Collegiality, Friendship, and the Value of Remote Work

Philosophers have not paid much attention to the impact of remote work on the nature of work and the workplace. The overall aim of this paper is to contribute to further debate over the value of remote work by focusing on one important dimension of it – the effect on collegial relationships. I distinguish two types of collegial relationships. On the one hand, there are what I call “Kantian collegial relationships”, which have been outlined in a recent account by Betzler & Löschke. These are colleagues who acknowledge and respect each other as equals, and provide relationship goods in a fair and distributive manner. On the other hand, there are what I call “collegial friendships”. In contrast to Kantian colleagues, collegial friends are emotionally supportive and show appreciation for each other for who they are, not just because they are equals. Both types of collegial relationship must be taken into consideration if we want to fully understand why we value our colleagues. I then show that there are reasons to believe that remote work will have a chilling effect on collegial friendships. Remote workers must interact online, which undermines forms of self-disclosure that are necessary for intimate relationships to form, like collegial friendship. Conversely, I suggest that remote work, for exactly the same reasons, will have a positive effect on Kantian collegial relationships. The proliferation of remote work forces us to consider whether this tradeoff is acceptable.


Introduction
Remote work 1 will have a significant impact on workers' ability to form and maintain collegial relationships, yet there has been little philosophical investigation of "remote collegiality" 2 . This is surprising because of the overwhelming evidence of a positive relationship between the quality of collegial relationships and meaningful work. Collegial relationships are a determining factor for job satisfaction, organizational commitment, well-being, and flourishing (Colbert, Bono, and Purvanova 2016;Morrison 2002;Morrison 2004). A philosophical inquiry into the nature and value of remote collegial relationships is therefore necessary if we want to determine the value of remote work overall.
In the following I address this gap in the philosophical literature. I begin by asking what collegial relationships are and why they are valuable. These are questions that, like remote collegiality itself, have not received much attention in the philosophical literature. One exception is a recent account by Betzler & Löschke in this journal (Betzler and Löschke 2021). They define 'colleague' as an equal in terms of work content, institutional affiliation, and level of responsibility. To account for the value of collegial relationships, Betzler & Löschke propose a "relationship goods" account, which has two aspects. First, it specifies certain goods that only colleagues can provide by virtue of being equals, such as assisting one another in work-related tasks or providing feedback and recognition of each other's skills and abilities. Second, the relationship goods account identifies certain reasons that colleagues should have -and understand themselves to have -to generate those goods.
Betzler & Löschke offer a quasi-Kantian view of these reasons: colleagues should provide relationship goods because they respect and acknowledge each other as equals. I will therefore call these "Kantian collegial relationships".
In my view, it is not necessary that a good colleague must always be a Kantian colleague. A good colleague can also be what I will call a "collegial friend". These are colleagues that, like friends in general, take themselves to have reasons to provide relationship goods because of their emotional involvement and appreciation of who the other person is, not primarily because they acknowledge them as an equal. But unlike regular friendships, collegial friendships are contextual -the goods they involve are distinctly collegial and related to what makes the friends equal. I argue that we must include collegial friendship in our concept of collegiality to fully account for the value of collegial relationships.
I then ask how remote work is likely to affect collegial relationships. I argue that the impact of remote work on collegial friendships will be largely negative. Online interaction, which is the only way for remote colleagues to interact, distorts and inhibits self-disclosure in ways that prevent intimate relationships, like friendship, forming. Because online interaction is characterized by a high degree of voluntary and verbal self-disclosure, it is difficult to pick up on more subtle nonvoluntary and nonverbal cues that reveal intimate information and lead us to assume appropriate emotional attitudes toward the other. The weakened 1 Remote work, as I understand it in this paper, is work that is exclusively remote. This is not the case for all remote workers. Many remote workers are better described as "hybrid workers": they work part time remotely and part time in the office. However, while hybrid work is currently the norm, more and more companies are making a transition to permanent remote work (Smith 2022). It is likely that remote work, as understood here, will account for a substantial number of jobs in the future. 2 The same goes for other workplace relationships, like remote employer-employee relationships, which are unfortunately beyond the scope of this paper. ability to become intimate means that remote colleagues are likely to adopt a dispassionate attitude, akin to what Hubert Dreyfus calls "leveling" (Dreyfus 2009).
I suggest that for exactly the same reasons that remote work has a chilling effect on collegial relationships, it will have a positive effect on Kantian collegial relationships. Kantian collegial relationships do not depend on intimate knowledge, nor do they require a passionate attitude toward the other. Remote work can therefore lead to the flourishing of Kantian collegial relationships. The main accomplishment of this paper is to force the recognition of this tradeoff when substituting remote work for in-office work.

What are Collegial Relationships?
As noted in the introduction, there has not been any sustained philosophical account of collegial relationships. The closest have been inquiries into the nature of what we might call "egotistic collegial relationships". These are relationship that, in line which the description, are between self-interested people who work together. What has interested philosophers is the possibility, or not, of egotistical collegial relationships realizing certain moral values, like solidarity, reciprocity, or justice 3 . But while this line of inquiry is certainly worthy of philosophical attention, it does not get us any closer to what collegial relationships are, in and of themselves. Self-interested people exist outside of business, and there are people in business who are not self-interested. While egotistical collegial relationships point us in the direction of collegial relationships, what the philosophical literature has been lacking is a conceptual clarification of what it means to be a colleague in the first place.
Betzler & Löschke's account in this journal is, as far as I know, the first systematic philosophical analysis of collegial relationships. As a starting point, Betzler & Löschke define a "colleague" as someone who is an equal in certain work-related ways, specifically in terms of work content, institutional affiliation, and level of responsibility (217-218) 4 . This explains why we usually do not call bosses or those who work for different businesses colleagues, though exceptions arise at the margins -journalists may consider one another colleagues if they work the same beat, even if they work for different newspapers, and academics may consider those who specialize in the same field colleagues, even if they do not share the same level of responsibility. However, the paradigmatic colleague is someone who is an equal in all three respects: they do similar work for the same organization at the same level in the workplace hierarchy. For the most part, it is the paradigmatic colleague I will have in mind when I use the term "equal" throughout the paper.
What is notable about 'colleague' as a concept is that being an equal is not sufficient for someone to truly qualify as a colleague. In contrast to other workplace relationships, like coworkers or teammates, being a colleague heavily implies that one adopts certain behaviors toward one's colleagues. In addition to being an equal, a colleague must also be collegial. If a colleague is "uncollegial" (215), whatever that means in practice, they are in some sense not a colleague at all 5 .
Betzler & Löschke claim that collegiality has two dimensions. The first dimension specifies the goods that colleagues can provide. These depend on the ways in which they are equals. Sharing medical knowledge is a good that is particular to colleagues in the medical establishment, and the esteem associated with the Pulitzer Prize is related to the fact that recipients are chosen by other journalists. Even though people who are not doctors can theoretically share medical knowledge, and people who are not journalists can show their acknowledgement of those who excel in the field, the possibility of realizing these goods and the value attributed to them often depend on a basic equality between those sharing them. Of course, a great many collegial relationship goods are not as lofty as these. Most goods provided by colleagues are more mundane, such as helping one another in the office or providing feedback on the work done by others. But in a similar fashion these goods are possible and significant because they are provided by colleagues 6 .
The second dimension of the relationship goods account concerns the reasons colleagues should have for being collegial 7 . While the goods mentioned have value, providing them is not in itself sufficient for colleagues to have a valuable collegial relationship. For example, a manipulative colleague who is only looking after her own interests may be as helpful as any other colleague, but being manipulative is not what it means to be collegial. This is because, as Betzler & Löschke explain, "relationships between colleagues are valuable" only "if the participants take themselves to have special reasons for generating these goods" (219). Clearly, a manipulative colleague is not acting for the reasons we think a good colleague should have. What, then, is the nature of those special reasons?
According to Betzler & Löschke, there is a close relationship between the kinds of goods colleagues can provide for each other qua equals, and the special reasons they have for providing those goods. In their view, colleagues should "relate to each other in secondpersonal ways that reflect an acknowledgement of their roughly equal status" which "not unrelatedly […] suggests that one considers oneself to have special reasons to give that person special treatment" (216). As I understand Betzler & Löschke, what they propose is a Kantian account of collegiality 8 . By this I mean that the special reasons for being collegial are generated by abstracting from personal concerns and values and being instead grounded in values that are shared by virtue of being equals. It is the acknowledgement that one has a certain set of features in common with another person that generates special reasons to provide relationship goods. For example, Betzler & Löschke claim that "[c]ollegial recognition might become less valuable when one's colleague is also one's friend, since a bias might come into play which leads to the recognition being based more on personal concerns than 5 Empirical research suggests that people understand 'colleague' as a "dual concept", meaning that they take it to have an irreducibly normative dimension as well as a descriptive one (Reuter et al. 2020). 6 For more examples of the goods colleagues can generate and discussion of why they are valuable, see Betzler & Löschke (2021). 7 This view evokes a standard conception in the ethics of social relationships. There is wide agreement that social relationship generate "associative duties" which are duties related, in one way or another, to the nature of the relationship itself. Disagreement concerns the source of associative duties. For an overview of positions see Gheus (2022). See, however, Nyholm and Smids (2020) for a defense of a performative account of the value of collegial relationships. 8 I do not suggest that Betzler & Löschke's account as a whole is Kantian, only that their account shares certain features with (an orthodox interpretation of) Kantianism. on professional standards" (223). Personal concerns, presumably, are concerns we have that are unrelated to being equals and therefore cannot generate special reasons for being collegial. Insofar as they generate a reason, it has little to do with collegiality. Consequently, I will use the term "Kantian collegial relationship" when referring to Betzler & Löschke's conception of collegiality.
It does not follow that a Kantian colleague can never accommodate the personal concerns of another, however. An exception is acting upon a concern which is not related to a collegial relationship good in the proper sense. For example, we have some colleagues with whom we prefer to tell jokes, discuss movies, or engage in small talk, but it does not follow that we value them more as colleagues. The reason is that these are goods that are typically incidentally related to the collegial relationship -we share jokes with some colleagues because we happen to be around them, and they happen to share our sense of humor. This is unproblematic since the Kantian account is only relevant when it comes to explaining the value of collegial relationship goods -that is, those goods that we can generate only by virtue of and because we consider each other equals 9 .
I find the Kantian interpretation of the relationship goods account plausible in some respects. It explains the value of a great many collegial relationship goods. For example, when journalists stand up for each other, it matters that they value each other as equals. Every journalist, wherever their journalistic freedom is suppressed, deserves the same amount of recognition and solidarity. The same goes for many of the more mundane ways in which colleagues generate relationship goods. For example, a colleague who is helpful only to a handful of colleagues is a bad colleague. Unless she has a very good reason to discriminate, she is wrong not to acknowledge and respect her colleagues as her equals.
The Kantian interpretation is implausible in other respects, however. This is because it denies an aspect of collegiality which, I suspect, many would agree is part of what it means to be a good colleague. I believe there are two senses in which we call someone a good colleague, in a Kantian sense and in a non-Kantian sense. For example, we sometimes call someone a "close colleague". But what we mean by "close" is (usually) not that they are more equal to us than other colleagues. On the contrary, some colleagues are close even though they are relatively unequal compared to other colleagues. What we mean is that they are close to us in a way that is similar to how a friend or a loved one is close to us -they are someone who care about us in a special way which is only tangentially related to us being equals. Although the relationship has professional characteristics, it also has certain personal ones. Hence, I will argue in what follows that we must include what I will call "collegial friendship" as another way in which someone can be a good colleague.

Collegial Friendship
To get a sense of what I mean by the term "collegial friendship", consider the relationship between Paul and Susan. Paul and Susan are roughly equal in work-related ways -they have similar work duties, work for the same employer, and are at the same level in the corporate hierarchy. Both consider themselves to have good reasons to be collegial toward everyone at work because they acknowledge and respect them as equals. However, Paul's upbringing has imprinted in him a strict Protestant work ethic, and so he finds Susan's more relaxed 9 I thank reviewer #1 for making this point. approach to work inspiring. He therefore prefers to help Susan in work-related tasks because he finds that it helps him see the ways in which he is an overachiever. For example, he notices that she never works overtimes unless it is absolutely necessary, which inspires him to emulate her. Susan, for her part, admires Paul's skills, even though they are only slightly above average. In particular, she finds it amusing that he is great at formulating sentences but frequently misspells common words. Because of this, Paul is the one she most likes to give feedback to, and it has become a routine for them to make light of misspellings that are funny or absurd. Nevertheless, their friendship has yet to extend beyond helping each other in work-related ways. Susan does not think it appropriate to come to Paul when she has a deep personal crisis, nor does Paul ask Susan to drive him to the airport or help him move.
Paul and Susan are colleagues according to one aspect of the relationship goods account -they provide collegial relationship goods for one another, such as helping each other and providing feedback on their work. Unlike colleagues who share jokes or discuss movies, Paul and Susan provide relationship goods that are distinctly collegial -goods that they can provide only by virtue of being equals. However, their reasons for doing so are non-Kantian. They, as Aristotle describes the paradigmatic form of friendship, "have this attitude because of the friend himself" (1999, 1156b, 13) 10 . The reasons they have for providing relationship goods are only tangentially related to their equality. It is not just that they recognize that they have professional concerns in common, though that may be part of it. It is that they have an emotional involvement directed at certain features of the other -Susan's personal perspective and Paul's quirks respectively -that for them generate reasons for providing relationship goods.
One might be tempted to characterize the relationship between Paul and Susan not as a collegial friendship, but as a relationship between colleagues who happen to also have a friendship. This is what is implied in the standard definition of so-called "workplace friendship": "an informal and voluntary relationship between two or more individuals in a workplace, who are interested in one another as whole individuals, not just as colleagues" (Zarankin and Kunkel 2019). This definition, when taken at face value, implies that workplace friendships 'transcend' the collegial relationship -Paul and Susan are interested in one another as 'whole' individuals, not 'just' as colleagues. However, Sias & Cahill's pioneering study of workplace friendship suggests another interpretation. Rather than the workplace being the place "where friendships evolve", they propose that "the context [i.e., workplace] is part of the friendship process itself" (Sias and Cahill 1998; see also Bridge and Baxter 1992). This is another, and in my view plausible, way of looking at Paul and Susan's relationship 11 . What makes them colleagues is not tangential to their friendship, it is a part of it. It determines what it means for them to appreciate one another for who they are, and what it means to be good toward one another 12 . Yet, what they have is clearly distinct from a Kan-tian collegial relationship. They are equals, and they do things for each other that they can do only by virtue of being equals, but they do not do it because they are equals -they do it because they are collegial friends.
Being a collegial friend is therefore another way of being a colleague. Collegial friends provide relationship goods for each other that they can only provide because they are equals, but the relationship is more intimate and emotionally involved and this is part of what it means for them to have this relationship. The question we must now ask ourselves is whether collegial friendships are valuable. Is it good to be a collegial friend, or is it more like being a manipulative colleague? Betzler & Löschke seem to think that collegial friends act for the wrong reasons. At the very least, one should strive not to be a collegial friend in instances where one must choose between being a collegial friend or a Kantian colleague. There are, as Betzler & Löschke put it, "reasons not to be the best possible colleague" (223), since it introduces a bias that in their view devalues the collegial relationship. I believe there are two problems with this view.
For one, this view fails to appreciate the value of friendship. Friendships are not just nice to have, but, as Aristotle reminds us, "most necessary" (Aristotle 1999(Aristotle , 1155a to live a meaningful life. A world without friends, would be a world of cool, thin, and somewhat disinterested relations. We would have acquaintances, colleagues, neighbors… but with whom would we share our lives, our joys, our innermost feelings? With whom could we be ourselves, spontaneous, open and fully trusting? (Annis 1987, 351) So, assuming that friendship is valuable, collegial friendship must be of some value. Against this, it might be objected that the workplace is not the appropriate place to form friendships. One could argue that collegial friends should only conduct their friendship outside of work, since such friendships introduce biases and preferential treatment which is contrary to the values of collegiality. This would presumably preserve the value of friendship without compromising collegial relationships. The problem with this response is that it downplays the importance of work in our lives. Work, as Schwartz (1982) argues, is a part of life 13 , not incidental to it. Meaningful work is constitutive of a meaningful life. Therefore, since collegial friendships can be a valuable part of meaningful work, it is reasonable that collegial friendships ought to be pursued, not just tolerated.
Second, there does not appear to be any a priori reason that being a Kantian colleague is better than being a collegial friend. Sometimes it is better to be a Kantian colleague, but at other times it is better to be a collegial friend. For example, prioritizing a collegial friend is wrong when deciding on who should become employee of the month. The nomination should be based on an objective assessment of skills and virtues, not personal fondness for a colleague's style of work. By contrast, receiving praise from a collegial friend for becoming employee of the month is more valuable than praise from a Kantian colleague. It feels more special to receive praise from a colleague if the relationship has a degree of intimacy.
Collegial friendships do raise important ethical considerations, however, as Betzler & Löschke correctly point out. There are instances where the duties of Kantian collegial rela- 13 It is an open question whether work ought to be a part of life. I do not think so, but it does not change the fact that it is for the majority of people. Furthermore, it is not a contradiction to claim that collegial friendships are valuable while also claiming that work is not. Indeed, pursuing collegial friendships is one way to alleviate the burden of having to work. tionships conflict with those of collegial friendships 14 . But the potential for conflict can be overstated. We take on several roles in our life, and it is part of what it means to be a social being to be aware of these roles and navigate around them. The way we solve this in practice is not by creating a hierarchy of roles. Instead, we rely on experience and practical wisdom to make judgements about which roles are appropriate to take on in any given situation (Cross, Borgatti, and Parker 2002). What we want is to "harmonize" (Sherman 1987, 592) the ends of Kantian collegial relationships with those of collegial friendship, not to lionize one or the other. This can be facilitated at the organizational level as well by having routines for solving conflicts and promoting a virtuous organizational culture. While situations may arise where there is no clear answer, we should not thereby conclude that it is best to stay away from either kind of relationship.
In conclusion, I maintain that collegial friendships must be included in our concept of collegiality. It explains one of the ways that we commonly refer to another as a colleague, and why we value our relationship with them. And although collegial friendships are not always ethically unproblematic, the ethical concerns they raise are not, in my view, such that we should discourage them altogether. There are, to paraphrase Betzler & Löschke, reasons for being the best possible colleague. I now turn to the question of what it means to form collegial relationships when working remotely.

Remote Work and Collegial Relationships
With a few exceptions like telemarketing and customer service, it is only in the last decade that remote work has gone from a fanciful notion to a practical possibility for large swathes of the workforce. The adoption of remote work has been accelerated by the pandemic and will have a major impact on the nature of work going forward (Gartner 2020). Among the many consequences of remote work is that it disrupts the 'normal' way collegial relationships are forged. Instead of socializing in the office, remote colleagues must socialize online. In this section I argue that this will have a chilling effect on collegial friendships. I also suggest that, for the same reasons, remote work could have a positive effect on Kantian collegial relationships.
I assume that, based on current trends, the way remote colleagues interact is through a work-related social media platform, like Microsoft Teams, Facebook Workplace, or Slack 15 . These platforms have a similar architecture. They are made up of channels and sub-channels -at the top level encompassing the organization as a whole and at the lowest level only including a few people. Within these channels, members can communicate either via chat messages or video calls. Usually, there are some channels that are mandatory, as at the organizational or team level, but there is also the opportunity to create one's own channels, on the condition that others are willing to join. While it is likely that remote colleagues will want to connect on other social media as well, like Facebook or Instagram, I will assume that other social media are more likely to play a supportive, rather than primary, role in establishing and maintaining collegial relationships.
Surveys suggests that remote workers have fewer collegial friends than those who work in an office (Olivet Nazarene University 2019) and that remote collegial friendships take longer to develop and have a higher risk of failure (Schinoff, Ashforth, and Corley 2019). One barrier to forming remote collegial friendships is that remote workers interact mostly when they have a work-related reason to do so. Therefore, the casual exchanges that are a natural part of how people get to know each other are fewer and farther between. However, as remote work became more common during the pandemic, numerous strategies were developed to allow for more socializing between remote workers, such as marking off time for informal socializing before and after video meetings, creating channels for sharing stories, jokes and memes, and encouraging workers to have one-on-one interactions. As remote work becomes even more ubiquitous, it is likely that these strategies will be further developed and optimized.
I want to suggest that remote work will nevertheless have a chilling effect on collegial friendships 16 . This is because there are barriers to forming collegial friendships that are not easily overcome, even if strategies are developed to facilitate them. This has to do with how online interaction distorts and limits self-disclosure of nonvoluntary and nonverbal cues that play an important role in the formation of friendship.
There are two ways that nonvoluntary and nonverbal cues are important in a budding friendship. One has to do with the disclosure of personal and intimate information. Being intimate involves knowing certain things about one's friend that others do not 17 . However, what a friend ought to know is not primarily a set of facts, like biographical information, but information that is highly context sensitive. For example, in addition to knowing how a friend is doing in general, they should also know how they are doing right now, as the following example illustrates: I notice […] my friend is anxious in confined spaces, in crowded places, or when her ex-partner is in the room [therefore] I will, for example, be more attentive to my anxious friend when her ex-partner enters the room, or try to lighten up the situation with a joke or some strategy of distraction or just discreetly get her out of the room (Cocking and Matthews 2000, 227).
Clearly, I would not be a good friend if it is only after her ex-partner has left the room that I notice my friend's anxiety. It is even worse if she has to tell me explicitly. As a friend, I should be able to pick up on her anxiousness even if she does not disclose this to me directly, but it is impossible to do so other than by relying on nonvoluntary or nonverbal cues.
The second role played by nonvoluntary and nonverbal cues in friendship is to cause reactive attitudes 18 . Reactive attitudes are attitudes and emotions that refer to others, such as feelings of love, compassion, anger, or guilt. These attitudes -or more specifically their intensity -are enhanced when informed by nonvoluntary and nonverbal cues. Although verbal and voluntary disclosure can cause reactive attitudes, nonverbal and nonvoluntary cues are significant because they have a deep psychological impact on how we react to others. A 16 Here I draw on arguments put forth by Cocking and Matthews (2000); Matthews (2008); Cocking (2008) and, to a lesser degree, McFall (2012). 17 Cocking and Kennett (1998) call this "the secrets view" of intimacy. For critiques of the secrets view, see Cocking and Kennett (1998); Inness (1992); Reiman (1976). The view I adopt is influenced by these critiques. 18 The term "reactive attitude" was coined by Strawson (1962). Here I rely on the interpretation by Turp (2020) on the role of reactive attitudes in friendship. smile elicits more feelings in us than hearing someone tell us they are happy, and our feeling of compassion is at its strongest when we see the pain someone is going through. What is important is therefore not just disclosing intimate information, but also the form in which it is disclosed. While there is nothing in principle that stops us from disclosing intimate information voluntarily and verbally -we can tell our friends we are anxious or that we are happy to see them -the issue is that being told intimate information has a different and less intense effect on us than when we infer it from more subtle cues. In short, nonvoluntary and nonverbal cues are the most "powerful elicitors" (Turp 2020, 274) of the kinds of emotions friends should have toward each other.
The role that nonvoluntary and nonverbal cues play in friendship is a problem when trying to form friendships online. Nonvoluntary and nonverbal cues are distorted and limited when interacting online (Walther and Parks 2002). Online interaction, in general, engenders "a far more carefully constructed picture of […] oneself than otherwise would be possible in nonvirtual contexts" (Cocking 2008, 125). This is especially true in text-based interaction which consists mostly of deliberately written speech. Video calls do offer a broader range of cues, like body language and the auditory qualities of speech, but is limited in comparison to face-to-face interaction, for example by circumscribing movement, haptics, and general appearance (more than the top half of the body). Environmental cues are also curtailed. In video calls, the background tends to remain static, and the camera can only capture so much. For example, disturbances outside the view of the camera will only be seen by the one the camera is facing. These attributes of online interaction therefore distort and limit the very cues that we rely upon to form intimate relationships.
Let us imagine how online interaction could have a chilling effect on the formation of friendship between Paul and Susan. Imagine that Paul gets to know that Susan has a rather carefree attitude toward work during a video call. Upon learning this, he tells Susan that he has grown up with a strict work-ethic and that he wishes he was able relax more. Susan, in response, gives Paul tips on how to become more laid-back. Although there has been an exchange of the same information as in the example from the previous section, it will take more for Paul to become truly inspired by Susan. For him to come to appreciate Susan's attitude, it is best for him if he can see Susan in action -for example by seeing how she does not read her e-mails the instant they arrive in her inbox, or how she is able to shrug off criticism. Paul can only be inspired by her words, not emulate her actions. The kind of intimacy required for friendship, one where Susan's character is disclosed directly to Paul, is foreclosed.
Regarding Susan, we can imagine that she is amused by Paul's misspellings and therefore wishes to give him feedback. However, her efforts to bond with Paul fail because it is difficult for her to show that she is being ironic when pointing out his mistakes. She knows that her demeanor does not signal a critical attitude, but since they are in a video call, she worries that Paul will mistake her irony for seriousness. She is afraid that Paul will be resentful, rather than amused and playful. An important reason for Susan's failure to elicit in Paul the attitudes and emotions she wishes to is that she is not able to disclose those cues that would otherwise evoke them.
These examples illustrate how online interaction negatively effects the formation of a collegial friendship between Paul and Susan. Note that the constituents of a collegial friendship are there, just as they were in the example from the previous section. The problem is that they are not allowed to unfold, and so the ease with which their friendship would other-wise develop is stunted. But this does not mean that Paul and Susan cannot have a valuable and fruitful collegial relationship. There is nothing that stops them from assisting or help each other out in other ways. What it means is that it is less likely that they will come to do so because they are emotionally involved and appreciate the other for who they are. In other words, their relationship is more likely to develop into a Kantian collegial relationship. This is what Dreyfus (2009) predicted in his inquiry into the effects of the internet on people. According to Dreyfus, the internet tends to "level" 19 our concerns for the world and others. Rather than paying attention to the qualitative differences between people and things, we see their differences primarily as a matter of quantity. Dreyfus attributes leveling to the way information flows on the internet, which decontextualizes and juxtaposes information from disparate sources. The Internet therefore promotes the ideology that "[n] othing is too trivial to be included [and] nothing is so important that it demands a special place" (ibid., 79), which in turn encourages people to favor dispassionate reflection over emotional engagement and general opinion over focused attention. The only way to escape leveling is by committing to an 'ethical' or 'religious' response where a determination is made of what constitutes the important and the unimportant 20 . Relatedly, to form a remote collegial friendship we must commit ourselves to get to know and be there for the other. We must be diligent in acquiring intimate information and allow ourselves to develop the right emotions. This requires significant time and effort, since we are barred from accessing the nonvoluntary and nonverbal cues that we would otherwise rely on 21 . The relatively high cost of the commitment to form a remote collegial friendship, combined with the relatively high risk of it not coming to fruition, is therefore likely to deter many workers from making the investment in a deeper collegial relationship.
Interestingly, the reasons why remote work has a discouraging effect on collegial friendships are exactly the same ones that mean it may have a positive effect on Kantian collegial relationships. Those in a Kantian collegial relationship are not dependent on being intimate with the other, they must only be acquainted with how they are equals. The latter is facilitated by channels that, by their very nature, bring together those who are equals, e.g., the channel for a team working on a specific project or the channel for the marketing department. There is no need for someone who is a member of a channel to guess what the goals of the other members in the group are, or how to contribute toward them. Furthermore, it is a benefit for Kantian collegial relationships that online interaction generates a somewhat dispassionate attitude to others. It is not good for Kantian colleagues to get too emotionally involved, since it can lead to biased and preferential treatment. So long as their dispassionate attitude toward one another does not lead to apathy 22 , it is possible for Kantian collegial relationships to flourish among remote workers. 19 Dreyfus gets the term "leveling" from Kierkegaard who uses it in reference to the way that the burgeoning press of his time promoted, he believed, a species of Enlightenment reason which "levels all distinctions", see Kierkegaard (2001). 20 For the nuances between the ethical and religious responses in this context, see Dreyfus (2009, 83-88). 21 Briggle proposes a model of online friendship based on 'penpalship' where, precisely by utilizing the distance and deliberateness offered by online interaction, online friends can "submerge the friendship to greater depths, or as a brighter beam of light with which to explore aspects of our selves" (Briggle 2008, 77). This is an interesting possibility worthy of further exploration in the context of remote work. 22 It is well established that remote work can lead to feelings of isolations and depression (Van Zoonen and Sivunen 2022). Remote work is therefore not unambiguously beneficial for Kantian collegial relationships.
There is more to be said about how remote work will affect collegial relationships, especially Kantian collegial relationships. Clearly there are other factors that determine how remote collegial relationships develop, such as organizational culture and the nature of the work. Given that the mass adoption of remote work still lies ahead of us, it may seem presumptive to make sweeping claims about the future of collegial relationships. Let me therefore emphasize that what I have proposed is that one aspect of remote collegial relationships -the effects of online interaction on self-disclosure -is a factor that workers and businesses must consider when thinking about the future of collegial relationships. If not addressed in some way, there are reasons to believe that collegial friendships will suffer in the long term and that it will be more fruitful to encourage Kantian collegial relationships.

Conclusion
The aim of this paper has been to explore the nature and value of collegial relationships, and how they will be affected by remote work. I have argued that a comprehensive account of collegiality must include collegial friendship as one of its aspects, as it is part of being a good colleague. Then I argued that remote work is likely to have a chilling effect on collegial friendships because of the limited disclosure of nonvoluntary and nonverbal cues. I also suggested that, for the same reasons, it is possible that Kantian collegial relationships will flourish in a remote environment.
The results of this paper are consequential for the philosophy of work, but even more so for the philosophy of remote work, which has yet to receive the philosophical attention it deserves. Although the paper only scratches the surface of what there is to be said about remote work, I believe it shows that philosophers have a lot to contribute to the assessment of a phenomenon that will have significant effects in the short term and far into the future. Since remote work has an effect on core values, like that of collegial friendship, a deeper understanding of the ethics of remote work is necessary. To that end, I hope that this paper will be a source of inspiration.