The Curious Case of the Excellent Gossiper

My main aim in this paper is to examine whether gossip should be categorized as an epistemically valuable character trait. Gossip satisfies the necessary and jointly sufficient conditions for an acquired character trait to be classified as an intellectual virtue under the responsibilist understanding of the concept of virtue. The excellent gossiper is (i) motivated to acquire epistemic goods through gossiping, (ii) reliably successful in acquiring epistemic goods through gossiping, (iii) competent at the activity of gossiping and (iv) good at judging when, with whom and what to gossip. Nonetheless, I show that the excellent gossiper inflicts (knower-initiated) epistemic wrong on others. The excellent gossiper comes to intentionally acquire another person’s private information (e.g., their sexual preferences) without their consent. This leaves virtue responsibilists with three options: (a) resist my argument that gossip qualifies as a responsibilist virtue and/or that excellent gossiping inflicts epistemic wrong, (b) bite the bullet and argue that the intellectually virtuous agent sometimes inflicts epistemic wrong on other agents intentionally, (c) develop a no-wrong principle that disqualifies gossip from being categorized as an intellectual virtue.

have also developed virtue-based accounts of gossip (see, for example, Niekerk, 2008;Robinson, 2016;Alfano & Robinson, 2017). Still, there are very few studies examining gossip through the framework of virtue epistemology. 2 This is an important gap in the literature of virtue epistemology, especially since, as I am going to show, the phenomenon of gossip has significant implications for epistemological theories of virtue.
Following the introduction of the concept of virtue in epistemology in 1980 (see Sosa, 1980), scholars working in analytic epistemology have accorded a prominent role to the concept of intellectual virtue and employ it in a wide range of epistemological topics (such as epistemic justification, knowledge and the Gettier problem). Broadly put, scholars working in virtue epistemology are divided into two camps: (i) virtue responsibilism and (ii) virtue reliabilism. The former camp (e.g., Baehr, 2007;Baehr, 2011;Code, 1987;Zagzebski, 1996) understands intellectual virtues as acquired and enduring traits of character that aim at epistemic goods like truth and knowledge (Baehr, 2016, p. 86) while the latter (e.g., Greco, 1993Greco, , 2010Pritchard, 2005Pritchard, , 2013Sosa, 1991Sosa, , 2007 conceives of intellectual virtues as faculties of the mind that are instrumental in maximizing truth over error (Sosa, 1991, p. 25). Virtue responsibilists consider character traits, such as open-mindedness and intellectual tenacity, as intellectual virtues while virtue reliabilists categorize under the concept of epistemic virtue cognitive faculties such as excellent vision and memory. 3,4 For the purposes of this paper, I focus on the responsibilist project since (as I will show in section 2) gossip qualifies as an intellectual virtue under the responsibilist understanding of the concept of virtue. Despite their attempts to categorize and characterize all major intellectual virtues (see, for example, Roberts & Wood, 2007;Hazlett, 2012;Watson, 2015;Battaly, 2017a;Ross, 2020), virtue responsibilists have yet to examine whether gossip is an epistemic virtue. Either that, or they do not consider that gossip is a character trait of epistemological significance. However, this is a notable gap in the literature irrespectively of the reasons as to why gossip has been overlooked by virtue epistemologists. Significantly, I will argue that excellent gossip is not an intellectual virtue despite the fact that it satisfies all the requirements that, according to virtue responsibilists, a concept must encompass in order to be categorized as an intellectual virtue. This is because the excellent gossiper commits an epistemic injustice -they come to acquire epistemic goods which they are not warranted to possess. I argue that these cannot be the actions (or motivations) of an intellectually virtuous agent.
In the section that follows, I show that gossip satisfies the necessary and jointly sufficient conditions for an acquired character trait to be classified as an intellectual virtue under the responsibilist understanding of the concept of virtue. The excellent gossiper is (i) motivated to acquire epistemic goods through gossiping, (ii) reliably successful in acquiring epistemic goods through gossiping, (iii) competent at the activity of gossiping and (iv) good at judging when, with whom and what to gossip. In the third section of this paper, I show that the excellent gossiper inflicts (knower-initiated) epistemic wrong on another agent. The excellent gossiper comes to intentionally acquire another person's private information (e.g., their sexual preferences) without their consent. I argue that my two main arguments (i.e., 1. that gossip qualifies as an epistemic virtue and 2. that the excellent gossiper inflicts epistemic wrong on another agent) put pressure on the virtue responsibilist position and leave responsibilists with three possible options: (a) resist my argument that gossip qualifies as a responsibilist virtue and/or that excellent gossiping inflicts epistemic wrong, (b) bite the bullet and argue that the intellectually virtuous agent sometimes inflicts epistemic wrong on other agents intentionally, (c) develop a no-wrong principle that disqualifies gossip from being categorized as an intellectual virtue. I conclude that the third reply is the best option. The no-wrong principle excludes epistemic wrong from being the intentional and/or frequent result of an act stemming from an intellectual virtue and fits well with the virtue responsibilist understanding of the concept of intellectual virtue.

Defining Gossip
It would be useful to start with a short description of gossip before proceeding to examine whether it is an epistemically valuable character trait. Briefly put, gossip requires the presence of two parties and a verbal exchange about a third party which is absent. As Lind et al. (2007) point out gossip involves a triad: (i) the speaker, (ii) the hearer and (iii) the subject. To illustrate this, consider the following example: John and Mary gossip about George's sex life (e.g., his choice of sex partners) while George is absent. Note, however, that in acts of gossip the roles of speaker, hearer and subject are not rigid. For instance, the hearer may assume the role of the speaker and share a piece of gossip about the subject and/or the two parties engaging in gossip might decide to switch subject (e.g., stop talking about George's sex life and start gossiping about Anne's cleaning habits). 5 Gossip does not simply involve an exchange of information. As Alfano and Robinson (2017, p. 475) point out, gossip is '…supposed to be juicy': it cannot be common knowledge and '…typically seems to involve some sort of norm violation, either by the subject or the speaker'. For instance, telling someone that 'Athens is the capital of Greece' or that 'team X won the football match yesterday' is not considered gossip -it is not some juicy fact about a specific subject. Of significant importance is also Robinson's (2016) distinction between non-evaluative 6 and evaluative gossip. Robinson argues that evaluative gossip involves evaluations of the subject (e.g., 'Joshua is a cheater'; 'Mark has a poor taste in clothes') while in non-evaluative gossip (e.g., talking about someone's sex life; talking about someone's financial situation) no evaluation is made by the speakers (either positive or negative) and the subject's behavior is not considered wrong. In addition, in non-evaluative gossip, the speaker violates norms by discussing taboo topics (e.g., sex) while in evaluative gossip it is the subject that has violated norms -e.g., aesthetic norms, moral norms, cultural norms (see also Alfano & Robinson, 2017).
While most scholars focus on evaluative gossip and praise its positive role as a deterrent of norm violation and as empowering oppressed groups (see, e.g., Beersma & Van Kleef, 2012;Alfano & Robinson, 2017), I focus on non-evaluative gossip (i.e., gossip which does not involve normative evaluations of the subject). I argue that, in the vast majority of cases, 7 nonevaluative gossip inflicts wrong on the subject (see, for example, McAndrew's, 2014 analysis of the negative aspects of gossip -e.g., employed as a tool for revenge, manipulation and exclusion) and hence cannot be thought of as having a positive impact to society. Consider, for example, again the case of John gossiping to Mary about George's sex life. John trusts 8 that Mary will not tell anyone that she heard the gossip from him. Nonetheless, Mary goes on to share the gossip with Kate (i.e., George's boss). Kate disapproves of what she hears about George, cannot hide her disgust about George's choice of partners, and ends up making his professional life difficult (e.g., demoting him). 9 This is an example of how non-evaluative gossip can inflict non-epistemic wrong on the subject -namely gossiping to someone about a third party can negatively shape their perception of that person. Still, in this paper, I focus on the epistemic consequences of gossip and argue (in section 3) that, besides non-epistemic wrong, non-evaluative gossip 10 quite often also inflicts epistemic wrong on the subject.
It is also important to note that several scholars studying gossip have suggested (in a direct or indirect manner) that gossip is a speech act predominantly carried out by female-identified individuals 11 (see, e.g., Tannen, 1990;Bergman, 1993;Chesler, 2001). 12 I do not believe that it is accurate (or fruitful in terms of studying the epistemological significance of this phenomenon) to associate gossip with a specific gender. In this respect, I am following Adkin's (2007Adkin's ( , 2017 and Alfano and Robertson's (2017) approach and study the phenomenon of gossip as gender neutral. Even if I am wrong, and gossip is indeed a speech act in which female-identified individual engage more often than their male counterparts, this does not affect my analysis of the phenomenon in question. I am approaching the issue from a virtue epistemology perspective, and in this regard the gender of the gossiper is not relevant to the discussion at hand (at least not directly).

Gossip: Motivation, Success, Competence and Judgement
The disposition to gossip is greatly influenced by the socio-cultural environment of the agent (see e.g., Baumeister et al., 2004). The tendency to gossip is an acquired trait of character. One is not born with an innate disposition to gossip -one acquires and develops it through time. There are agents who engage in gossiping whenever the opportunity arises and there are agents who refrain from it at all costs. Agatha Christie's Miss Marple is a well-known example of an excellent gossiper. 13 Gossip is Miss Marple's main tool for solving murders. Through gossip she acquires juicy information about suspects -information which would not be accessible to her otherwise. In the remainder of this section, I proceed to show that gossip satisfies all the conditions of intellectual virtues identified by virtue responsibilists and accordingly should be classified as a responsibilist virtue.
Virtue responsibilists agree that motivation is an integral and necessary condition for the presence of intellectual virtues. They argue that for an agent to possess intellectual virtues she needs to have a strong rational desire for epistemic goods (such as truth, knowledge, understanding). For instance, Zagzebski (1996, p. 167) notes that epistemic virtues are characterized by the agent's 'motivation to have cognitive contact with reality'. Also, Roberts and Wood (2007, p. 305) consider a strong rational desire for knowledge -which they define as an intellectual virtue that they call 'love of knowledge' -as necessary for the presence of (all other) intellectual virtues. Similarly, Baehr (2016, p. 87) notes that 'A subject S possesses an intellectual virtue V only if S's possession of V is rooted in a "love" of epistemic goods'.
I argue that gossip satisfies the motivational component of intellectual virtues. The excellent gossiper engages in acts of gossip out of a strong desire to acquire epistemic goods about other agents (e.g., their friends, partners, neighbors, coworkers) -her actions stem from her virtuous epistemic motivation. 14 Consider, for instance the example of Mary who is an excellent gossiper. Mary gossips with John about George's sex life out of her desire to know the truth about George's sex life. Her motivation to gossip stems from her epistemic desires. Still, not every agent has the motivation to engage in gossip. For instance, in contrast to Mary, Christin has no desire to engage in gossip since she is not interested in learning juicy information about other agents. Christin lacks the motivation to gossip and thus cannot possibly be an excellent gossiper 15,.16 Still, some virtue responsibilists (e.g., Baehr, 2011Baehr, , 2016Zagzebski, 1996) argue that the motivational component is not sufficient for an agent to possess intellectual virtues. For instance, according to Zagzebski (1996, p. 137), the intellectually virtuous agent is not only motivated by their desires for epistemic goods but is also reliably successful in acquiring the objects of their intellectual desires -viz., epistemic goods such as truth and knowledge. I argue that, besides the motivational component, the excellent gossiper also satisfies the success component of intellectual virtues. As Adkins (2017, p. 3) notes, 'gossip is a path to knowledge'. For instance, Mary (who is an excellent gossiper) is reliably successful in acquiring juicy truths about other people through gossip (e.g., information about George's sex life). In contrast, there are agents (non-excellent gossipers) who, despite being motivated to acquire epistemic goods through gossip, are usually unsuccessful in acquiring juicy facts about other people (this can be attributed to poor judgement -e.g., gossiping with the wrong person).
However, not everyone agrees with Zagzebski (1996) that the intellectually virtuous agent is reliably successful in acquiring epistemic goods. Some scholars consider Zagzebski's success component of intellectual virtues as too demanding of epistemic agents. Significantly, according to Baehr (2011, pp. 123-126), reliability is not a necessary component of intellectual virtues. Baehr agrees with scholars such as Montmarquet (1993, p. 20) that in adverse conditions and 'abnormal' conditions (e.g., in a demon-deceiving world), an intellectually virtuous agent would not be reliably successful in apprehending epistemic goods. Scholars opposing Zagzebski's success component of intellectual virtues argue that in such epistemically unfriendly conditions, the agent is intellectually virtuous despite the fact that they are unable to acquire epistemic goods on a reliable basis. Baehr (2016) discusses the competence component of intellectual virtues as an alternative to Zagzebski's (1996) success component. According to Baehr (2016, p. 91), 'S possesses an intellectual virtue V only if S is competent at the activity characteristic of V'. For example, the agent who possesses the intellectual virtue of inquisitiveness is competent at the activity characteristic of this virtue, namely good questioning (Watson, 2015). Although I tend to agree with those scholars who oppose the reliability condition of intellectual virtues, my aim in this section of the paper is not to argue in favor of either side of the debate. Rather, my aim is to show that gossip satisfies all necessary conditions of intellectual virtues identified by virtue responsibilists (and hence should be classified as an epistemic virtue). Thus, I argue that gossip does not only satisfy Zagzebski's (1996) success component of intellectual virtues but also Baehr's (2016) competence condition. The excellent gossiper is competent at the activity characteristic of this character trait: i.e., gossiping. For instance, Miss Marple is highly competent at engaging in acts of gossip -although not always reliably successful in acquiring epistemic goods through such speech acts.
Baehr (2016) also discusses the judgement component of intellectual virtues. According to him, 'S possesses an intellectual virtue V only if S is disposed to recognize when (and to what extent, etc.) the activity characteristic of V would be epistemically appropriate' (Baehr, 2016, p. 92). The excellent gossiper satisfies the judgement criterion of intellectual virtues. Firstly, the excellent gossiper is good at judging when to engage in acts of gossip. For instance, Mary who is an excellent gossiper is good at judging that John is unlikely to share any juicy information about George's sex life with her when George is nearby. The excellent gossiper is also good at judging what to gossip about and with whom. For instance, Mary is good at judging that, unlike George, Christin has no desire to engage in gossip and hence refrains from attempting to gossip with her. She also knows that John is happy to gossip about George's sex life but is quite unwilling to share gossip about his own sister. Unlike the excellent gossiper, the non-excellent gossiper may be motivated to engage in gossiping out of their desire for epistemic goods, may be competent at the activity of gossiping itself, but may nonetheless lack the judgement to know when, with whom or what to gossip about.
All in all, I have shown in this section that gossip 17 possesses all the necessary components that, according to virtue responsibilists, an acquired character trait must encompass in order to be classified as an intellectual virtue. I have argued that the excellent gossiper is (i) motivated to acquire epistemic goods through gossiping, (ii) reliably successful in acquiring epistemic goods through gossiping, (iii) competent at the activity of gossiping and (iv) good at judging when, with whom and what to gossip. Thus, it seems reasonable to conclude that gossip is a responsibilist virtue.

Gossip Is Not an Intellectual Virtue: Two Objections
Despite the fact that I have argued that gossip satisfies all necessary and sufficient conditions for a character trait to be considered an intellectual virtue, some readers may still remain unconvinced. For instance, one could argue that gossip is not an epistemically valuable trait: the excellent gossiper does not learn 'important' truths through gossip (see, e.g., Roberts & Wood, 2007). 18 For instance, gossip does not lead one to acquire important truths such as the chemical composition of oxygen but only to possess 'petty' information such as that Chris cheated on David with someone from work. Still, one could counterargue that the value of truths for the agent are highly dependent on the context. For example, finding out that Chris cheated on his husband might give the agent the means to defend herself (e.g., threaten to expose him to his husband) when being the victim of extortion from Chris -a means of defense which would not be otherwise available to the agent.
One could also note that gossip is an unreliable source of epistemic goods even in normal epistemic environments (e.g., in non-demon worlds) -information acquired through gossip may be inaccurate or even outright false (Meeker, 1999 -see also Gelfert, 2013). For instance, it might turn out that John was lying to Mary about George's sex life (i.e., George never cheated on his husband). John's motivation for lying can be due to a number of different factors (revenge, jealously, simply disliking him, etc.). It is also quite possible that John spreads false information about George's sex life without knowing it -i.e., John genuinely believes that George cheated on his husband. However, I argue that the excellent gossiper is good at judging when someone is purposefully spreading false rumors 19 in order to wrong another person. The excellent gossiper also knows that information acquired through gossip is quite often unreliable and thus requires double-checking (e.g., by gossiping with another person who might also know about George's sex life).
Nonetheless, it should be noted that one does not need to benefit from epistemic goods acquired through gossip, nor do they need to be reliably successful in acquiring such goods, in order for gossip to be classified as a responsibilist virtue. Significantly, the person who lacks the intellectual virtue of gossip has less information about her environment -she is in an epistemically worse condition that the excellent gossiper. The excellent gossiper acquires through gossip epistemic goods which 18 This section is informed by an objection often raised against virtue reliabilists -i.e., that their accounts of intellectual virtues include the acquisition of frivolous epistemic goods. For more on the distinction between frivolous 'low-grade' epistemic goods (e.g., the knowledge that I had spaghetti for dinner on May the 8th, 2004) and important 'high-grade' epistemic goods (e.g., knowledge of how to perform CPR) see Sosa (2003) and Pritchard (2021). 19 For example, Miss Marple knows when someone is lying: "(Mr. Rafier) 'Yes, it's quite right, but I don't know how you knew' 'Well, it's the way you insisted on the point,' said Miss Marple, 'I have a certain experience of the way people tell lies' (Christie, 1964, p. 277). would otherwise be inaccessible to them -gossip improves their net epistemic condition. Hence, it seems that gossip is an epistemically valuable character trait.

Gossip Is Not an Intellectual Virtue
In the previous section, I argued that gossip is an intellectual virtue under the responsibilist understanding of the concept of epistemic virtue -i.e., gossip satisfies all the conditions that, according to virtue responsibilists, a trait must meet in order to be categorized as an epistemic virtue. However, in the remainder of the paper, I proceed to show that gossip is not actually an intellectual virtue: the excellent gossiper intentionally inflicts an epistemic injustice which wrongs the subject and those cannot possibly be the actions (or motivations) of an epistemically virtuous agent. I begin this section by presenting and discussing Fricker's (2007) analysis of epistemic injustice. I then move on to argue that the excellent gossiper inflicts (what I call) knower-initiated epistemic injustice on the subject. I conclude this section with the argument that virtue responsibilists need to add a no-wrong principle to their list of conditions of intellectual virtues in order to exclude traits such as gossip from being categorized as epistemic virtues.
Miranda Fricker (2007) has recently introduced and discussed the concept of epistemic injustice. According to her, epistemic injustice occurs when an agent is wronged in their capacity as a knower. Fricker focuses on two kinds of epistemic injustice: (a) testimonial injustice and (b) hermeneutical injustice. 20 According to Fricker (2007, p. 10), in testimonial injustice the speaker 'receives an unfair deficit of credibility from a hearer owning to prejudice on the hearer's part'. For example, testimonial injustice occurs when a police officer does not give credence to a witnesses' testimony, although they have concrete evidence suggesting that they should, because they are negatively prejudiced towards her (e.g., because of her gender). 21 On the other hand, in cases of hermeneutical injustice, the agent suffers epistemic injustice because they have 'a significant area of their social experience obscured from understanding owning to prejudicial flaws in shared resources for social interpretation' (Fricker, 2007, p. 148). In order to illustrate her conception of hermeneutical injustice, Fricker (2007, p. 7) gives the example of a victim of sexual harassment who cannot make sense of their experience or communicate it to others because it took place 'prior to the time when we had this critical concept'.
Building on Fricker's (2007) conception of epistemic injustice, I argue that the excellent gossiper inflicts epistemic injustice on the subject. The gossiper intentionally comes to acquire epistemic goods which they are not warranted to possess. For example, through gossiping with John, Mary intentionally comes to acquire details about George's private life -e.g., his choice of sex partners -without his consent. To give another (and more elaborate) example, consider the case of a house cleaner called Nick who gossips with other members of staff (e.g., the cook) in order to learn the most intimate circumstances of his employers (e.g., he learns that his employer Kirsten is making a lot of money). By coming to the possession of such juicy and personal information without the consent of the subjects, both Mary and Nick cause epistemic injustice to them. It might be more accurate to call the kind of epistemic injustice caused by Mary's and Nick's actions knower-initiated epistemic injustice. This would denote that, unlike Fricker's (2007) examples of testimonial and hermeneutical injustice, in the case of the excellent gossiper the epistemic injustice is inflicted by the knower. In other words, it is not a wrong done to someone in their capacity as a knower, but an epistemic injustice inflicted by the knower on another agent because the knower intentionally comes to acquire intimate information about them without their consent (i.e. intimate details about a person's private life: e.g. their sex life, their income). This kind of epistemic injustice can be categorized as '… a wrong that a knower perpetrates as a knower' (Pohlhaus, 2017, p. 14).
Note that the kind of epistemic injustice that is caused by gossiping inflicts a specific kind of epistemic wrong on the subject. There is epistemic wrong inflicted on the subject every time another agent acquires the subject's private information without their consent. This is an epistemic wrong, as opposed to a non-epistemic one, because the subject is, without their consent, deprived of their right to decide as to who should have access to such information and hence are wronged by another knower the moment this other knower comes to obtain these epistemic goods. Contrarily, If the subject has consented to sharing their private information with another agent, then there is no epistemic wrong inflicted on the subject by this agent knowing their information. But note that the latter is very rarely, if ever, the case when it comes to gossip -one does not ask for the subject's permission to gossip them. 22 Gossip does not need to have non-epistemic consequences in order to be considered as a non-virtuous trait. For instance, in the house cleaner example discussed above, imagine that Nick sells the juicy information he acquires. By spreading such information, he inflicts non-epistemic wrong on the subjects (e.g., harms their reputation). But even if Nick keeps the gossip to himself (and hence does not inflict nonepistemic wrong on the subject), he still inflicts knower-initiated epistemic wrong on them. He has intentionally come to acquire information about their personal lives without their consent.
Also, note that non-epistemic wrong inflicted by gossip does not require that the gossiper is successful in acquiring truthful information about the subject (e.g., they may spread false rumors about them). On the other hand, epistemic wrong hinges on the acquisition of truthful information. There is no epistemic wrong in acts of gossip that do not result in the acquisition of another person's personal information. If Mary is unsuccessful in acquiring information about George's sex life, then she has not wronged him epistemically -i.e., she has not come to possess personal information about George without his consent. Related to this is the idea that the excellent gossiper, because of their competence in the activity characteristic of the gossip, is more likely to inflict epistemic wrong on the subject than the non-excellent gossiper. Intentionally acquiring another person's private information without their consent is a necessary and sufficient condition for knower-initiated epistemic injustice. 23 Putting forward an argument that relies on the idea that there are certain epistemic goods which epistemic agents are not warranted to possess might strike some readers as dangerous and 'anti-epistemological'. One could argue that such a position could be used as a precedent to argue that there are many epistemic goods which knowers are not entitled to possess. I am sympathetic to such worries. Still, I want to stress that my conception of 'unwarranted' epistemic goods is restricted to agents' personal and private information from which, had it been public knowledge, society would not benefit. My argument relates to the principles of information privacy which informs privacy laws (see, e.g., EU's General Data Protection Regulation and the Data Protection Directive). It is the kind of information protected by such laws that I have in mind when talking about knower-initiated epistemic injustice.
Lastly, one might note that the paper focuses predominately on the effects of gossip on the agent or the subject, and not on the effects on the community. This relates to Kawall's (2002) critique of virtue epistemology as lacking with respect to the community and is not an objection that is unique to my argument. Still, one could insist that more needs to be said on this topic given that gossip is a communitydefined practice and part of my critique rests on community transmission. We can identify cases, for example, where acts of gossip inflict epistemic wrong on the subject but also produce epistemic (and non-epistemic) benefits for the society as a whole. For instance, in many cases of public revelation of serial sexual harassers, the identity of the victims is also often disclosed without their consent or awareness -or again relevant information is imparted and readily allows for the identification of the victim in small communities. But it is also clear that it is the public revelation that finally leads to repercussions for the harassers. Accordingly, we can on the one hand acknowledge the benefits of gossip for the community while on the other hand recognize that such community benefits also bring about a cost to the individual, that is epistemic injustice for victims who had their identity and/or private information publicly revealed without their consent. 24 Note that even if in certain (limited) cases the benefits of gossiping outweigh its negative consequences, it still remains that excellent gossiping inflicts epistemic wrong on the subject -viz., the subject is deprived of their right to decide who should have access to her private information. 23 Nonetheless, it should be noted that additional wrong is often inflicted on the subject by the use or redirection of gossip. The spreading or use of gossip, even under the most charitable of circumstances, is problematic and irresponsible and that is because there is no accounting for the direction it will take or the way it may be used. For instance, in the house cleaner example discussed above, Nick cannot possibly know (or reliably predict) how the other party (i.e. other members of staff) is going to use and/or with whom they are going to share this new information that they have acquired through gossiping with him. 24 It could even be argued that gossip is not the right path to public revelation in such cases precisely because of the absence of control over its spread and use in/by the community (it is virtue irresponsible, despite best intentions). Public revelation is important but needs to be done respectfully and with the consent of the victims.
The fact that the violation of this right can sometimes yield positive (epistemic and non-epistemic) outcomes does not undo the fact that it is an epistemic wrong inflicted on the subject.

Gossip and the no-Wrong Principle
Thus far in this section, I have argued that the excellent gossiper inflicts epistemic wrong on the subject intentionally. Still, as I have shown in section 2, gossip qualifies as an intellectual virtue under the responsibilists' understanding of the concept of virtue. This leaves virtue responsibilists with three main options: (a) resist my argument that gossip qualifies as a responsibilist virtue and/or that excellent gossiping inflicts epistemic wrong, (b) bite the bullet and argue that the intellectually virtuous agent sometimes inflicts epistemic wrong on other agents intentionally, (c) develop a no-wrong principle that disqualifies gossip from being categorized as an intellectual virtue.
First of all, virtue responsibilists could resist my initial argument (section 2) according to which gossip qualifies as an intellectual virtue under their understanding of the concept of virtue. If they succeed in doing so, then they will have shown that their conditions for epistemic virtues do not allow for an intellectually virtuous person who inflicts epistemic wrong intentionally. A proponent of virtue responsibilism could also argue against my position according to which gossip inflicts epistemic wrong on the subject (section 3.1). This would defuse my argument since, irrespectively of whether gossip is an intellectual virtue or not, they would have succeeded in showing that gossip does not cause epistemic injustice. 25,26 Still, I believe to have sufficiently shown that both my arguments hold under scrutiny. I have anticipated and discussed objections to my argument that gossip qualifies as a responsibilist virtue (section 2.3) and to my view that the excellent gossiper inflicts epistemic wrong (section 3.1).
Another option for virtue responsibilists is to bite the bullet and concede that intellectual virtuous agents sometimes inflict epistemic wrong on other agents intentionally. I am skeptical of such theoretical move and believe it greatly weakens the value of responsibilist virtues: if virtuous activities inflict epistemic wrong, then why do we consider them virtuous? Arguing that activities characteristic of certain intellectual virtues could inflict epistemic wrong is also inconsistent with some key arguments of virtue responsibilists. Significantly, Zagzebski (1996, p. 104) argues that intellectual virtues '…are qualities that deserve praise for their presence, and blame for their absence' (see also Baehr, 2011, p. 27;Battaly, 2017b, p. 226). But there is nothing praiseworthy in intentionally acquiring another person's private information without their consent. Moreover, Roberts and Wood (2007, p. 155) argue that intellectual virtues are valuable because the virtuous agent is driven to acquire '…knowledge that enables human life and promotes human wellbeing rather than knowledge that degrades and destroys'. Still, as I have already argued, the kind of knowledge that excellent gossipers acquire through acts of gossip inflicts epistemic injustice (and hence qualifies as knowledge that 'degrades and destroys').
The third available option for virtue responsibilists is to explicitly add another condition to their list of integral features of intellectual virtues. I call this the nowrong principle of intellectual virtues: X possesses an intellectual virtue Y only if X does not frequently and/or intentionally inflict epistemic wrong (including knowerinitiated epistemic wrong) on other agents through the activity characteristic of this virtue.
I believe that the no-wrong principle is the best way for virtue responsibilists to handle cases such as the excellent gossiper. It highlights the fact that for a person to possess an intellectual virtue, they must not inflict epistemic wrong through the activity characteristic of this trait. Hence, the no-wrong principle disqualifies the excellent gossiper, who intentionally inflicts epistemic wrong on other agents through the activity characteristic of gossip, from being considered as intellectually virtuous. Note that although virtue responsibilists mention the importance of acquiring knowledge that does not degrade and destroy (e.g., Roberts & Wood, 2007, p. 55), they have not explicitly stated that not inflicting epistemic wrong is an integral feature of epistemic virtues (see, for example, the list of conditions discussed by Zagzebski, 1996 andBaehr, 2016). The no-wrong principle makes explicit that epistemic traits such as gossip (whose characteristic activity always inflicts epistemic wrong if carried out skillfully) should not be classified as intellectual virtues.
Notice that the frequency with which an agent inflicts epistemic wrong as well as their intentions are quite significant. According to the no-wrong principle, the virtuous agent does not frequently and/or intentionally inflict epistemic wrong through the activity characteristic of virtue Y. This allows for exceptional and sporadic cases in which the agent possessing Y inflicts epistemic wrong unintentionally. For example, consider the example of Sarah who possesses the intellectual virtue of curiosity. 27 Sarah, through the activity characteristic of this virtue, unintentionally comes to possess another person's private information without their consent. In such a case, Sarah does not cease being an intellectually virtuous agent -she came to possess juicy information about another person unintentionally and this is not something that happens to her frequently through the activity characteristic of the virtue of curiosity. 28 Not only is the no-wrong principle efficient in defusing cases such as the excellent gossiper, but one could even go as far as to argue that incorporating the no-wrong principle in their conception of epistemic virtues would give an edge to virtue responsibilists over virtue reliabilists. On the one hand, the virtue responsibilist understanding of intellectual virtues readily allows for the inclusion of a no-wrong principle (one could even argue that it already seems to imply it to a certain extent). On the other hand, it would be difficult (if not impossible) for virtue reliabilists to include such a condition to their conception of intellectual virtue -for example, excellent vision is a virtue for them irrespectively of whether the agent possessing it inflicts epistemic wrong (e.g., spying on someone). Still, one could argue that virtue reliabilists are not interested in excluding epistemic wrong from their conception of intellectual virtue. Nonetheless, even if this is the case, explicitly incorporating a no-wrong principle in responsibilist virtues shows that the responsibilist conception of virtue is more demanding than the reliabilist one. It also consolidates the idea that it is much more praiseworthy for an agent to possess responsibilist virtues than reliabilist ones.

Concluding Remarks
My main goal in this paper was to examine whether gossip should be considered an epistemically valuable character trait. Initially (in section 2), I argued that gossip qualifies as an epistemic virtue under the responsibilist conception of virtue. The excellent gossiper is (i) motivated to acquire epistemic goods through gossiping, (ii) reliably successful in acquiring epistemic goods through gossiping, (iii) competent at the activity of gossiping and (iv) good at judging when, with whom and what to gossip. However, I proceeded to argue (in section 3.1) that the excellent gossiper inflicts knower-initiated epistemic wrong on the subject. The excellent gossiper intentionally comes to acquire another person's private information (e.g., their sexual preferences) without their consent. I concluded (section 3.2) with the argument that virtue responsibilists have three ways to resolve the tension that arises from the conclusions of my two main arguments (i.e., 1. that gossip qualifies as an epistemic virtue and 2. that the excellent gossiper inflicts epistemic wrong). They could resist either, or both, of my arguments; they could bite the bullet; or they could explicitly include a no-wrong principle to their list of conditions for a trait to be considered an epistemic virtue. I argued that the third option is by far the better choice: the no-wrong principle excludes epistemic wrong from being the intentional and/or frequent result of an act stemming from an intellectual virtue and fits well with the virtue responsibilist understanding of intellectual virtue.