rule. For example, the tenacious scientist's hard work might yield experimental data whose significance is only recognized after her passing. responsibilist conceptions of epistemic virtue vary in the importance they place on effects and motives (Battaly, 2014).
The present way of characterizing epistemic virtue is sufficiently broad to allow for the possibility of other-regarding epistemic virtues. Jason Baehr writes, for example, that: Intellectual virtues can…be oriented toward the epistemic good or well-being of others-they can be aimed at others' acquisition or share in the epistemic goods. (2011, p. 216) In short, some properties are epistemic virtues principally in light of their effects on the epistemic prospects of parties other than their possessors. Thus, Kawall (2002) takes traits like honesty, sincerity, integrity, and creativity to be other-regarding epistemic virtues. Like Kawall, Linda Zagzebski is principally concerned with responsibilist virtues. She recognizes a category of 'teaching virtues' that principally serve to improve the epistemic condition of others (Zagzebski, 1996, p. 114).
While Kawall and Zagzebski are both primarily concerned with responsibilist virtues, the broad conception of epistemic virtue described above leaves room for other-regarding epistemic virtues in a more reliabilist spirit. Kawall hints at the possibility of such virtues, writing that: [W]e could take more specific skills or methods which would allow one to lead others to true beliefs in various fields of inquiry to be intellectual virtues. (2002, p. 262) Other-regarding virtues in a reliabilist vein would not be reliable belief-forming faculties, but competences that primarily contribute to reliable belief-formation among others. I develop this point at greater length in Sect. 3.
For the present, the crucial point is that recognition of other-regarding epistemic virtues opens the door for a category of epistemic vices that are defined as such primarily in virtue of their relations to parties other than their bearers. Whereas epistemic virtues (aim to) promote the epistemic flourishing of their bearers and/or others, epistemic vices (aim to) promote the epistemic languishing of their bearers and/or others. Such languishing might consist in worsening of the epistemic condition, as in cases of deception, or in avoidable failures to make epistemic improvements.
Let us pause here for a brief terminological aside. Kawall and other authors use the term 'other-regarding' to capture those virtues oriented toward the epistemic flourishing of individuals other than their bearers. I prefer the terms 'inward-facing' and 'outward-facing' for two reasons. 5 First, the term 'other-regarding' is potentially confusing in the case of epistemic vice, as some epistemic vices involve lack of regard for the epistemic flourishing of others. Second, the term 'other-regarding' suggests an attention to particular parties. However, some paradigmatic manifestations of epistemic vice are not directed at any particular party, Consider, as an example, the individual that anonymously posts misinformation online, hoping thereby to spread general confusion but without intending to deceive any particular person. According to the approach taken here, such a person likely displays outward-facing epistemic vice, even if their deceptive practices are not directed at any particular party.
I have thus far argued that, given a broad conception of epistemic virtue and vice that treats these as properties that contribute to or aim at epistemic flourishing and languishing, respectively, some properties may be epistemic vices in virtue of their effects on or aims with respect to parties other than their bearers. This proposal requires both clarification and defense. Beginning with the former, it must be noted that the present proposal does not claim that all properties that promote epistemic languishing are epistemic vices. It is consistent with the present proposal that epistemic vice is not strictly defined in terms of its effects or aims. Some might argue, for example, that epistemic vices are properties for which the bearers are responsible (Zagzebski, 1996). Such a proposal is controversial insofar as lacks of competences of the reliabilist type arguably amount to vices, even if the subject is not responsible for them. For example, poor vision is sometimes regarded as an epistemic vice (Battaly, 2020, p. 23). Construing such properties as epistemic vices is controversial, as some epistemologists prefer to regard them as mere cognitive defects (Cassam, 2019). According to a less controversial proposal, vices are relatively stable features of their subjects. So, for some examples, the muddled reasoning capacities, blurred vision, and intellectual recklessness that might result from excessive alcohol consumption are not themselves epistemic vices, even if excess consumption itself is reflective of some vice. I do not intend here, or anywhere in this paper, to provide a full account of epistemic vice. The preceding examples serve instead to emphasize that there is very plausibly more to epistemic vice than its effects and aims with respect to epistemic states. Now to the defense. It might be thought that the present case for outward-facing epistemic vices is too quick. While this conclusion falls out easily from the sort of broad descriptions of epistemic virtues and vices discussed above, such a description is merely elliptical for a longer description that would rule out outward-facing epistemic vices. Or so one might argue. For example, one might think that epistemic virtues are those characteristics that (aim to) promote the intellectual flourishing of their possessors, and that epistemic vices are those characteristics that (aim to) promote the intellectual languishing of their possessors. Such a view might be supported by the notion, suggested for instance by Zagzebski (1996, p. 167), that a desire for contact with reality is central to epistemic virtue.
This proposal faces three key difficulties. First, while Zagzebski suggests that some virtues involve the subject's motivation to be in cognitive contact with reality, she quickly adds that other virtues-including inventiveness and originality-involve a motivation to advance knowledge for humanity in general (1996, p. 167). Second, at least in the case of moral virtues, it is clear that some virtues are principally selfregarding and some are principally other-regarding (Kawall, 2002). For example, temperance is plausibly principally self-regarding while compassion is principally other-regarding. Similarly, moral vices are sometimes principally harmful to their possessors and sometimes principally harmful to others. For example, gluttony principally harms its possessor, while callousness principally harms others. Third, there are good reasons specific to both reliabilist and responsibilist virtue epistemologies to think that some epistemic virtues and vices are principally identifiable by their effects and aims with respect to others. Virtue reliabilists maintain that what makes something an epistemic virtue is, at least in part, its conduciveness to the production of epistemic goods (Kelp, 2020, p. 4). Epistemic vices might be defined, in part, in terms of deviations from this conduciveness. In this way, some property of a subject being an epistemic good is derivative of its effects on the production of epistemically valuable states. But nothing in this description of epistemic virtues and vices suggests that the only relevant epistemic goods are those enjoyed by the possessor of the virtues and vices in question. There is thus room, at least in the absence of any compelling reason to narrow our understanding of the epistemic virtues and vices, to recognize the existence of outward-facing epistemic virtues and vices in a reliabilist vein. Moreover, given the broadly consequentialist structure of virtue reliabilism, and the impersonal valuation of goods that is typical of consequentialisms, there is reason to doubt that any such narrowing would be in the spirit of virtue reliabilism.
Virtue responsibilists tend to define epistemic virtues in terms of motivations, rather than-or in addition to-effects. For example, Zagzebski takes epistemic virtues to involve both reliability and proper motivation (1996, part II). What makes the epistemic virtues epistemic, is the motivation toward epistemic goods like true belief, knowledge, and understanding. However, nothing in this description requires that the epistemic goods in question be enjoyed by the subject of virtue. Plausibly, the epistemically virtuous agent is motivated to produce true belief, knowledge, and understanding, even if these goods are not enjoyed by the agent herself. Indeed, the placement of great value on knowledge is arguably especially epistemically virtuous when the agent is not principally concerned with having knowledge for herself. Thus, in the spirit of the example above (see fn. 3), we can easily imagine a scientist principally motivated to produce knowledge for future generations (cf. Zagzebski, 1996, p. 183), and such a scientist would seem to be especially epistemically virtuous. Just as proper motivations may be central to responsibilist epistemic virtues, improper motivations-or the simple lack of proper motivations-may be central to responsibilist vices. For example, some forms epistemic recklessness plausibly involve a lack of proper motivation toward knowledge. But one might also be improperly or insufficiently motivated with respect to the epistemic condition of others. In this way, some responsibilist epistemic vices may be principally outward-facing. Beginning in Sect. 3, I discuss outward-facing epistemic vices of the reliabilist and responsibilist varieties.

Testimonial ineffectiveness
I begin with an outward-facing epistemic vice that consists in a lack of competence, as opposed to a faulty motivation. In this way, the vice in question is more akin to what might be called a reliabilist vice, as opposed to a responsibilist vice. This way of dividing up the vices is not universally practiced, as some virtue reliabilists join responsibilists in treating traits of character as epistemic vices (Baehr, 2011, p. 55). Still, where virtue epistemologists have devoted prolonged attention to epistemic vices, they have often recognized a distinction between vices construed principally in terms of negative effects and vices construed principally in terms of negative motivations (Battaly, 2014). I consider vices of the former variety in this section, and of the latter in Sect. 4.
The vice in question is best understood by first considering a counterpart virtue. The virtue in question is a competence for producing true beliefs in an audience through testimony, given that the testifier possesses relevant true beliefs. In lieu of a better name for this virtue, I will call it testimonial effectiveness. This admittedly unwieldy title helps to convey that many factors go into the possession or lack of the virtue in question. Consequently, there are many ways of succumbing to the corresponding vice of testimonial ineffectiveness. One might be testimonially ineffective because one struggles to make one's testimony clear, because one has difficulty projecting an appropriate degree of confidence in one's assertions, 6 because one has difficulty modulating one's assertions according to the background(s) of one's audience, and so on.
Testimonial effectiveness is, like belief-dependent belief-forming processes (Goldman, 1979, p. 13), best assessed in terms of its conditional reliability. One may be highly testimonially effective but still struggle, as a consequence of the sorts of inwardfacing epistemic vices discussed above, to produce true beliefs in one's audiences. Alternatively, one might be entirely free of inward-facing epistemic vices and yet, as a consequence of testimonial ineffectiveness, struggle to produce true beliefs in one's audience. Especially unfortunate agents might be both testimonially ineffective and subject to inward-facing epistemic vices.
Even if one is free of inward-facing epistemic vice, one's beliefs are consistently true, and one is testimonially effective, this does not ensure that one's audience will form true beliefs. Most obviously, testimonial effectiveness must be supplemented with the motivation to promote true beliefs in others. I take up the issue of motivation in the next session. But even a well-motivated, competent communicator with largely true beliefs may fail to promote true beliefs in others. This is because the reliable transmission of true beliefs makes demands of both testifiers and receivers. It is for this reason that successful belief transmission can be understood as a kind of joint activity (cf. Greco, 2021, ch. 3). Failings on the part of the audience may lead to the failure of this activity. Consider, for example, a teacher whose clear and compelling lectures pass with little notice from students distracted by their smartphones. Consider also cases of epistemic injustice, some of which may involve audiences who unjustly devalue the testimony of certain asserters, and consequently fail to form true beliefs on the basis of testimony 7 (Fricker, 2007).
There is a case to be made for recognizing a range of different epistemic vices within the broader category of testimonial ineffectiveness. First, one might take there to be different vices associated with specific ways in which testifiers tend to fail to produce true beliefs in their audiences. For example, some communicators struggle to present their claims with the confidence required for audiences to believe them-they may instead show misleading signs of uncertainty. Some communicators struggle to express their claims clearly. One might argue that failings like the inability to assert with confidence or with clarity are individual epistemic vices. I have no case against this proposal, except to suggest that, just as one's deductive reasoning may be unreliable for various reasons, one's testimonial ineffectiveness may plausibly be attributable to a range of more specific limitations. Insofar as general properties like poor deductive reasoning are treated as epistemic vices, there is reason to treat the broad property of testimonial ineffectiveness as a single vice. Still, nothing of importance here hinges on this commitment. An alternative case for recognizing different epistemic vices within the broader category of testimonial ineffectiveness hinges on the fact that assertions can take many forms and can occur in a variety of settings. One might be an effective communicator in person, but not over video chat. One might be an effective speaker, but an ineffective writer. And so on. Questions thus arise concerning the conditions under which an individual counts as testimonially ineffective. For example, is an individual who is highly reliable in communicating in face-to-face contexts, but who is intimidated by technology and thus ineffective via text message and video chat testimonially ineffective? Here we face questions akin to those that give rise to the generality problem for reliabilist (Feldman, 1985;Goldman, 1979) and other (Comesaña, 2006) epistemologies. Similarly to how belief-forming processes may be individuated more or less narrowly, testimonial effectiveness and ineffectiveness might be understood broadly or differentiated narrowly according to medium and circumstance. I will not attempt to make the case for any degree of granularity at which to understand these properties here.
For the present, what is most important is that individuals may vary dramatically in their testimonial effectiveness. Testimonially effective individuals, provided they are possessed also of the inward-facing epistemic virtues that promote their own true beliefs and the outward-facing responsibilist epistemic virtues that motivate them to communicate effectively with others, will tend to promote the epistemic flourishing of their communities. In contrast, those who have true beliefs and are motivated to share them, but who are testimonially ineffective, will regularly fail to promote the epistemic flourishing of their communities.

Responsibilist outward-facing vices
In Sect. 3, I discussed the vice of testimonial ineffectiveness, which may or may not be separable into a group of narrower epistemic vices. Testimonial ineffectiveness resembles reliabilist virtues and vices in that it is defined principally in terms of its effects or, more exactly, the effects it would have given a suitable audience and true beliefs on the part of its subject. In this section, I turn to outward-facing vices in the responsibilist vein. As with testimonial ineffectiveness, it will be useful to introduce these vices alongside counterpart virtues.
Compared against the reliabilist epistemic virtues, the responsibilist epistemic virtues are less centered on effects and more centered on motivations. The degree of difference here is a matter of controversy. Zagzebski (1996), for instance, takes even the responsibilist epistemic virtues to have a reliability component. Yet it is open to the responsibilist to take the epistemic virtues to center on a motivation for true belief, knowledge, or some such epistemic good, without adverting to any effectiveness in bringing about these effects (Montmarquet, 1993). For present purposes, what matters is that, whether or not responsibilist virtues have a reliability component, the motivation component is central to such virtues.
In the remainder of this section, I will discuss a series of outward-facing responsibilist vices. To begin, it will be helpful to consider Baehr's (2010) discussion of epistemic malevolence. As Baehr understands it, epistemic malevolence consists in opposition to the epistemic good as such. According to Baehr, opposing the epistemic good as such amounts to treating the epistemic good as an enemy, where this is consistent with opposing the epistemic good for instrumental reasons (2010, pp. 191-192). Epistemic malevolence may be personal-as in the case of an individual who dedicates herself to the epistemic languishing of some target (Baehr, 2010, pp. 206-207). Epistemic malevolence may also be impersonal-as in the case of someone opposed to knowledge generally, and not to any particular individual's possession of it. Baehr's distinction between varieties of epistemic malevolence is independent of the inwardfacing/outward-facing distinction made here. One might in principle be personally opposed to one's own possession of knowledge or some other person's possession of knowledge. Impersonal opposition to knowledge cannot be neatly sorted into either the inward-facing or outward-facing category, as I have construed them here. Most importantly for present purposes, epistemic malevolence involves a kind of antagonism toward the epistemic good that need not be involved in the outward-facing responsibilist vices I consider below. Instead, many forms of outward-facing epistemic vice involve an indifference toward, or insufficient concern with, the epistemic good.
With this in mind, let us turn to the familiar virtue and vice combination of honesty and dishonesty. Kawall (2002) includes honesty among his list of other-regarding epistemic virtues. Construed as an epistemic virtue, honesty involves a motivation to help others believe the truth or, at a minimum, to not cause them to believe falsehoods. Honesty is sometimes treated as a moral virtue, and it is an open question whether and if so in what sense the moral and epistemic virtues of honesty are distinct (cf. Zagzebski, 1996, p. 148). I will not attempt to settle this issue here. For present purposes, the crucial point is that, whatever its relation to the moral virtue of honesty, there is an outward-facing epistemic virtue of honesty. Corresponding to this epistemic virtue is the epistemic vice of dishonesty. One who is dishonest, as a matter of intellectual character, is not properly motivated with respect to the epistemic conditions of others. Dishonesty may in principle be rooted in epistemic malevolence. For example, Baehr (2010, p. 208) takes the Cartesian deceiver to be a paradigmatic example of epistemic malevolence, at least supposing its deceptive behavior is rooted in opposition to the epistemic good. Dishonesty need not be based on epistemic malevolence, however. For example, those agents that lie instrumentally, and do so without adopting any animosity toward epistemic goods or the would-be possessors of these goods, manifest dishonesty without epistemic malevolence. Whether or not dishonesty is rooted in malevolence, one who is dishonest cannot be counted on to support successful epistemic communications, at least in the absence of external incentives.
Consider, next, what I will call communicative conscientiousness and communicative recklessness. Elsewhere, epistemologists have discussed recklessness construed as an inward-facing epistemic vice, and have contrasted this with conscientiousness (Bland, 2022b;Sosa, 2019). Recklessness, so construed, has to do with the formation of one's own judgments. For present purposes, we are concerned with a communicative form of recklessness. The communicatively reckless agent, in contrast to the communicatively conscientious agent, communicates without proper regard for the likely effects of his communications on the doxastic states of others. Thus, the communicatively reckless agent may, for example, discuss false rumors and gossip without intending to mislead his audience, but also without proper concern for the possibility that his audience will be misled nonetheless. I will have considerably more to say about communicative recklessness in Sect. 5, where I discuss online epistemic communication.
Let us turn to a third virtue and vice pair whose relation to the epistemic states of others may be less clear. Consider first the virtue of patience. Patience can take many forms but, as is especially clear to parents to teachers, some forms of patience involve the valuing of others' epistemic states in contexts of epistemic communications (Baehr, 2011, p. 216). Let us call this form of patience communicative patience. The communicatively patient teacher is willing to devote considerable time and energy to students that struggle with the material being taught. The communicatively patient parent is willing to answer a seemingly endless series of why questions from a curious child. The vice of communicative impatience, in contrast, may incline its bearer toward the cessation or degradation of communication. The communicatively impatient teacher may become frustrated with a student's lack of progress and may give up or, less dramatically, may reduce the effort put into communicating effectively. The communicatively impatient parent may cease to provide detailed answers to a child's questions, and may instead attempt to divert the child toward other forms of entertainment. In the terms introduced in Sect. 3, a lack of communicative patience may compromise an agent's testimonial effectiveness.
Related to the virtue of communicative patience is the virtue of communicative creativity. While the epistemic virtue of creativity has received attention elsewhere, including as an other-regarding epistemic virtue (Kawall, 2002, p. 270), I have in mind a distinct virtue of creativity in communication. The communicatively creative agent is willing, and perhaps able, to generate novel ways of clarifying matters for their audiences. A communicatively patient teacher may fail to be communicatively creative. Think, for example, of a teacher who repeatedly explains a concept in the same way to a struggling student. Such a teacher is likely communicatively uncreative. A communicatively patient and creative teacher would instead at least seek alternative ways of explaining the point with which the student is struggling. In this way, communicative patience and communicative creativity complement one another.
Let us consider one further virtue and vice pair. The epistemically sensitive agent is willing to take into account the epistemic positions of the audience in order to shape more effective assertions. Like communicative patience and creativity, epistemic sensitivity is an asset to parents and teachers (cf. Zagzebski 1996, p. 114). But epistemic sensitivity may be especially important in other contexts, notably including political ones. Some politicians are more willing than others to shape their messages in ways that resonate with diverse audiences. Similarly, some private individuals are more willing than others to present their views in ways that appear reasonable to political opponents. In this way, epistemic sensitivity is of particular practical importance in the present moment, where concerns about polarization abound. In the US context, for example, it has been noted that hostility between Democrats and Republicans has deepened despite some degree of convergence concerning key issues (Talisse, 2019, p. 160). This suggests that some forms of political polarization in the US context are partly due to failures of communication. Such failures may be due in part to the epistemic insensitivity of partisans. Insofar as individuals are unwilling to express their views in ways that resonate with individuals in other parties and communities, communication is likely to be unsuccessful in such contexts.
The preceding overview of outward-facing epistemic vices is far from exhaustive. Drawing on Kawall's (2002) pioneering discussion, we might add to this list the vicious counterparts of sincerity and integrity. Similarly, taking inspiration from Zagzebski (1996, p. 114), we might add the vicious counterpart(s) of the virtue of intellectual candor. In addition to some of the outward-facing epistemic virtues mentioned above, Baehr mentions carefulness, precision, and fair-mindedness (2011, p. 216), qualities that plainly have vicious counterparts. Existing work on the moral virtues might inspire further additions to the list. To take just one example, one can imagine a distinctively intellectual form of generosity (cf. Roberts & Wood, 2011, ch. 11). Such a trait is possessed, for example, by some public intellectuals that devote their time and energy to improving the public understanding of those matters on which they are experts. In contrast, a lack of generosity might eventuate in the hoarding of true beliefs among certain populations. Beyond additions to the list, other modifications might include reorganizations to treat some of the abovementioned virtues and vices as variants or special cases of others. The aim here is thus not to provide anything resembling a definitive list of outward-facing intellectual virtues and vices of the responsibilist variety. Rather, the aim is to show that, insofar as epistemic virtues and vices are construed in terms of the motivation or lack thereof toward epistemically valuable states, responsibilists can make room for a broad range of outward-facing epistemic virtues and vices. I hope also to have shown in this section how outward-facing epistemic virtues and vices of the reliabilist and responsibilist varieties complement one another. Zagzebski writes that "the motivation to know leads to following rules and belief-forming procedures known by the epistemic community to be truth conducive…" (1996, p. 167). In other words, responsibilist virtues promote true beliefs because such virtues drive their possessors to form beliefs in reliable ways. Similarly, I have suggested, those with outward-facing epistemic virtues of the responsibilist variety will at least try to develop the competences required for what I have called testimonial effectiveness. As we have seen, success in doing so may require time, effort, and attention to characteristics of the audience. To this we might add that testimonial effectiveness, in some contexts, may require the development of skills with various forms of communication, including verbal, written, and more recent online forms. Insofar as the epistemically virtuous agent is motivated to promote the epistemic flourishing of others, such an agent will be motivated to acquire such skills.
Whereas the possessor of outward-facing responsibilist virtues will be motivated to promote the epistemic flourishing of others, the possessor of outward-facing responsibilist vices will lack such motivations and may even, as in the case of the dishonest agent, be motivated to promote others' epistemic languishing. In this way, the outward-facing epistemic vices, over and above their inward-facing counterparts, may contribute to epistemic dysfunction in both small-scale relationships and among larger communities.

Vice in online epistemic communication
Thus far I have focused on the epistemic vices as they bear on testimony. But testimony is just one way in which agents can contribute to the epistemic flourishing or languishing of others. In this section, I discuss two forms of communication that increasingly shape our epistemic environments but which, to this point, have received little attention from epistemologists. I aim to show, first, that the long-standing focus of social epistemologists on testimony threatens to overlook some important ways in which agents shape the epistemic environments of others and, second, that there is a role for vice epistemology in addressing this lacuna.
Consider, first, the act of information sharing. For present purposes, I stipulate a narrow definition of information sharing roughly as the passing on of information in a form other than the words, expressions, symbols, or gestures of the sharer. On this narrow definition, when a speaker verbally summarizes an article, this does not count as information sharing. The same speaker passing on the article itself in either a physical or electronic form does amount to information sharing, however. Information sharing is better captured by ostension than definition. Information sharing may involve the transfer by one party to another of a physical object, like a book, news article, or photograph. In recent years, information sharing has largely occurred online and has been facilitated by the infrastructures of various media outlets and social media platforms. For example, Facebook allows for existing posts to be shared, and Twitter allows for existing posts to be Retweeted. Additionally, many online media outlets encourage the sharing of their content by including buttons on their web pages that simplify the sharing process. For present purposes, I will understand information sharing broadly enough to include both reposting online content so as to broadcast that content to a (potentially) wide audience and the forwarding of content through private channels like E-mail and direct message. As a final note on our target, I will not treat information sharing as a success concept. In familiar cases, we might say sharing occurs only if some party accepts the offerings of another. For example, I have only shared my lunch with you if you accept some of it-otherwise I have merely offered to share. When it comes to information sharing, however, I will allow that sharing may occur even if the shared content does not find a receptive audience. Thus, for example, I will assume that Retweeting an article counts as sharing it, even if no one else reads the article, and indeed even if no one else sees the Retweet. Some epistemologists have recognized the epistemic significance of information sharing via social media and thus have recently offered insights into distinctive, epistemically-significant aspects of such sharing. Regina Rini (2017), for example, argues that online information-sharing practices have led to epistemic dysfunction, in part, because the norms governing such practices are unclear. In particular, it is unclear whether information sharing in various online contexts signals the sharer's endorsement of the correctness of the information in question. Neri Marsili (2021) has argued that Retweeting does not amount to endorsement, but instead resembles quotation in some respects and indication in others.
While the norms of online information sharing are unsettled, two points ought to be uncontroversial. First, individuals sometimes use online sharing functionalities to spread information that they deem credible. Second, regardless of the intentions of the sharer, the reception of shared information may shape the attitudes of recipients. Suppose, for example, that I share a post from a political figure pushing what I consider to be an absurd take on climate change. Even if I do this because I find the post amusingly misguided, and expect some of my followers to feel the same, I thereby amplify that content and potentially introduce it to audiences that will find the post persuasive. Moreover, given the different ways in which individuals interpret shares, some individuals may interpret my share as higher-order evidence that I deem the post credible. In short, the sharing of information may negatively influence the doxastic states of others.
The epistemically virtuous agent will be attentive to the potential for acts of information sharing to influence the doxastic states of others. The epistemically vicious agent, in contrast, may share information without proper concern for its effects on others. To illustrate, consider some epistemically vicious information sharing activities. One aiming to improve one's standing with members of a given political group might share out-of-context clips of representatives of an opposing political group. Doing so might reflect a form of dishonesty on the part of the sharer, but might instead reflect communicative recklessness. To better illustrate this distinction, consider next a social media user whose principal aim is to gain more followers. Knowing that extreme content tends to drive engagement, this social media user combs through online extremist forums in search of especially reprehensible or disturbing content. The social media user then posts this content to Twitter alongside accompanying condemnations. This user need not have any deceptive intentions. The user's condemnations of the content in question may be sincere. Even so, the user described effectively amplifies extremist content in a way that may shift some audience members' attitudes toward extremism. If the user is unmoved by this fact, the user thereby exhibits communicative recklessness without dishonesty.
Thus far in this section I have discussed how outward-facing epistemic vices, especially communicative recklessness, may shape information sharing online. Online information sharing resembles other forms of information sharing. As, I have noted, for example, sharing information online bears some similarity to passing along a physical article. I now consider a form of communication that, I think, is relatively discontinuous with earlier forms of communication.
It risks cliché to note that the internet connects people. Still, some of the epistemically-significant ways in which the internet connects people, and the relevance of vice epistemology to these ways are underappreciated. In addition to sharing, online platforms often offer various ways of interacting with content that, in one way or another, shape the epistemic environments encountered by others. Consider some examples. On Twitter and Facebook, posts and accounts that one "likes" might, as a consequence, appear in the feeds of one's followers. On YouTube, liking or viewing a video contributes to its popularity, thereby increasing its chances of appearing on the homepage or in lists of recommended videos. One factor in the TikTok recommendation algorithm is the rate at which videos are watched from beginning to end. In this way, details of one's viewing habits may subtly influence the epistemic environments encountered by others, in part by communicating that certain content is worth consuming. These examples are just a few of the myriad acts, aside from testifying and sharing information, by which individuals can and do influence the epistemic environments encountered by others. 8 In the online world, even if one is not communicating with others, one's actions might communicate to others the value of certain content.
The ability to influence others' epistemic environments without testimony or information sharing is not strictly due to the nature of the internet. One's decision to buy a book for oneself or to watch a particular news program might contribute to the book or program's popularity, thereby subtly influencing the shape of others' epistemic environments. However, while online actions such as those I have highlighted here are not wholly discontinuous from various offline actions that have long been available, three important facts make it worth highlighting the former set of actions in the present context. First, online actions tend to have immediate epistemic effects. Liking a post now immediately increases the chances that others will encounter it, for instance. Second, given the role of algorithms in content distribution, one's online actions can shape others' epistemic environments without further human intervention. Thus, when one watches a YouTube video, this automatically increases the chances that others will see it. Finally, the shaping of others' epistemic environments through one's online actions is comparatively ubiquitous. A broad range of online activities-including clicking on or scrolling all the way through articles, liking posts, following accounts, and viewing and commenting on videos-as well as the sequences in which one performs these activities, influence the epistemic environments encountered by others.
The ways in which actions like liking posts and viewing online videos shape the epistemic environments encountered by others are unlikely to be top-of-mind for most internet users. Where users are entirely oblivious to this fact, engagement with misleading content in a way that shapes the epistemic environments of others need not reflect any epistemic vice on the part of the agent. However, if a given agent is aware of the effects of engagement with misleading content on the epistemic environments of others, but engages with such content nonetheless, this may reflect a sort of outwardfacing epistemic recklessness. Suppose, for example, that I find conspiracy theories entertaining and thus tend to follow conspiracist Twitter accounts and watch conspiracist YouTube videos and suppose that I know that these actions are likely to have some impact, however small, on the sort of content that others encounter. Being wholly unmoved by this fact would reflect a sort of epistemic recklessness. This is not to say that engaging with misleading content online is invariably epistemically reckless. One might, for example, regard engagement with such content to be sufficiently epistemically valuable to compensate for potential epistemic ill-effects upon others. Additionally, one might argue that any epistemic ill-effects upon others that are traceable to an individual's engagement with misleading content is the responsibility of the platforms whose algorithms translate engagement with misleading content into increased prominence for that content. One might, in short, think that identifying engagement with misleading content as epistemically reckless lays too much blame at the feet of individuals, and not enough at the feet of platforms and the corporate entities that design them. If such issues sound familiar, it is likely because such issues have close analogues in environmental ethics. There are substantial questions in that domain concerning the responsibility of individuals for large-scale problems like pollution and climate change. Notably, ethicists have found it useful to apply virtue concepts to such issues (Jenkins, 2016;Williston, 2015). Likewise, the tools of virtue and vice epistemology can promote a better understanding of the extent of individual responsibility for degradation of epistemic environments (cf. Bland, 2022a;Levy, 2022). Concepts of virtue and vice may be especially useful in thinking about how the mundane actions of individuals produce small and diffuse effects that, jointly, influence epistemic environments on a grand scale.

Concluding remarks
Attention to the now well-established fact of our epistemic inter-reliance has produced momentous shifts in epistemology over the past few decades. Thus far, most epistemological theorizing has considered this point from the perspective of those in a position of epistemic dependence. For example, considerable attention has been paid to the issue of how individuals should respond in cases of expert disagreement (Coady, 2006;Goldman, 2001). Yet the prevalence of epistemic inter-reliance implies that individuals are often at least causally responsible for the beliefs of others. That individuals are causally responsible for the beliefs of others is relatively obvious in cases of testimony and information sharing, but less so in cases of the algorithmically driven effects of individuals' actions on the epistemic environments encountered by others. Attending to the outward-facing epistemic vices helps to clarify what can go wrong, on the side of the responsible agents, in such relations of epistemic inter-reliance. In this way, a vice epistemology that admits of outward-facing epistemic vices, and especially a vice epistemology attentive to the myriad ways in which individual acts shape epistemic environments, allows for a more comprehensive social epistemology.

Conflict of interest
The author has no relevant financial or non-financial interests to disclose.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/ by/4.0/.