Trajectories of middle-class belonging: The dynamics of place attachment and classed identities

Considering residential choice

Various considerations – practical and social – constitute residential choice. Among the practical considerations are the supply of suitable dwellings in a particular location and economic constraints, what is and is not affordable to the household. Neighbourhood selection may thus be bound by family circumstance (income, position in the life course) and need, particularly in relation to the size and type of property sought. It may also be influenced by proximity and/or ease of access to place of work for one or more household member.

The social dimensions of residential choice make explicit its relationship to social position (Savage et al., 2005). These include concerns over social reproduction as well as lifestyle, taste and aesthetics. In particular, it has been well-documented that schooling is often a key consideration within residential choice, with many middle-class families including considerations over proximity to high-performing schools within these even if this means significant trade-offs in relation to housing (Bridge, 2003). Similarly, the social environment offered by a neighbourhood may be considered on the grounds of whether it is an appropriate place to raise children (Halfacree, 1994; Karsten, 2003), a measure which often includes judgements on the social and ethnic mix of a particular location. In these respects, neighbourhood provides the context for the accrual of social and cultural capital, playing a role within the development of social identities (Forrest and Kearns, 2001).

Together with other factors – for example neighbourhood aesthetics and physical environment – such social considerations are part of the constitution of certain neighbourhoods or neighbourhood types as culturally significant to the middle classes. In this manner, particular imaginings – the ‘place-in-the-mind’ (Butler and Robson, 2003) that these occupy – are valorised and upheld by middle-class residents in the conduct of their daily lives and their rhetoric of residential choice (Benson and Jackson, 2013).

The value that households place on these different dimensions of residential choice is by no means homogeneous; it is by no means the expectation of all households that a neighbourhood fulfils their aspirations across the social fields of housing, education, consumption and neighbourhood. Indeed, there might be significant compromises and trade-offs made by families as they prioritise particular fields over others, constrained by their circumstances and the resources that they have available to them (Bacqué et al., 2013; Bertaux and Gotman, 1993; Bridge, 2003; Watt, 2009).

In this rendering, residential choice may be considered as a ‘function of the situation’ (Michelson, 1977: 362), not entirely satisfying long-term residential aspirations, but good enough for now. Furthermore, such priorities may change alongside transformations within the household, notably at times of family formation and retirement. As Bertaux-Wiame (1995) explains, residential choice should be considered as the interplay of resources, family capital and points of reference in terms of residential models, e.g. the types of neighbourhoods that they have previously lived in.

However, it is also clear that decisions about where to live are not only derived from rational calculations, but may also contain consideration over what particular environments might offer. Residential choice thus represents the coming together of spatial and social identities; considerations over where to live and which dimensions of residential choice to privilege are significant to social identity. It is not only the case that housing and residential environment are representative of the self (Rapoport, 1981), they also play a role within social identities. As Savage et al. (2005: 207, original italics) argue:

… residential space is a key arena in which people define their social position … [O]ne’s residence is a crucial, possibly the crucial, identifier of who you are.

Neighbourhoods offer opportunities in relation to the accumulation of various capitals, economic, cultural and social. Residential choice, as Savage et al. (2005) stress, reflects the lifestyles and consumption practices of middle-class residents, replacing the desire (or need) to live within close proximity to the place of work. Such opportunities may be rendered through the presence of particular institutions, for example schools, access to which, at least in the case of state education, is place-based, or through the other people who already live there, what Butler and Robson (2003) label ‘people like us’.

But it is also the case that the characteristics of a neighbourhood may elicit claims to distinction and thus be considered as a form of cultural capital; indeed, as May (1996) argues, the ability to live with difference within a multi-ethnic neighbourhood may be mobilised by some middle-class residents as a symbol of their distinctiveness (see also Jackson and Benson, 2013). As Butler and Robson (2003: 11) clearly demonstrate in their comparison of neighbourhoods in London, the exact offer of a neighbourhood may vary; in pursuit of particular residential aspirations, middle-class households deploy ‘stocks of capital – cultural, economic and social – in different configurations depending not only on their resources but also their aspirations for the kind of community they wish to live in’.

Beyond this, however, there is a need to bring together the different forms of capital that a neighbourhood offers, and convert these into symbolic capital. What this requires is that the capitals offered and accumulated by residents are recognised and legitimated by their peers. In this respect, neighbourhood – as with any other social fields – works to position people within the middle classes only when others confer their status (Bourdieu, 1984). This is indicative of the process by which middle-class claims to belonging are validated.

Belonging

The relationships that people have with their places of residence are often framed through the language of belonging, place affinity or place attachment. Such conceptualisations draw attention to the affective bases of people’s relationships to their residential environments (see Altman and Low, 1992; Shumaker and Taylor, 1983), the psychological bonds that people develop in relation to their places of residence.

As the literature on this theme demonstrates, such bonds may develop through long-term, perhaps lifelong residence, but may also be present in the absence of this. Reflections on place affinity and attachment nevertheless emerge as central to processes of identification. For example, Feldman (1990, 1996) argues that belonging may be maintained through the ‘continuity of residential experience’ (1990: 186), irrespective of residential mobility; in other words, belonging emerges through identification with a particular type of residential environment, the development of ‘bonds that relate the identity of a person to the identity of a type of settlement’ (Feldman, 1990: 222). In other words, experiences of past residential environments generate a preference for particular types of neighbourhood (see for example Blaauboer, 2011; Feijten et al., 2008) and the meaning that its residents place upon it (Clark, 2009).

This demonstrates that residential environment has a role to play within residential choice. Although this account highlights that continuity may be significant within residential mobility, it does not give a sense of the importance of scale, whether the settlement types that individuals feel a particular affinity for are on the level of street, neighbourhood or city. In addition, this approach does not reflect on the relationship between identity and belonging.

Bourdieusian approaches to belonging allow for the closer examination of how identities inform residential choice and are reconstructed through these. In particular, studies adopting this approach have emphasised the relationship between place attachment and classed identities (notably Butler and Robson, 2003; Savage et al., 2005). In these understandings, residential mobilities are telling of the processes by which identities are created spatially, shedding light on how individuals negotiate, context and position themselves at the intersections of geographical and social space (Clark, 2009; see Bourdieu, 1984).

Similarly to Feldman (1990, 1996), Savage et al. (2005) position residential choice within the contexts of personal biographies, demonstrating how people present such choices as ‘congruent with their lives’ (Savage et al., 2005: 203), confirming ‘a sense of who they are’ (Savage et al., 2005: 53). The focus on the middle classes reveals their capacity to choose locations where the available fields – housing, education and consumption infrastructure – fit as closely as possible to habitus, resulting in a sense of belonging. They use the term ‘elective belonging’ to denote the moral ownership over place that the middle-class residents claim through their ability to choose. This apparent freedom to choose, even when constrained, is a marker of class difference and middle-class status (see Savage, 2000; Skeggs, 2004). As Savage (2010) stresses, this form of belonging is distinct from ‘dwelling’ in place; the focus of these residents is on living in a suitable environment for ‘people like us’, privileging the symbolic meanings of their place of residence. If your habitus does not match up to the available local fields, then one option – given access to the right resources – is to move.

Butler and Robson (2003; see also Butler, 2002) adopt a similarly Bourdieusian framework in their account of the ‘metropolitan habitus’ – the predisposition for living in a global city with global connections. This draws attention once again to the proclivity for particular types of residential environment and their offerings. The questions of scale that I found unanswered in Feldman’s (1990, 1996) work are in part resolved by Butler and Robson’s (2003) recognition that within the ‘metropolitan habitus’ lie mini-habituses, that represent affinities for particular neighbourhoods; ‘important distinctions can be drawn amongst those living in the city, for whom different areas take on different meanings and associations that attract potential residents and then act on those who are settled there’ (Butler and Robson, 2003: 9). This model for understanding belonging means that residential choice can be understood both in terms of a taste for urban environments as well as at the level of particular neighbourhoods within the city, taking account of the opportunities/constraints that these offer in terms of ‘fit’ across a variety of different social fields.

As Bridge (2003) reveals in his account of onward residential mobility of gentrifiers, belonging needs to be plotted in relation to the changing priorities of households. For example, habitus may adjust to prioritise and accommodate the requirements of family members in relation to education, resulting in the development of new preferences for residential environments that are seemingly at odds with prior choices. Watt (2009) further highlights the importance of space in his account of selective belonging, demonstrating how spaces of belonging may be delimited and localised by middle-class residents to exclude areas of lower (cultural) value. In this manner, they claim moral ownership over their places of residence by investing them with positive value. In this manner, both Bridge (2003) and Watt (2009) give a sense of belonging as in process across time and space.

This paper builds on these Bourdieusian understandings of residential choice and mobility among the middle classes, to stress that belonging results from the dynamic interaction between the neighbourhood (field) and habitus (see Callaghan, 2005) (de)generated as the individual moves through time and space. It takes as a starting point the recognition that habitus is in process, mutable and adaptable (Bourdieu, 1977, 1992; see also Hillier and Rooksby, 2005) to argue that residential mobility offers the possibility of transforming habitus while also being structured by it, and highlighting the interactions of the social and the physical environment within this process.

Neighbourhood ‘fit’

As I demonstrate in this section, many respondents in the study, when asked about why they had chosen to live in their current neighbourhood, presented it in an almost matter-of-fact way: this was the type place that people like them lived or that they had always lived in neighbourhoods of this type. In many ways this reflects what Ærø (2006: 114–115, emphasis added) refers to as an ‘innate disposition of place … Here the person does not choose, because where to live is a matter of convention and tradition’. The choice over where to live appears to be remarkably unreflexive and it is almost second nature to live in a particular type of residential environment. Indeed, in support of these points, respondents often linked their choices to prior residential experiences, highlighting their histories of living in similar environments as a way of supporting their claims to having the knowledge of how to live in their current neighbourhood (see also Blaauboer, 2011; Feijten et al., 2008).

Among respondents in the commuter belt, there was a repeated suggestion that experiences of growing up in the countryside had given them a taste for the rural. It was common for them to refer to themselves as ‘country people’. They valued the proximity to nature, the ability to walk from your door into the open countryside, and being able to see the fields and woodlands from the window. Their lifestyle preferences also reflected this, as they recalled being ‘outdoor people’ who enjoyed walking in the Surrey Hills and who were keen to participate in local activities and organisations. As Emma, the mother of two pre-school age children, residential choice was a case of ‘following patterns that you always followed’. ² This was representative of the way that many of our respondents in this environment articulated their residential choice; the taste for living in the countryside was presented as being deeply ingrained, and was largely unreflexive and, to a certain extent, taken for granted.

Similarly, respondents in Peckham and Balham described themselves as ‘city people’, in this respect, giving a sense of the metropolitan habitus (Butler and Robson, 2003):

I come from … I’m sort of a city girl … we’ve never thought, ‘my god, I’ve had a child. we’ve got to move to the far suburbs or the country because they won’t be safe here’. In fact, we’ve always liked the city thing and we’re kind of proud to have our children in the city … I just remember when my kids were at nursery school, sort of reception class and things at Clapham Junction, lots of their friends were disappearing all the time … and they would go out to the suburbs – partly to get a bigger home … but it was also that somehow it [London] was about a bad place to be. (Hilary, Balham resident)

These narratives demonstrate an affinity and commitment to a particular type of residential environment, broadly ‘country’ and ‘city’. Such accounts therefore illustrate how changing residence, following Feldman (1990), can be understood as ‘continuity of residential experience’ (Feldman, 1990: 186) whereby individuals identify with a particular type of residential environment.

The lack of reflexivity within these residential trajectories was described clearly by Bill, an American artist aged in his 50s, who had lived in London for a long period of time:

I have always lived in London, and I have followed the pattern of almost every other Londoner, I’ve always lived in the South – or most of Londoners who live in the North always live in the North, so I’ve lived in Clapham, I’ve lived in Lewisham, I’ve lived in Kennington, and now I live in Peckham.

This gave a sense of long-term residence and experience inscribed on the body, such that residential preference had become a habit, an unquestioned part of his habitus, reproduced through his repeated residential choices.

Other respondents additionally stressed that there were certain normative residential practices that were common among ‘people like us’, drawing attention to the shared nature of this feel for place and particular residential trajectories through the city. Nowhere was this sense more clear than in Balham, where respondents had relatively local circuits of residential mobility: 28 of the 38 respondents originated in the south of England, and had lived predominantly in neighbourhoods in southwest London such as Putney, Clapham and Fulham. This sense of trajectory was made particularly clear by Laura, a lawyer aged in her 30s, who identified a ‘natural’ residential trajectory that people like her follow, ‘the sort of Clapham-ish area is more the sort of Home Counties girls like us that grew up in Surrey and moved to London, so, in a weird way, we fit this’.

This sense of a normative trajectory was also prominent within the accounts of some exurban residents, as though moving out to the countryside was a longstanding aspiration and social achievement:

I’d moved down south 25 years ago, and lived in Hertfordshire, then moved into London, and then gradually as people tend to do, is you find a partner, you live in London and like the bright lights, and then you gradually move further out into the rural areas. (Tom, Effingham)

This habitus for a particular trajectory – moving from the city to the countryside – equally framed ideas about future residential mobility, clear among Balham residents, with predictions of further movement west and south coinciding with their children growing older.

These residential moves provide some insights into the time–space trajectories of these middle-class individuals (see Bridge, 2003). Beyond this, however, they illustrate clearly how residential mobilities are explained through the value and hence meaningfulness placed on particular neighbourhoods (or residential environments) and residential trajectories by certain middle-class groups (Butler and Robson, 2003). Culturally valued by these middle-class groups, the ability to live in such neighbourhoods can be considered as a source of symbolic capital and is therefore significant to social status and identity. The ability to move through these trajectories is made possible by access to resources, in particular economic capital from working within particular occupations, but also from prior property investments.

In Peckham, the most ethnically and socially mixed of the neighbourhoods in the study, neighbourhood fit was articulated by drawing attention to previous residential practices, with respondents either highlighting how they had lived in other ethnically mixed neighbourhoods in London or had lived abroad, specifically Africa and the Middle East, often working in charity and non-profit sectors (see also May, 1996). It was particularly notable that these respondents stressed that they felt ‘at home’ in Peckham precisely because of the local ethnic mix, which, not unproblematically, they claimed reminded them of their previous residential experiences:

I’m very happy because of course … it’s rather like being back in Yemen or Afghanistan! (Peter)

I’d spent a year living out in Africa … so for me it was quite entertaining to come here and go ‘oh, this is just like Africa’. (Henry)

Respondents thus presented their prior residential experiences as equipping them for living in a multi-ethnic neighbourhood. It should be noted, however, that while on the surface this identification with the multi-ethnic dimensions of the neighbourhood appears to indicate continuity with past experience, such continuity depends upon experiences of migration. An earlier experience of changing habitat brings with it the possibility of having changed habitus as the individual responded to their changed environment (Friedmann, 2005). In this case continuity is made possible by the adaptability of habitus to changed physical and social environments.

Drawing out the continuities with their past residential experiences, whether this related to residential environments, trajectories or the ability to live with difference, respondents presented the fit between habitus and their place of residence and claimed belonging. Presenting residential choice as almost second nature, these narratives reveal what Bourdieu and Wacquant (1992: 19) describe as ‘prereflexive, infra-conscious mastery’ of the field, whereby the practices of these middle-class residents maintain their (social) position without conscious effort.

Residential trajectories and social mobility

The following examples make more explicit the relationship between residential and social trajectories, and thus draw attention to the spatial contexts of social status. On the one hand, it becomes clear that reflections on the current place of residence narrate a sense of social mobility – look at where we came from and where we have ended up. On the other hand, residential trajectories may map a sense of ambivalence when people realise that their current residential circumstances are at odds with their long-term residential aspirations. In contrast to the accounts presented in the previous section, the reflexive residential biographies presented here demonstrate the entanglements of residential mobilities in the making and remaking of classed identities, reflecting the process of becoming or staying middle class.

The idea that a neighbourhood was ‘more upmarket’ than where they had come from was a central way in which respondents articulated the sense that their residential mobility reflected a change in their classed identity. This was as true in the gated community of Oak Tree Park as it was in Berrylands, a quintessential 1930s suburb.

I think it was that it was more upmarket than where I’d come from, and I was quite pleased about that I suppose, because you felt like you were making progress … Sunbury, there it was a development of 12 identical houses … and here we had a house that was different to everybody else’s, that we’d chosen to build … Yes, I felt we were on our way. (Wendy, Oak Tree Park)

I knew Berrylands for being a very posh area …so I never ever dreamt I would be able to maybe even live in Berrylands… I’ll be a posh woman, be a posh middle-class person living in Berrylands! (Joy, Berrylands, emphasis added)

For respondents in Oak Tree Park, rather than conforming to the middle-class backgrounds that their current assets might indicate, respondents here came from diverse class backgrounds, often explaining how they had ‘made it’ through their employment trajectories, with their properties in the Park mobilised to support this. The exclusivity of the neighbourhood and the distinctiveness of the housing (no two houses within the Park were the same) were presented as further confirmation of their social mobility.

In Berrylands, the story was somewhat different. Respondents equally drew on their residential histories to support the claim to social mobility, but these were often remarkably unassuming. Berrylands was presented as ‘more upmarket’ with better quality housing and larger gardens than they could previously have afforded. As Joy makes clear, in her mind it had been inconceivable that she would be able to live in such a neighbourhood, demonstrating that this had exceeded her residential aspirations. Andrea, in her 40s and married with one daughter of primary school age, similarly expressed this sense; as she explained, Berrylands was posher than the suburb in the Midlands that she had grown up in, a place in which she felt she would no longer be able to live. Therefore, her experiences of living in this residential environment seemed to indicate a change in her habitus.

However, Andreas’ evaluations of the neighbourhood were overshadowed by the fact that her parents-in-law felt that their son had traded down in terms of neighbourhood:

[Husband’s] family, they’re in a neighbourhood which is even posher. They say, ‘Oh dear it’s not a very good address, surely you could trade up?’ And we go, ‘But why? There’s only three of us, we’re not going to need any more space, we’re close to the school’. ‘Well when [daughter] goes to senior school perhaps you can move to a better address’.

As this example demonstrates, discussions of how residential histories reflect upward social trajectories were not the only discourse at play within the respondents’ narratives. The following examples demonstrate a sense of ambivalence about what residential trajectories reveal about their social mobility. It is clear therefore that middle-class residents are concerned about what people will understand from their residential choices, highlighting an acknowledgement of the relationship between social and geographical space (see Burrows and Gane, 2006), and how cultural and/or economic capital are conveyed in neighbourhood and housing selection.

Indeed, in Berrylands, there were several respondents who seemed somewhat bemused about how they had ended up there, articulating a sense that this residential move was at odds with their habitus. Tim, aged in his 50s, married with three children, explained that before they had come to look at the house 18 years ago, he had had a preconception of the suburbs as not being a place that he particularly wanted to live in:

I think I was privately dreading moving into this house. I remember seeing the advert which [wife] was very keen on … and I remember looking at the picture of the house and it just looked so suburban I just thought it would be awful moving into a mock Tudor. So I wrote burble saying no not this one, let me out, this is so boring …

Living in Berrylands still did not quite fit to his habitus, even though he had reconciled himself to this choice. The mock Tudor house was mobilised as a representation (contra his childhood home) of how his home was not a marker of his success, and placed him ambiguously within the middle classes. As he recalled his childhood home in the Kent countryside, he explained:

… it probably never really occurred to me that I wouldn’t live somewhere similar … I probably had always assumed that I’d end up in something like that, forgetting the bit about having to work hard and be successful to pay for it all. So I think it’s probably taken me a little while to realise that this is it, this is where we live. This is fine.

As with Bridge’s (2003) respondents in Bristol, living in Berrylands was a compromise. Despite his habitus not quite fitting to the locally available field of housing, it allowed him to privately educate his children, thus privileging social reproduction. However, it was also the case that the neighbourhood was affordable and within easy reach of his place of work. In this respect, Tim’s account reveals how complex decisions over where to live can be, particularly in cases where constraints on resources might lead to the prioritisation of particular social fields.

For other respondents, such as Emily, a nurse aged in her 30s who was renting a flat in Peckham, the lack of fit between the current place of residence and her sense of where she had grown up (and therefore would fit), gave way to the sense that her residential trajectory would in time lead her back to the type of environment that she had grown up in:

I sort of grew up in a particular place, and I think eventually I sort of find myself wanting to go back to that … but it was countryside and it was … a village really. With a close community … I think I just miss the idea of communication being easy and not having to plan lots of things in advance, everything can be a lot more spontaneous … kind of just feeling more ownership of your surroundings is important.

This highlights her emotional attachment to a different type of residential environment, in particular the ease of social relations. What this demonstrates is that current residential choices may not completely match residential aspirations. This might be the result of particular constraints on residential choice (Bacqué et al., 2013; Bridge, 2003). It might also be, as Michelson (1977) reminds us, that these are long-term aspirations to be fulfilled at some stage in the future; more to the point, the fact that these have not yet been realised does not necessarily give rise to dissatisfaction.

Transforming habitus through residence

There was an additional possibility that in cases where habitus did not match to the neighbourhood, residents may adapt their habitus to the place of residence. This is a converse of what I have shown elsewhere, where the lack of fit between neighbourhood (as field) and habitus, may drive active interventions as respondents attempt to shape the neighbourhood in their own image (i.e. to make it fit to their habitus) (see Benson and Jackson, 2013), the success of which is dependent on the position of these middle-class residents within the social field of the neighbourhood.

This reconciliation might require the subtle adjustment of pre-existing knowledge to match the particularities of the neighbourhood, while in others it is a steep learning curve. Living in a particular environment and engaging with it, can thus enable the generation of a combination of embodied and objectified cultural capital:

… if you’ve not been brought up in the country per se, is, I wouldn’t necessarily call myself a real nature lover in terms of I don’t really understand nature, because I’m not really educated in it in that respect, but [son] and I particularly love the nature. We have deer that come in, we’ve had owls that have been here … last year and they’re just back, which we’re delighted about, and I’ve just been trimming up the lavender for the bees next year, and all that sort of thing. We really love that, and I think after you’ve been here a few years you really get to appreciate that, that nature that’s really on your doorstep, and I think it helps having children because they like it, you can engage them in that sort of thing, and hopefully nurture them a little bit in those ways. (William, West Horsley)

The knowledge of how to value the neighbourhood (in ways that match up to those of other middle-class residents) and how to live within it can position people within the social field. The lack of such knowledge in others, in a field characterised by processes of social distinction may thus be presented as evidence of lack of fit, social and spatial identities mismatched, and hence questions over belonging (see Benson, 2011). William’s account demonstrates how through bringing his son up in this environment he hoped to give him the opportunity to develop this knowledge so that it became second nature; an opportunity that he had not had. However, as William reveals, the experience of living in a neighbourhood may result in developing a feel for place. As Feijten et al. (2008: 142) argue, ‘… having lived in a place may also change the awareness of and attitudes towards the type of residential environment it offers’.

In this sense, by living in a neighbourhood, individuals engage in a process by which they come to embody understandings of how to live in a neighbourhood, becoming attuned to living in those environments in ways that are, nevertheless, mediated by imaginings and representations of place. This can be understood, in Bourdieu’s (1992) terms, as an emerging logic of practice. For example, Sarah, who had been living in Effingham for 22 years explained ‘… one thing I found really difficult to get used to, to start with was the fact there are hardly any streetlights here’. As she continued later in the interview, ‘we walk the dog at night with torches … you just accept it as part of life really, part of the way it goes’, adjusting her habits to the new environment.

This notion of learning the neighbourhood and how to live in it was also clear among some respondents in Peckham who drew attention to how their initial sense of the wider area as threatening and overwhelming had given way to a more nuanced understanding of local life:

I have been very surprised by how I have now turned my way of thinking completely around… I’d be a bit wary of going round the back streets at night time. There are a lot of men who hang around, they’re African men and they are hanging around … and I just realised … it’s just ordinary people having a nice time. I could feel a bit threatened by all that … Peckham has surprised me, I mean it’s no more dodgy than anywhere else actually … I mean everyone’s just going about their normal business getting on with their lives … (Linda, emphasis added)

This process of generating a feel for place did not happen overnight; it could take a long time. Linda, a divorced woman aged in her 50s had additionally explained how her relationship with London had changed over time; as a lonely young mother, she had hated London but after 30 years of living there, she had come to love it. The quotation above shows that she had developed a deeper understanding of her neighbourhood through living in it, valuing the social mix on offer in a multi-ethnic neighbourhood.

Certainly for long-term white middle-class residents in Peckham, whose choice had been framed by a range of factors that included the low-cost of property in the area, there was a commitment to living with difference and hopes for social equality at the level of the neighbourhood. Such accounts of residential choice resonate with the educational choices of the middle classes reported by Reay et al. (2011). As these authors stress, the new middle classes reject the schooling systems of the established elite, choosing to educate their children in local state schools. Nevertheless, it becomes clear that in cases where progressive politics guide these choices, middle-class privilege is not overcome through school choice. Ingram (2012: 309) in her review of Reay et al. (2011) refers powerfully to this as the ‘(im)possibility of ethical school choice’.

Although progressive politics may frame some middle-class residents’ aspirations, ethical residential choice within a global city marked by ever-increasing property values seems as equally impossible as ethical school choice. As more and more middle-class households move into the neighbourhood, drawn to its proximity to other desirable locations, the neighbourhood as social field is gradually restructured, with longer-term residents who maintain a value for social mix left holding a smaller stake within the resulting power struggles.

What this demonstrates is that over the course of residence, the rules of the game might change. This is perhaps most well-documented in the case of gentrification, whereby working-class populations are displaced/replaced by middle-class gentrifiers (Glass, 1964). The case of the neighbourhoods in the study shows that shifts in the population of a neighbourhood can lead to a sense of lack of fit among long-term residents.

This is most evident in the case of long-term residents of Oak Tree Park, who find the ongoing shift from residential park to gated community disconcerting. As they regularly cite, the changes to the neighbourhood wrought by the development of large houses and the installation of gates on individual properties are at odds with their reasons for choosing to live in the park and their vision of the neighbourhood. Across all neighbourhoods, the repeated emphasis from long-term residents that they would not be able to afford to buy a property in the neighbourhood today, is one indication of how social fields – in this case field of housing – may change over time. Within this altered field, long-term residents may find themselves displaced from their previous position of privilege, and losing some of their symbolic power within this field.

It is also clear that people might not want to generate a feel for place, and maintain instead ambivalence about their neighbourhood. This is particularly so in cases where residents do not feel that the neighbourhood adequately reflects their sense of themselves.

Colin and Susan, a couple with two young children who lived on the edge of Berrylands, explained how they had initially moved to the area as a temporary measure while they got to know London a bit better. They had now stayed there for two and a half years. On the one hand, the neighbourhood offered good schools for their children and garden space that they would not otherwise have been able to afford in London, a factor that they were not prepared to compromise on. On the other hand, they described moving out to the suburbs as ‘a bit weird and a bit unsettling’, their relationship to the neighbourhood presented as ‘fragile’. They were not surrounded by like-minded people and felt that this was not the neighbourhood where their ‘tribe’ lived nor where the consumption infrastructure that they valued was located. While they felt that they were doing the best for the children in relation to social reproduction, and they had found a property that fit closely to their housing aspirations – unlike many of our respondents in Berrylands, which is known for this 1930s aesthetic, Colin and Susan lived in a flat in a Victorian conversion – they did not believe that the neighbourhood represented their social and indeed classed identity.

Contra Michelson (1977), Colin and Susan were not satisfied for now with their residential choice but neither were they in the position to move to an environment that better ‘fit’ their habitus and residential aspirations. This is a somewhat unusual case that demonstrates some people lack a sense of belonging to their place of residence, expressing ambivalence in its place. Of course, it is possible that the adaptation of habitus documented above has just not taken place yet because of the short period of their residence in the neighbourhood. Another interpretation might be that there are no guarantees that by living in a neighbourhood, belonging may be generated. In this respect, the (re)generation of habitus to fit to place may not always be achieved.