Accounting for care within human geography

Human geography has experienced a burgeoning interest in care Despite this, the more radical potentials of thinking with, and through, care remain largely unexplored In this paper, we critically examine one such potential, asking how care might facilitate a substantial rethinking of practices of research and analysis within human geography We argue that care does not simply name practices of social reproduction or emotional attachment, but is a distinct mode of ethics, both visible in the social world and capable of informing academic practice We ask what it means to recognise everyday accounts as acts of care, and to analyse these same accounts through an ethic of care where knowledge, action, relating to others, and the shaping of ethical commitment are inextricably intertwined While, typically, everyday accounts are seen as about some sort of underlying meaning or dynamic, we suggest that such accounts need to be understood as parts of efforts to navigate and re-make social worlds We unfold our argument by first tracing how care has been understood and analysed within human geography as a shifting and situated social practice Building on, but moving beyond, such approaches, we examine social worlds as ?matters of care?, where everyday understandings, and the potential for action and ethical commitment are not only continually negotiated, but are staunchly kept open to new possibilities Through the close reading of extracts from in-depth interviews with first-time parents in the city of Oxford, UK, we illustrate how care offers a committed practice of knowing and relating within research We argue this approach provides new ways of thinking about geographical research, where primary research, analysis and scholarly narratives are all implicated in the remaking of everyday worlds which, in turn, reveal a new terrain of political potentiality

committedly feminist geographers, feminist theory and philosophy have yet to be taken up seriously by mainstream human geography. In this paper, we endeavour to make the case for care not simply as a topic of study, but as a core ethos for human geography, with a particular focus on what it means to conduct empirical research and analysis with care. We term these practices of care-ful research and analysis "a hermeneutics of care." However, as we will argue, interpreting the socio-material world carefully necessarily entails moving beyond the textual practices typically implied by the term hermeneutics, in order to make interpretation a matter of active, ongoing ethical entanglement. Drawing on recent research with first-time parents in the city of Oxford, UK, during a period of marked urban austerity, we illustrate how care-ful analysis not only attends to the profusion of everyday labour put into sustaining and remaking worlds, but also becomes entangled in such efforts.
We begin by tracing how care has been conceived within human geography. Human geographers have developed a sophisticated understanding of care as an entangled and often pervasive social practice. At the same time, however, we highlight how these existing approaches have also worked to constrain the relevance of thinking on care, by imagining care largely as a topic to be studied, or as a standard for critique, rather than approaching care more expansively, as an ethic. Building on these approaches, but attempting to reach beyond them, in the third section, we trace an expanded notion of an ethic of care, beyond that which has broadly been used within human geography, by returning to the work of key thinkers from feminist philosophy, and science and technology studies. We trace how such work challenges us to resist the quick reduction of the accounts we study into "evidence" and evidence into theory, and instead to "stay with the trouble" (Haraway, 2016) within such accounts, so as to be able to recognise them as acts of "persistent tinkering" over time (Mol et al., 2010), situated within and acting on the world. In turn, we trace how this understanding of our informants' accounts leads us to a different understanding of hermeneutics, where analysis is not simply something that occurs after the research encounter but is also an inescapably ongoing, interactive, and politically consequential process, achieved between researchers and subjects collectively. In the fourth section, we unpack this argument empirically, showing how the narratives shared by first-time parents not only reflect ongoing efforts to reshape their world, but also inevitably draw us, as researchers, into such efforts. In the concluding section, we consider the broader implications of this argument not only for how we understand care within human geography but also for how we conduct geographical research more generally.

| GEOGRAPHIES OF CARE
Over the last 20 years there has been a reinvigorated interest within human geography in understanding care in socio-spatial terms (Conradson, 2003a(Conradson, , 2003bMcKie et al., 2002;Parr & Philo, 2003;Popke, 2006). As Parr (2003) notes, much of the previous attention given to care was within medical geography, where care was understood largely as a formal or semi-formal institutional practice, and where the primary questions were those of access, inequality, and governance. Certain domains of care, such as parenting or home-making have enjoyed longer histories of attention, but have been predominantly framed as discrete domains of inquiry, rather than being treated as instances of care (Blunt 2005;Blunt and Varley 2004;Holloway 1998). Furthermore, for the second half of the 20th century, feminist critiques of patriarchy and heteronormativity within human geography inadvertently served to discourage sustained attention to more conventional domains of social reproduction such as the family, and to the spatial dynamics of social reproduction more generally (Valentine, 2008). The 1990s and 2000s saw this inattention shift with the publication of an influential set of texts within political philosophy that sought to recover care as a feminist concern (Clement, 1996;Fraser, 1997;Gatens, 1998;Held, 2005;Kittay, 1999;Sevenhuijsen, 1998;Tronto, 1993). Reviewing the "flourishing" body of geographical studies on care and caring today, Power and Hall frame of literature as collectively "driven by questions of who and where in the provision of care" (2017, p. 2). The question of "who" cares and "where" care occurs, however, must first start with a basic understanding of what care is. We argue that in human geography, even as understandings of care have grown more sophisticated, care has consistently been imagined either as a topic of study, or as a normative ideal, rather than as an ethos for doing geographical research more generally.
In general terms, care has been defined "as the proactive interest of one person in the well-being of another" (Conradson, 2003a, p. 451, drawing on Silk, 2000. This interest may be practical, facilitating processes of social reproductionas in the acts of feeding, therapy, changing nappies, or watering a plantor emotionalas in the feelings of care we may experience towards children, characters in a story, or pets. As these examples suggest, care can encompass both human and non-human others, and practical and emotional forms of care neither presuppose nor preclude each other. This understanding is woven into our everyday language; we frequently talk both about caring for others, as well as caring about them. In what follows, we review how care has been understood within human geography around three prominent conceptualisations of care: Milligan and Wiles' work on "landscapes of care," the work of Bowlby and her collaborators on "carescapes"/"caringscapes," and a range of work on an "ethic of care." As Milligan and Wiles (2010) note, the term "landscapes of care" is not their own coinage, but one that has emerged from health geography. Milligan and Wiles synthesise these uses into a way of thinking about care geographically. Starting from a definition of care as "the provision of practical or emotional support" (2010, p. 737), and drawing a distinction between practical "caring for" and emotional "caring about," they argue that geographers ought to pay greater attention to how these two forms of care are distributed, supported, and governed. They argue that care is shaped not only within physical space but also across social and emotional space, with each form of spatiality producing new relationships between the up-close, interpersonal experience of care and the broader ideological frameworks and public policies that shape this experience. Collectively, they refer to the ways in which care is distributed and shaped across these different forms of space as "landscapes of care." Similar approaches to Milligan and Wiles include Holloway's (1998) work on "moral geographies of mothering," which traces how "local childcare cultures" are shaped by the distribution of support and resources, and how mothers from different backgrounds are addressed within policy. McDowell et al. (2005) build on this by extending situated understandings of the meaning, ethics, and possibilities given to mothering to the work of household social reproduction more generally. Similar approaches have also been taken by Boyer et al. (2017) and Mckie et al. (2001) in their work on the changing gender dynamics of household care, or by Parr and Philo (2003) on rural mental health. Uniting these are considerations of how care is made possible, or indeed re-made, across different spatial relations.
If the approaches above trace how care becomes possible in different ways across space, Bowlby and her colleagues are keen to emphasise that care changes form across time as well. In early work (McKie et al., 2002), they call for particular attention to be paid to how care unfolds through spatio-temporal processesranging from "planning" or "worrying" to "shifting patterns of work," community, or familial involvement that enable or constrain care (2002, p. 915). They refer to the collective ways in which care is practised, organised, and governed across time as "caringscapes." In later work, Bowlby (2011) pushes this understanding of caringscapes/carescapes 1 further, arguing that care is fundamentally a mobile practice in that the emotional attachments, practices, and routines, socio-material forms of support, and forms of politics and ideologies that collectively shape care are all processes that not only change but come to be realised across time and across the life course.
This understanding of care as a mobile practice, reconfigured and realised across time and space, resonates with more recent approaches which conceive of care in terms of "assemblage"that is, in terms of a heterogeneous, contingent, and shifting configuration of relations (cf. Deleuze & Guattari, 2004). For instance, recent geographic scholarship on parenting has framed the practices and experiences of parenting, or of family life more generally, as matters of assemblage, produced through a shifting set of relationships between parents, children, and a range of environmental factors, such as strangers, policy-makers, online forums, the weather, or the design of the pavementall of which come together to remake the meanings, feelings, and practices associated with parenting at any given moment (Boyer & Spinney, 2016;Clement and Waitt, 2017;Clement and Waitt, 2018;Jupp, 2013;Pedersen & Lupton, 2018;Price-Robertson & Duff 2019;Robinson, 2018).
Finally, human geographers have also drawn more directly on feminist philosophical work on care as an "ethic." In broad terms, this approach treats caring "not so much an activity as a way of relating to others" (McEwan & Goodman, 2010, p. 103). Ideas around ethics of care have been deployed at different levels of analysis. For example, Conradson (2011) highlights the work of social geographers on care in a globalised world and understandings of care as a transformative ethic in relation to care and responsibility for distant others. Some scholars have used this idea descriptively, as a way of highlighting the ethical dimension of relationships marked by practical or emotional caring (Askew, 2009;Barnett & Land, 2007;Brown, 2003;Day, 2000;DeVerteuil & Wilton, 2009;Enticott, 2016;Evans & Thomas, 2009). Others, however, have deployed the idea of an ethic of care as a critical framework, to trace the absence of caring commitments, as measured against an ideal of a world sustained through caring interdependence. In this vein, the idea of an ethic of care has often been used to criticise marketised or austere configurations of care, which are seen as failing to live up to caring ideals (Barnes, 2011;Green & Lawson, 2011;Haylett, 2003;Murray & Barnes, 2010;Smith, 2005;Till, 2012). Meanwhile, others have used the idea of an ethic of care in a more meta-theoretical way, suggesting that it offers us a template for how we might conduct research, formulate theory, teach others, or even run universities (Greenhough & Roe, 2010;Lawson, 2009;McDowell, 2004;Raghuram et al., 2009). Tronto's work (1993Tronto's work ( , 2010 provides the most widely reproduced formulation of care as an ethic, understood as "a species activity that includes everything that we do to maintain, continue, and repair our 'world' so that we can live in it as well as possible. That world includes our bodies, ourselves, and our environment, all of which we seek to interweave in a complex, life-sustaining web" (1993, p. 103, originally in Tronto & Fisher, 1990, p. 40). Building on this, Tronto (1993) emphasises four interlinked principles for what she refers to as "adequate" care: an attentiveness to the experiences and needs of others; a stance of (collective) responsibility towards these needs; competence in meeting the needs of others; and a dynamic of mutual responsiveness between care recipients and care givers. Taken together, these four principles might be understood as starting from an acknowledgement of human interdependency with others, and then as formulating an ethics that acknowledges, supports, values, and even deepens this interdependency. This approach has been taken up descriptively and critically by a range of geographers to interrogate the politics of how worlds are sustained (see for example, Brown, 2003;Evans & Thomas, 2009;Popke, 2006;Till, 2012).
Collectively, even while tracing the entanglements produced by care across space and time, the approaches above envision care either as a social practice or ideal (of caring "about" or "for" others) that can be studied empirically, or as a standard for critique. However, a different understanding of care can emerge from highlighting an underappreciated dimension of Tronto's work: her insistence that the four principles of care remain integrated. Here, an "ethic of care" not only entails valuing social reproduction, or championing an ideal standard of caring labour, but also a mode of knowing, acting, responding to others, and shaping ethical commitments, as a singular, situated practice, where these dimensions acquire value only in relation to one another. In human geography, this understanding of care not simply as a social practice or as a concept but as an ethic has been taken up by Lawson (2007Lawson ( , 2009, who argues programmatically for geographers to take up a "critical care ethics for our epistemological, ontological, methodological, and daily life practices (as professionals and citizens)" (2007, p. 3). She advocates for geographers to attend epistemologically to how knowledge is an inescapably situated and committed practice; analytically to how particular accounts of the world are produced; critically to how certain accounts, entailing particular configurations of value, caring-labour, and responsibility, gain dominance over others; and meta-theoretically to how our own academic labour might foster or break commitments of its own. In what follows, we build on Lawson's integrated and encompassing approach to an ethics of care and work emphasising the socio-spatial entanglements of care (cf. Bowlby 2011;Milligan & Wiles, 2010) in exploring how we might account for the experiences of our research subjects in a care-ful manner.

| A HERMENEUTICS OF CARE
Collectively, geographers have come to highlight the dynamic nature of care. Care as a practice as well as the subjects care invokes, entangles, and shapes are not stable or discrete entities but ones that unfold across both space and time and through a range of shifting relations. In this section, we ask what it would mean for geographers to base our research and analytic practices on such an understanding of care, moving care beyond simply being a topic of research or a standard for critique towards informing a disciplinary ethic. Here, we focus on practices of research and analysis, of how geographical data are collected and "read," not because we think these are the only domains in which geographers might think and act with care but because, as we will argue, the possibilities for taking up an ethic of care depends fundamentally on how we account for the world and our position within it. To unpack this fundamental role, and to outline a potential hermeneutics of care, we turn to work in feminist philosophy and in science and technology studies.
To produce an account of the world, Latour (2005) argues, is also to shape it, by suggesting a particular configuration of relations to which other actors will respond. Latour distinguishes between "matters of fact" and "matters of concern" (2004,2005). Matters of fact encompass those worldly affairs where things can be taken, more or less, for granted. Most of us trust that keys will unlock doors, that the traffic lights will change when they ought to, and that breast milk will nourish infants and not make them ill. Matters of concern, on the other hand, entail affairs where outcomes are less certain, and are subject to negotiation. Whether something is a matter of fact or a matter of concern, however, depends on perspective and circumstance. For traffic planners, the changing of lights is a matter of ongoing concern. Meanwhile, parents may find that matters of fact become matters of concern when their babies refuse to eat, develop reflux or, for whatever reason, mothers are unable to breastfeed. All accounts involve particular configurations of both facts and concerns. Callon and Latour (1981) argue that power often relies on the construction of matters of fact that are widely recognised and responded to as such. When matters of concern emerge, then, some actors may prioritise resolving these (back) into matters of fact. However, other attitudes towards matters of concern are also possible. For example, Puig de la Bellacasa (2017) imagines a certain disposition towards matters of concern, where they are treated as "matters of care." This entails a commitment to avoiding easy resolutions, in favour of attending to the wide range of interdependencies and entanglements that surround particular matters of concernincluding those with actors whose voices and interests are harder to discern within dominant accounts. Haraway (2016) has referred to this as "staying with the trouble." Doing so requires holding practices and understandings open. Parents, for example, continually face shifting and demanding dilemmas around sleep, feeding, changing, work, and exhaustion. These are often met with exigent solutions, ranging from disposable nappies to hired help for those who can afford it. To stay with the trouble is not to reject the need for such exigency, but to acknowledge these solutions are provisional, and to continue to search the grounds of daily life for other ways of attending to the wider entanglements of parenting.
Puig de la Bellacasa's call to treat matters of fact as matters of concern builds on a broader history of thinking on care, both within the material semiotic tradition in which she works, and in feminist philosophy. Mol (2008) contrasts the logic of care to the logic of control. The logic of control approaches the social world as if it is largely made up of matters of fact, where the same interpretations and ways of acting can achieve similar outcomes across varied contexts. The logic of care, however, attends to the situated particularity of life, and to active efforts to inhabit the world. For instance, with diabetes, seemingly straightforward advice given to diabetics to "control" their blood sugar, by injecting insulin or eating at given times, may be thwarted by workplace norms that treat insulin use on construction sites as a sign of weakness and fallibility, or that see eating during corporate meetings as showing a lack of control and leadership. The logic of care attends not only to the need for insulin or for sugar, but to the broader entanglements within which these needs emerge and within which they must be negotiated. Moreover, the logic of care recognises that these entanglements are likely to change overtime, and as such recognises care as involving "persistent tinkering in a world full of complex ambivalence and shifting tensions" (Mol et al., 2010, p. 14).
This continual unfolding resists the reduction of affairs to matters of fact. While the logic of control risks objectifying recipients of care, by treating them as fixed entities amenable to standardised interventions across time and space, the logic of care is oriented towards the continual and situated co-production of subjecthood (Samanani, 2017). As Arendt (2013Arendt ( [1958) suggests, this continual unfolding is marked by "natality" and "plurality," meaning that it is through these ongoing negotiations that new ethical possibilities emerge into the world, capable of making the world other than it already is. In turn, because the practice of care attends to the particularities of different contexts, and because in doing so it transforms subjectivities not just for carers but for care-givers as well, attuning them to particular, situated relationships, care necessarily entails ethical choice and commitment; one cannot care for everything at once, and the understandings and subjectivities fostered through different practices of care may prove incommensurable (Joks and Law, 2017). Raghuram (2016) traces the ways in which understandings of care have mostly been derived from practices in the "global North." In doing so, she argues that scholars of care must be cautious not to treat care as a generic ethic, based on experiences in the "global North" or otherwise, but as an emplaced and committed way of relating, complicit in the production of particular political worlds.
Underlying the "persistent tinkering" of care, argues Sevenhuijsen (1998), is a distinct mode of interpretation; care as a mode of ethical judgement is simultaneously a mode of hermeneutic judgement as well. As she puts it, "I thus see the practice of ethics as a narrative and textual practice. Social practices of moral deliberation can be interpreted as forms of 'storytelling,' in which signification, evaluation, and judgement are intertwined" (1998, p. 28, references omitted). This can be illustratively contrasted to what are sometimes referred to as "the hermeneutics of suspicion" or to "critical reading" (for a comparison of these framings, see Felski 2011)a way of reading social texts that starts with the assumption that such accounts are shot-through with the workings of power and repression, and that critical readings ought to work to uncover and trace such forces. Here, we can return to Latour ( , 2005, who highlights two prevalent issues with these "suspicious" approaches to reading and analysis: the first that they tend to rely on pre-given analytical categoriessuch as, for instance, gender, motherhood, or capitalismwhich, with their theoretical heft, tend to dominate explanatory narratives, making it difficult to discern the ongoing negotiations that may surround them. Second, he argues that critical approaches deconstruct particular accounts of the world that, however fraught, provide maps to understanding and action, without constructing new alternatives to replace them; for all the talk of constructivism, most work under its banner stops at deconstruction. We need to reconstruct, argues Latour, invoking Haraway, because it is only through the construction of accounts that we are able "to protect and to care" (2004, p. 232). These two critiques are intertwined: it is by suspending our most familiar categories of analysis, he suggests, that we create space for new constructive accounts to emerge, and it is through the emergence of such accounts that we might formulate constructive alternatives to the current order of things.
To attend to matters of concern as matters of care, then, keeps us resolutely within the territory where social possibilities are continually generated and reworked. Doing so reveals a world not of stable entities and categories, but one that is continually remade through ongoing ethical action across the terrain of everyday life. This ethic of care demands close and sustained attention to the ongoing production of meaning, and of the possibilities for acting alongside others, within particular situated contexts. Simultaneously, it demands that we look beyond this everyday terrain, to trace the wider entanglements that shape it, and to hold these entanglements in generative tension with everyday commitments.
When it comes to human geography research, then, an ethic of care provides a very different orientation to how we collect, read, and analyse the accounts provided to us as "data." While there are notable, self-consciously political exceptions such as Participatory Action Research (see Kindon et al., 2007) and feminist approaches (see Rose, 1993), human geography is conventionally dominated by an understanding of research where the process of gathering data is imagined as one where accounts are gleaned from participants and then subsequently analysed. As such, it is only later, at the stage of analysis when accounts are deemed to be "about" something, such as normative gender dynamics, the affective experience of discrimination, or the weight of austerity on everyday lives. Thus, even within many self-consciously activist traditions, it's only following this later analysis that the political stakes of accounts become apparent, and action becomes possible. These themes emerge as we sort and narrativise the data, combining moments into new mid-level accounts that, in turn, connect up with high-level theoretical explanations. For instance, to explain themes of normative gender expectations, we might turn to Foucault's theories of discourse and governmentality (cf. Foucault, 1991). Regardless of the approach taken, the accounts we are given are parsed, sifted, and then partly reconstructed as part of a broader argument. Rather than speaking for itself, data require accounting for. Moreover, it is only once accounts are deemed to be "about" something that ethical stakes then emerge, and ethical frameworks (be they "care" or otherwise) can be applied. Collectively, this process deploys what Mol (2008) describes as the logic of controlwith the exception that here it is not bodies but utterances (or other exchanges) that are treated as interchangeable, equally amenable to particular methods of analysis and to theoretical framing as all the other exchanges in the dataset. Across this process, exchanges are progressively translated into matters of fact, until, by the conclusion, the analyst can say, with confidence, what everything was really "about" (Law, 2004).

| ACCOUNTING FOR CARE
In contrast, approaching data with an ethic of care entails seeing data not as something to account for but as already part of an active process of accounting, where participants strive to describe, interrogate, and reimagine the world. Meaning in such accounts is neither fixed nor encompassing, but emergent and situated, involved in negotiating particular matters of fact and concern. In turn, this collapses the distinction in time and space between research and analysis, instead locating analysis as an inextricable dimension of the exchanges that make up research encounters. Here, the work of analysis emerges initially from actively engaging with, and taking a stake in, social worlds and in doing so emphasises the ongoing politically charged negotiations and labour that sustain the world.
This perspective need not entail jettisoning our critical or theoretical apparatus, but it does require approaching them from a different angle. On the one hand, a self-conscious ethic of care remains committed to including those voices that might not be prominent within given accounts and yet may nonetheless remain deeply implicated in them. For instance, if a parent asserts that they manage fine, with little outside help, and if we read this account as a political and ethical claim on the world, then we might also ask what sorts of interdependencies are being actively denied or re-framed through this utterance. And yet, staying with an ethic of care also suggests that these instances of exclusion or silencing do not override the original account, but complicate it. Rather than a critical stance that reduces social dynamics to matter-of-fact instances of exclusion, marginality, or inequality, a caring critique stays with the trouble, tracing the messy ways in which accounts are entangled and negotiated against one another, in a plural and ever-changing world.
To unpack and illustrate this argument, we turn to a series of research encounters, emerging from seven months of field research and interviews over the course of 2018. Here we do not approach parenting as a singularly caring domain. Rather, we approach parentingand first-time parenting in particularas a moment where many new and challenging matters of concern open up, inviting care as one possible response. The core of this research was comprised of in-depth, multi-part exchanges with 16 new parentstwo fathers and 14 mothers. 2 These exchanges involved elements including a semi-structured interview, collaborative mapping exercise, and the open-ended discussion of a daily parenting diary, which collectively focused on exploring the everyday experiences of new-parenthood. In addition to this, however, we made active efforts to be a part of the ecosystem of early-years services, relationships, and policy-making within Oxford. Farhan volunteered regularly at one of the city's busiest early-years drop-in sessions, while Jennie reconnected with friends, neighbours, networks, and volunteers she had met through her own experiences as a new parent in the city. Both of us visited and helped out with other services, took part in processes of urban policy design and review, interviewed service providers and policymakers, and spent time with parents in different everyday spaces. All participants were recruited through our involvement in these different places, events, and relationships. These wider experiences afforded us a situated perspective on parenting in Oxford at a time of austerity, which helped inform our exchanges with parents themselves.
For us it was first and foremost these exchanges themselves that pushed us towards understanding care as an everyday analytic practice. Interested in the impacts of austerity on how parents experienced the transition to parenthood, we approached these exchanges attempting to take care as a topic. Yet, while participants certainly provided plenty of commentary on parental care, there was something about the unfolding, provisional nature of these accounts, as well as the ways in which they implicated our presence as researchersin questions, moments of reflection, and discussionswhich suggested that the very work of constructing accounts may have been a part of the process we were interested in, in the first place. In what follows, we focus closely on three exchanges with three different research participants, to unpack three interrelated points: that everyday worlds are replete with matters of concern, that the speculative navigation of such matters of concern represents an essential part of caring labour, and, crucially, that this navigation inescapably implicates us as researchers, shaping different possibilities for knowledge and action within the research encounter itself.
The first of these exchanges takes place with Evie. Earlier in the interview, Evie had discussed a WhatsApp group 3 of local mums she belonged to, as well other parent groups, as a mixed blessing, offering both advice and reassurance but also feeling like a potential source of judgement and anxiety. Leading into this exchange, she puzzles over why some people voice such strong judgements and opinions around infants: Evie: People are really evangelical about so many things to do with, to do with babies and children. Researcher: Why do you think that is? Evie: I don't know, I think it's often the people who have more than one child who are way more easy going about most things, because they realise it's so much down to personality and the baby and stuff and also that if you don't do what's deemed, well what's thought of like, you know the best thing, how you make a difference. I don't know like -I don't know. I guess it's like you feel like suddenly you've got quite a lot of power because you make all the decisions for someone else. Researcher: 'I'm going to shape this.' Evie: And make themyeah. Do you want to come here? Yeah. And you feel like you're being kind ofyeah that's right. Come on stand up. Oh. And, I don't know, you think like, oh your baby will start to be known as the one who's -I don't know. I remember one of my friends once describing it as like performance anxiety with her son because she felt like whenever she would take him to like a social thing, he might be a bit cranky and then she felt likeoh dear. She felt like people wouldn't like him, or people would think that he was really problematic child and it was like, no he's just going through a bit of a kindoh no, would you like your drink? Bit of a kind of like tricky stage or something. So, I don't really suffer from that -I'll just get his drinkas much. As in like performance anxiety as much but he is on the whole fairly straight forward so, I don't know. I mean, yeah.
In describing parenting as holding "a lot of power" Evie acknowledges a class-inflected norm where parenting is a process of actively shaping your child through "making decisions." Likewise, she goes on to acknowledge the weight of social judgement, and the relation between this judgement and how children behave. Finally, although she starts off by expressing anxiety around such judgement, she ultimately shrugs it off by positioning her baby as well-behaved, and as meeting standards of judgement as "fairly straightforward." A gendered dimension is likewise implicit in the fact that she speaks only of herself, in the singular, and of other mothers, as involved in shaping children's development. Holloway (1998;Holloway and Pimlott-Wilson, 2014) argues that different approaches to childcare play a key role in reproducing class distinctions. In her study of two Sheffield neighbourhoods, middle-class mothers would performatively position both themselves and their young children as embodying norms of autonomy, individual capability, and socially recognisable merit. When children's behaviour became more challenging, it was quickly recast as anomalous or even indicative of potential, and often used to make a case for additional support. In contrast, working-class mothers were more likely to focus on personality over development, and to treat challenging behaviour as an inevitable part of childhood. Since 2010, a series of heavy budget cuts in the UK have led to a significant withdrawal of early-years services. In this context, scholars have highlighted how the middle-class ideal of the independent parent(s), producing responsible, enterprising, and "capable" children has become pervasive in both policy and popular discourse as a governmental ideal against which parents are judged, supported, and shaped (de Benedictis, 2012;Holloway and Pimlott-Wilson, 2014;Jensen, 2012;Jensen and Tyler, 2012;Jupp, 2013;McDowell et al., 2006;Vincent et al., 2010). Evie's account can certainly be read critically in these terms. However, while class and gender dynamics are clearly at play within Evie's account, to declare that this account is necessarily "about" these familiar themes can serve to occlude other interpretations. In particular, it can obscure how, for Evie, class and gender are not pregiven or fixed commitments but dimensions of a broader experience she is working to navigate, uncertainly and experimentally.
Beyond human geography there are now well-established traditions of acknowledging the interactive nature of research encounters, and adopting bottom-up approaches in the analysis of everyday accounts, yet there is a notable lack of engagement with such work in the discipline. For instance, in anthropology, the "writing culture" turn moved to emphasise the "poetics" of everyday life and experience, as something that would always exceed the grasp of analytical categories (e.g., Abu-Lughod 1991;Clifford and Marcus, 2010;Marcus, 2007). In social psychology, discourse analysis emphasises the variability of interviewee talk and the importance of understanding interviews as interactions. For example, Potter and Hepburn highlight how during interviews people make available issues of stake and interest where "they may respond to what others say as based on particular interests, and they may manage issues of interest in their own talk" (2005, p. 15). Likewise, sociologists Holstein and Gubrium (2004) use the concept of the "active interview" to draw attention to not only what meaning is produced but also to how its production occurred. They advocate methods that are "sensitive to both the real in everyday life and the representational practices of participants" (2004, p. 16), recognising both interviewer and subject as involved in producing particular narratives. These traditions point to Miller and Glassner's contention that "dominant discourses are totalising only for those who view them as such; they are replete with fissures and uncolonised spaces within which people engage in highly satisfying and even resistant practices of knowledge making" (2004, p. 126). Approached this way, what is striking about this account is the ways in which Evie enrols the voices of those around her, as well as her own infant himself in trying to navigate the embodied experience of parental anxiety evoked by others. Whether on WhatsApp, or out on the streets, she clearly has a sense that others are prone to expressing strong judgements around parenting. What remains a question for her is how she ought to relate to these judgements.
In considering Evie's account as an unfolding negotiation, we might start by noting how she begins by challenging the gravity and validity of these judgements, by enrolling an image of second-time parents "who are way more easy going" and self-assured. And yet a sense of anxiety persists as she emphasises her uncertainty ("I don't know") and speculates on the impact her own actions will have on her child. Exploring this sense of anxiety, she starts in the second person, connecting her anxieties to those of others. Here, she sympathetically cites her friend's idea of "performance anxiety" to imply her own feelings when encountering the judgement of other parents. Invoking this anxiety in the voice of her friend, however, allows her to simultaneously distance herself from this anxiety, as she shifts into first person, and asserts "I don't really suffer from that"giving this assertion weight by reflecting that her baby is "fairly straightforward." Even so, she ends on a note of scepticism, declaring once again "I don't know." Analysed this way, Evie's account is less about the straightforward reproduction of classed or gendered identitiesthough these identities remain at stakeand more about navigating a range of public expectations, as well as the embodied experience of being judged, as active matters of concern whose political implications remain unresolved, in relation to her own experiences as a parent. This navigating is an ongoing process that even in the space of a few sentences can move back and forth between anxiety and assurance, or affiliation and disavowal.
The more we spoke with parents, the more we began to sense that this process of figuring-out marked by indeterminacy and ambivalence was an act of care in and of itself. As Miller (2005Miller ( , 2010Miller ( , 2017 has illustrated, first-time parenthood is a time of marked transition and uncertainty. Not only do familiar social norms and categories often cease to resonate, but everyday relations can become fraught with uncertainty as babies grow and respond in new and unpredictable ways, and as other relations shift around them over time. An important theme in Miller's work, however, is the efforts parents put in to re-inhabit or re-establish familiar norms and understandings. Her interviews are often spaced out over many months, and these wider temporal horizons play a role in how parenting is discussed. By contrast, we suggest that when focusing in on the day-to-day experience of parenting, parents not only face sustained uncertainty but actively inhabit it. Whether out of necessity or out of choice, parents' relationships with their children were often ongoing matters of concern, something where easy resolutions were not forthcoming, forcing parents to "stay with the trouble" (Haraway, 2016). Hansa illustrates this experience through her struggle to work out when to return to work, and what sort of care to arrange: Hansa: Um, so we're just … I'm in the process of just figuring that kind of stuff out at the moment and experimenting with different things … Researcher (looking at her daughter, who is both screeching and smiling): Um, ha. Hansa: um, I think it helps to know if Sita has a pattern in the day, so there was a period where she was having two or three hour naps in the afternoon and that was a huge chunk of time where I could just [baby makes noises] focus on my research … Researcher: Mm Hansa: and my partner doesn't expect for me to be putting the laundry out and doing things around the house during that time … Researcher: Right. Hansa: … during that time, he expects me to be doing my research and then we share those tasks when he comes home from his work, so that helps … Researcher: Hmm. (Towards her daughter, who is laughing as she picks up and presents various toys) Um ha.
Hansa: (watching her daughter): … um … Researcher: Ha! Hansa: … and I think that that was good but when it changes, when things change and that's not how she's sleeping anymore … Researcher: Right. Yeah. Hansa: … then there's a period of transition and figuring out what the new routine is and there have been weeks where it just hasn't worked and there's been no routine and I think that's been very challenging … Researcher: Mm. Hansa: … and that's when I've started to explore other options of childminders and things (Her daughter squeals loudly) Researcher: Hmm? Hansa: … but as I'm exploring those things she'll settle into a new routine and I'll get used to that um … (turning to her daughter who is handing her a toy and squealing) Thank you Sita. So, I think one of the things that I've had to learn is just patience that even in the moments where it feels impossible to juggle all those things to just wait a couple of days and see what the new pattern is and know that it will emerge and there will be another way around these things. Researcher: Okay. Hansa: I think that it is getting to a point where I need some extra support with childminding or nursery or something and that's going to be very challenging because of just how attached she is, and I took her in for a settling in session where she was there for an hour without me yesterday and she cried the whole time … Researcher: Oh! Hansa: … so, yeah, that will be interesting Researcher: Hmm. Hansa: … and I think for me raises questions as well about just what care-giving looks like … In Hansa's exploratory back-and-forth, she makes a series of discursive moves that serve to resist a single definitive account. Throughout the extract, care emerges as less about pre-defined or pre-valued practices, and more about "exploring" and continual "figuring out"or what Mol et al. (2010) would call "constant tinkering" in a shifting world. While here too it is certainly possible to produce an interpretation that critically foregrounds the operation of power, in gendered, classed, or other terms, it is also possible to understand this account as actively negotiating a situated set of concerns, in which gender and class are implicated but not determinate. For example, Hansa states how her partner "expects" her to prioritise her own research. In doing so, she enrols her partner's voice to authorise their division of time and roles. This sense of "being authorised" is used as a narrative device in presenting/managing/inhabiting specific, normative, roles. In other words, Hansa's account is marked by her own reflexive attempt to navigate "between the lines," as a form of caring labour, which might open up new political possibilitiesnew ways of valuing employment, doing gender and family, or valuing care labour.
However, if the weaving of accounts forms an essential part of caring labour, then this quickly entangles others addressed within such accountsincluding researchers themselveswithin the daily politics of care. Beyond parents and children, a range of other actors were evoked in the speculative accounts parents shared. Like Evie's friend, these other actors offered certain 'affordances' (Keane, 2016), through their presence, attention, replies, actions, and commitments (caring or otherwise) against which parents tried, experimentally, to situate themselves in different ways. The affordances of such others may help in the production of a caring account or may resist it. This holds not only within longstanding relationships, but also within the research encounter itself, as a final exchange with Neha illustrates.
In the interview, she and Farhan are discussing the diary she has kept. The interview starts out upbeat, warm, and humorous, marked by Neha's affection for her son. However, as she and Farhan go through the diary, she begins to sound less positive. Her son has gone through a patch of poor sleeping and this, coupled with her ongoing struggle with anxiety, has left her feeling drained, judged by other parents, on edge, and struggling to leave the house. In the interview, she speculates nervously as to whether this isolation has been harmful for her son's development, and for her as a parent. As the interview continues, this anxiousness begins to dominate the interview as she repeatedly questions whether she is a good parent, or whether she is coping at all. Farhan becomes increasingly concerned for Neha and that his line of questioning might be exacerbating this anxiety. In what follows, the two of them collaboratively steer the research encounter back to its more light hearted and upbeat beginnings: Neha: Yeah the last few days have been a bit miserable in the old diary. Researcher: So yeah, I guess the perspective sort of changes how you see the day then, what counts as getting out or what counts as achieving something or … Neha: Yeah absolutelywhenyeah you're right 'cos there is achieving all day long you know. He's learning stuff so I'm giving him that space to like learn how to pull himself up like if I was constantly out I don't know walking or playgroups and plonking him down and not letting him do his stuff than maybe you knowmaybe this is what we need to do this this last few weeks. Researcher: This is Mason [her son] learning. Neha: Yeah exactly you know maybe this is right but yet I'm putting so much pressure on myself to do it differently and more better -'more better?'better. Researcher: 'More better?' We can do 'more better.' Neha: Yeah. Create a new language.
It is a well-established argument that there is no "neutral" way to conduct a research encounter (Silverman, 2004). However we present ourselves, and however we engage with our interlocutors, we become actors within the accounts they present back to us. As Holstein and Gubrium (2004) argue, this means that the interview is necessarily an "active" encounter, involving the co-production of accounts and experience. Richardson (1990) distinguishes between two particularly important narratives that can emerge in interview encounters: the "cultural" story and the "collective" story. Cultural stories are constructed with close reference to dominant norms, while collective stories unpick or play with established norms in order to allow new voices to emerge. In this instance, Farhan became increasingly aware that to simply question Neha about her diary, while leaving her to interpret and respond to the material in it alone, was working to elicit something like a cultural story, where her increasingly anxious narration emerged against an imagined ideal of good parenting. In contrast, although the collective registerthe voice that resisted this anxiety with affection and humourwas much less present in the interview, Farhan made attempts to open opportunities for the two of them, together, to cultivate this register with care, and bring it to the fore. This care is first offered by Farhan tentatively ("I guess") but when it elicits a positive response and further elaboration from Neha herself, he responds more definitively ("This is Mason learning"), committing to this alternative account, as best as he is able.
The fact that these anxieties could be resisted, if only momentarily, through the shared construction of a different sort of account, highlights how researchers are inescapably implicated in the production of their interlocutors' accounts and worlds. Our process of research was filled with such moments, where the joint construction of accounts opened up new, intertwined possibilities for action and knowledge. There was Lisa, who contrasted the "really low" feelings that came from struggling to breastfeed, and being chided by nurses for her baby's low weight, to the care and affirmation she received from a local breastfeeding support service, which left her feeling like "a super woman mum." The story she was telling was a linear one of dejection and anxiety being replaced with a sense of empowermentbut then something caught, and she started to cry. Negotiating this moment happened not so much out loud, but physicallya short pause, a warm smile, and a quiet "I get it." Sitting together for a moment and then her getting up to refill her water bottle. All this opened up a space where we could double back and dwell on how both care and anxiety were felt and lingered as embodied qualities, often overlapping, often persisting despite attempts to insist otherwise. Or else there were the moments at drop-in play groups where Farhan met with quiet, guarded parents, uninterested in being interviewed, whose children clung shyly nearby. Here, joint acts of careteaching a child to play the xylophone, or reading a story togetherbecame the basis not only for parents opening up, but for sharing particular concerns around the struggles of connection, both with children and with other parents. Or, again, there was our participation in a municipally funded workshop, discussing the funding of early-years services in the wake of austerity, where our reflections on the sense of isolation parents voiced fed into new discussions about how to make spaces accessible and welcoming in a deeply unequal city. In all these instances, gestures of carespoken or enacted shift the narratives being assembled in meaningful and revealing ways.
Although dominant traditions of "reading between the lines" insist on attending to nuance and indeterminacy, this attention is typically imagined as an observational and textual practiceas a matter of attending to and attempting to evoke the world as it is. As we have argued elsewhere (Lenhard & Samanani, 2019) for anthropology, and as methodological texts (e.g., Silverman, 2004) suggest is likewise the case for social psychology and sociology, such traditions may advocate participation as a mode of observation, but they resist forms of participation that are seen as exerting influence or acting in a transformative manner on research subjects or their worlds. In contrast, approaching research within an ethic of care recognises not only that all knowledge is situated, but that knowing is an inescapably entangled process of attention, analysis, (inter)action, and commitment, 4 with live political stakes, where different approaches to this process will produce different forms of knowledge and political outcomes. As such, to postpone analysis until after the research encounter is to forgo participation in the analytic, world-making labours of our interlocutors, and in doing so is to simultaneously foreclose and underappreciate such efforts.
Returning to the exchange between Neha and Farhan, however, what is apparent is that the fragility of the alternative account they construct together is related to a set of broader entanglementswith mental health services, parenting groups and other parents themselvesalso implicated in how Neha accounts for and practices care. While these entanglements may have helped impose a 'cultural' story onto Neha's account, it also mattered that Farhan occupied a particular position within this context himself (through volunteering, etc.); this enabled him to respond to Neha's anxieties about her own struggles, other parents' attitudes, or the inclusiveness of available services from a situated position of existing commitment and recognition. In shaping the responses he was able to provide, this situated position opened up new forms of knowledge, eliciting not only Neha's anxieties but her optimism and determination as well.

| CONCLUSION: TOWARDS A CARING GEOGRAPHY
Recognising the broader entanglements within which care is located, many feminist scholars have made a point of situating the ethics of care alongside other ethical stances, such as a liberal commitment to freedom and equality (Fraser, 1997;Fraser & Honneth, 2003;Tronto, 1993; but see Kittay, 1999;Sevenhuijsen, 1998) because they recognise that care is laborious, consuming, and ultimately finite, that the ethical potential of care is inescapably entangled in a wider range of relations, and that care and carers in turn need forms of support beyond the caring relation. In this paper, we have argued for a hermeneutics of carea practice of analysis that not only attunes itself to the everyday practices of care that pervade the world, but that, in doing so, reimagines the relationship between observation, analysis, commitment, and responsiveness as a single, entangled process within research encounters. In reviewing existing geographical approaches to care, we have explored how geographers have traced and theorised the entangled nature of care, but have also highlighted how much of this work has conceived care as a topic of study or a critical standard, rather than taking care up as an ethic for making geographical research itself more entangled. We have argued that approaching care as an ethic for research entails staying with the trouble within matters of concern, resisting reducing accounts to familiar "matters of fact" and instead approaching analysis as a situated, inter-subjective form of attention, commitment, and responsiveness. We have argued that such a hermeneutics is likely to make forms of knowledge and action available to human geographers that would otherwise remain invisible and under recognised. All practices of reading produce their own forms of knowledge and potentials for action, and so, alongside Tronto, Fraser, and others we recognise that care should not necessarily be the only ethical paradigm through which we approach the world. Nonetheless, we want to conclude by highlighting what the broader adoption of a hermeneutics of care might imply for human geography.
Attending to accounts carefully matters deeply. While many may see academic labour as the work of unpicking accounts, getting at their underlying truth, and connecting these with larger ideas that bear social relevance, a care-ful perspective acknowledges that much of this labour is already being done all around us. Seeing the world as actively in the process of being made and remade through this everyday labour offers a very different perspective on how we trace broader trends, relations, and patterns, and how we assess which stories matter. It begs the question of why we, as researchers, so often see normative stories or prevailing patterns of power as the "point" of accounts, when accounts are filled with so much elseincluding, often, resistance to or reworkings of these dominant formations. In turn, the question of how we determine what within accounts countswhat is signal and what is noisecomes into focus as a political question, rather than one of accessing an "objective" truth, grounded in the question of what connections we are committed to tracing and reproducing (Strathern, 2004).
These choices feed into a broader politics. Many human geographers are deeply committed to transformative political projects of various guises. It matters, however, where we locate these transformative projects, and how we envision transformation in relation to the worlds in which we live. Feminists have long highlighted the devaluation of the labour done by a myriad of actors to maintain, sustain, and reproduce the world. If we acknowledge that everyday accounts are full of caring labourof ways of navigating the challenges of the world that ultimately reproduce or transform it, on the basis of active, conscious effortthen what does it mean if we only locate our politics above and beyond such everyday efforts? If politically active geographers feel their primary task is to identify and theorise systems of dominance, by pulling traces of these systems out of everyday accounts that actively do so much more than simply, directly reproduce such systems, then where do we locate the possibilities for resistance or transformation? Alternatively, if, following thinkers such as Gibson-Graham (1996, we think of the everyday as a site where alternative values and political configurations are already being realised, how might we keep these possibilities in focus and build a new, broader politics out of them? Arguing in favour of the latter vision of politics, and drawing on Stengers' Cosmopolitics (2010, 2011), Haraway warns: "we cannot denounce the world in the name of an ideal world." Rather, "decisions must take place somehow in the presence of those who will bear their consequences" (2016, p. 12). A hermeneutics of care allows us to stay with the trouble politically, not only in attending to the fraught and complex ways in which dominant orders are navigated and negotiated in everyday life, but in valuing the richness and possibility of the everyday as well, as the basis of a different sort of politics.

| EPILOGUE: CARE IN THE TIMES OF COVID-19
Coronavirus is a health crisis of unprecedented proportions. To date, the COVID-19 pandemic has claimed over 700,000 lives across the globe, with this figure continuing to rise. When British Prime Minister Boris Johnson and his Health Secretary Matt Hancock contracted COVID-19, Michael Gove, a Senior Cabinet Minister, led a press conference on 27 March 2020 claiming that the virus "does not discriminate." A few weeks later the Director General of the WHO, Dr Tedros Adhanom Ghebreyesus, echoed similar sentiments. However, as Rebecca Solnit points out: "Coronavirus does discriminate, because that's what humans do … people who face racism, sexism and inequality are more likely to get sick. Taking care of each other starts with understanding this" (2020, n.p.). The impact of this virus has been markedly uneven. Diverse intersectional inequalities have been both exposed and exploited by the pandemic. Since writing this paper, we have both witnessed and, at times, become entangled with such inequalities. For example, the irony is not lost on us that one of us attempts to write these very words as a female researcher on a fixed-term contract in the midst of HE job cuts and recruitment freezes while being a single parent caring for two young children at home. Meanwhile, we have interlocutors, who have become friends and collaborators in the care and community sectors, who are struggling with reduced funding and virtualised services, with the effects compounding inequalities. In certain moments we have joined in with efforts to navigate these pressures. Yet what are the broader implications of considering accounts of care in the wake of COVID-19?
As we have spoken with, listened to, and engaged with friends, neighbours, colleagues, and others we encounter in our everyday lives, such engagements have served as a further reminder of the need to examine "social worlds" as matters of care with everyday understandings being continually negotiated whereby uncertainty and processes of figuring things out are acts of care in, and of, themselves. As the inevitable, yet incredibly necessary and important, onslaught of research considering the broader socio-economic impacts of COVID-19 begins, there is a risk that we approach care as a quantifiable goodas a resource that certain communities or demographics can possess more or less of, and where justice is simply a matter of distribution. The radical, often-far-reaching, uncertainties of the present moment demand more than this. In this paper we have argued for a vision of care as "an entangled process of attention, analysis, (inter)action, and commitment" that not only attends to pre-identified needs but opens up new possibilities for knowledge and action. For those tracing this transformed, anxious, hurting world, we urge for notions of care to be adopted that do not presume it knows what the concernand so the solutionis. Instead everyday accounting of matters of concern that reach beyond neatly defined issues need to be taken more seriously and with it, the everyday unfolding of possibilities, open to collaborative nurturing, that might answer back to the far flung uncertainties of this moment. ENDNOTES 1 Bowlby herself distinguishes between these two similarly named terms by noting that caringscapes refers to "individuals' organisation of their caring activities in time-space," while carescapes refers to the configuration of "available resources and services which themselves alter in response to shifts in the economy, government policy, and social and technical change" (2011, p. 2,112). Nonetheless, these two concepts are clearly related, and Bowlby frequently refers to them as suchdeploying the formulation "caringscapes/carescapes." The summary presented here is of the overall picture produced by bringing these two analytical concepts together. 2 Our focus here is more on the interpersonal dynamics of particular research encounters than on the group of parents as a whole. 3 A social messaging app oftenas in this caseused for group conversations. 4 Which we would map onto Tronto's four elements of careattention, competence, responsiveness, and responsibilityrespectively.