Reconceptualising play: Balancing childcare, extra-curricular activities and free play in contemporary childhoods

This paper rethinks geographical approaches to children ’ s play in the Global North. One narrowly conceived strand of past research considered the erosion of outdoor free play, but overlooked children ’ s views of alternative play environments (e.g., out-of-school activities). A separate thread examined the feminisation of employment and growth in childcare, but investigations into children ’ s play in care-based environments are rare, and explorations of their ability to balance different play landscapes are lacking. This paper builds on insights from both lines of research, and challenges their deficiencies, to map out a more broadly conceived geography of play. Drawing on rigorous quantitative and qualitative research with middle- and working-class parents and children, the paper examines the socially differentiated reconciliation of neighbourhood, extra-curricular and care-based play in contemporary British childhoods. The analysis highlights stark class differences in the balance between these play environments. Middle-class children ’ s elective engagement in extra-curricular activities results in reduced outdoor free play, but they have little control over the total time they play in care environments. Working-class children have greater experience of playing out, but are underscheduled as they have less than desired access to play in structured activities and wraparound care. In conclusion, the paper argues that geographers must: (1) move beyond the romanticisation of free play and recognise children ’ s right to participate in a diversity of playful spaces; (2) embrace an intergenerational approach which recognises that play is not a matter for children alone – adults can facilitate as well as limit play, but reliance on women ’ s unpaid labour makes play a feminist issue; and (3) uncover play ’ of work, to forge a new agenda for research on a more broadly conceived geography of children ’ s play. From the literature on play, we take an enduring focus on the eroding geographies of outdoor free play but also recognise the need to address vital lacunae in our understanding of children ’ s experiences of play in structured activities. This is augmented by our reading of the feminist literature on social reproduction, which highlights omissions in our knowledge about children ’ s play in care environments, and reveals the absence of debate about the accommodation between different aspects of play within children ’ s lives. In examining the balance between children ’ s play in different environments we are mindful of both the need to foreground children ’ s right to play and the necessity that we consider how play is implicated in social reproduction. The aim of the paper is therefore to trans-form geographical understandings of children ’ s play through an exploration of the class-differentiated reconciliation of childcare-based, extra-curricular and free play in contemporary British childhoods.


| INTRODUCTION
Geographies of Children, Youth and Families (GCYF) emerged as a vibrant field of research in the latter parts of the 20th century, and have since undergone a period of rapid growth and institutionalisation within the architecture of the discipline (Holt, 2011). This paper draws together two early and enduring concerns within the field: the changing landscapes of children's play and shifts in the terrain of family life associated with the feminisation of the labour force and the growth of childcare provision. Previous research into the changing complexion of play has emphasised a decline in children's independent use of outdoor space in the context of fears about traffic and stranger danger (Alparone & Pacilli, 2012;Holt, Lee, Millar, & Spence, 2015;Valentine, 1997). This literature has quite rightly concentrated attention on those freedoms which have been lost to children, but it has failed to give commensurate consideration to young people's attitudes to newly emerging play landscapes, such as organised clubs and activities (Karsten, 2005). Another, largely separate, strand of literature has examined women's increased labour market participation and the development of non-familial childcare which has made this socio-economic transformation possible (Busch, 2013;Gallagher, 2014). This literature is largely adult-centric, for example considering parents' work-life balance strategies (Holloway & Pimlott-Wilson, 2016;Ward et al., 2010), with little consideration given to how these societal shifts impact on children's everyday lives (Harden, Backett-Milburn, Maclean, Cunningham-Burley, & Jamieson, 2013;Smith & Barker, 2004). This paper builds on insights from these two separate strands of research, and challenges their individual deficiencies, through an integrated examination of the combined impact of these processes on the spatialities of children's lives. Specifically, the paper explores how these two trends are reshaping the topographies of contemporary British childhoods, and traces the importance of social class in shaping the reconciliation of: free play; organised clubs and activities; and play in childcare environments in children's everyday lives.
The paper begins below with a review of the literature on the changing geographies of play and family life, before the aims are defined and the methodology outlined. The central sections of the paper examine different children's attitudes to play in diverse environments and their views on an appropriate balance between these. In conclusion, the paper moves beyond loss-oriented understandings of the nature of children's play, and adult-centric understandings of work-life balance, to produce a rounded conceptualisation of play that affords researchers the analytical space to scrutinise the interlinked impacts of changing social, economic and political forces on the lived experience of present-day childhoods.

| GEOGRAPHIES OF PLAY AND SOCIAL REPRODUCTION
Play in the English language means to "[e]ngage in activity for enjoyment and recreation rather than a serious or practical purpose" or to "[t]ake part in (a sport)" (Oxford, 2017). The first of these societal emphases on pleasurable participation is that most commonly reflected in social science definitions which cast children's play as: "enjoyable"; having "no extrinsic goals"; "spontaneous and voluntary"; and involving "active engagement" (Garvey, 1990, p. 4). In Geography, this broad-ranging notion of children's play is implicitly spatialised through the dominance of research on outdoor free play. This enduring tradition (Hart, 1979) gained fresh impetus at the turn of the 21st century as concern grew about the eroding geographies of children's street play in the Global North. Valentine's (1996aValentine's ( , 1996b research was conceptually important in highlighting discursive shifts which led to children being labelled as out of place in public space, whether that is as "little angels" at risk from stranger and traffic danger, or as "little devils" who might disrupt adult order. Numerous academics have amplified these arguments, demonstrating that children now face greater restrictions on their independent ability to play, hang out and travel through public space than in previous generations (Holt et al., 2015;Loebach & Gilliland, 2016;Witten, Kearns, Carroll, Asiasiga, & Tava'e, 2013). The corrosion of children's capacity to access outdoor play opportunities is somewhat differentiated by social group, with middle-class, female, younger and minority ethnic children likely to face greater restrictions (Karsten, 2005(Karsten, , 2015Mackett, 2013;Tucker & Matthews, 2001) in a context where more parents are chaperoning or chauffeuring their children through the urban environment (Carver, Timperio, & Crawford, 2013;Schwanen, 2007).
The societal tendency to idealise outdoor play as naturally beneficial for children (Carrington, 2016;Palmer, 2015;see Verstrate & Karsten, 2015) has echoes in researchers' insistence that this decline of outdoor free play must be challenged because the activity and environment cultivates positive social, mental and physical development in a childhood world beyond adult intervention (Gray, 2011;Loebach & Gilliland, 2016;Skar, Gundersen, & O'Brien, 2016). Most geographers regard precise measurement of play's developmental impacts as beyond their disciplinary turf (Holloway, 2014), but innovative interdisciplinary work with public health specialists is tracing the physical benefits of calorie-burning, outdoor free play to children's physical well-being in the context of rising childhood obesity (Mackett, 2013;Oliver et al., 2016). What is noteworthy in a discipline long shaped by the public/private dichotomy (Women and Geography Study Group, 1997) is that children's reduced independent access to public space has not prompted a commensurate wealth of studies on the reverse side of this coin, the rise of domestic play environments. These opportunitieswhether that be an explosion of trampolines in family gardens or rapid developments in indoor gaming and media cultures (Crowe & Bradford, 2006;Thomson & Philo, 2004;Witten et al., 2013) have received less attention by geographers than they warrant.
Outdoor free play, or cocooning in an inevitably permeable domestic sphere, are not the only alternatives, however, when we think about children's play. Historical research points to the provision of "rational" or "morally" uplifting forms of play provided through philanthropic or religious organisations. Gagen (2000Gagen ( , 2004, for example, highlights how the playground movement in the USA sought to acculturate urban, immigrant children into appropriately gendered American identities through the provision of social and physical play opportunities. Similarly, Mills (2013Mills ( , 2015 has examined the historical aims and practices of the Scouts and The Jewish Lads Brigade and Club as they provided structured play and social opportunities designed to aid young people's development into worthy future citizen-subjects. Uniformed organisations such as Rainbows/Brownies/Guides and Beavers/Cubs/Scouts continue to offer young people structured play environments outside their own home, and these now sit within a mixed economy of state, private-and third-sector provided extra-curricular activities for children encompassing sporting and diverse cultural opportunities (Holloway & Pimlott-Wilson, 2014).
In wider society, these "rational" forms of play are increasingly seen to be good for young people as they offer them the opportunity to develop physical and social skills while also contributing, albeit indirectly, to increased educational attainment (Bradley & Conway, 2016). Their rapid growth in recent years reflects both state support for the provision or signposting of such activities through schools (Cummings, Dyson, & Todd, 2011) and contemporary intensive parenting cultures in which middle-class mothers and fathers seek to ensure that their children have enjoyable and productive childhoods during which they develop social and cultural capital that will be valuable in adult life (Karsten, 2015;Lareau, 2002;Vincent & Ball, 2007). In Geography, Katz (2008, p. 11) argues in broad-brush terms that this leads to the phenomena of the "overscheduled child", as parents concerned about political-economic futures in the Global North cram children's lives with extra-curricular activities designed to bolster their future success. Research with parents suggests her argument has some merit, although it also highlights significant emphasis on children's current safety and happiness, alongside concern for their future social and economic well-being (Holloway & Pimlott-Wilson, 2014). What is largely missing from this picture, however, is an analysis of what this proliferating form of play means to children (Karsten, 2005). This is a serious lacuna which this paper will address. This emphasis on clubs and activities stretches geographical conceptions of play in the Global North, and broadens the spatial lens from streets and parks to the institutionalised public spaces, such as community halls, leisure centres and playing fields, where these take place. This expanded vision of what count as "playful spaces" is vital (Thomson & Philo, 2004, p. 111), but we not only need to incorporate the diverse spaces of children's play into our research, we must also be mindful of why such playful spaces matter. Geographers have tended to reflect on this issue in one of two ways, both of which inform our agenda in this paper.
On the one hand, geographers argue that these spaces matter as children have a right to play. There is an unmarked assumption in the discipline that play is a natural and essential component of childhood development (Harker, 2005). Global policy initiatives such as the United Nations Convention on the Rights of the Child (UNCRC) (United Nations, 1989), and the turn-of-the-century paradigm shift that saw children being conceived of as social beings rather than human becomings, have resulted in play being explicitly framed in GCYF as a child's right (Holloway, 2014;Matthews & Limb, 1999). However, geographers' emphasis on children's right to play has not emerged evenly across research into different play environments. Significantly, it is strongly apparent in research on children's free outdoor play, where geographers are attuned to the philosophical gold standard of playwork that play should be free of charge, freely chosen by children, and participants should be free to leave (Brown & Patte, 2013). By contrast, it is absent from research on extra-curricular activities, as geographers valorise free play over structured activities (Aitken, 2001;Katz, 2008), despite the fact that both forms of play are defined as a child's right by Article 31 of the UNCRC (United Nations, 1989). We challenge this paternalist bias towards free play in the discipline, and strategically conceptualise the child as a bearer of rights in this paper (Holloway, 2014), in order to explore different children's desire, and ability, to exercise their entitlement to access varied forms of play.
On the other hand, playful spaces have been conceived of as sites of social reproduction. Aitken and Herman (1997, p. 63), in a theoretical account inspired by Winnicott and Flax, identify play's progressive potential as a space in which "liberatory notions of justice and difference may develop". However, in accounts that combine theoretical engagement with empirical research, Katz (2004), drawing on Vygotsky and Benjamin, and Harker (2005), on Massumi and Deleuze, both argue that while play is important in the emergence of child subjectivities, the potential this offers for progressive social change is rarely realised. In Katz's (2004, p. 101) words, play in her study produces "more of a tendency toward stability than toward its undoing". These theorisations have yet to be fully explored through empirical research into primary-schoolaged children's outdoor free play or structured activities in the Global North; nevertheless, they are important in shaping the agenda in this paper. In framing child subjectivities as contingent constructions, these authors move beyond the liberal subject of rights-based approaches to play and emphasise instead the importance of play as a sphere through which (usually unequal) regimes of power can be reproduced. There is undoubtedly potential to adopt this perspective at the level of the individual street game or structured activity. Here, however, we scale up the approach to consider social reproduction across play environments, and mostly notably the ways classed power emerges in, and is reproduced through, children's participation in diverse landscapes of play.
The decline in children's outdoor free play, and its partial replacement by organised play environments, is, however, only part of the story of changes in the nature of children's play. The feminisation of the labour force since 1970 means a growing proportion of women now work through their childbearing years, and their employment rates climb once children reach school age (Office for National Statistics, 2013). This increase in the numbers of women employed has raised a variety of issues for geographers and other social scientists. First, it has prompted concern that the emancipation of middleclass women in the workforce is being bought at the expense of local working-class, or indeed migrant, women who are employed on low rates of pay in the childcare industry (Busch, 2013;Pratt, 2003). Second, in a context where the feminist literature points to women's double shift in paid and unpaid work (Hochschild & Machung, 2003), attention is being given to the ways parents, and most notably women, manage their work-life balance (Jain, Line, & Lyons, 2011;Teo, 2016). The question that remains neglected in Geography is what these shifts in the nature of family life mean for children (Harden et al., 2013;Pimlott-Wilson, 2012), and crucially, in respect to this paper, what this means for their geographies of play (Witten et al., 2013).
The consequences that women's enrolment in the labour force alongside men has for children's play depends on the type of childcare parents choose to use. Much early research on childcare focused on care given by individual women (Gregson & Lowe, 1994), but work on collective care environments became increasingly common as political shifts in society have led to nurseries being viewed as the proper place for young children's education (Gallagher, 2013). In the case of primary-school-aged children, after school clubs are increasingly important places in the play landscape (Smith & Barker, 2004). Their growth in the UK has been driven both by parental demand and the characteristic roll out neoliberalism of a workfare state (Training and Development Agency for Schools, 2007) that seeks to promote parental employment as the route to economic security and social inclusion. Our research has explored their differential value to social reproduction in middle-and working-class families (Holloway & Pimlott-Wilson, 2016), but it is vital that we also ask what the growth of these spaces means for the changing nature of children's play. The paucity of attention given to children's play in these environments reflects geographers' casting of children as the objects, rather than the subjects, of care in debates about working parenthood and social reproduction. This is a partiality that the paper seeks to redress.
It is not sufficient, however, simply to respond to this body of work by adding another play environment to our list for investigation. The literature on the feminisation of the labour force has forged understanding of adults' work-life balance (including research that explores children's understanding of this [Pimlott-Wilson, 2012). Recognition that adults must reconcile different aspects of their lives should prompt us to question how children accommodate contrasting parts of their everyday worlds. Depending on their family circumstances, children may have access to childcare-based, extra-curricular and/or outdoor free play opportunities, and it is not only their experiences in these environments but their ability to balance their time between them that matters. This paper will therefore explore both children's experiences of play in care environments and their abilities to reconcile diverse play landscapes in daily life.
In sum, this paper draws on an innovative combination of insights from literatures on children's play and social reproduction, and an appreciation of the deficiencies within these bodies of work, to forge a new agenda for research on a more broadly conceived geography of children's play. From the literature on play, we take an enduring focus on the eroding geographies of outdoor free play but also recognise the need to address vital lacunae in our understanding of children's experiences of play in structured activities. This is augmented by our reading of the feminist literature on social reproduction, which highlights omissions in our knowledge about children's play in care environments, and reveals the absence of debate about the accommodation between different aspects of play within children's lives. In examining the balance between children's play in different environments we are mindful of both the need to foreground children's right to play and the necessity that we consider how play is implicated in social reproduction. The aim of the paper is therefore to transform geographical understandings of children's play through an exploration of the class-differentiated reconciliation of childcare-based, extra-curricular and free play in contemporary British childhoods.

| RESEARCHING CHILDREN'S PLAY
This investigation centres on Hortonshire 1 in the English Midlands. Children live in a mixture of large urban, smaller urban and rural communities (but not, given their shire-county location, in inner-city areas). Residents have access to professional employment within commuting distance, but traditional working-class employment has been hit by deindustrialisation. In class terms, the schools vary widely above and below the national free school meal (FSM) rate of 19.2%, but ethnic diversity is low in this disproportionately White British area (Department for Children, Schools and Families, 2009;Department for Education, 2013).
The research involved a questionnaire survey designed to elicit information from parents about their rules around free play; their children's participation in extra-curricular activities; and their use of childcare. The survey involved middle-and working-class parents with children in Years 2 (ages 6-7) and 6 (ages 10-11) in 18 primary schools (8 low FSM rates; 10 high FSM rates) (n = 321; middle class 160; working class 161; response rate >40% in both areas). Pupils were given a paper survey to take home to their parents and return in a sealed envelope. We designated parents as middle class if their children attended a school with an economically advantaged intake (FSM < 2.5%) and their household's primary wage earner was in managerial/professional occupations. We defined parents as working class if their children attended a school that draws from more financially impoverished communities (FSM > 30%) and their household's primary wage earner was in a routine/manual occupation (or had never worked/was long-term unemployed). This two-fold approach allowed insight into the family's individual and community context (Irwin & Elley, 2011). Parents with younger and older children were included as previous research demonstrates that age matters to children's spatial range (Hillman, Adams, & Whitelegg, 1990). The survey data were subject to chi-square analysis to compare middle-and working-class patterns, as well as differences between age groups. Significance was measured at the 5% level and results are reported in Table 1. Full numerical details of the tests are footnoted in each section for rigour; invalid tests (where >20% of cells have expected count <5) have been excluded.
This quantitative information from parents was combined with qualitative data from children in Years 2 and 6 in four primary schools (two low FSM rates; two high FSM rates). Children were interviewed alone or, if they preferred, in small groups of two to three children. In total, 73 children participated in the study: interviewees were evenly split in class and gender terms; two-fifths were in Year 2 and three-fifths in Year 6. Children were asked about their attitudes to and participation in: free play; school and community-based extra-curricular activities; and wraparound childcare. Research ethics are important in all studies. In this instance, parents were given the opportunity to opt their children out of research and children's individual consent was also sought (through verbal project descriptions, age-appropriate participant information sheets and consent forms). Children were also told they could withdraw at any point. Confidentiality was assured unless safeguarding concerns were raised. Pseudonyms were allocated, and identifying information removed, to protect interviewees' anonymity. All interviews were fully transcribed and subject to qualitative coding in a branching tree system with NVivo being employed to facilitate data slicing (Bazeley & Jackson, 2013). Nodes which reflected pre-existing and emergent research themes were analysed for dominant and counter themes and interpreted in the light of, and used to inform, the academic literature.

| Free play and family time
Geographical interest in children in the Global North was stimulated by a decline in their access to public space (Valentine, 1996a(Valentine, , 1996b, and today's research is heavily influenced by concern about the loss of this idealised play environment. As Loebach and Gilliland (2016, p. 575) argue: "[i]ndependent neighbourhood play and exploration have become endangered childhood experiences, posing significant challenges to healthy child development". Our quantitative research confirms that just under half of parents are comfortable with their 6-7-year-old children playing out immediately in front of their house, and only around a quarter say they allow their children to play in the street as far as a parent can see (see Table 1). 2 Parental comfort with the distance children can range increases significantly with age: most parents allow children aged 10-11 to play outside their house and in their street; around half allow them to go out of sight around the block; and roughly a quarter are happy to let them roam more widely in their neighbourhood (Schoeppe, Duncan, Badland, Rebar, & Vandelanotte, 2016). In the contemporary context, these data confirm the long-standing association between children's age and their ability to roam (Hart, 1979), though environmental factors such as the busyness of local roads continue to shape individual parents' decisions. In historical perspective, they demonstrate a sharp decline in children's ability to occupy and move through public space (e.g., in 1971 over 70% of 7-year-olds went to school unaccompanied) (Hillman et al., 1990;Woolley & Griffin, 2015). Indeed, much of today's movement is not fully "free range" (Hart, 1979, p. 46) as children may be: directly observed playing in front of the house; obliged to tell their parents timings/destination if going further afield (perhaps taking a mobile phone, though ownership is far from ubiquitous); and required to move as part of a group of friends or siblings (or with a dog).
In this respect, our research confirms the picture of sweeping change in the nature of children's play that has prompted much discussion in the discipline. Our qualitative interviews and focus groups with children add nuance to these otherwise bald statistics. The primary concern within the geographical literature is with the decline in outdoor free play, and the commensurate rise in indoor play in the "electronic bedroom", as well as the growth in structured activities (Witten et al., 2013, p. 222). Although this conceptual distinction between outdoor and indoor play has been important in academic analysis, many children describe a seamless mix of indoor and outdoor play, incorporating physical, social and technological activities into their after-school routines (Loebach & Gilliland, 2016 In terms of outdoor play, it is apparent that gardens as well as the street offered children highly valued play opportunities (Thomson & Philo, 2004). Many gardens are now better provisioned with play equipment than in the past: the cost of swings and slides has fallen, but it is the rapid rise of the trampoline (an estimated 250,000 of which were sold in the UK in 2014 alone [Royal Society for the Prevention of Accidents, 2015]) that emerges most clearly in the accounts of middleand working-class children who enjoy this energetic play. Moreover, it is striking that while class-differentiated parents gave statistically similar accounts of their rules, younger working-class children were much more likely to discuss playing out in their neighbourhood than were their middle-class counterparts: This apparent mismatch between similar parental rules and diverse child practices is in part explained by greater middleclass participation in extra-curricular activities and childcare, which are discussed in the subsequent sections. The one statistically significant class difference in respect to outdoor free play is that fewer middle-class parents will let their 10-11-year-old children play out in front of their house (see Table 1). Our qualitative data reinforce this point and show that some middle-class children, far from pressuring their parents to increase their spatial range (Valentine, 1997;Woolley & Griffin, 2015), are too fearful of strangers and traffic to play out when allowed to do so. It would be easy to argue that these children have simply internalised the moral panic which sees children as at risk in public space, but as Pain (2006) makes clear, it is important to consider the material basis of their fears. In this case, these include the responsibility to keep younger siblings safe from traffic on busy roads and local incidents of children being harassed by men in vehicles. Looking forward, some of these boys and girls also reject the notion that being free to hang out in public space is a desirable goal. Once again, it would be possible to dismiss their view as the product of a wider social discourse that labels young people as out of place in public space (Valentine, 1996a(Valentine, , 1996b; however, if as a sub-discipline we listen seriously to children, a rather different view emerges. Specifically, primary-aged children not only prefer to do something positively enjoyable rather than pass time hanging out, but also choose to inhabit pleasant public spaces rather than marginalised locations. These girls' decision to swim in a leisure centre with friends can be understood as a choice facilitated by their unrecognised class advantage (as they, unlike other young people, have parents who can afford to pay), rather than an opportunity missed because of their age: Hayley: I'd rather make something out of the life than hang around on the streets. Carys: And do something positive. . .where you're living instead of hanging around and getting into fights and stuff. Hayley: When we come home from swimming [at a pool in a less well-off neighbourhood] there'll be like loads of teenagers hanging about, there'll be some where the youth centre is, by the leisure centre. . .

Carys:
And it looks really boring just standing there shivering. . .Why don't they just go in instead of standing out in the cold? (MC-Y6) The emphasis placed on children's play in public space in the wider literature (Holt et al., 2015) also has an interesting relationship with the importance of family, which emerges through our qualitative research with children. It is clear that sibling relationships-and, in working-class communities in particular, extended-family networks with cousinsare important in children's experiences of playing out: What is evident in this quote, however, is not only the importance of child relatives, but also the value children place on intergenerational relationships. The geographical and wider literature tends to romanticise play as a children's world to which adults do not have access, and to consider parents only in terms of the limits their choices place on children's right to independent play (Holt et al., 2015;James, Jenks, & Prout, 1998). However, children in this study not only think parents should have rules; they are also keen, especially in busy middle-class families, to ensure their families have dedicated cross-generational leisure time: Abigail: I said to Mum why don't we have like a family Sunday where we go on a walk with the dogs and we've been doing that and it's really good because we never get any time together. This demonstrates that although it is vital to place declining independent outdoor play on the geographical agenda (Schoeppe et al., 2016), we must avoid conceptualising play solely in these terms and recognise that intergenerational forms of sociable leisureincluding family outings (Karsten & Felder, 2015) and time together at homealso matter to children.

| Clubs and activities
In contrast to the decline of free play (Gray, 2011;Holt et al., 2015), children's cultural, sporting and uniformed clubs have grown rapidly in size and scope during the 21st century. These instrumental and institutional play spaces have a long history (Gagen, 2004;Mills, 2013), but their recent expansion has been stimulated by consumer demand, state policy and, in some parts of the sector, entrepreneurial initiative (Holloway & Pimlott-Wilson, 2014;Karsten, 2015;Katz, 2008;Vincent & Ball, 2007). Our quantitative research (see Table 1) 4 demonstrates that middle-class children participate in significantly more extra-curricular activities than their working-class counterparts: for example, 50% of middle-class children engage in five or more activities per week, compared with only 10% of working-class children. Nevertheless, it is crucial to note that over 75% of working-class children also participate in one or more extra-curricular activities each week. The popularity of these activities has a profound impact on children's use of leisure time. Middle-class children in particular often have extra-curricular activities on multiple evenings a week, and on occasions, multiple activities in a day: These activity patterns have largely been explained in the wider literature through reference to class-differentiated parenting logics. Lareau's influential thesis, for example, suggests that middle-class parents are disproportionately likely to foster children's talents through extra-curricular clubs in a process of "concerted cultivation" (Lareau, 2002, p. 748), while Vincent and Ball (2007, p. 888) have emphasised efforts to produce a "Renaissance Child" through a mix of sporting and cultural activities (see also Karsten, 2015). Concern has been raised that this results in the overscheduling of children (Katz, 2008), with some psychologists identifying a causal link between the decline of free play and the rise of psychopathology in young people (Gray, 2011). This point is contested, however, by others who provide evidence that intensity of participation in organised out-of-school activities is positively associated with psychological flourishing, civic engagement and educational attainment (Mahoney & Vest, 2012). Involving young people as subjects rather than simply as objects of research significantly enhances our understanding of this phenomenon. What is most notable is that children do not regard themselves as overscheduled. They not only consider participation in multiple clubs a normal part of everyday life (with only one extremely busy child in our study suggesting that he did too much), but also feel that it is children themselves, in the context of adult support and encouragement, who make the choices about which, and how many, activities they do (Randall, Travers, Shapiro, & Bohnert, 2016). They thus feel in control of the balance between organised activities and other forms of play. Their individual choice of activitieswith large numbers of children trying a diversity of clubs before settling on those which suit them as a personunderpins enthusiasm and satisfaction with extra-curricular activities. Quite simply, they continue with those they like and leave those that they do not enjoy (Randall et al., 2016). As Megan (MC-Y6), one of the busiest children in this study, argues in respect to her parents: "they encourage me. . .they don't force me to do anything. . .I decide. . .if ever I was tired. . .of doing them then I'd just say right I don't want to do this anymore". In sum, rising participation rates do not simply reflect an ethos of concerted cultivation among middle-class parents, but also the importance of what we might term a process of elective engagement among middle-class children.
Elective engagement in multiple clubs and activities was not, however, an option open to many working-class children. This was not because working-class parents have a radically different cultural logic of parenting (cf. Lareau, 2002): in this study they are significantly more likely than their middle-class counterparts to think that organised activities are better for children than playing out (57% cf. 27%). 5 Rather, as children were painfully aware, structural inequalities limit their access. Some activities are unaffordable at point of access: Samuel: I would like to do violin but they're too expensive for both of us to do it, it would cost a lot of money. Maybe when my sister [who does it now] leaves school. (WC-Y2) Alice: [Y]ou said you would start going to dancing didn't you? Amber: I lied. Alice: You lied? Amber: . . .My Mum can't afford that much money.  In other instances, the cost of travel to activities can be prohibitive. As Sebastian (WC-Y6) explains: "I used to play for a football team but. . .my Dad got sick and tired of paying for the taxis". The consequence is that many children are reliant on activities that are provided for free or at very low cost in school or their immediate neighbourhood by the state or voluntary sector (Bennett, Lutz, & Jayaram, 2012). These opportunities mitigate inequality in access, but the number and range of clubs is insufficient to meet child demand and provision is becoming increasingly precarious as austerity bites. This means children are unable to establish their own preferred balance between structured activities and free play. Moreover, even nominal charges for these subsidised services can prevent participation: Loss of outdoor play is not the issue here; rather, the research evidence raises the spectre of the socially-excluded, underscheduled child, as these working-class children are unable to exercise their right to access recreational activities (United Nations, 1989). Middle-and working-class children's evident enthusiasm for organised clubs and activities warrants consideration. Children's rationale, the findings suggest, is twofold. First, clubs are seen as an opportunity to have "some fun with my friends" (Henry Y6-MC). One concern raised about the decline of outdoor free play is that it reduces children's opportunities for ludic sociability and the associated developmental and mental health benefits (Gray, 2011). Our research demonstrates that the extra-curricular activities provide an alternative source of social play. For working-class children, the friends with whom they spend time are more likely to be their classmates as many of their activities are provided through school or in the immediate locality. Middle-class children, however, find that access to activities in diverse locations means they not only get to play with friends, they are able to develop wider social capital (Bourdieu, 2008) with like-minded individuals: Carys: I think you get more friends. . .who like enjoy the same things as you and have got like a match with the personality as you. . ..Because if you're at school there might be people who like the same things as you, but then if you go to a club with the thing that you like doing you know everyone's really liking the same things that you [do].

(MC-Y6)
Second, extra-curricular activities are seen by children to be part of a healthy lifestyle. Many clubs offer children the opportunity to be physically active:

Isabel:
It's good because you get to go on the beam and you get to go on the monkey bars. (MC-Y2) Lucas: I think it's better than just sitting in the house [you are] getting more energy and training with your friends (MC-Y6) James: [We] run about most of the time! And we do penalty practice, we have these little practice matches. (WC-Y6).
Children not only enjoy these energetic activities, but also recognise and value the impact they have on their corporeal health and fitness. In addition, middle-class children also experience mental health gains: achievements in extra-curricular activitieswhether that be progressing to the deep-end in swimming or performing well in sporting competitionswere an evident source of pride and self-esteem for many (and no concerns were raised by children about the performance ethic associated with some of these form of play [Holt et al., 2015]). This particular mental health benefit is less available to children in working-class communities: here subsidised provision reduces inequality by giving children a taste of different activities, but this is often offered short-term (e.g., to a specific year group or for a half-term) meaning children are unable to develop skills through long-term participation. Nevertheless, this is still valued by children, especially those who live in families and communities where anti-social behaviour is a problem, and who see clubs as a positive alternative to playing out in their neighbourhood:

| After school clubs
After school clubs (ASCs) developed rapidly as environments for children's play in 21st-century Britain. The state strategically expanded its role in social reproductionre-designating primary schools as spaces for childcare as well as educationin order to increase working parenthood and reduce social exclusion. Structural changes such as the feminisation of the labour force have important consequences for children's play as they reduce informal supervision of free play because fewer women are able to act as "eyes on the street" (Jacobs, 1961), and fewer children are able to play out because they, especially in the younger age range, spend time in childcare while their parent/s work (Holt et al., 2015;Witten et al., 2013). However, it is vital to note that the feminisation of the labour force, and indeed the provision of this service, has in practice been shaped by class divisions (Cummings et al., 2011;Holloway & Pimlott-Wilson, 2016;Smith & Barker, 2004). This means that the consequences of these wider structural shifts will not necessarily be the same for children from different class backgrounds. Table 1 6 demonstrates that in this study a third of middle-class families use ASCs, but that uptake is significantly lower among younger working-class children. ASCs in middle-class schools, which can be run either on a commercial or not-forprofit basis, offer working families childcare from the end of school until early evening (commonly 3.30 p.m.-6 p.m.). In previous generations, these children would have left school at the end of the day with many then playing out in their neighbourhoods (or at home) in the late afternoon. Although their parents might theoretically still allow them to play out, the play environments to which they have access have undoubtedly changed, with ASCs offering physically active outdoor and creative indoor play opportunities, such as: informal sport; den building or tactical games; traditional outdoor toys; free play; indoor games and toys; and craft activities. Children's presence in these ASCs contributes to the broader reduction in levels of unsupervised outdoor play that has prompted academic and social concern (Carrington, 2016;Loebach & Gilliland, 2016): however, our research with children, especially those in the younger age ranges, demonstrates that they value and enjoy the play opportunities these care environments offer: I: [W]hat is it like at ASC? Ethan: Well it's really cool because in the summer we go outside. . .We go on the field and we have like, we get some water pistols. . .

Aidan:
And we get to squirt each other with a water pistol. Ethan: And it's really fun. . . even in winter [we go out] but not on the field. . .we get to draw and get to like Aidan: . . .do like cards (MC-Y2) Jake: It's good fun. . .. we could go outside and play basketball and stuff. . .. it's mostly if the weather's OK. (MC-Y6) The earning potential of parents in working-class areas means they are much less likely to be able to afford this provision; however, differences in usage rates are lower than they would otherwise be because some schools in low-income neighbourhoods strive to run ASCs at nominal cost (e.g., £1), despite pressures on educational budgets under austerity, in order to provide children with an hour's post-school play opportunities. Children may use these clubs because their parents work, but many with non-working parents such as Paige (WC-Y2) choose to attend because they enjoy the provision offered: "I like coming to ASC because I get chocolate bars and all sorts, we get colouring, we get games, we get animals, we get cars, we get chocolate biscuits". Children's palpable enjoyment of ASC is noteworthy. Their growth in middle-class areas in particular stems from an increase in working parenthood rather than an imperative to respond to children's play needs (Holloway & Pimlott-Wilson, 2016). These labour-market-induced changes in young people's everyday environmentswhich do indeed contribute to reductions in outdoor free play in local streets and parks (Holt et al., 2015) are welcomed by children, especially those in the younger age ranges. Indeed, it is clear that some children who do not attend ASC find the prospect attractive: However, children across the age spectrum qualified their enthusiasm for ASC with a proviso about frequency of participation. For example, Henry (MC-Y6) likes ASC "because you get to do activities and play with your friends", but agrees with his mother that he should not go every day as "it gets a bit boring after you've been for 4 days a week". Equally, Jackson (MC-Y2), who goes two nights a week, would be "a little bit sad" if he went every night, as "I like to go swimming sometimes". These time caveats reflect children's enthusiasm for collective play environments, but also their desire to have varied lives which mix different play opportunities during the course of a week. This desire for balance is important, but, unlike structured activities, decisions about the volume of time spent in childcare are much less likely to be in the hands of the child.

| CONCLUSION
This paper marks a step change in research on geographies of children's play. Our point of departure is geographical interest in the decline of outdoor free play and children's growing exclusion from adult-controlled public space (Alparone & Pacilli, 2012;Schoeppe et al., 2016;Valentine, 1996b), but in exploring this issue we insist that equal consideration is given to children's views on other fast-developing forms of play such as clubs and activities (Karsten, 2005). We then bring this expanded interpretation of what count as "playful spaces" (Thomson & Philo, 2004, p. 111) together with the feminist research about rising maternal employment (Gallagher, 2014;Holloway & Pimlott-Wilson, 2016;Ward et al., 2010). This focuses attention on burgeoning play opportunities in collective childcare environments, and elucidates the need to consider how children balance diverse play landscapes in their daily life. This intellectual thrust is empirically grounded through rigorous quantitative and qualitative research with middle-and working-class parents and children that explores their class-differentiated reconciliation of childcarebased, extra-curricular and free play in contemporary British childhoods. In conclusion, we consider three valuable, and related, contributions the research makes to geographical understandings of play. First, the research counters the geographical tendency to idealise outdoor free play as the most authentic, natural and developmentally wholesome way for children to play, expounding instead a more broad-ranging definition of playful spaces, and strategically asserting children's right to access diverse free, structured and cared-based play environments. Geographical debates about the decline in children's outdoor free play and independent mobility have, almost without exception, articulated a quixotic desire to halt or reverse this process in order to protect children's rightful access to public space and to outdoor play that promotes physical and social development (Holt et al., 2015;Loebach & Gilliland, 2016;Schoeppe et al., 2016;Witten et al., 2013). We support this goal; but our strategic conceptualisation of children as bearers of rights, and to whose views we should listen, also causes us to question the overwhelming emphasis placed on this within the discipline. Our research demonstrates that some children enjoy playing out; however, many also move seamlessly between indoor and outdoor play in different environments. Gardens, now often better provisioned than in the past, are seen to offer both enjoyable and vigorous play. Equally, children revel in play in bedrooms with traditional toys/games or electronic gadgets that can bridge the public/private divide. Moreover, many children now have the opportunity to play in public spaces beyond the street, for example in organised clubs and activities. Children's outdoor play is not simply in decline because of parental fears about traffic or stranger danger; it is also diminishing because many children, unlike their post-War counterparts who were sent out to play due to a lack of other opportunities (Karsten, 2005), now have other places they might choose to go. As a discipline, we need to continue to support children's rightful access to free play in public (and private) space, but we must also recognise and assert children's rightframed within the Article 31 of the UNCRC (United Nations, 1989)to access recreational activities in institutionalised public spaces such as community halls, leisure centres and sports fields.
Second, the focus on more diverse play environments foregrounds questions about intergenerational relations, and highlights the need to consider adults' varied involvement in "children"s' play. Children's outdoor play that is "freely chosen and directed by the participants" (Gray, 2011, p. 444) is often assumed to be a matter for children alone, in contrast to adult-organised activities, and geographers have attached significance to "maintaining portions of children's lives that are not organized and institutionalized by adults, which are not. . .arranged through "little league" tournaments" (Aitken, 2001, p. 17). We value both forms of play, and find the dualist distinction concerning adult involvement too simplistic. On the one hand, free play does require adult input. Reductions in informal supervision of neighbourhood space by mothers, who are increasingly likely to be in paid work, is one factor contributing to parental unwillingness to allow their children to roam free (Witten et al., 2013), and the decreased presence of other children then compounds the issue (Holt et al., 2015, p. 81). Outdoor play is free for children at the point of access, but it has always depended, to some degree, on unpaid maternal labour to monitor it. The import of this is that access to play is not simply children's right (United Nations, 1989), it is also a feminist issue, and we need to find ways to put "eyes on the street" that do not simply belong to unremunerated women. On the other hand, scrutiny of children's views shows that children and adults do not necessarily operate in an antagonistic formation and that adult involvement in children's play can be a positive good. For example, some children in this study actively carve out spaces for intergenerational recreation within the family. Equally, adults facilitate and resource extra-curricular activities for children, and children may enjoy this play, which challenges them to learn new skills, promotes their physical health and extends their social networks. In future research, geographers need to ensure that we neither construct children's play as the rightful preserve of young people alone, nor construct adult involvement solely in terms of supervisory limits (though these can be important), instead teasing out the consequences for both generations of their involvement in play.
Third, the findings demonstrate the importance of play as a sphere for social reproduction (Aitken & Herman, 1997;Katz, 2004), as children's class identities emerge in, and are reproduced through, the childcare-based, extra-curricular and free play spaces to which they have differential access. To expand, our research demonstrates that middleand working-class children have a different relationship with the play-work benchmark that play should be free to access and freely chosen, with children free to leave (Brown & Patte, 2013). Middle-class children have considerable ability to reconcile their own play landscapes. They are not overscheduled as previous research has implied (Katz, 2008); rather their elective engagement in paid-for extra-curricular activities, which can aid the reproduction of their middle-class status (Katz, 2008;Vincent & Ball, 2007), alongside reduced outdoor free play, reflects their desired accommodation of these different play environments. Their presence in wraparound care is more complex: many children enjoy this play environment on a part-time basis, they freely choose from a range of play activities while there, but they are not free to leave. Parents' work commitments place the accommodation of this play landscape outside the individual control of children (even as they simultaneously benefit from the impact their parents' employment has on their own middle-class position).
Working-class children, by contrast, spend considerable time in the very kind of free play that is venerated in the literature (Gray, 2011). However, they are underscheduled in that they want greater access to extra-curricular activities than they have, and some feel they miss out on the play opportunities that wraparound childcare can fulfil (Smith & Barker, 2004). Our point here is not that children's time ought to be organised; rather, it is that these children have insufficient control over the balance between different play environments. Some forms of play they covet are not financially free, their low family incomes mean they cannot therefore freely choose to participate in them and lack of access to alternative play landscapes means they are not free to leave play on the street. Indeed, structural cuts to education and local government budgets in the age of austerity are making their access to diverse play environments ever more precarious. In sum, class is crucial in shaping the reconciliation of children's play environments, and children's differential access to these play environments helps reproduce this unequal regime of power. This not only matters in the present as working-and middle-class children experience very different childhoods, but also shapes young people's futures through their unequal opportunities to develop embodied social and cultural capital in different play environments.
Children's play landscapes are changing in the Global North and as a discipline we need to broaden our conceptual understandings of what constitutes play if we are to understand the lived experience of contemporary childhood. Future research must consider inequalities in children's ability to reconcile different play environmentswhether that be in terms of class, as explored here, or along other axes of social difference and locationin a geography of play that neither unconsciously romanticises outdoor free play nor vilifies its structured and care-based alternatives, and that recognises the importance of adults and children in constructing ludic landscapes. participants at the International Symposium on Knowledge and Space (Heidelberg), the Nordic Geographers Meeting and the 5th International Conference on Geographies of Children, Youth and Families for their insightful comments.