Community, solidarity and care through data? An ethical analysis of the interpersonal dimension of self-tracking

Abstract

This paper discusses the interpersonal dimension of self-tracking technologies from the standpoint of Dewey’s pragmatist ethics. Users of self-tracking routinely exchange data with others, interact through social features embedded in their tools, and form communities focused on the sharing and discussion of data. I employ Dewey’s notion of transaction to discuss how self-quantification impacts users’ perception of others and how it mediates interpersonal relations. In Dewey’s ethics engagement with others is a fundamental part of moral life and individual flourishing can only be achieved in social contexts and through the extension of care towards other people. The ethical ideals and epistemic concepts present in pragmatism allow me to argue that despite claims that self-tracking technologies lead to the formation of communities, pro-social instincts exist prior to the adoption of the devices and self-tracking actually inhibits the formation of mutually uplifting social bonds. In my discussion, I focus on the degree of epistemic distance that makes genuine engagement with others difficult, limits to transactional agency that prohibits users from deciding how they are perceived by others, and the reductive and instrumental nature of interactions embedded in the social features of self-tracking technologies. Moreover, I also analyse some political aspects of community-formation in the context of self-quantification and criticise the ownership structures of data and infrastructure that limit communities’ political agency.

1 Introduction

In this paper, I use John Dewey’s concept of transaction to analyse the interpersonal relations arising through self-tracking.¹ Although collection and evaluation of quantified personal data may seem to be a predominantly solitary activity, users of self-tracking technologies routinely engage with others in direct and indirect ways, willingly and unwillingly, as well as knowingly and unknowingly. Many self-tracking projects involve the sharing of data with friends and strangers alike, discussion of results in communities, and comparisons between users through leaderboards or other social features embedded in popular self-tracking apps.

At the same time, even those who do not deliberately attempt to self-track in a social manner, commonly end up intertwined with other people and their data (Boddington 2023). The algorithms employed in self-tracking aggregate individuals’ data into larger data sets and use this collectively-generated resource to provide personalised services. Consequently, a user looking at numbers and recommendations on their device indirectly engages with countless other individuals whose personal information contributed to the displayed results in various ways. And while it might be often difficult or even impossible to trace the influence the other (often anonymous) users have on individual self-tracking projects, any self-tracker is also connected to the developers who have a much more direct influence on the practice. The self-perception of self-trackers is partially shaped by the developers’ decisions, and by merely turning on their self-tracking devices, users open themselves to normative evaluations in light of the standards set by the makers of their chosen technology.

The literature does not provide conclusive findings and a clear evaluation of the interpersonal dimension of self-tracking (see Wieczorek et al. 2022 for a review). Although authors discuss self-quantification in the context of solidarity and community formation (e.g., Barta and Neff 2016; Kristensen et al. 2021; Sharon 2017; Sharon and Zandbergen 2017), discussions of self-tracking as a narcissistic or individualistic pursuit are also common (Morozov 2013, p. 233; Lupton 2016a; Sharon 2017). At the same time, self-tracking technologies are also criticised for reducing the other to numbers or promoting an instrumental view of social relations (i.e., treating others only as a useful reference point for one’s own pursuits, see Gabriels and Coeckelbergh 2019, Lupton 2016b, or reducing interaction to “liking” the activity reports of others, Kent 2018).

These ambiguities can be explained by two related reasons. First, self-tracking is a diverse practice picked up in numerous contexts and for a variety of reasons (Lupton 2016b). Authors providing conflicting assessments of the interpersonal dimension of self-quantification often examine altogether different aspects of the phenomenon. Some critical voices focus on institutional tracking and the internal features of the technology such as comparative leaderboards (Gabriels and Coeckelbergh 2019; Kent 2018), whereas more positive analyses (e.g., Barta and Neff 2016 or Kristensen et al. 2021) look at bottom-up user communities oriented towards the exchange and discussion of data. Arguably, the most nuanced discussion of the social dimension of self-tracking was presented by Sharon (2017) and her article discusses a much wider variety of self-quantification practices than most other work, albeit still focuses only on the context of healthcare.

Second, a large portion of the literature is descriptive and presents empirical research done on user and patient communities that utilise self-tracking devices. Even if authors identify values embedded in and expressed by such communities (e.g., solidarity), they do not necessarily seek to analyse them in more detail and do not assess them from a normative standpoint. In this sense, differences in the evaluation of the different aspects of self-tracking practices may result from ambiguous or conflicting understanding of key values and normative ideals underpinning social relations.

This paper seeks to contribute to the literature by examining a broad variety of self-tracking practices from a perspective of a single, well-articulated ethical perspective. I believe that Dewey’s pragmatist ethics provides good grounds for discussing and evaluating the interpersonal dimension of self-tracking practices. In addition to providing ethical guidance resulting from an understanding that ethical life is inherently social and should be assessed as such, Dewey’s philosophy integrates epistemic considerations into moral deliberation (as discussed in detail by Medina 2013, p. 81). This is particularly important in the context of self-tracking technologies as the production and exchange of information is a fundamental part of the social relations they mediate.

Over the course of this paper, I apply Dewey’s ethics to analyse different aspects of the interpersonal dimension of self-tracking. Previously, I explored how Dewey’s philosophy can be used to examine the impact of contemporary technologies on users’ behaviour (Wieczorek 2023) and demonstrated how his notion of the intelligent habit challenges our understanding of empowering aspects of self-tracking technologies as they pertain to individual users (Wieczorek 2024). This paper builds on this work by employing Dewey’s notion of transaction to critically interrogate how the ideas of community, solidarity and care relate to self-quantification. Over the course of this paper, I apply Dewey’s ethics to analyse different aspects of the interpersonal dimension of self-tracking. In Sect. 2, I summarise the main concepts guiding my analysis, namely transaction, ethical ideals of continuing, inclusive growing and democracy, and the notion of a data double. In Sect. 3, I discuss the impact of self-tracking-mediated transactions on the growth of individual users. In Sect. 4, I analyse self-tracking as enabling or inhibiting users’ care for (unspecified) others. In Sect. 5, I focus on self-tracking as impacting solidarity and (Deweyan) democracy. I conclude by recapitulating my arguments and highlighting how they can be related to the broader technological landscape. Although I adopt self-tracking as a main reference point, I argue that Dewey’s ethics and social philosophy offer valuable lessons on how the notions of community, solidarity and care can be problematised in the context of a wider range of data-intensive technologies.

2 Theoretical background

While on a basic level, the pragmatist notion of transaction is very similar to that of interaction, Dewey deliberately uses a distinct term to highlight the continuous and fluid nature of interpersonal relations (see Sullivan 2001 for an in-depth overview). For Dewey, others are an inseparable part of an individual and their experience since transactions with them are constitutive of one’s behaviour, character and circumstances. In pragmatist terms, life and all its parts (behaviour/habits, circumstances, problems that need to be addressed, the tools for addressing them, etc.) are inherently social (see Dewey 2008, pp. 322–328; Pappas 2008, p. 217). Sullivan (2001, pp. 15–16) illustrates this notion by referring to a metaphor of a stew: while it is certainly possible to distinguish individual ingredients placed in the pot, each of them irreversibly changes and influences others in various ways – to the point that a carrot found in a stew has qualities (e.g., taste) far removed from the qualities of a raw carrot, even if it is still recognisably the same vegetable. In this vein, we cannot just stop transacting with others at will and thus get rid of all the residue these transactions have left upon us. In Deweyan terms, interpersonal relations continuously shape us in an irreversible way, and we do not always have the final say about their content and outcomes.

In my view, this notion provides an apt account of the relationships with others arising in the context of self-tracking. The practice is inherently social, users do not have full control over the relations in which they engage, and they are simultaneously deeply affected by these relations. Self-tracking and the relations it mediates intimately shape users’ habits and character.

However, the notion of transaction is not merely useful for describing the interpersonal relationships arising between different parties involved in self-tracking practices. It has a clear normative dimension and it can help in evaluating the relationships arising through self-tracking. Dewey’s ethics places particular emphasis on two interrelated ethical ideals – that of continuing, inclusive growing (see Fesmire 2003), and democracy (see Pappas 2008).

Growing can be defined as an intellectual and practical attitude aimed at the ongoing improvement of one’s experience and character (hence the notion of continuing growing). Moreover, in Dewey’s philosophy, individual flourishing and the good life are intimately connected with a concern for others and their well-being, which entails that morally desirable behaviour should also account for and promote the needs of others (hence the notion of inclusive growing). In this sense, the ideal of continuing, inclusive growing entails that individuals’ habits and actions should contribute to their own and others’ well-being and ameliorate the quality and variety of their lived experience.

To assess whether transactions with others mediated by self-tracking meet or fall short of this ethical ideal, I extend discussions found in pragmatist epistemology, particularly in the work of Shannon Sullivan (2017) and Jose Medina (2013). In her work, Sullivan defines epistemic injustice as a situation in which “the speaker isn’t allowed to epistemologically transact with the world in ways that enable her own as well as others’ flourishing” (Sullivan 2017, p. 210). I build on her ideas to pinpoint situations in which self-tracking technologies push users into transacting with others in ways that impact their own as well as others’ flourishing.²

For this purpose, I rely on Medina’s (2013, p. 50) ethical-epistemic principles of engagement and equilibrium which require, respectively, that our epistemic practices involve arecognition and confrontation with competing worldviews, and that differing points of view are adequately taken into account in deliberation (even if only to be rejected as mistaken). These principles enables me to assess whether individuals are properly recognised by others in transactions mediated by self-tracking and whether this recognition is adequately extended to all the participating parties.³ Consequently, my aim is to analyse whether transactions with others mediated through self-tracking (1) contribute to the growth of particular users, or (2) contribute to the growth of the people with whom a user is transacting. In this way, I attempt to distinguish the impact of transactions both on specific people engaging in self-tracking as well as on others who might be (indirectly) affected by the practice.

Moreover, I turn my attention to Dewey’s notion of democracy, which can be characterised less as a political system, but more as a cooperative approach to the solution of shared problems (see Dewey 2008, pp. 348–349; and Pappas 2008 for an in-depth analysis). In this way, I want to highlight the political dimension of interpersonal relations arising through and mediated by self-tracking. Drawing on Dewey’s (2016) notion of a public, I analyse whether the sociality embedded in self-tracking enables users to band around shared interests and advocate for the recognition of these interests in the political sphere. I discuss instances in which self-trackers form a community focused on specific shared problems (e.g., collection of health data) or in which participation in different self-tracking practices could lead them to the recognition of shared interests (e.g., tracking as shared labour). On this basis, I evaluate how self-tracking enables or inhibits their solidaristic and democratic pursuits, and whether it is possible to speak of duties and obligations towards a community as arising within the self-tracking sphere.

However, before I proceed to the next section, I want to provide several notes on the objects of transactions mediated through self-tracking. Although the practice and associated technologies regularly hint at interactions with other people, the level of access (in epistemic and practical terms) to these people is not at all clear.

Existing research on self-tracking widely refers to the notion of a data double, first introduced by Haggerty and Ericsson (2000) and popularised in the context of self-tracking by Ruckenstein (2014). According to Ruckenstein, self-tracking does not necessarily provide users with more knowledge about themselves and their everyday behaviour. Instead, it would be more accurate to say that a self-tracker engages with a data double of themselves – a likeness created out of data that resembles the user in many ways but, crucially, does not possess all their characteristics and often differs in some regards.

Self-tracking technologies can be inaccurate in collecting data, and where data is missing, they flesh out a user’s profile by relying on inferences, which often turn out to be mistaken (Crawford et al. 2015). Moreover, specific data points might cease to be relevant in a given context, for example, when a user’s physical characteristics change over time (Neff and Nafus 2016). Consequently, it is inaccurate to claim that I am represented by the data available through my device. Rather, the data is a representation of my data double, a digital doppelgänger, who in some instances can effectively serve as my replacement, but is a separate, if related, entity to myself.

This problem may seem overblown at first glance, but research suggests that users engaging with data doubles imbued with characteristics not possessed by the users themselves may end up with a distorted image of themselves. Vegter et al. (2021) argue that it is akin to looking at a funhouse mirror, except that users are likely to privilege the distorted image over their direct sensory experiences. This is a good example of the phenomenon of datafication which plagues users’ engagement with self-tracking technologies and which is widely discussed in the literature (see, e.g., Kreitmair and Cho 2017; Sharon 2017). Users are often inclined to treat the quantified, device-mediated representations as more accurate (more “scientific”) than the testimony of their own senses.

The concept of a data double is especially important for this paper as the transactions with others mediated through self-tracking are epistemically suspect for similar reasons. It is difficult to determine whether these transactions should be seen as engagement with particular people or merely with their data doubles. At the same time, even if users were aware of this distinction, they would most often be unable to determine the degree of distance between individual people and their datafied representations, especially when the transactions with them occur primarily in the digital sphere. Moreover, as data doubles are bound to contain only a fraction of the information pertinent to a specific person, they are problematic also from the perspective of equilibrium. It is doubtful whether the information that is most crucial to the transaction is always adequately weighed and represented when data doubles enter into play.

Consequently, the discussion presented in this paper needs to be qualified. The transactions with others I discuss may often be better characterised as transactions with data doubles. Such indirect contact may not lead to genuine care for others or feelings and performance of solidarity. I try to account in my analysis for the degree of epistemic distance between the transacting individuals, but this is not always possible – data doubles are by their very nature confusing and often blend in with our ideas about the user they represent/copy. Ultimately, while self-tracking allows for the creation of new social bonds and enables otherwise unavailable transactions, these social interactions are problematised by the aforementioned epistemic problems. Crucially, this is not something that can be resolved in my analysis and the above considerations are only meant to highlight this issue and its importance for this paper. The tension between qualitative phenomena and their quantitative representations is central to self-tracking and impacts all aspects of the practice, including its interpersonal dimension.

3 Impact of transactions on the individual

When using self-tracking devices, people are routinely exposed to information and metrics that fall beyond the scope of their everyday experience. Even when self-quantification is used in a familiar context such as running, data supplied by other users on in-app comparison tools like leaderboards and challenges is bound to differ from the data that an individual is able to produce on their own. Other users’ results may include values not attainable to the individual in question and serve as a source of motivation, or be obtained in unfamiliar contexts, potentially offering insights into new (to the user) ways of engaging in a given activity. Cross-comparisons can also draw users’ attention to metrics they previously ignored and thus allow them to focus on the features of their activity that they have not previously considered relevant. Finally, exposure to others’ data can help users contextualise their own information and activity. Many of the standards endorsed through mainstream devices are far from attainable for a large subset of users and people routinely struggle to meet thresholds presented as “normal” (Lupton 2016a; Nissenbaum and Patterson 2016; Sharon 2017). Exposure to similarly “deficient” results achieved by other people can put any perceived shortcomings in perspective and help users deal with feelings of inadequacy as well as re-evaluate and readjust their own practices in light of new insights (Barta & Neff 2016; Sharon and Zandbergen 2017).

These aspects of data sharing can have a positive impact on users’ self-tracking practices and, at first glance, seem to be in line with the Deweyan ideal of continuing growing, as well as Medina’s principles of engagement and equilibrium. Exposure to new forms of data can be seen as enriching the individuals’ perspective and broadening the scope and quality of their experience. Running data sourced from others may merely improve the efficiency and variety of one’s workout, but in contexts such as mental health or sleep tracking, it could have a noticeable impact on users’ overall wellbeing. Still, even the less pronounced or less impactful improvements can be thematised in the context of user empowerment and increased agency. The added insights derived from shared or aggregated data may increase users’ ability to shape their everyday behaviour and potentially counteract the universalistic tendency of many self-tracking devices, as discussed in the previous chapter. Ideally, data sourced from others should contribute the necessary context that is not provided by default through recommendation systems and user interfaces, and thus help people approach their self-tracking practices in a more critical manner. Although a significant degree of data literacy and practical wisdom would be necessary to actually reap the benefits derived from shared or aggregated data, users engaging with others may have more tools at their disposal than those tracking in a solitary manner.

Members of the Quantified Self movement have long argued that those willing to share and discuss their data and the insights it provides are better equipped to take control of their habits (Ruckenstein and Pantzar 2017). The movement’s events focus on direct engagement between different self-trackers who disclose their data, as well as its interpretations. By design, the speakers are expected to receive valuable feedback, while others are provided with new ideas and grounds for comparison. Recently, empirical research into the interpersonal dimension of self-tracking has demonstrated that users who discuss and share their data reap palpable benefits – relations with others help them deal with negative feelings resulting from their self-tracking practices, they find it easier to interpret the metrics, and they aid each other in staying motivated enough to continue their projects of self-quantification (Barta and Neff 2016; Kristensen et al. 2021).⁴

Additionally, metrics are a powerful tool of expression. They enable individuals to make sense of their experience and communicate it with others. A runner quantifying the features of their exercise may later refer to their data to share their accomplishments and receive encouragement and praise. More advanced metrics may allow individuals to discuss some features of their everyday life in all kinds of circumstances, especially when they lack other hermeneutic resources for doing so. For example, a mental health patient tracking their mood (even using a very basic and reductive numerical scale) could refer to this data in discussion with friends, family and medical professionals without necessarily having to engage in a detailed explanation of their mental states. In this sense, self-tracking devices could be seen as communicative partners, allowing users to better understand themselves and communicate such understanding to others (Li et al. 2021). When treated as a communicative shorthand, self-tracking data can help individuals make up for epistemic deficiencies or even respond to some epistemic injustices.⁵

However, the reality of self-tracking is much more complicated. I agree that direct interactions with others mediated by or centred on data can be epistemically and developmentally beneficial to individuals, but I argue that this is less of a feature of the data and the practice and rather just a likely outcome of generally uplifting transactions. It would be more accurate to say that such positive transactions arise alongside or in reaction to self-tracking and not thanks to it (or that they merely use self-tracking as a tool), especially as many of the relationships between self-trackers described in the literature seem to aim at addressing some of the deficiencies of the practice (e.g., the difficulty in making sense of the numbers or the propensity of users to develop negative feelings in relation to their results). As illustrated by Kristensen et al. (2021) and Schwennesen (2019), the relationships of care arising between different people involved in self-tracking are often meant to repair the elements of the practice that do not function according to users’ needs and expectations or to ensure that the users are not negatively impacted by its shortcomings. Following such arguments, care should not be seen as a feature of self-tracking but a reaction to the technology and its deficiencies (especially as care is a significant or even primary aspect of our interpersonal interactions, see, e.g., Noddings 2003). Attention from and towards others can help users make better use of self-tracking technologies, but, as I argue, such benefits should be attributed to our ability and propensity to engage in mutually uplifting transactions.

Consequently, I argue that when evaluating the interpersonal dimension of self-tracking, we should look at the kinds of transactions that are embedded in the design of the devices and are often arising outside of users’ initiative. In this sense, the proper subject of analysis in this paper should be transaction with (typically unknown) others through technology and in-built social functionalities. After all, people can bond even over predominantly solitary and non-social activities, such as fishing, but it would be a mistake to take the existence of fishing associations as a sign of an inherently pro-social characteristics of fishing. For this reason, I want to discuss transactions as mediated by the practices, rather than transactions occurring around or alongside the practice, as the latter are generally likely to arise due to our social nature (at least inasmuch as my pragmatist perspective would make me assume).

I doubt whether transactions with unknown others (e.g., on an anonymous leaderboard) can be classified as engagement in the sense I put forward in this paper.⁶ The degree of distance as well as fragmentariness of the exchanged information make it difficult to recognise the encountered person as a unique individual with their own needs, desires and qualities (see Villegas-Galaviz and Martin 2023 for a discussion of moral distance in the context of ethics of care). Recognition of a singular trait (often relatively minor in the context of overall character and identity, such as physical performance) is not enough to serve as the basis of an uplifting constitutive bond of the kind discussed by Sullivan. Data shared through self-tracking can have bearing on individuals’ flourishing or their ability to consciously shape their character, or as Dewey puts it, “decide what kind of person one wishes to become” (see Dewey 1957, p. 217). However, this is not an outcome of actual transactions with others (rather, simply the use of data divorced from its original context). Individual users can directly benefit from the input and labour of others (e.g., when user data is algorithmically aggregated and used to offer better recommendations), but the use of infrastructure and products resulting from this labour is not the same as engaging with the labouring people in a mutually constitutive manner. Similarly, using a bridge may be described as a transaction with one’s environment and is certainly habit-producing but it would be absurd to interpret it as a transaction with the people who built that bridge.

On the other hand, exchanges of self-tracking data problematise information flows, reducing individuals’ ability to project their identity and shape how it is perceived by others.⁷ Users have to contend with being encountered by others through the lens of their metrics, which in some instances entails conforming to or struggling with the distorted and reductive datafied representations of themselves. Other self-trackers primarily or even exclusively encounter a data double of a given person and perceive that individual through the lens of characteristics that the person may not necessarily possess or wish to disclose (see Gabriels and Coeckelbergh 2019 for a different version of this argument).⁸

In this sense, self-tracking can have a harmful impact on individuals as it reduces their transactional agency – users have little ability to negotiate the terms at which they interact with others. At the same time, the prominence of data doubles in these transactions is problematic from the perspective of engagement and equilibrium. The interaction with an individual’s datafied representation effectively replaces direct and in-depth engagement with that individual as their unique characteristics are not adequately recognised (instead, their perceived and recognised identity is determined by what has been captured by data). It also reduces the diversity of perspectives and information within the transaction,⁹ privileging the narrow range of information that can be described with data and that is typically collected and shared through self-tracking technologies. Arguably, transactions mediated through self-tracking are impoverished both in epistemic and ethical terms, at least according to the pragmatist view that an individual’s ability to transact with the world is foundational to their growth and, consequently, their ability to flourish. Self-tracking introduces a degree of distance between users that casts doubt upon the possibility of treating the practice as a medium for meaningful and uplifting transactions – the influence of technology constrains users’ ability to freely shape their roles in interpersonal relations and reduces the variety of identities they are able to adopt/express. In the most egregious cases, it could press upon them social roles far removed from what they desire (e.g., when data about a person results in them being judged or treated in a way incompatible with the identity they actually hold or aspire to).

Finally, I have some concerns about the kind of sociality promoted through the social features typically embedded in self-tracking technologies – namely leaderboards and social media highlights. Although these functionalities enable users to celebrate their individual performance and receive recognition from others for some accomplishments, they could be criticised as encouraging boasting or even a narcissistic obsession with one’s activity. They seem to promote the opposite of virtues such as humility, moderation or honesty. Instead, the set of dispositions instilled in users through these functionalities could be labelled under the umbrella term of digital smugness.¹⁰ Leaderboards and social media highlights lack direct social interactions and connect users with others in an impersonal, comparative and competitive way (see Gabriels and Coeckelbergh 2019, and my discussion in the next section). Consequently, they are an environment that invites users to indulge their pride and vanity, while reducing the chances that their ego will be checked by others. The character promoted through the social features embedded in self-tracking technologies is an egocentric and vainglorious one.

4 Impact of transactions on others

As the notion of growing promoted in Dewey’s pragmatism is meant to be inclusive, our habits should not only be oriented towards the amelioration of our experience, but also that of others. In the simplest of terms, good habits should make everyone better off, not just the person who possesses them. It is then prudent to take a look at the dispositions towards others developed through self-tracking as well as their moral relevance.

As already noted in the previous section, exposure to data from other people can help users combat the universalistic tendencies of self-tracking and expose them to a wider range of perspectives and experiences. This can be beneficial not only for the development of better habits but could result in a greater awareness of the diversity of needs and circumstances of other users. In particular, members of the QS movement claim that the sharing and discussion of data makes them more mindful of the specificity of individual self-tracking projects and prompts them not to treat their fellow self-trackers with a one-size-fits-all approach characteristic of the algorithms and recommendation systems embedded in mainstream technologies (Nafus and Sherman 2014; Ruckenstein and Pantzar 2017). This increased awareness has been discussed in the literature as potentially leading to care for others and as a foundation for the formation of communities centred around specific self-tracking practices (Barta and Neff 2016).¹¹ In fact, empirical research has largely supported these claims, although it has to be noted that the majority of the studied groups formed in an offline environment – in this case, self-tracking seems to be more of a pretext for the establishment of an overall beneficial relationship rather than a tool for the creation of friendships (Sharon and Zandbergen 2017; Kristensen et al. 2021).

At the same time, many self-tracking practices have a distinct orientation towards others and this is particularly evident in medical contexts. People monitoring their symptoms (e.g., blood glucose levels in the case of diabetes) often share their results online and discuss them with strangers in the hope of finding valuable insights that could facilitate the management of the disease. Even users who do not stand to directly benefit from the social forms of tracking often decide to share their data if they see it as contributing to the common good. For example, some self-trackers share their metrics with medical researchers in the hope that they will eventually be used to further medical knowledge or help in the development of new treatments. While it is arguably not the main motivation behind self-tracking for most users, concern for the wellbeing of others can be an important element of quantification practices and it should be recognised.

On the other hand, it is easy to question whether such attitudes to data sharing are actually a feature of self-tracking, and to what extent they should be considered a form of transacting with others in the pragmatist sense. First, self-tracking seems to be merely making it easier for people to follow their altruistic instincts, especially as the cost is relatively minor or non-existent. Sharing medical data through a self-tracking app usually takes as little as pushing a few buttons and there are no immediately discernible downsides to the users, unlike with acts such as organ donation. Since public health is generally considered a common good, I find it doubtful that users become interested in its advancement merely through their engagement with self-tracking – the functionalities of the device simply make it possible to contribute in yet another, often quite frictionless manner.

Second, while certainly commendable and important, mere contribution of data for the benefit of unknown others does not entail engaging with them in the full sense. In pragmatism, personal interaction and a cooperative approach to the solution of shared problems are a foundation of social life and directly contribute to the growth and wellbeing of the involved people. Certainly, the sharing of data for purposes such as the advancement of medical research fulfils some of our obligations towards others but does not contribute to the creation of personal bonds and the formation of communities—in most instances, users in no way engage with others by passively contributing to the data sets and they certainly do not engage with those who might reap the benefits of that contribution in the future (e.g., patients undergoing treatments developed with the help of data). Moreover, this kind of data sharing is not cooperative, at least in the Deweyan sense. Users pool resources and delegate actual work and responsibility to a trusted third party rather than coming together through shared work and deliberation. In this sense, the sharing of data is more akin to the mechanisms of representative democracy (which Dewey recognised as important but insufficient, see Honneth 1998) rather than the cooperative democracy entailed by Dewey’s ethical ideals.¹² A political entity that can arise through the sharing of data would not constitute a public in the Deweyan sense, as it would passively delegate the furthering of its interests to third parties instead of getting directly involved in political action and advocacy.

It is true that many self-tracking practices are oriented towards others (through data sharing and other means, see for example the attempts to quantify attention devoted to others described by Nafus and Sherman 2014) and express some instincts of care, but it does not mean that self-tracking itself is an inherently pro-social and caring practice, despite what some seem to be suggesting. It is far more likely that self-trackers use the tools at their disposal in a social and caring manner. The habits prompting them to do so are not products of self-tracking but exist independently of the practice and are merely operationalised through it (as they would be in all kinds of contexts).

In fact, I argue that some features of self-tracking are particularly problematic from the perspective of sociality and care. In many ways, the devices actively prompt users to transact with others in ways that are not in line with the ideal of continuing, inclusive growing—that is, the transaction does not prompt users to extend adequate concern over others. As I demonstrate over the remainder of this section, this could be largely blamed on the distance between interacting users and the impersonal nature of datafied interactions.

It is doubtful whether self-tracking provides a meaningful way to encounter other people and not their data doubles. The experience of the COVID-19 pandemic has clearly demonstrated the importance of in-person interactions for our overall wellbeing. The indirect interactions offered by self-tracking devices can, at best, be an extension of traditional engagement with others and not a viable alternative to it. In other words, it would not be possible for me to offer others through self-tracking the kind of attention that would constitute adequate engagement and recognition—in terms of direct contact, self-quantification offers even less than a Zoom call and can only accompany rather than replace other forms of interaction. Moreover, even if self-tracking can allow us to engage with others at all, this engagement would be limited chiefly to other users (and arguably to similarly active users). It seems that the practice can facilitate transactions only with a narrow community, if at all (see Sharon 2017).

At the same time, encounters with data doubles are problematic from an ethical standpoint. Translation of a person’s characteristics into numbers reduces that person to a definite set of metrics, which mirrors the problems already discussed in the previous section. Despite my best intentions, data will largely determine my knowledge of and attitudes towards the person I encounter (assuming I can only refer to data and not, e.g., direct interactions), which might limit that person’s ability to shape and adopt their chosen identity—they might struggle or decide to conform with how self-tracking makes others perceive and evaluate them.

Moreover, self-tracking might have the adverse effect of closing me off from the specificity of others’ experience, especially if that experience is not or cannot be (easily) quantified. This has important implications for the overall shape of relations towards others fostered through self-tracking. As some identities and perspectives are not easily captured through data, self-tracking technologies cannot express the diversity of users’ needs, desires and circumstances. This raises doubts about the purported empathy-increasing potential of self-quantification. While it is certainly the case that paying attention to what does not fit within the practice can make users more mindful of the situations of others, self-tracking itself limits the range of perceptible features and characteristics. In this case, any potential benefits are the results of savvy users working against the default makeup of the technology. Here again, the reactive notion of care discussed by Kristensen et al. (2021) seems particularly apt – users feel a need to work together because of self-tracking, but not thanks to it.

Additionally, the digital smugness discussed in the previous section is certainly problematic in line with the ethical ideals found in Dewey’s philosophy. As noted by Gabriels and Coeckelbergh (2019), the sharing of self-tracking data in the form of leaderboards and comparisons encourages users to treat others instrumentally—as yardsticks useful only for the evaluation of one’s individual performance. This is hardly in line with the notion of inclusive growing and is certainly not cooperative. Rather, it promotes anti-social, individualistic instincts that are easy to condemn from the pragmatist perspective. The social features embedded in self-tracking apps work predominantly in the interests of an individual and encourage that individual to value others only insofar as their input is relevant to one’s own experience with the app or device.

Finally, current literature does not discuss the duties and obligations towards others that arise alongside self-tracking practices. It would be certainly possible to advocate in favour of a duty to share data, especially for altruistic purposes such as medical research, but the current political-economic landscape of self-tracking makes me wary of all forms of data sharing. The circulation of data has an extractive character and lies in the interests of technology companies rather than the users (see also Ajana 2018, as well as Couldry and Mejias 2019 who demonstrate how contemporary capitalism “colonises” more and more aspects of life by employing a wide range of technologies that enable unfettered extraction of data and profit). It directly transfers power from users and increases the companies’ ability to generate profits (without adequately compensating those who created the data in the first place). For this reason, I find it more prudent to discuss the obligations and responsibilities towards others arising in relation to dispositions and behaviour mediated through self-tracking.

In my view, it is hardly possible for users to assume responsibility for the dispositions and ideas created through self-tracking technologies that impact other people (e.g., those discussed in this section, such as the reductive perceptions and attitudes fostered through quantification, or the instrumental treatment of others encountered through leaderboards). This is not to say that they should not be expected to do so. In pragmatist terms, people have a duty of ensuring that their habits are beneficial to the growth of others and enable the satisfaction of the needs and desires of others. The wellbeing and flourishing of others should be an important consideration during the process of dramatic rehearsal and should co-determine the habits considered as viable in the given circumstances – this is, in essence, the meaning of the ideal of continuing, inclusive growing.

However, habits¹³ instilled through self-tracking are hardly reflective and users have a limited ability to shape the dispositions they adopt with the help of their apps and devices – rather, they are often nudged towards behaviour determined by the designers (Lanzing 2019). This is not to imply that the blame for potentially harmful habits instilled through self-tracking should fall exclusively on the developers. Rather, the limited reflectivity of habits produced through self-tracking makes it impossible or at least highly difficult for users to fulfil their interpersonal obligations as they cannot fully respond to their responsibilities regarding the impact of their habits on others. Nevertheless, they still ought to bear some responsibility even for the habits of which they are not fully aware or which they cannot fully control. Since self-tracking reduces our ability to transact with others in ways that are beneficial to them, the ideals of continuing, inclusive growing and democracy should make us think twice about picking up self-tracking in the first place and encourage us to change the terms in which self-tracking technologies render others and make us encounter them. As I noted in the introduction to this paper, transacting with others carries certain ethical obligations and the opaque and inflexible nature of self-tracking technologies does not absolve users from the duty to meet them.

5 Transactions, democracy and solidarity

In this section, I discuss whether self-tracking can serve as a platform or a tool for cooperation. If self-trackers are able to recognise the interests and problems they face together, self-tracking can be understood as a public-making tool in the Deweyan sense and serve some democratic and solidaristic purposes.¹⁴ Even though in the previous section I was sceptical that genuine engagement and community can arise from self-tracking on its own, the practice can still be beneficial in specific contexts.

Members of the Quantified Self movement as well patients tracking together to facilitate the management of their disease certainly use self-tracking to tackle specific problems, potentially even forming something at least resembling a public by meeting to discuss and further shared interests. Moreover, the aggregation of data, even if not cooperative in the strictest sense, can contribute both to the public and the individual good (e.g., by resulting in more accurate recommendations or supplying researchers with valuable information). In this sense, despite being ambiguous or even harmful in the context of the most basic transactions between individuals, self-tracking can still provide discernible benefits to specific communities or the public when used as a tool or platform for cooperation. However, as I demonstrate in this section, some of the features of self-tracking are incompatible with Dewey’s notion of democracy as they limit users’ ability to shape and direct the socio-political initiatives they mediate.

First, despite the interpersonal dimension of self-tracking, the deployment of quantification technologies has been often (and accurately) criticised as individualising responsibility (see Wieczorek et al. 2022). According to this view, the attainment of some socially valuable goals, such as the provision of healthcare and promotion of public health, is reframed with the help of self-tracking as predominantly within the domain of individual efforts rather than being a prerogative of the wider community or the state. In this sense, self-tracking can be positioned within a larger trend that positions individual health and behaviour within the domain of market logic and consumerism (Lupton 1997). This tendency is present in all facets of self-tracking and the practice has been justly criticised for promoting and replicating a neoliberal worldview (Ajana 2017; Lupton 2016a). Users are encouraged to take greater responsibility over their habits and manage different metrics precisely because specific behaviour changes are imagined as within the power of an individual and achievable through local interventions. Consequently, a large part of the discourse surrounding self-tracking seems to be effectively diminishing the value of cooperation, at least on the socio-political level. The wider institutional and material contexts and their influence over the tracked metrics are rarely mentioned outside of critical and academic articles. The self-tracking community (e.g., the QS movement), as well as the marketing surrounding the practice discusses the individual both as the agent and the subject of change. At most, the sharing of data is mentioned as a way to optimise and coordinate different individual efforts, but initiatives in which self-tracking is used to advance shared or political goals in a cooperative manner are scarce and motivated by factors external to self-tracking. For example, Lupton (2016b) discusses the use of self-tracking data on mobility in the context of urban activism – people use the data on their walking and cycling patterns to motivate policy changes and influence how urban infrastructure is constructed. However, even this example portrays activists who treat self-tracking as another tool available in their pursuits – their pro-social and cooperative activity has been facilitated by self-tracking, but not established through the practice.

Second, self-trackers banding together for a shared purpose face a significant limitation as they possess little control and insight into the infrastructure that makes their cooperation possible. A community or a public using self-tracking as one of its platforms for communication and activity is largely dependent on the whims of the developers, whose interests often differ drastically from those furthered by the users (as technology companies are chiefly motivated by profit). Bottom-up initiatives started with the help of self-tracking devices might quickly find themselves at a loss while trying to backup vital data or glean how some crucial features and metrics are communicated and distributed among the user base. Some scholars have previously argued that the community arising through self-tracking might be akin to communities arising in the workplace since communal tracking is in many respects similar to shared labour—by tracking, users produce a resource that is later aggregated and exploited by the developers and third parties (see Till 2014).¹⁵ However, in many countries workers are capable of achieving some institutional independence from the company structures by forming a union (thus also gaining some ability to influence the organisation of and decisions within their workplace). Currently, users of self-tracking technologies have no institutional means for increasing their influence over the infrastructure on which they depend¹⁶ and there are no legal tools they could use to have their interests recognised. In many respects, self-tracking technologies are governed in a despotic manner and “the power of the people” has little bearing on the decisions surrounding the infrastructure of self-tracking.¹⁷

This lack of political and material sovereignty of the communities formed through self-tracking is problematic also in the context of the ambitions of technology companies to reach beyond their traditional domains and have a bigger influence over various aspects of our lives. Self-tracking is now commonly being used in contexts such as fitness, healthcare, insurance and the workplace. In addition, elderly care (e.g., through fall detection), education (e.g., in PE classes and mental health counselling), childcare (e.g., child-tracking apps meant to increase safety, see Gabriels 2016) and many other areas are currently being targeted as “standing to benefit” from the introduction of some self-tracking tools.¹⁸ The saturation of different aspects of our lives with untransparent and undemocratically governed self-tracking tools would reduce the influence we currently have in these contexts and limit our ability to self-organise to cooperatively answer the problems arising within them. While on the surface the narratives surrounding self-tracking – data-sharing, community of self-trackers, democratisation of healthcare – sound appealing, they are, in fact, suspect and could be described as a form of solidarity-washing of these technologies. As noted by Ajana (2018), greater circulation of data and embedding of technology in different walks of life are predominantly in the interests of technology companies who prey on our pro-social attitudes to make these features more palatable and uncontroversial.

6 Conclusion

Over the course of this paper, I applied Dewey’s notion of transaction and the ethical ideals found in his philosophy to evaluate the interpersonal dimension of self-tracking technologies. Overall, I argued that even if many communities have formed in connection to self-quantification, this does not necessarily mean that the practice and associated technologies positively contribute to social relations. My pragmatist analysis helped me argue that desirable interpersonal developments can be attributed to users’ pre-existing attitudes, and their needs and desires to engage with others in a mutually uplifting manner.

In fact, many of the features of self-tracking technologies negatively impact social relations. The degree of epistemic distance between users resulting from the existence of data doubles makes it more difficult to meaningfully engage with others, while the reductive nature of data affects how people are perceived through their metrics and limits their transactional agency – the ability to negotiate the terms at which they engage with others. Moreover, elements such as leaderboards and social media integration promotes a comparative and instrumental view of social relations, potentially reducing others to a set of numbers with which to compete on the way to one’s own self-improvement.

I also noted that self-tracking is problematic from a political standpoint. Many communities rely on the collection of sharing of quantified data but their lack of control over the infrastructure that enables these features makes them reliant on powerful companies and reduces their political agency. The progressing integration of self-tracking into more contexts (e.g., insurance, workplace, care) could ultimately reduce the communities’ ability to make decisions about vital elements of their life and environment.

If we follow pragmatism in evaluating the content and not the mere existence of interpersonal references embedded in self-tracking, the purportedly pro-social and community-forming aspects of the technology quickly show their darker side. The shortcomings of existing tools of self-quantification make them an obstacle to genuine and uplifting transactions and ultimately negatively affect the interpersonal relations arising in connection with self-tracking practices.

At the same time, I want to stress that this article is not merely a critique of self-tracking. I already hinted that technologies of self-quantification are merely one of the many tools that encourage greater circulation of data, often with the purpose of enabling its extraction by for-profit entities. Stories about community, solidarity, and care feature prominently in self-tracking’s marketing and similar claims can be found in the public-facing communication of social media companies, online video game developers, or digital health providers. Dewey’s ethics and social philosophy allow us to critically interrogate the nature of social relations mediated by such technologies, and, as my analysis of self-tracking showed, highlight the problematic assumptions and objectives behind them. In this sense, I hope that my arguments will contribute to the ongoing debate on the sociality of technology and provide novel philosophical tools for an examination of technologically mediated transactions and their ethical impact.