Apprehending digital hostility and online abuse: Feminist care ethics in/and digital ecologies

Introduction

Online abuse, trolling, internet pile-ons, disinformation and misrepresentation, cyberbullying and harassment of other users – otherwise known collectively as digital hostility – represents a key social problem for the 2020s, given the centrality of digital communication as a setting for information, exchange, democratic participation, minority belonging and everyday cohabitation. Although the roots and antecedents of problematic online behaviour and offensive content have been part of the digital project since the early internet of the 1990s, the acceleration in instances, experiences and perpetration of digital hostility has increased substantially over the past decade in the digital platform era (Khosravinik and Esposito, 2018). Increasingly recognised as a ‘crisis’, digital hostility is now becoming a matter of public inquiry, proposed and passed legislative responses, and calls for better, more serious regulation of digital platforms as both facilitators and mediators of hostility and hate speech.

The concept of digital hostility represents three contemporary, 2020s phenomena: (i) the ‘massification’ of online adversities in which large numbers of perpetrators participate in hostile activities; (ii) a persistent focus on measuring abusive content in terms of recognisable ‘hate speech’ leaving lesser incivilities and adversities as unregulated despite the fact they may be just as injurious to some users; and (iii) a continuing cultural perception that online unethical acts are the work of ‘bad individuals’ or ‘bullies’ targeting some unfortunate users with problematic content, rather than as participants in a widespread damage to and toxification of digital culture as an information ecology we all inhabit and participate in as social actors. Recognising online abuse as harmful to both other users as well as damaging to the digital ecology opens the possibility of asking what cultural changes are necessary to underpin and advance the existing – and limited – interventions, regulations and protections. That is, if we are to follow practices of prevention and education as recognised strategies in other areas of feminist insight on crime and harm, it is essential to begin by asking what kind of care ethics could underpin strategies that actively call upon users to curtail their harmful speech or behaviour.

Although well-meaning proposals for better regulation, education, platform moderation and intervention have been made, a failure to recognise digital practices as core cultural settings grounded in interdependency, belonging and engagement has fostered responses and intervention techniques that are overly individualist, oriented towards risk reduction rather than the facilitation of environments of care, and framed by liberal-humanist tolerance perspectives that over-emphasise a right to free expression rather than a responsibility for caring behaviours. This liberal individualism, alongside the assumption that victim-survivors should self-manage their care (such as by withdrawing from digital platforms as avoidance), eschews the creative possibilities that emerge from exploring the problem of digital hostility as one that can be apprehended through a feminist care ethics and by seeking to recognise online harms not as risks and individual perpetrations but as both violence among users and violence to the digital ecology.

This paper explores some of the creative opportunities afforded by an approach to the issue through a feminist care perspective. I will begin with a brief summary of some of the key issues emerging in digital hostility and online abuse to demonstrate how it can be apprehended as a social problem or crisis that warrants an ethical responsiveness grounded in care in order to produce or re-figure a non-toxic digital culture that remains a necessary space for contemporary liveability. I will follow this with a summary of core precepts and recent thinking in feminist care ethics as understood through an orientation towards their utility for addressing problems of digital hostility. I will then work through two key concepts for care in digital context: care of the user as an embodied subject of interdependency, and care for the digital ecology as a site of necessary liveability. The aim here is to draw on recent feminist thinking on care, community, interdependency and liveability in order to open alternative, critical frameworks for a more ethical response that may help to justify behaviour change, education (including the incorporation of caring practices into digital literacy curricula) and regulatory measures that oblige platforms to foster communicative settings of care rather than adversity.

Digital hostility as violence

This paper draws on research funded by the Australian Research Council Discovery Scheme . The project aims to provide a comprehensive account of Australians’ experiences of online hostility, abuse, trolling and extremist hate speech, by analysing diverse user, moderator and stakeholder experiences and attitudes towards addressing online abuse as a socio-cultural problem. The project is presently undertaking empirical data collection with everyday users, key stakeholders, platform moderators and policy-makers to better understand the attitudes and innovations in relation to online abuse, much of which to date indicates the substantial increase in instances of online hostility as well as growing public, policy and scholarly interest in the topic as a contemporary social problem or ‘crisis’. However, the present paper is focused primarily on supporting the theorisation and analysis of empirical research by critically exploring the potentialities and pitfalls of feminist care ethics as a first principle for cultural responsiveness, and for approaching the development of new and alternative perspectives for intervention, support, regulation and education responses.

Core in this approach is understanding digital hostility as violence. This includes not just hate speech – which is better recognised as a cause of injury – but also includes the forms of problematic online behaviour that have generated social concern but remain often tolerated by platforms, regulators and free speech advocates. These include trolling (posting to cause deliberate upset, often repeatedly), doxxing (sharing private details such as addresses or workplaces to embarrass a user), harassment (persistent use of a digital technology to cause harm, embarrassment or injury to someone), pile-ons (mild shaming or call-out of a user but in such great numbers that it generates injury), and other kinds of adversarial communication. Violence is, of course, difficult to identify in ways that can garner consensus (Butler, 2020: 5). Liberal-humanist and institutional attempts to set a tolerable limit to violence often focus on regulating interpersonal violence while ignoring and thereby supporting ingrained state and institutional violence (Brown, 2006: 106). Although digital hostility does not occur in the context of direct physical violence, we can understand it in a cursory way to begin with as violence perpetrated upon a victim-survivor; certainly, there is enough evidence that being shamed online by a multitude in a Twitter pile-on, being subject to humiliating misrepresentation online, and being persistently trolled has led to physical harms and, in more extreme cases, some victims have found their lives are not liveable and chosen suicide as the only available mechanism for ceasing the intolerable emotional pain of such harms (Thompson and Cover, 2022; Kang et al., 2022; Park and Kim, 2021). In other cases, digital hostility has disrupted a sense of belonging and thereby a sense of identity stability, increasing a cluster of marginalisations from social participation (Cover, 2023a). However, digital hostility has also produced spontaneous and innovative networks of mutual and supportive care, indicating the ways in which many users see it as a matter of violence, and as a problem that requires, at least, community-based survival strategies (Author, 2022).

Existing remedies, whether regulatory (legislative, platform policy), interventional (moderation, artificial and machine-learned policing), or preventative (cybersafety education), tend to be stymied by approaches that eschew care, ethics and community in favour of risk-based approaches that individualise violence rather than perceive it as a broader cultural problem. That is, there is an ongoing tendency to individualise the problem of digital hostility as one that can identify bad actors and perpetrators who can be warned, penalised, have their content subject to moderation or removal, or be banned by platforms. This minimalist, risk-averse approach by platforms and legislators ignores the responsibilities that digital platform stakeholders may have not only to prevent users from being harmed (rather than intervene subsequent to harm) but also to provide and sustain a setting for cultural communication that is non-violent and non-adversarial in form. This problem is exacerbated by a continuing tendency to identify digital hostility by content rather than by behaviour, as seen in the promotion of legal practices of defamation as an approach to regulating online abuse. This reduces the ability among moderators to recognise how repeated but relatively mild trolling or shaming perpetrated by very large numbers en masse may harm a subject as much as criminal hate speech, position them as an ungrievable subject, or lead to intolerable levels of shame that make for an unliveable life (Thompson and Cover, 2022). Additionally, digital hostility as a public concept and the remedies proposed fail to recognise how participation in digital platforms, online interactivity and other aspects of contemporary digital culture is increasingly experienced as a necessary form of cultural and identitarian practice and may not always be a transactional choice. Public pedagogies that provide everyday advice to users experiencing hostility to withdraw from online connectivity or to pivot online engagement towards smaller, closer-knit social networks (Ordoñez and Nekmat, 2019) tend to ignore how such avoidance may not be a caring solution to those whose identities are built on digital participation (Cover, 2023b; Light, 2014). Although new scholarly assessment of remedies are beginning to emerge that attempt to complexify the discourse of potential remedies beyond individualised, content-focused moderation and withdrawal strategies (e.g., Marwick, 2021; Matias, 2019), there is substantial further work needed to develop approaches that place care at the centre of interactive digital communication as a core necessity for digital communication, belonging and community participation – rather than an added bonus – underpinned by an ethical obligation to prevent violence and thereby support and enable liveability.

In these respects, it is possible to argue that the available remedies and public discourse on digital hostility are limited by a failure to recognise violence (adversity, trolling, abuse, shaming and hate) as an increasingly central experience of contemporary digital communication (Cefai, 2020). To say so is not, however, to suggest that digital hostility is technologically determined by the medium of communication, nor to suggest that all online interaction is violent. Rather, it is to argue, as Mark Andrejevic (2020: 61–62) suggests, that the pathologies of the contemporary digital setting result from media and institutional infrastructures that misrecognise the ‘forms of interdependence that underwrites our society’. This, then, is to begin to recognise digital hostility not only as a form of interpersonal violence, but also as the kind of violence that takes place when cultural, institutional and infrastructural supports fail to provide the necessary conditions for interdependent users to participate with liveability and safety from harm as liveable lives (Butler and Worms, 2023: 31–32). Here, turning to care as a formation that not only seeks to make an ethical case against violence, but can also actively mechanise those ethics by calling for a cultural shift for more protective infrastructures presents one critical option for apprehending and addressing digital hostility.

Feminist ethics of care

In the next three sections, I will aim to make the case for why an ethical framework of care informed by feminist critical theory provides a valuable lens for apprehending the violence of digital hostility and for providing remedies in the forms of reconfigured digital communicative spaces as an alternative to the nascent individualist, risk-averse and self-managing forms of intervention and prevention. There are several different approaches to care that warrant an assessment for their suitability as an ethical response, and I will unpack some of these in this section before discussing two important points derived from critical care perspectives: first, that a restoration of the body to perceptions of digital communication helps us understand the extent of violence and the ethical obligations for care settings that sustain users as corporeal subjects; and second, how understanding violence not as interpersonal but as damage to what I refer to as the digital ecology can help reframe the mechanisation of care as directed towards providing ethical infrastructural supports for liveability, including across digital platforms and digital culture.

Across the past two decades, much of the scholarship on the deployment of care has included care's value in responding to violence – particularly violence against women and minorities. Some of the more valuable work in this space has been based on Foucault's (2005) reading of Greek, Hellenistic and Roman philosophy and Christian spirituality in which he discerned articulations of care for others as concomitant with care of the self. In the practice of caring for the self, it is care for the other – other people – that is indispensable in achieving the goal of mastery of the self (Foucault, 2005: 127). In caring well for the self, one seeks to care for others whereby the practice of the self and subjectivity is seen always as part of an ethical social relationship (164). Although ancient and Hellenistic philosophies of care were historically dislodged in European culture by approaches to reason credited to Descartes (1968: 14), it is the sustained domination of ethics by European Enlightenment culture and particularly liberal-humanism that continues to dominate perceptions of ethics and selfhood into twenty-first-century modernity. The ancient model of the care of the self and others articulated by Foucault provided one point of injection into developing new models for the ethics of non-violence, allowing scholars, practitioners and service providers to thereby posit an anti-violence ethics built not on moralistic models which emphasise the ‘do nots’ (Albury et al., 2011: 346) but, instead, on practices of decision-making and negotiation across interpersonal relationships in ways which seek to minimise harm.

The interest in a Foucauldian approach to care thus helped anti-violence advocacy to dislodge the (less useful) North American care perspectives modelled on maternal care for infants, which inaccurately presumed care was a natural and endemic disposition of women that might be encouraged and practised more widely among others (Butler and Worms, 2023). For example, Moira Carmody's (2003, 2005) significant contribution to anti-violence theorisation expanded substantially on Foucault to develop further an ethical framework that called upon perpetrators of violence to consider the care of the self through the care that must be directed towards the other, namely, women victim-survivors of that violence. Carmody made a powerful case for a feminist care framework as the grounding for the development of ethical behaviours among men that might overcome the risk of violence towards women. In this framework, recognising the vulnerability of a victim-survivor is an anti-violent act of caring for the other, whereby violence towards the vulnerable other is unethical partly in its perpetration but also in that it perpetrates violence against the self. For Carmody, then, an ethics of the care of the self in and through care for others overcomes the problem of highly individualised approaches to the prevention of violence which assign responsibility for managing risk on individual actors perceived as more likely to perpetrate violence, and promoting the use of avoidance strategies among victim-survivors including particularly minorities and women perceived as more likely to be subject to violence (Carmody, 2003: 200). In this framework, care perspectives are drawn upon to call for a primary inculcation of anti-violent ethical practices as a form of prevention in which all parties, including bystanders, have responsibility. To instil such an ethics, Carmody (2003) argued, requires widespread education in order to ensure that all parties engaged actively in developing ethical relations do so by building on critical reflections on the implications of their behaviour on others and thereby on themselves. While a valuable contribution to the array of feminist care ethics, the key question remains not why violence occurs, but what interpretive frames prevent the recognition of the victim-survivor as vulnerable (Cover, 2014; Butler, 2009) – in not being able to address this aspect, violence as normative is sustained. Additionally, the deployment of this particular model of care retains a focus on violence as that which occurs principally in interpersonal relationships, leaving aside therefore the structural, institutional and infrastructural settings that either fail to prevent violence or could better provide care as anti-violence.

A more critical approach to care has been developed within feminist theorising, however, in ways which draw back the lens on how care might function socially, and thereby differing from the Foucauldian framework by disavowing the need to understand care in the context of interpersonal relations between individuals or in the understanding of care as having utility in the mastery of the self (Lupton et al., 2021). Significant to this critical approach has been the incorporation of social practice models, both drawing on and extending post-structuralist approaches. For example, in the work of Joan Tronto (2013), care is shifted away from a matter of individual or personal responsibility – regardless of how or to whom care is directed as an obligatory anti-violence in relations – and towards the social, cultural, economic and infrastructural supports that encourage connection: a movement away from ‘caring for’ or ‘caring about’ to ‘caring with’ as a framework for the political mobilisation for transformations that prevent violence, harms, deprivation or indignities (The Care Collective, 2020). In furthering Tronto's approach, the Care Collective (namely Andreas Chatzidakis, Jamie Hakim, Jo Littler, Catherine Rottenberg and Lynne Segal) reactivated feminist approaches to care and ethics for contemporary settings attentive to the need to overcome the individualisation of care, responsibility and harm management encouraged in neoliberal settings. They began by pointing to the ways in which the idea of the good and responsible citizen utilised in other anti-violent care perspectives is rooted in an over-emphasis of individualism and individual responsibility that draws on the 1980s brand of neoliberalism of the United States and United Kingdom (The Care Collective, 2020: 12). The collective proposed, instead, that the lack of caring cultures and ways-of-being witnessed across the array of personal harms, population health disparities, inequalities and injustices is the source of violence. The only viable solution to such violence of the present era, then, is a progressive ‘model of “universal care”: the ideal of a society in which care is placed front and centre on every scale of life’ and at the core of every infrastructural support necessary for communities to operate genuinely (The Care Collective, 2020: 19).

Although not exclusive of either maternal models of care or the Foucauldian care of the self/other, this critical approach encourages a re-focusing away from both interpersonal violence and care as interpersonal relations in order to consider the role that social institutions, cultural practices and support infrastructures and languages can play in preventing violence, including the violence of making some lives less liveable than others, and some subjects less grievable than others. This critical feminist approach to care (and violence) has substantial value, then, for addressing digital hostility as a normative form of violence in contemporary communication – not because it seeks violence prevention, but because it provides ways of understanding the communication environment as one which could be an infrastructural care support, and because it incorporates an appreciation of the a priori interdependency among all subjects (or digital users) as primary and prior to the construction of the individual, thereby providing powerful reasons why digital settings as infrastructures themselves have a responsibility to produce a care setting that seeks to preclude violence, including the violence of digital hostility. Such an approach is supported by the post-structuralist theorisation of interdependency in Judith Butler's (2020) recent work as I demonstrate in the next two sections in connection to understanding digital settings as both a site for vulnerable bodies and an infrastructural ecology for liveability.

Care of the body in interdependency

Unlike the early cyberculture discourses of the 1990s, we no longer think about digital communication as a setting ‘hived off’ from the body. Some aspects of early cyberculture drew on fictional accounts to describe and celebrate the early internet as an environment of separate, disembodied ‘space’ in which we interact at the level of the ‘mind’ while leaving behind a redundant ‘body’ (sometimes derogatively described as ‘the meat’). This notion of the disembodied digital user drew, of course, on the mind/body dualism that was derived from the seventeenth-century philosophy of René Descartes (1968) and maintains some continuing influence in twenty-first-century cultural understandings of subjectivity. Cultural studies had already pointed to the conceptual construction of the mind/body myth, pointing to some of the ways in which it had been constitutive of discourses that relegate the body to an inferior position (Grosz, 1995). Arguably, the genealogy of a mind/body dualism that retains a spectral influence on how contemporary digital culture is understood could be apprehended as a reason for some of the digital hostility at play: online abuse, trolling and other harms are perhaps perceived by some perpetrators as non-violent because they are not recognised as having a direct violent impact on a body.

The body, of course, is absolutely endemic to digital communication – there is no disembodied user. Rather, digital practices are always about the body, whether that is the body that types responses, the body that clicks, the body that historically has come closer to the technology through the touch-screen, the virtual reality devices and the wearable technologies, the body that is photographed in shared selfies, the bodies that are seen on-screen, the body that reacts with the full array of affects to the interaction encounters experienced online, the body whose health and physical responses are tracked by wearable devices, the body that – knowingly or unknowingly – performs free labour for digital platforms through the attention economy, or the body that interacts with generative AI in one-sided creative conversations (Author, 2023b). Indeed, one might ask what sort of care practices digital culture deploys if it is able to sometimes ‘forget’ the body by way of downplaying the physical harms of digital hostility.

Judith Butler's intervention into ethics in the late 2000s added to the growing corporeal perspective in feminist critical and cultural studies by pointing to the fact of corporeal vulnerability as one of the key shared experiences of human subjectivity, that is, because we are corporeal subjects we are vulnerable and precarious from the beginning of life and prior to subjectivity, we depend on others for corporeal survival and, of course, that vulnerability is not shared evenly or democratically, leaving some subjects more vulnerable than others (Butler, 2004, 2009). The primary vulnerability of the body to the violences of harm or loss is a valuable perspective as both a grounding and a goal for a feminist care ethics directed towards non-violence in digital settings.

However, it is important to be very careful with how we figure corporeality when we bring concepts of violence and concepts of digital culture together, as it is easy to make the mistake of seeing violence as not only interpersonal but also inter-corporeal and thereby miss the opportunity to consider digital hostility as genuinely violent or, indeed, to sustain attention on the role of institutional and structural violence. As Butler (2020) points out, liberal-humanist approaches to violence have typically used the physical blow to the body as the measure of what counts as violence, seeing that which does not take the ‘literal’ form of a blow as either not violent or violence of a lesser kind. This has the unfortunate impact of failing to recognise the embodied character of all violence, including structural, institutional, linguistic, discursive, marginalising and shaming that not only take their toll on the body but are also experienced bodily even when no physical blow is witnessed (Butler, 2020: 1–2, 137–138). It is important to recognise first that all digital communication is always bodily from the beginning and there is no such thing as a disembodied user of digital media (Author, 2023b), and second, that violence is always embodied even when it is not marked by the physical blow. Rather, a care ethics that draws on Butler's approach to corporeality starts with recognising that all subjects are vulnerable, that such vulnerability is shared, and that any register of violence is thereby unethical.

Some of Butler's recent works have sought to address violence further by driving their definition of the vulnerable body into the framework of human interdependency to ground an ethical claim to obligatory non-violence: like vulnerability, interdependency is an a priori condition of embodied subjectivity and this point buttresses the ethical obligation not to perpetrate, accept or foster violence. Acknowledging their indebtedness to a longer trajectory of feminist writing, and noting the links between their ethical approach and Tronto's practice models (Butler and Worms, 2023), Butler argues that the radical individualisation of bodies that marks liberalism from Hobbes, Locke and Mill to contemporary rights assertions has been pivotal in preventing an effective intervention into violence. Butler's repositioning of the body as that which is neither an expression of the individual or collective will (Butler, 2020: 197) but is ontologically constituted in dependency and interdependency provides a feminist care ethics with a critique that demonstrates that violence is not merely perpetrated on individuals but harms the collective interdependent subjectivity itself. Dependency, they argued, has often been written out of the foundational myths of natural human life (Adam, Robinson Crusoe, the independent, self-sustaining individual whose dignity is marred by a social contract and the force of law). This presents an argument that recently marked so much rejection of COVID-19 care practices such as isolation, distancing and vaccination, all of which were designed to protect the vulnerable in interdependency and because of that a prior interdependency of human life (Cover, 2023c; Lupton et al., 2021).

Counter to individualism and the myth of the natural, autonomous, self-subsisting man, Butler argues that ‘no body can sustain itself on its own’ (Butler, 2020: 49), never has, but is, rather, always defined from the beginning by an emphatic lack of self-sufficiency. Recognising interdependency does not mean eradicating distinctiveness or boundaries, personal responsibility or agency as subjects, but acknowledges that such distinctiveness and boundaries are constituted in the interrelationships of all bodies that cohabit the world (Butler, 2020: 16–17). Our subjective interdependency across the lifecycle is infused in our everyday embodied lives from how food and water are made available to our psychic dependency on various relationalities, to the technologies that mechanise digital communication. Recognising our interdependency leads to an ethical framework: if violence (interpersonal, structural, linguistic) is between corporeal subjects, then there is already a social relationship founded in global interdependency which obliges non-violence because any violence is directed first to that interdependency and only secondly to the individuated subject (Butler, 2020: 9).

Butler's contribution to a feminist ethics of care is grounded in the fact that all embodied subjects are constituted in interdependency with other subjects, thereby warranting care for that intersubjectivity that precedes us and constitutes our own subjectivity. Once we acknowledge that all digital communication is embodied from the beginning, and that violence in the form of digital hostility, online abuse, trolling, shaming and harassment always have embodied consequences, and that as corporeal beings we share both vulnerability and interdependency before we are subjects, we are able to recognise an ethical obligation of non-violence, and that means an obligation on digital spaces, platforms, companies, regulators and others to participate in providing an infrastructure built on caring not merely about differentiated vulnerable others, but about all subjects in that interdependency.

Care of the digital ecology

It is tempting to limit a discussion of feminist critical care approaches to digital hostility to a consideration of the body and by fortifying this ethics by listing (as a kind of evidence) the many instances in which individualised bodily harms have been experienced by those who have been subject to digital hostility. However, the mechanisation of an ethics in digital settings calls upon us to ask not just about preventing violence between corporeal subjects because they are corporeal, but also about what happens when violence damages the supportive ecology that sustains the possibility of communication with liveability. This is to consider the significance for care ethics that emerges from understanding care as formed through and directed towards appropriate institutional, social and discursive ‘supports’ that enable interdependent lives to live as ‘liveable lives’. Indeed, Butler recognises the intersection between the significance of infrastructural supports in their own ethics and the supportive care frameworks outlined by Joan Tronto (Butler and Worms, 2023: 36). Infrastructures are not cultural formations that exist alongside dispositions to care; rather, to be ethical, they must be caring infrastructures; indeed, care as infrastructures, and infrastructures of care. In the context, then, of digital hostility, ethical platforms, regulatory frameworks, policies and technical mechanisms of communication and online interactivity must have care at the centre; they must be care as platforms and platforms of care in ways that enable equitable interdependency to flourish such that all lives (users) are able to flourish as liveable lives.

By invoking ‘liveability’ here, I am pointing again not to interpersonal kindness among digital users, but to that fact that liveability in interdependency involves relying on the supportive social, communicative, linguistic and cultural structures that enable people to lead liveable lives, which means lives not marked by subjection to violence. Importantly, liveability does not mean maximised individual happiness; rather, it implies a state of being that is not subject to violence and life that is grievable, in other words, a life that is subject to care. Indeed, for Butler, the key ethical question for contemporary care involves asking what ‘intersubjective institutional conditions … would include conditions for the possibility of cohabitation such that life could become increasingly livable for populations deprived of that possibility?’ (Butler and Worms, 2023: 30). In this context, for Butler, non-violence is not about the absence of adversity, but about the capacity to cohabit in interdependency sustained by social structures that ‘affirm and … support the institutional conditions that support interdependent lives as part of a community’ (Butler and Worms, 2023: 32). To understand the digital setting as one that might become non-violent in order to support interdependent liveable lives – to care – is, then, an understanding that must start by perceiving digital culture as an array of settings, practices, regulations and norms that operates as an ecology.

An ecology is usually defined as the mutual relations of organisms with their environment and with one another. In anthropologist Margaret Mead's (2017) definition, an ecological setting includes the geographical, human and nonhuman groups that provide the conditions for sustainability and cultural evolution. If bodies are both vulnerable and interdependent as irreversible conditions of subjectivity, and if care is founded on providing support, then our attention must be upon the ecology in which that care is sustained such that lives can be led as liveable lives, that is, lives not subject to violence. And that requires paying attention to a diverse array of ecologies whether geographical, local, global, climate-related, in terms of institutional and universal healthcare, and in the context of our contemporary forms of digital communication.

To consider what an ecology might mean in terms of fostering a setting for liveability as non-violence, it is useful to turn to the late Lauren Berlant's (2022) posthumous work On the Inconvenience of Other People, in which she addressed some aspects of the ‘infrastructural commons’ as a way of exploring the meaning of a concept of ‘we’ (an alternative framing of interdependency). Infrastructures and ecologies have typically been apprehended through separate registers – Donna Haraway (1990), for example, argued that an ecology was best perceived as a science of complex spatial relations that subsequently enables an infrastructure to be thinkable. However, Berlant deliberately entangles the concepts of support, infrastructure, culture and ecology by describing social and supportive infrastructures as relational and ecological processes that sustain worlds, even though what we witness in contemporary culture is a failure that comes from attempting to hive off one from another and deploy infrastructures to sustain only some subjects (Haraway 1990: 146), thereby making liveability an uneven or sometimes inaccessible condition of survival.

By considering the infrastructural care supports as a necessary and inevitable component of any sustainable ecology, we are able to ask an important set of critical questions for a feminist care ethics oriented towards resolving the violence of digital hostility:

In what ways does the practice of digital hostility toxify the ecology and leave it non-supportive and unable to sustain liveability?

Such questions become starting-points for a genuine care intervention into the digital ecology.

Thinking violence as always a destruction of the ecology rather than harm to an individual body puts corporeal life at the very centre, because it is a violence to the ecology in which all bodies in egalitarian interdependence may be socially supported for liveability. Every act of violence is violence to the shared ecology that the infrastructure’s interdependent bodies rely upon for that interdependency and to sustain liveability. Every act of violence is an act that damages the ecology, and this is witnessed perhaps most clearly in the permanency of adversarial and abusive content, in disinformation that leaves continuing traces in searchable content, and in the inequitable, haphazard and overly individualised practices of reporting problematic content and having that content moderated. The ecology, as a shared life-affirming social, economic and psychic infrastructure is already one considered toxic. To be dependent implies vulnerability: all subjects (or users) are vulnerable to the toxification of any ecology, and when that toxicity reaches a certain level it fails to sustain liveability, and does so in inequitable ways such that some lives are made less liveable and more precarious than others – an outcome that, itself, further toxifies the ecology.

By re-focusing a feminist ethics of care, understood critically in the context of the necessary supports for liveable lives, we are better positioned to overcome the problematic discussions of which individual instances of digital hostility are violent, to which individual the violence was done, and about which perpetrator is responsible for the violence. Rather, we are able to focus on the demand for those key stakeholders in digital ecologies to consider their obligations to provide a non-toxic, non-violent supportive ecology for the populations that must, today, live their lives in and through digital settings. We are constituted as much by digital communication as any other form, and we are thereby bound to it, and obliged to care for that digital ecology as the basis for interacting with others in a sense of interdependency and care.

Conclusion: cohabiting a digital commons of care

Addressing online harms through state regulation for either purposes of penalisation or behaviour normalisation has proven difficult primarily due to the interjurisdictional complexities (Vincent, 2017) with victim-survivors in one country, perpetrators in another and platforms in a third. Digital literacy education has remained focused on young people and (individualised) risk and safety (Hartley, 2009), while platform interventions and moderation have remained limited due principally to the perceived benefits to platforms’ business model that adversity online brings (Wark, 2019) and the known costs of increased platform moderation (Roberts, 2019). Although there is an increasing public mood for addressing problematic digital communication (Flew, 2021), it remains very much the case that the lack of a shared understanding in the cross-sectoral environment of policy, legislation, platform regulation, education, behaviour and norms of communicative practice is a key impediment to addressing the problem as a problem. This is where, then, a feminist ethics of care is offered as a way of shaping the discourse around digital harms in terms of finding the commonality in the question: what kind of (digital) work might we want to cohabit?

In this conclusive section, I would like to make a few remarks drawing on the concept of cohabitation as a way of asking what kinds of digital ecology might be more sustainable of liveability, more capable of care, and more of a setting to care about than the current platform culture marked by a mode of production in an attention economy that not only eschews care for users, but actively profits from the generation of adversity and other forms of hostility as the most effective means of creating traffic and therefore profit. By pointing to the fact that digital spaces, much like local and global geographic settings, are cohabited, this implies a further obligation for non-violence and care among those who cohabit and may do so in the future. That is, a digital ecology implies a culture of unwilled proximity and unchosen cohabitation with others in platforms, forums and other digital tools. In that respect, just as in other corporeal frameworks, that unwilled and unchangeable fact that is prior to any subjectivity obliges an ethical orientation of non-violence – ‘not to destroy any part of the human population or to make lives unlivable’ in Butler's (2012: 24) words. It is not just that we cohabit the digital ecology as an infrastructural support but, in Butler's terms, the digital ecology has become ‘an immanent feature of life itself’ (Butler, 2020: 197–198) and thereby warrants an ethical orientation. When the digital ecology is subject to violence in the forms taken by digital hostility, then, the ecology is toxified, damaged or in disarray – just like any other ecology that can no longer support liveable lives when toxified. To say this is to reposition both violence and care as that which must be understood not in interpersonal terms, but in terms of an ecology that includes all the interdependent, mutually supportive embodied subjects, the infrastructures, institutions and discourses that make it sensible and recognisable as a cohabited setting.

This is not to suggest that the subjects of hostility (perpetrators and victim-survivors) are eradicated from the conceptual map of understanding violence and care in the digital ecology; rather, it is to say that all subjects who cohabit digital culture must be perceived as equally vulnerable, equally grievable, equally capable of being harmed by the toxicity of adversarial, abusive or trolling behaviours in spite of the diversity of those cohabiting the ecology. To put it another way, ‘cohabitation implies an affirmation that one finds the condition of one's own life in the life of another where there is dependency and differentiation, proximity and violence’ (Butler, 2012: 130). Understanding the digital ecology not only as an infrastructural support for contemporary liveability and social participation but also as one cohabited today, and likely to be cohabited by those subjects yet to come, calls upon us to recognise it not as a service we pay for through the transaction of attention or the provision of data to digital companies, but as a space that is cohabited commonly and that like other ecologies demands detoxification by calling on cohabitants to care for it as a commons.

Although the platformisation of digital communication and interactivity shifted internet culture away from being one sometimes described as the digital ‘commons’ of the Web 1.0 era in favour of a corporatised post-capitalist subsumption of attention and extraction of user time (Flew, 2021; Terranova, 2022; Wark, 2019), the digital setting operates nevertheless as an ecology enabling subjects (users) to engage in the necessary communication, exchange, participation and accessibility to support living a (contemporary) liveable life, a life of sociality. Feminist care frameworks as I have been describing them in the context of digital-corporeal users, digital practices and a digital ecology prompts us to make the ethical call for a global repositioning of digital culture as a commons, or at least to be treated with the collective care that, traditionally, is warranted by a commons which supports interdependent liveability and shared use (Hardin, 1968). This is not to make an idealistic claim that the ownership of digital platforms should be handed over to a global collective of users (Lazzarato, 2004). Rather, it is to argue that since it is the site of liveability, since the digital ecology sustains contemporary liveability through fostering social participation, and since its toxification affects not only its users as corporeal subjects but all sorts of other common institutions in everyday life, regulation, education and shared attitudes towards the digital ecology must replicate those attitudes accorded to other commons such as shared use of public land and the demand not to toxify the shared resources of liveability. As Hardt and Negri (2009) have noted, the corruption of the idea of a commons has refigured community bonds as adversarial and detrimental rather than socially supportive, and we might claim that the subsumption of the digital commons by platform economies has done likewise within digital culture.

This is to make a call that a care ethics in the context of digital hostility is not a demand for kindness among users, civility or restrained speech, but the much larger task of detoxifying and sustaining the digital ecology as that which we all share. It is also to suggest that individualised, risk-averse practices of curtailing digital hostility are unworkable in non-commons settings governed by the ‘sovereignty of the profit motive’ (Eagleton, 2011: 270). Indeed, by refiguring the digital ecology as a commons encourages care in three respects: (1) caring practices digital interactivity in which violence towards one user is perceived as a violence towards the interdependency of all users; (2) caring about the common ecology as a place that must sustain liveability rather than serve as a challenge to it; and (3) care about the future generations of users to whom all this content – informative and harmful; creative and toxic – will eventually be bequeathed.

If care is about enabling infrastructural supports for liveability (Butler and Worms, 2023; Tronto, 2013), and if supports are obliged to ensure equitable access to liveability, and if such liveability includes an appreciation of the interdependency of all subjects (Butler, 2020), then the mechanisation of care here is about a demand for policy, regulation, ownership practices and interjurisdiction frameworks that restore the digital ecology to one of a commons where – at least – the support towards an equitable grievability of all subjects is guaranteed. This is to create what the Care Collective (2020: 51) refer to as a ‘sharing infrastructure’ founded in a sensibility of public community space, which is digital space in this context.