Empathising in online spaces

ABSTRACT

This paper aims to better understand and account for potential difficulties in empathising with each other in online spaces. I argue that two important differences between online and in-person communication are both to do with what information comes across in equivalent interactions. Firstly, there are ways in which less information comes across in online interactions (both consciously and unconsciously). Secondly, agents have greater control over what information comes across in online interactions. I argue that these differences can cause problems in terms of whether online spaces are as fruitful ground for empathising with each other. This is because the information that gets left out is often what’s important for building up a fuller and more vivid picture of what the other person’s experiences are like, such as identity-markers or vulnerabilities. But despite this, I end on an optimistic note, because we will become more practiced at reading between the lines, as well as adapting and using media more creatively.

KEYWORDS:

• Empathy
• the internet
• online communications
• understanding
• internet ethics
• philosophy of technology

1. Introduction

Any new technology – especially that which ends up making a substantial difference to how we live – is likely to face its fair share of suspicion. But just as we might think Socrates was wrong to hand-wring about the practice of writing,¹ so too are some people wrong in their worries about the increasing number of interactions that take place online. Furthermore, there’s a dimension of inequality that we should bear in mind: historically, the panic that comes with new technology was sometimes used to discourage members of certain groups from making use of it – such as when electricity or trains were decried for bringing danger to women and children.² Online spaces can be incredibly valuable to members of oppressed groups, in no small part due to their ability to bring together communities of marginalised people, and their ability to allow such members to interact with others in a safer medium.³ We shouldn’t fear change just because it’s change; sometimes things will get better.

But, as long as we’re cautious, this shouldn’t prevent us from looking into whether there are some justified concerns about such online spaces, and whether those concerns can be overcome.⁴ The aim of this paper is to better understand one worry in particular: whether online interactions tend to be less fruitful grounds for the participants to empathise with one another. I intend both to validate this worry to an extent, and then to optimistically show how things might not be as bad as they first seem, and that we have reason to believe they’ll continue to improve.

I will do this by arguing that two of the differences between online and offline communications are that, in the former, (1) interlocutors often have access to less information about each other when they interact and that (2) interlocutors each have more control over what information is conveyed. Both discrepancies, I’ll show, are the source of some of the worries we might have about empathy – since much of the kind of information that doesn’t come across is potentially relevant for understanding (and getting a more vivid picture of) what that person’s perspective is like. This is particularly interesting, since control over our information – when we have it – is such a benefit to our online interactions in other ways. Finally, I’ll end with an optimistic note: one about how we are already adapting to better empathise online, and how we are likely to continue to do so.

Section 2 begins by introducing the idea of empathy in more detail, in part by giving three examples of interactions in which empathy can be valuable. Each of these three cases will have a version which occurs online, and a version which occurs in person. In Section 3, I propose a metric for understanding why some circumstances can be better foundations for empathy than others: in each of the three examples, I show how less information may be conveyed in the online versions of the communication compared to the in-person versions. Section 4 is where I refine the problem – looking specifically at the kinds of information that interlocutors usually either can’t or won’t voluntarily share online. I suggest that if there’s a reason to be wary of empathising in online spaces, here’s the best place to find it.

2. Empathy and online spaces

Many people who have concerns about how we interact with one another online will worry about something broader than just whether online spaces are fruitful grounds for empathy. For instance, they might also worry about a number of similar and related forms of fellow-feeling, such as how the online spaces affect our ability to care for each other, to sympathise with each other, or to treat each other with respect. Although in this paper my focus will be on empathy, I expect that the conclusions can be generalised more widely.⁵

Empathy, insofar as I am interested in it, is a way of coming to share in another person’s experiences.⁶ It won’t always be something we do intentionally, or for any good reason, or that we always want to do. It is a phenomenon where our minds go some way towards matching the experience of another person, providing us with an insight into what an experience might be like for them.⁷ I might succeed in empathising with Carmella when I see that she’s angry with her husband by recreating a version of that anger in my own mind, while still recognising it as anger that I am feeling for (and with) her rather than on my own behalf. Another person might fail to empathise with Carmella by mistakenly taking her to be sad, instead of angry, and so recreating the wrong sort of feeling in their own mind. Successfully empathising with someone can give us an insight not just into an experience on its own, but what the experience is like for a specific person.⁸

For now, I will continue my investigation of empathy by listing three possible cases, and showing how each of these cases might work in an online space and a physical space.

The referendum (in person)

Bec and Jasper have never met before, but are both at the house of a mutual friend. The mutual friend starts a conversation about the results of a local referendum. Bec says that they hope the referendum is successful. Jasper hears this and has the opposite views about what he would like to happen. Jasper expresses this out loud.

The referendum (online)

Bec and Jasper have never met before, but are both on the same social media platform and share a mutual friend. The mutual friend posts on social media about the results of a local referendum, in a way that allows for their friends to leave comments and have a discussion. Bec leaves a text comment, saying that they hope the referendum is successful. Jasper reads this and has the opposite views, so expresses them via a further comment on the post.

The commiserating (in person)

Suppose that Dave has recently been offered a job in a city that his friend Jade used to work in. Although Dave was initially excited about the prospect, he’s since realised how bad the rental market in the city is, and how little he will be able to afford. He meets up with Jade at a pub to complain about his situation to her.

The commiserating (online)

The same set of circumstances apply to Dave and Jade, but instead of being able to vent his frustrations in person, Dave vents to her via an online messaging service. She’s online, and replies in real time with written messages.

The interview (in person)

Rose is going to interview John in person. John arrives at the room on time, and sits across from Rose as he answers her questions about his suitability for the job.

The interview (online)

Rose is going to interview John over an online video calling platform. John arrives at the online room in time, and both Rose and John appear on their own and each other’s screens. They each have a working microphone and camera, and John answers Rose’s questions about his suitability for the job.

I have picked out three examples in which we might be worried that the online alternative to the in-person version is less fruitful ground for the participants to empathise with one another. This is so that we can better understand whether such worries are legitimate, whether there might actually be trends that make empathy more difficult that occur in the online cases that happen because they are online, rather than in person.

Bec and Jasper haven’t met before in either case, but a political discussion on social media is one where it might be harder to see things from the other person’s perspective than if the same discussion were had face-to-face. Jade is in a good position to empathise with Dave because she’s been in a similar situation before – but when he’s only venting to her over text messages she might not fully be able to pick up on the depth of feeling in the same way as she would if he were in front of her. And even though Rose’s main objective isn’t to empathise with John, John might be right to worry about coming across less well, being understood less, and being less easy a target of Rose’s empathy, if he’s not interviewing in person.

There are multiple ways to worry about empathy in online vs in-person communications. One simple version of the problem might be if some spaces are less likely to bring about any empathy at all. Perhaps, for one person, interacting with others online might leave them struggling to view their interlocutors as being people with experiences, motivations and perspectives, whereas in physical interactions that same person might have no problem in coming to understand others in that particular way. But this might not cover many cases; empathy seems more likely to be scalar than binary: it’s something we can achieve to greater or lesser extents in different situations. If one of my precariously employed colleagues is struggling with money, perhaps I’ll be able to empathise with them even more than some of the more senior members of staff in my department, who haven’t ever had comparable experiences. That’s not to say that the more senior professors can’t empathise at all – they can understand well enough that it’s a bad situation to be in, and they’ve certainly faced some troubles in their lives, even if there are differences. But perhaps in this situation I’m able to get more of the details right, and feel them more strongly in the way that the target of my empathy feels them. So although sometimes the worry might be that online spaces prevent us from empathising, in plenty of other cases the worry might be that we aren’t able to empathise as well with the person in question as we would’ve done in the same interaction in person.

I find it useful here to think about a distinction Kagan makes about different kinds of belief. Kagan, interested in what moves us to act morally, makes a distinction between ‘pale’ and ‘vivid’ beliefs:

We must distinguish between two ways that beliefs can be represented in the mind: a belief can be vivid or it may only be pale. Pale beliefs are genuine beliefs, and they may well be deeply entrenched beliefs, but they are displayed to the mind in such a way that the individual does not fully appreciate their import.⁹

The description of some beliefs as being either vivid or pale is useful when thinking about why some instances of empathy may be more accurate than others. It’s not enough to know a list of propositions about what the target of empathy feels – the empathiser must in some sense share in the affective side of the mental content as well.¹⁰ We can think of other analogies here – take, for instance, the difference between how a bright and realistic painting of some breakfast might really get you in the mood for some food compared to a picture that is vaguer, blurrier, or duller in colour. The former might make your mouth water more, and cause you to feel like you can almost smell the aroma coming off the plate, in a way that the latter picture doesn’t. Or for those who aren’t sighted, a more accurate soundscape might better bring on stronger emotions than one that is less detailed. These same differences can also account for how accurately we are able to empathise with one another – whether we might have a pale imitation of the target’s experiences or a stronger, more vivid one.

3. Information

Given that sometimes we can do a better or worse job of empathising with one another, what are some of the conditions that can influence us? In this Section I will first discuss these conditions generally, and then look specifically at the online vs in-person cases. I propose that one feature of communication that can make for a more fruitful ground for empathy is the amount of information conveyed by the potential target. I use ‘information’ here broadly, and as a term of art, and one that should cover a lot of different kinds of detail that a person might pick up. Just like the colour and brightness of a painting might affect how it moves us, so too might a range of features of communication affect how we understand the perspective of the person behind it. Not just the words being used, but also the way the words are said, the gestures, the body-language, etc.

Let us think about some reasons why, in ordinary cases, a person might do a better job of empathising with someone else. Firstly, the latter might be particularly good at using language to explain what their perspective is like. They might have a good vocabulary and a good memory, and a lot of time in which to tell their story. They can describe a range of different things about their experience and paint a full picture of what’s important for empathising with them. All other things being equal, simply telling us more about what their experience is like might make it easier for us to empathise with them. We have more information, and so it’s easier for us to re-create their experience in our own mind.

That’s not to say that getting more information across successfully through words is always going to come down to using more of them. A second person might be able to give us an even clearer understanding of their perspective more succinctly, if the words are chosen well. Think about people who are good at using poetic language, for example, to give a more detailed impression of a situation. Others can convey a lot through use of shared points of reference, or shared ideologies or values.

In other cases, the easiness with which we can empathise with a target might be affected by how much we can relate to them or what they’re going through. In those cases, a small amount of conversation or a small number of words can tap into a much larger and more detailed understanding. More information might come across because of what we already know, as well as just what we’re able to see or hear about the other person directly. A person doesn’t need to say much for me to be able to empathise with their difficulty with the rental market, because I’ve had similar struggles recently, and so I already have more detail to hand about what that’s like. If I were trying to empathise with their difficulties in skiing, I might need to hear a lot more from them to be able to reconstruct their experience in my mind. Similarly, if two people are facing the same problems, I might find it easier to empathise with the person who comes from a more similar background to my own, or who shares more of my own identity traits, either because we might have more in common in terms of the way we react to the same events, or because I (correctly or not) infer that we do. Of course, this can often be problematic – leading to biases and differential treatment to those from different groups. But these are examples of what things can sometimes make empathising easier, not an account of what should make it easier.

Proximity is also sometimes listed as a reason why we might find empathising easier. This, too, could be understood in terms of information in a few different ways. Partly the difference here could be explained in terms of what we’re able to see about the person – the closer they are to us, the better we can take in their expressions, their hesitation, their body-language, etc. Partly we might just feel a sort of closeness and kinship from people who are closer to us, and as a result find it easier to make the right sorts of connections in our minds. In both cases, the proximity means we’re in a better position to receive and understand the information in a way that helps us to recreate a more accurate representation of the experience in our own minds.

I don’t mean for any of these suggestions to sound new. Ilyes shows us that many of these are covered in Hume’s account of the mechanisms of empathy, for example.¹¹ She says that for him, the process of coming to empathise with another person is about associating the ideas we have about their own experiences with the stronger and more vivid impressions we have about our own lives. The stronger these associations, the stronger and more vivid those ideas become – even becoming like impressions that we might experience directly.

… the stronger the relation betwixt ourselves and any object, the more easily does the imagination make the transition, and convey to the related idea the vivacity of conception, with which we always form the idea of our own person.¹²

She goes on to say that for him, we find these associations (and therefore the more accurate, vivid instances of empathy) easier when there’s ‘resemblance, contiguity in time or place, and cause and effect’.¹³ That is, we find it easier to empathise when we can see more resemblances between ourselves and the targets, (as before, information can be easier to pick up on if we are more familiar with it already), or when we can make connections between the ideas and impressions through their being otherwise closely or causally related. I don’t take this to be a rival way of understanding the more or less favourable conditions of empathy to my own, rather I hope that the lens of information can simply be one way through which we can view and understand these other mechanisms.

Let’s think back to The Referendum, The Commiserating, and The Interview. Particularly in the first two of these examples, it’s easy to see ways in which less information comes across in the online versions of the cases when compared to the in-person versions, and how that might provide a worse foundation for empathy. In The Referendum, Jasper and Bec have never met before. When they’re having a discussion in-person, they each have more information to go on about who the other person is – whether or not this is something they consciously think about. Not just the name they choose to use on social media and maybe a small avatar to represent them, but a whole host of other features about the way they each want to represent themselves – their gender presentation, their style, their accent, etc.¹⁴ This information might help form their understanding of each other’s perspective. The fact that they share a mutual friend might also be more vivid to them – perhaps something they’re reminded of more easily because of the way the physical space of their friend’s house takes up more of their surroundings than it would do if it were confined to a screen.

In The Commiserating, Dave and Jade already know each other well, but there are still plenty of ways to convey information in person that don’t apply over an instant messaging service. Dave can’t see Jade’s expressions or hear her voice, and might miss out on the way that her sadness is physically affecting her. Seeing her describe her feelings over text vs hearing her describe it in person, while watching her speak, might be like the difference between the brighter and more colourful vs the duller and less colourful representations of breakfast. There’s some sense in which the content is the same – the words Jade is using in both cases are the same, and the items in the breakfast are the same in both pictures. But one picture, and one of the methods of communication, provides additional detail that we might not explicitly be thinking about but that can still help to either inform our picture of what the person’s experience is like or help us to feel it more accurately.

In The Interview things are a little less clear-cut. This is because the first two cases are using text (or text plus avatars and other basic social media information) instead of in-person speech, and so these cases are the easiest for pointing out examples of what information might be missing in the online cases. I’ll have more to say to diagnose problems here in the next section. But of course, there’ll still be ways in which the information given in a videocall will be less than that in person. Rose won’t see all of John, and she won’t see him as clearly as she would in real life. The image of him will be smaller, less detailed, and take up less space in Rose’s vision.

These examples all come with caveats. It’s certainly not the case that more information will always make it easier for us to empathise with someone – perhaps knowing more about what someone is going through, for example, will only highlight the ways in which their own perspective is unfamiliar and different to what we’d expect. But one of the tasks of this paper is to examine whether any worries about empathy in our online interactions are justified – and that includes recognisable trends or tendencies, even if they’re not necessary or sufficient conditions.

Svenaeus seems to raise this sort of worry about empathising on social media. He concedes that there are times when text alone can provide a rich context for empathy, but he’s critical that this would apply to a lot of our conversations online. He says,

The point could be raised that people have enhanced their social understanding of other human beings by way of reading biographies, historical and fictional stories for hundreds of years. This is true, but it also needs to be taken into account that novels are typically richer and more complex in nature than social-media stories are. They allow for more complex personalities and ambivalent stories to be developed in which the reader may empathize and identify with more than one person and in ways that are less single minded.¹⁵

Biographies and stories have much more space for additional words and context, so arguably make more reliable fruitful ground for empathy than social media stories or discussion ever could. However, I think that Svenaeus is only making a partial diagnosis here. After all, as I discussed above, the number of words used in a description won’t fully account for how much information it’s able to convey from the target to the empathiser. Even when text is all that’s used, it’s sometimes possible to use fewer words to tap into a shared pool of knowledge or shared values, and get more information across that way.

I also think we have reason to be optimistic because of our adaptability and creativity when it comes to communicating in online spaces. The information discrepancies I’ve described so far tend to be contingent features about why less information is conveyed in the online versions of conversations, but that doesn’t mean that they’re necessarily the case, nor that they’ll be the case for very long after we become more practiced at using them. Sometimes it’s appropriate to describe details differently online to how they would be said in person, for example, to help make up for the information gap. Furthermore, even in the cases we’ve been thinking of as primarily text-based, we often don’t actually communicate simply with text, but with emojis, pictures, memes, ‘reactions’ to messages, ‘likes’, and other forms of endorsements that a variety of platforms can give us. Even in the videocalls, platforms will often provide tools that allow for supplementary emojis, reactions, etc.

So far I’ve tried find out why some instances of online communication might be less fruitful ground for empathy than their equivalent in-person counterparts. I’ve shown that many of the circumstances that make empathy easier can be understood in terms of the information conveyed between the target and the empathiser – including unconsciously. When some of that information fails to come across in online spaces, it can show why these might be disadvantageous for empathy even when the same words are being used to communicate. But I’ve also argued in defence of our online interactions, and shown that these aren’t necessary features of online communication – and we have reason to be optimistic about our increasingly competent and creative use of such technology.

In the next section, I’ll look at the same problem in a slightly new light, and give a better explanation for what might be going wrong in some online cases: not just a lack of information as such, but an increased amount of control over that information. I will show that this problem is harder – although not impossible – to overcome.

4. A more precise diagnosis

Next, I would like to focus on a particular sub-category of information that is harder to get across in online contexts. I would like to suggest that a significant potential problem for online spaces as fruitful ground for empathy is the increased control we have over the information conveyed online.

There are some ways in which this is an intuitive answer. After all, unlike with information generally, it’s less of an accidental feature of our online interactions that we have more control over the information that comes across. When we take part in an online conversation, we tend to deliberately choose our words, our reactions and the images of ourselves that are presented. When we’re represented by an avatar or a profile, we tend to have control over what it says about us.

But there are also some ways in which this diagnosis is quite unintuitive. There are many reasons why having control over own information is incredibly valuable. It’s important for when we have reasons to value privacy,¹⁶ when we want to explore new aspects of our identity, or hide aspects of it from those who might mean us harm, for example. The ability to pick and choose what information is available is sometimes necessary for our freedom and our safety. And on the face of it, it also seems like it would be helpful for empathising. If I have more control over the information that comes across, perhaps I can better portray myself in a way that lets others emphasise with me.

So why might this added element of control also pose a problem for our ability to empathise with one another? Because firstly, even when we consider a variety of additional ways that we can voluntarily share information online, there will often be some details that we’re not always going to want to or to be able to convey. And secondly, some of that information is going to be the kind that can improve the conditions for potential empathy.

Let’s look at our examples. In The Referendum, the format for the discussion online is focused more on the discussion itself, and less on the social aspect or the shared friends. In person, friends are often introduced to each other, and everyone has an opportunity to learn a certain amount about each other (explicitly or not) before people start getting to grip with political discussions. Even when nobody has been introduced, they can’t help but seeing things about one another that do that work – they are confronted with each other’s demeanour, stance, tone, the way they interact with the host, etc. It’s not that the social media equivalents don’t have ways for this information to come across online, but rather this information seems inappropriate to convey before jumping into a conversation.

In The Commiserating, Jade and Dave might be able to communicate a reasonable amount of information about their emotions, but there might be plenty of detail left out because they’re not aware of all of it. Our experiences and perspectives are regularly influenced by beliefs, desires and possibly even pleasure or pain that we’re not aware of at the time.¹⁷ Some of this will come through in ways we don’t have full control over – such as through body language or a tone of voice. Perhaps Dave hasn’t had the time to reflect on the embarrassment he feels about the situation, in addition to the frustration. Where he might therefore struggle to communicate this intentionally, these details might come across to Jade because of the way he speaks, holds himself and acts in the same conversation in-person.

The Interview is, I believe, a particularly strong case in terms of understanding what difference control makes over the conveyed information. Even though these conversations include live video – going further towards replicating the in-person experience – we have more control over our appearance than we would do in real life. We’re able to hide certain parts of ourselves (such as a nervous leg twitch) and adjust others (such as by sitting up straighter and holding our heads up higher when we see our own images). The image of himself that John puts forward to Rose is undoubtedly more curated than it would be in real life. And while this has advantages, it’s important to understand that this control might come with disadvantages as well.

Are these kinds of information really the kinds that are relevant for empathising with each other? I argue that they are. Online communications will exclude, for example, a lot of detail about people’s vulnerabilities. It seems plausible that information about our human weaknesses and our idiosyncrasies will often help us relate to each other. So too will features of ourselves that we don’t think will be relevant, or facts about ourselves that we aren’t even aware of. All of these aspects will be part of what makes the versions of our experiences and perspectives that we convey more vivid for our interlocutors.

The framework in this paper diagnoses a possible problem for our online communications, and it’s clear that there’s room for more work to be done. But I think it’s still possible to end on an optimistic note. Having increased control over what information is conveyed doesn’t mean that we’ll never be able to unconsciously convey things about our experiences and our perspective, just that it’s harder to do right now. But just as we’re adapting at being able to convey more information in new contexts, so too we’ll get better at reading that information – as well as reading between the lines.

5. Conclusion

This paper developed a framework for understanding some more or less fruitful conditions for empathy. The framework is limited – I’ve focused on a particular worry about empathising, and I’ve only looked at ‘good’ cases, where the parties are largely unprejudiced, and open to both sharing and receiving information about each other. Nonetheless, I think it has proven useful in understanding why some instances of online communications might be – particularly when we’re still learning how to use the platforms – less beneficial grounds for empathy. I argued for the relevance of one particular difference between online and in-person communications: the added control that we have over what information is conveyed. Contrary to what might’ve seemed true on the face of it, having increased control over what information comes across won’t necessarily make things better for empathy, but they might indeed make things worse. We don’t always know what information to put across, for example, and sometimes it would be socially inappropriate to communicate certain things intentionally.

But this is by no means a fatal blow for online communications. As we get better and better at communicating in these ways, so too we’ll get better at subtleties, and reading between the lines. To the extent that there are problems, I think we can eventually overcome them. And in the meantime, we can do a better job of understanding them.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Elizabeth Ventham

Elizabeth Ventham is a Postdoctoral Research Associate working at The University of Salzburg. Before this, she worked on empathy at The University of Liverpool. Her other philosophical interests include desire and practical reasoning.

Notes

1 This is in Plato, Phaedrus. The same comparison between Socrates and modern worries about technology can be found in, eg, Sullins (2022).

2 Some discussion of this kind of moral panic and its relation to technology is discussed by Genevieve Bell in a Wall Street Journal article, Rooney (2011). The use of technology by such groups has also sometimes drawn the ire of its inventors, as Feenberg points out: ‘The telephone, for example, was originally intended for serious business conversation. When women appropriated it to organize the social life of their families, engineers complained bitterly about the waste of their beautiful instrument’. (Feenberg 2017, 43).

3 For some more examples, Feenberg (2017) talks about the value of online fora for bringing together patients with the same fatal disease, and Vanden Abeele, de Cock, and Roe (2012) discuss the benefits for visually and hearing impaired adults.

4 There are plenty of worries about technology that this paper doesn’t address, which are still the topic of interesting contemporary philosophical debate. Take, for example, worries about the polarising nature of some discussions that might come from echo chambers and epistemic bubbles online (eg Nguyen 2020; Robson 2014; Jamieson and Cappella 2010).

5 I should note that the relationship between empathy and moral behaviour in the philosophical literature is not a simple one. Some influential papers have taken a sceptical stance about empathy’s value for morality at all, with Prinz (2011) being a paradigm case. Having said that, I take his views to be in the minority. Furthermore, many of the criticisms of empathy take issue with our ability to empathise broadly or widely enough, or with the right targets, rather than with the phenomenon itself. Examples of recent defences of such a relationship include Bailey (2022), Betzler (2019), Masto (2015), Simmons (2014) and Ventham (2023).

6 It’s worth noting that a lot of the existing philosophical scholarship so far on empathising online is about ‘empathy’ as it’s understood in the phenomenological tradition. As Osler describes it, the term there is ‘ … used to describe the fundamental way in which we encounter other embodied subjects and their experiences’ (Osler 2022, 146). As with the other forms of fellow-feeling listed, I take this to be a distinct but related phenomenon, and to be more specific I’ll focus on empathy as it’s currently understood in the analytic tradition.

7 This is put well by a number of empathy scholars. See, eg, Bailey (2022) who says that for empathy ‘I must retain a firm awareness of two facts: first, my emotion is responsive to a situation that is not actual for me, and second, it is responsive to a situation that is actual for the other’, and Coplan (2011) who says that (affective) empathy requires ‘other-oriented perspective-taking, and self–other differentiation’ (as well as matching, to some extent, the person’s affect).

8 Bailey discusses this in more detail, arguing that empathy is important for a certain kind of ‘humane understanding’ in Bailey (2022).

9 Kagan (1989, 283).

10 Some empathy scholars refer to this particular phenomenon as ‘affective empathy’. See, eg, Spaulding (2017) and Maibom (2017)

11 Hume didn’t use the term ‘empathy’, since the term wasn’t coined in English until later, but he seems to be talking about what we mean by it now.

12 (Hume 2007; quoted in Ilyes 2017, 103)

13 Ilyes (2017, 99).

14 It’s important to note again here that some identity features we present in person won’t always be sources of potential connection and empathy, but rather sometimes they’ll be markers that leave the targets vulnerable to prejudiced behaviour and discrimination. Just as in the positive cases, this process will often be one that happens unconsciously. See, eg (Brownstein 2019; Sullivan-Bissett 2014; Holroyd 2017). For some good discussion on difficulties people have in being receptive to members of oppressed groups, including in online discourse, see Habgood-Coote, Ashton, and El Kassar (forthcoming). A further look into empathy’s role here would make for interesting future work. For now, though, I will stick to trying to understand the positive cases.

15 Svenaeus (2021, 91)

16 See Steeves and Regan (2014) for more on the value of privacy online, particularly for younger people. For more about how increased control over our online presentation can affect authenticity and privacy, see Marmor (2020) and Véliz (2022).

17 Bramble (2013; 2020) in particular argues that experiences can even be pleasant or unpleasant for us without us being aware of that fact.