The public playground paradox: ‘child’s joy’ or heterotopia of fear?

ABSTRACT Literature depicts children of the Global North withdrawing from public space to ‘acceptable islands’. Driven by fears both of and for children, the public playground – one such island – provides clear-cut distinctions between childhood and adulthood. Extending this argument, this paper takes the original approach of theoretically framing the playground as a heterotopia of deviance, examining – for the first time – three Greek public playground sites in relation to adjacent public space. Drawing on an ethnographic study in Athens, findings show fear to underpin surveillance, control and playground boundary porosity. Normative classification as ‘children’s space’ discourages adult engagement. However, in a novel and significant finding, a paradoxical phenomenon sees the playground’s presence simultaneously legitimizing playful behaviour in adjacent public space for children and adults. Extended playground play creates alternate orderings and negotiates norms and hierarchies, suggesting significant wider potential to reconceptualise playground-urban design for an intergenerational public realm.

Heterotopia of Deviance (Foucault 1998): a space with strict rules, driven by society's anxieties occurring from the perceptions of what it is to be a child, and how one should play (See : Aitken 2001;Gagen 2000aGagen , 2000bGülgönen and Corona 2015;Holloway and Valentine 2000a;Loukaitou-Sideris 2003;Jarvis, Newman, and Swiniarski 2014;Rasmussen 2004;Singh and Gupta 2012;Solomon 2005). In societies permeated by a 'culture of fear' (Furedi 2002, viii), playgrounds function as places where 'vulnerable' children can be segregated for their 'protection'.
At the same time, the concept of heterotopia is used to explore the playground's potential, expressed and realized through the act of playing. Heterotopia offers a way to think about transgression, moving away from dualisms: Transgression, then, is not related to the limit as black to white, the prohibited to the lawful, the outside to the inside, or as the open area of a building to its enclosed spaces. Rather their relationship takes the form of lighting in the night which […] lights up the night from the inside, from top to bottom, and yet owes to the dark the stark clarity of its manifestation. (Foucault 1998, 74) While Foucault was more interested in institutionalized heterotopias, intended (in their majority) to preserve the existing status quo, 1 our approach reflects McLeod's (1996) stance: we explore how heterotopias form everyday life; or as we would like to call them, 'everyday heterotopias'. These are interwoven in the urban fabric, forming an integral part of people's everyday experience. Everyday heterotopias include the element of the familiar and the repeated rather than the exception and the festival and as such, are capable of empowering their subjects. Genocchio (1995) criticizes the overuse of the term and argues that by naming a heterotopic space as such, one deprives it of its heterotopic characteristics. While recognizing this position, we contend that this can also transfer heterotopia from the scale of the everyday into an abstract theoretical sphere with little practical spin-off.
For the purposes of this study we identify two core characteristics of heterotopia, of relevance to the playground space. First, as discussed above, deviant: the playground space is a heterotopia of deviance (Foucault 1998). Playgrounds can be understood as places created to house childhooda state of human life that is usually thought of as one that deviates from the 'normality' of adult life. Playgrounds are spaces of protection for the vulnerable, but at the same time spaces where the alternate orderings of play manifest. Playgrounds become one of the '"acceptable islands" (Matthews, Limb, and Taylor 2000, 63), "special" (Rasmussen 2004, 157) "proper" places' (Olwig and Gulløv, 2003, 101) that children are allowed to use, creating a network of temporary, dispersed 'housings', potentially engaging children with the public realm, yet at the same time segregating them from it. The totality that seems to emerge from Foucault's attempt to present heterotopias as completely different and distinct from their surroundings has been often criticized (Genocchio 1995;Saldanha 2008). In this study we first approach the playground space as a heterotopia of deviance in order to focus on this characteristic of difference. However, we then move beyond, in order to explore the limits of this difference.
Second, isolated yet penetrable: the playground has a 'system of opening and closing that both isolates it and makes it penetrable' (Foucault 2012, 267), transforming it into a kind of 'enclave'referring not only to the physical characteristics of space, but also to the social interactions taking place thereconstructed by the members of the heterotopia, consciously or unconsciously. Existing literature approaches the playground as a heterotopia without exploring its opening mechanisms or its reciprocal relation to its surroundings (Campo 2013;Kern 2008;Richards 2012;Vermeulen 2011;Wesselman 2013). Rather it examines playgrounds as spaces in themselves, without connecting them to their general context. Drawing on the outlined characteristics, this paper examines the interrelationship between fear and the practices associated with the playground space. The term 'fear' has been chosen in order to cover the expressions of paranoia and fear (Gill 2007) emerging from a 'culture of fear' (Furedi 2002, viii) which continues to gain ground: 'a generalised and insidious anxiety about safety that has found expression in fears for children' (Gill 2007, 14) as expressed in the playground space. How does fear manifest itself in the playground? and What are the associated implications for children's presence in public space?

Methodology and methods
There is a long tradition of using ethnographic methods to research the playground space (Blackford 2004;Corsaro 2003;Ferré, Guitart, and Ferret 2006;Mayeza 2017;Opie 1994;Thorne 1999;Willett et al. 2013), with a few studies specifically using ethnography to approach play and playgrounds as heterotopias (Gallan 2015;Low 2008;Richards 2012;Vermeulen 2011). Ethnography was here chosen as a means to focus on the practices of heterotopia, the ways the playground space is experienced as 'other' and its connection with what is considered 'normal' public space. The findings in this paper are drawn from a study that took place over five months in 2016 and 2017 in three playground sites in Athens, Greece. Approximately 100 h of observations were conducted in the playgrounds and their adjacent spaces, during the morning, afternoon and evening, both weekdays and weekends, by Alkistis. The data collection methods consisted of ethnographic observations, field notes, informal discussions and 61 semi-structured ethnographic interviews (Angrosino 2007;Aitken and Herman 2009), using 'theoretical sampling' (Ball 1990, 165). A reflexive journal (Punch 2012) complemented Alkistis' field notes, forming a comprehensive 'research biography' (Ball 1990, 170). A variety of visual mapping techniques were used to explore the physical and spatial characteristics of each space and their relationship to participants' behaviours. The theoretical framework of heterotopia guided the analysis process by engaging with ethnography's plurality and constant change informing the codes and making sense of the data.

Cases and context
Athens is a city that struggles with its limited and as a result, valuable, public space. Since 2008, Athens 2 has been undergoing a period of social transformation suffering the results of austerity; an 'ongoing humanitarian crisis' (Dalakoglou 2012). Greek culture is characterized by simultaneous extroversion and introversion towards local public space. Immediate neighbourhood space is no longer as familiar as it used to be (See: Athens' Oral History Groups 2016) while neighbourhood interactions have become infrequent. Many public spaces, in Mediterranean countries, are seen as potentially dangerousespecially in the eveningand are avoided by citizens. As Ferré, Guitart, and Ferret (2006, 173) argue: This affects people's daily lives, especially those people who for reasons of gender and age (women, the elderly) spend more time close to home. However, the summer months, especially, bring many people to piazzas and playgrounds. Typically placed in public piazzas or in leftover spaces in local neighbourhoods, playgrounds are seen as meeting points both by children and parents (Galani 2011;Ferré, Guitart, and Ferret 2006). 'Metaphors for cultural meanings' (Blackford 2004, 237), Athenian playgrounds are places where one can trace the forces structuring children's everyday lives.
The chosen playground sites were paradigmatic cases, that is 'cases that highlight more general characteristics of the societies in question' (Flyvbjerg 2001, 80) reflecting the typical neighbourhood play-space in Athens: fenced, municipality-provided playgrounds placed in public piazzas, abiding by the 'standardized playground' model (Doll and Brehm 2010;Solomon 2005). Typical play structures were see-saws, swings, slides, monkey bars, climbing structures and ropes, while soft 'carpet' was placed around the structures. We selected three cases, each located in a district representing a different socio-economic identity. 3 In this study we focused on the ethnographic research principles structuring a 'thick description' (Carspecken 1996). As Flyvbjerg (2001) argues: 'The advantage of large samples is breadth, while their problem is one of depth. For the case study, the situation is the reverse' (87). The purpose of this research was not to make comparisons between contrasting cases, but to study the typical playground population examining a range of patterns of behaviour.
The focus on minutiae, which directly opposes much conventional wisdom about the need to focus on 'important problems' has its background in a fundamental phenomenological experience, that small questions often lead to big answers. (Flyvbjerg 2001, 133)

Fear classifying space
One of the clearest and most strongly evidenced findings of this study was that people's experience when in the public realm was largely informed by an unwritten, shared classification of space. Within each observed site, particular areas and physical characteristics were observed to be associated with particular sets of behaviours and types of users. This classification was shared between the three cases, revealing broader societal norms regulating people's interaction with space: If you were supposed to enter the flowerbed, there wouldn't be bars around it. (Father, Dexameni) All piazzas had a clear distinction between children's and adults' spaces. Guardians would classify areas and elements as safe/not safe, for play/not for play, according to their designated use: Because they are safer there … It is there where the playing structures are. (Nanny, Dexameni) Despite the playground appearing to physically be part of public space (physically accessible to all users) it was not socio-culturally perceived as public space. Rather it emerged through both observations and interviews as a distinct space, classified as 'children's'. Separation and supervision are the two main attributes of 'proper' places for children's play (Olwig and Gulløv 2003, 2;Aitken 2001). The fencethe physical structure providing this separationemerged as having major importance: We control them better that way. (Father, Ilioupolis) Understood through the lens of heterotopia, the fence defines the porous limit of the playground enclave of deviancea physical indicator that this area accommodates alternate orderings in the normality of public space: Because children are more constrained that way … more secure. The fence is necessary for the children's safety. (Nanny, Dexameni) Even in the more radical approaches to the playground (Kinchin and O'Connor 2012; Lady Allen of Hurtwood 1964), the fence has endured as a prerequisite for its operation, acting as the boundary between child/adult, play/non-play space. When asked about the 'ideal playground', the quality of the play equipment was often mentioned by the parents second after the need for a fenced space that would make segregation and supervision easier. They often chose to visit smaller playgrounds for their perceived safety, despite offering fewer play opportunities. There was evidence that supervision had been internalized by the children (Blackford 2004) in a Foucauldian sense (Foucault 1991): We need the fence. Because without it we may get out chasing the ball and get lost. (Girl, Ilioupolis) It is interesting to note that children [even toddlers] were familiar with the spatial restriction of the playground in their everyday lives, recognizing this 'cut-out' from the public realm as their own. In Vyronas, for example, where the playground was visible from quite far away, it was observed that children and toddlers recognized the space, asking their parents to enter. The fence acted as a landmark in the city's landscape, indicating the 'rightful place' for play. Perceiving playground as merely a space with specialized play-equipment strengthened the belief that one should only play with this equipment, while all other games (e.g. ball games) and interaction should take place outside: We are playing here [outside] as inside is for those that want to swing. (Boy, Vyronas) Both adults' fears and the perceptions of children, as special 'deviant' beings were materialized in the spatial characteristics of the playground. Parents characterized as 'good' a playground that is safe, clean, fully paved with soft materials, containing low challenge equipment, no visual obstructions (thus allowing supervision) and adequately segregated from its surroundings. The fear culture (Gill 2007) has established strict playground safety regulations and standardisations in Greece (See: ELOT 2008; Ministry of Internal Affairs 2014, 2009) that are continually increasing, fuelled by increasing parental safety concerns and the fear of physical injurya trend similar to those noted in the UK and USA (Gill 2007): I was surprised to listen to a mother in Ilioupolis listing the technical regulations as established by the law. (Field notes) Children themselves tended to simply prefer challenging spaces where they could meet friends: There aren't any other children to play with in the playground, so I prefer sitting on the bench outside … . (Girl, Vyronas) I would prefer it if the slide was higher. (Girl, Ilioupolis) As places created to safeguard children, the playgrounds' function revolved around children's safety. Conforming to the 'proper' use of the play structures was often mentioned as the main prerequisite to avoid both conflict and accidents, strengthening the playgrounds' deviant character as a space promoting specific behaviours: They should play properly. So they will not get hurt. And that way, other children can play as well. They take turns on the slide. (Mother, Ilioupolis) Children, following the unwritten rules of the heterotopia of deviance would often scold each other, making sure they were using the structures 'properly': Not like that! You should play that way! (Girl, Dexameni). (See also : Zeiher 2003) As one might expect, perceptions of playgrounds as solely children's spaces excluded adult play with no adult able to give examples of their own play behaviours in the playground: For example, we wanted to play on the monkey bars, but I was ashamed to do so because I was expecting someone to say to me that there are children that want to play. (Father, Ilioupolis) The fear of 'I don't want to break anything' (Father, Ilioupolis) was present, defining the playground's character as a child's space and strengthening the heterotopic characteristic of deviance needing 'special equipment'.
Parents' fears did not welcome the presence of other adults in the playground: If it is a young man we would talk to him, we may say to him 'do you want something here?' … We will ask him … we don't want him to stare at the children. (Grandmother, Dexameni) Parents' interaction with the 'outsiders' acted as the opening and closing mechanism of the heterotopia of deviance. It is interesting to note, however, that while male adults were considered dangerous, older people or childless mothers were seen with sympathy and were allowed to use the playground: I would think that … ok. She came to see the children … Often the ones that do not have children of their own, crave to see children playing […]. (Grandmother, Ilioupolis) In line with parental perceptions, 'Others' themselves did not perceive the playgrounds as part of the public space, but rather as 'places only for children' (Blackford 2004, 232). The suspicions that parents had expressed about certain unaccompanied adults were, accordingly, felt by those adults: A grandfather in Dexameni argued that he wouldn't stay in the playground if he wasn't with the toddler as he would be embarrassed by the people looking suspiciously at him. (Field notes) These perceptions were observed to limit interactions and restrict access.
Transgressive play: beyond the fence Loaded with expectations, the playground space could not always fulfil its purpose as a safe playspace, often lacking the needed infrastructure. Play often acquired a transgressive character as people invented new ways of interacting with the space in order to compensate for its inability to accommodate the desideratum 'valuable' play. Play, bearing 'possibility', transformed the playground and questioned the norms of the Heterotopia of Deviance. What was considered 'normal' and 'proper' was challenged, with new orderings emerging: A father plays on the see-saw with four children. He balances the one side of the see-saw and the children try to balance on the other side of the structure. (Field-notes, Dexameni) A mother in Vyronas, self-conscious of the transgression of the playground norms, trying to justify allowing her son to jump in and out of the fence, said: They are bored … there aren't any play-structures.
A major finding of this study is that the fence's physicality often supported play to take place outside the playground. The absence of the fence was often mentioned by guardians as having the potential to restrict play: No, it's better that way. When they kick, they kick to the fence. If it was open […] the ball could hit a child in the playground. If the two areas were together, the older children wouldn't have anywhere to play. (Mother, Dexameni) In all three cases, the piazzas' infrastructure, especially those elements placed closer to the playground, compensated for the limitations created by the playground's prescriptive and age-specific structures. Children would exit the playground to play in the piazza, taking advantage of its infrastructure and its affordances, using these as an extension of the playground structures.
A boy runs from inside the playground to the green area, climbs a tree, jumps down the ledge, runs around in the piazza, climbs the ledge again, into the upper level green area. (Field-notes, Ilioupolis) A strange paradox emerged, what we here call 'the playground paradox', according to which the fence did not confine play inside the playground but actually supported transgression. The playground emerged as a space physically segregated, but not isolated from the public realm, with the physicality of the fence itself allowing games to transgress the playground boundary: A girl walks on the fence's ledge and starts shouting and waving to the people in the café across the street. They wave back to her. (Field-notes, Vyronas) Someone kicked the ball so high from inside the playground that it landed in the flowerbed!! The children kick it back in again laughing. (Field-notes, Dexameni) It is important to clarify that this kind of play behaviour was not observed in piazzas that did not include a playground. 4 In all three cases, play flowed from the playground space unifying the playground with its surrounding space and extending the play area, challenging the classifications of both piazza and playground.
Despite play taking place beyond the fence, the playground's supervision ring (Blackford 2004), fuelled by adults' fears, was observed to also transfer to the adjacent public space, usually placing parents on the margins of the piazza, supervising children who would play in the centre. Children, however, used boundaries in their games in order to explore and challenge safety limitations. Their play often revealed their perceptions about the enclosed playground space and in many cases it questioned or even reversed it: The children exited from the hole in the fence and were calling their parents sitting in the playground sitting area 'you are in a cage!!'. (Field-notes, Dexameni) The new play area supported the extension of children's agency into the public space. All kinds of games (ball, scooters, pretend play) occupied the whole piazza area, changing the space classifications: When children play in the green area [and] occupy the whole space, creating dust and noise, no one complains. This is an area for play, if someone doesn't want dust she could sit somewhere else. (Field notes: Vyronas) Children often projected an attitude of 'owning' the space in and around the playground: Boys mocked two old ladies after they scolded them for throwing a ball directly onto them, while the father murmured 'go sit further down'. (Field notes, Vyronas) At the same time, while perceptions of playgrounds as solely children's spaces tended to exclude adult play, public space, unrestricted by play-structures, accommodated a variety of play expressions: I prefer playing in the piazza: they chase me, I climb the tree et cetera. I don't play in the playground space though. I feel embarrassed. The play structures bias you towards child's play. (Father, Dexameni) The father did not stop the children, but encouraged them to get wet, indicating which beck of the green areas [outside the playground] was throwing out more water […] The children move in and out playing in both areas [piazza and playground] using the playground as their 'castle'. (Field notes: Ilioupoli) Adults playing more actively occupied and manipulated space with children. An intergenerational play area was created, catalyzed by the playground, but having a dynamic of its own. People taking advantage of the piazzas' affordances sustained a new, intergenerational play-space, distinct from the playground but in constant relation with it: Toddler kicks ball, ball goes to an old man sitting on the red bench. The boy gets closer, the old man kicks it back. The toddler kicks the ball again toward the old man. They play like that for 10 min. (Field-notes, Dexameni)

Conclusion: playground paradox; deviance and transgression
It becomes clear that examining the playground as a self-contained structure, isolated from its surroundings, greatly limits our understanding of this space. At the same time it is not easy to argue in favour of the playground's character as a place accommodating either fears or joy. The study reveals a space classification that reflects broader perceptions and relations between the play-space actors. Perceptions of the playground as children's space inform practices and interactions, limiting appropriation by adults and intergenerational play. In particular, adults' willingness to engage in play with their children was hindered by the playground classification as space for children.
The playground as a heterotopia of deviance, informed by fixed conceptualisations of childhood, both as a precarious stage in human life (Gagen 2000a(Gagen , 2000bOlk 2009;Zeiher 2003) and as a 'repository for hope' (Kraftl 2008, 82), was structured as 'children's space'occurring from and stimulating parental fearswith strictly prescribed use and safety standards, intended for a specific group of users. The physical border of the fence was considered essential for its function by the parents, as segregation was interwoven with the playground's purpose. Our observations revealed that it was the adults' anxieties that acted as the main opening and closing mechanisms for this heterotopia, dictating who could use the playground space, barring those considered suspicious. The findings suggest that parents' fears often hindered the playground from becoming what one might suppose it to bea space of play, experimentation and joyinstead restricting it to its function as a segregating heterotopia of deviance.
Paradoxically, adults' fears about children's safety and expectations of 'valuable' play in the playground space were observed to support its own transgression. The 'playground paradox'a phenomenon whereby the playground restricted play and interactions in its premises, but supported an intergenerational play area outside its limits is a key finding of this study. Extending the play area, play itself created alternate orderings, negotiating norms and hierarchies (Edmiston 2010;Sutton-Smith 1997) and accommodated 'joy' outside the playground boundaries. This study contributes to the reconceptualisation of the 'city as a playground' (Stenros 2014, 213) and playful cities (Alfrink 2014;Borden 2007;Donoff and Bridgman 2017;Stevens 2007;Vanolo 2018;Walz and Deterding 2014) placing traditional playground spaces in the centre of the debate. The temporal transformation of surrounding public space supported a reconceptualization of playground play as an inclusive, intergenerational, behaviour. When in play, children and adults here coauthored their identities and negotiated the established hierarchies, proposing 'alternative ways of being' (Radley 1995, 9).
The critical framework of Heterotopia allowed us to examine the interplay of both deviance and transgression in the playground space. Normalization and transgression practicesoften in the guise of self-regulating fear or embarrassmentwere often mentioned by both parents and children, making play both the cause of anxiety and socio-spatial transgression. The alternate orderings of the purpose-centred, oppressive heterotopia, paradoxically fostered alternate orderings that empowered their subjects. We draw connections with Loxham's (2013) study using the concept of heterotopia to argue that a nineteenth-century park was used both as a site of oppression and liberation for its users: Despite an intended discipline, the ordering here differed from that of normal society and paradoxically an unintended freedom to indulge in otherwise forbidden acts was provided. (566) In both studies, it is this negotiation at the scale of the everyday rather than a utopian conceptualization of liberation that bears the possibility of change; an everyday, unintentional utopianism (Gardiner 2004;Kraftl 2007Kraftl , 2009.
Perhaps the most striking finding of this study, therefore, is that the playground, a 'child's rightful space' with a prescriptive nature, encouraged an extended freedom for children's action outside its limits. This study adds to the literature about children's spaces of transgression. It moves away from conceptualisations about carved out spaces (Beazley 2000;Jones 2000; proposing children's direct engagement with the public realm. We suggest that the debate should not revolve only around ways to familiarize children with the public realm, but also include familiarizing the public realm with children's presence, practices and play. Extending the playground to the surrounding public space, this ethnography reveals the potential of the playground to bear the freedom to enact alternate orderings without being restricted by design intentions and expected behaviours. The playground, supporting and projecting children's agency into the public realm, can be transformed into a place that includes 'children's joy'. Using the space's affordances as an extension to the playground equipment, play was seen to temporarily transform surrounding public space into a continually negotiated space, which children were free to appropriate. This play area, importantly, emerged as a part of the public realm. These findings therefore lead us to ask: what can be done in order for the playground to contest the defining rhetoric of safety and instead engage more fully with 'children's joy'? How might the playground effectively expand the observed intergenerational and spatially transgressive play area to a more extended public realm, dissipating fears for and of children? We call for future research to examine this potential, exploring the possibilities of the playground and urban design in dialogue, to help alleviate playground fears and to reconceptualise the function of this space. Approaching the playground as a space through which children can act in 'adult' public space, this research suggests that the playground can afford social and spatial transgression, which extends into and shapes a new character of public realm. Notes 1. The school does not exclude individuals, even in confining them: it fastens them to an apparatus of knowledge transmission […] an apparatus of normalization of people (Foucault 2000, 79). 2. Chosen in part because of Alkistis' familiarity with the city since she grew up there.
3. Traditionally understood as lower, middle and upper-middle income areas. However, it is important to note that in austerity Athens, one cannot make clear-cut categorizations of municipalities according to their socio-economic status. This is because of both the effect of the 'Vertical social differentiation' (Maloutas and Karadimitriou 2001) and the fact that economic austerity has minimized the previously extended middleclass, altering the socio-economic structure of society. Any differences observed between the case studies cannot be clearly connected to their socio-economic identity. 4. Although these cases were beyond the scope of this research, Alkistis' experience as a resident of Athens confirms this.

Disclosure statement
No potential conflict of interest was reported by the authors.