Housing Studies Vol. 19, No. 6, 875–892, November 2004
Fortress UK? Gated Communities, the Spatial Revolt of
the Elites and Time–Space Trajectories of Segregation
ROWLAND ATKINSON & JOHN FLINT
Department of Urban Studies, University of Glasgow, UK
[Paper first received 27 October 2003; in final form 9 June 2004]
ABSTRACT Anecdotal evidence suggests that ‘gated communities’ are growing in
popularity. This paper uses empirical evidence to profile the location and characteristics
of gated development in England and details the relative integration of residents. The
paper also attempts to think through the wider theoretical and urban policy impacts of
gating. In contrast to the view that gated communities provide an extreme example
of residential segregation we go further and argue that the time-space trajectories of
residents suggest a dynamic pattern of separation that goes beyond the place of residence.
Gated communities appear to provide an extreme example of more common attempts by
other social groups to insulate against perceived risk and unwanted encounters. Patterns
of what we term time-space trajectories of segregation can thereby be seen as closed
linkages between key fields, such as work and home, which enable social distance to be
maintained and perceived risks to be managed by elite social groups. We conclude that
gated communities further extend contemporary segregatory tendencies in the city and
that policy responses are required which curtail the creation of such havens of social
withdrawal.
K
EY WORDS: gated communities, segregation elites
Introduction
Gated communities have begun to creep on to the policy agenda in the UK.
Gated communities can be defined as walled and gated residential develop-
ments that restrict public access and these attributes have often sparked debate
about their desirability or legitimacy in an open society (a forthcoming special
issue of Housing Studies will be devoted to these debates early next year).
Existing accounts of gated communities (hereafter GCs) have considered their
internal socio-legal foundation (Blakely and Snyder, 1999), governance (McKen-
zie, 1995) and the impact on their residents (Low, 2003). However, in the UK
very little is known about them though there has been critical commentary on
their undesirability should they increase in number (Minton, 2002). In this
context this paper has two key aims. First, to broaden our knowledge of the
nature and extent of GCs in England and, second, to provide an early challenge
to the idea that GCs represent a communitarian ideal or private choice which
lacks wider social repercussions.
0267-3037 Print/1466-1810 On-line/04/060875–18 2004 Taylor & Francis Ltd
DOI: 10.1080/0267303042000293982
876 Rowland Atkinson & John Flint
Extreme segregation in US cities has often been seen as a social problem
through which social disintegration, low affiliation and social conflict have been
engendered (Fainstein et al., 1992; Massey & Denton, 1993). Segregation has been
seen as a product of both income inequalities and discriminatory filters that
allocate people in uneven concentrations across the city. Weaker income inequal-
ities and an urban history which has broadly lacked such racial overtones partly
explain why segregation is generally lower in the European context (Friedrichs,
1998) and why it has received less political attention. However, while segre-
gation is generally lower in European cities, some writers have argued that the
costs of segregation are pronounced for poorer households, including the
possibility that living apart produces preconditions for countercultures with
norms and values that do not emphasise work or education (Musterd &
Ostendorf, 1998).
The ‘area effects’ debate has generally been concerned with the additional
impacts on life-chances that poor residents face through living in poor areas
(Atkinson & Kintrea, 2002, 2004; Ellen & Turner, 1997). However, what is notable
about existing treatments of segregation and concentrated poverty is that there
has been almost no consideration of the social costs imposed through elite and
middle-class groups living in similarly ‘ghettoised’ concentration. This paper
raises this issue, suggesting a shift in the focus of studies of segregation and a
much greater consideration of the dynamic flows of everyday life both within
and outside the field of residential interaction and lived experience. In short,
segregation needs to be considered both in its daily dynamism as well as its
static residential manifestations.
Segregation and Seclusion in the City
While various ‘defended’ territories exist in cities such as gang ‘turf’, ethnic
enclaves, gentrified neighbourhoods and areas of religious significance, gated
communities provide a force for exclusion in new and different ways to earlier
forms of residential patterning. A long-standing interest in segregation (Peach,
1996) stems from a general presumption that it is a negative phenomenon but
what is the implication of segregation if this is extended and played out in the
movements and patterns of social interaction in the city? In other words, what
if segregation can be conceived of as having a space-time trajectory as well as a
residential element? We suggest that the impacts of GCs are far greater than the
fact that the affluent might simply live apart from a wider ‘mass’. The absence
of the affluent from many aspects of city living creates a wider problem for those
concerned with the quality of civic spaces and institutions.
In general cities characterised by impermeability and residential location
based on key attributes of social difference (most often income or ethnicity) are
contrasted with a notion of the ‘good city’ in which Bridge & Watson (2000) see
‘open, civilising and democratic possibilities’. Everyday exposure to difference
has been argued to be a key aspect of socialisation and to participatory modes
of governance (Sennett, 1970, 1977; Amin et al., 2000). However, more recent
commentaries on urban form, infrastructure and networking (Graham & Marvin,
2001) have described the connected seclusion of spaces provided for hyper-
mobile upper income groups in the global urban context.
Fortress UK? 877
Graham and Marvin argue that as fortified residential and work spaces are
connected by private roads, private shopping streets and malls, closed circuit
television (CCTV) and transportation, new secessionary spaces have been cre-
ated which allow unfettered movement of the affluent in which a:
set of processes are under way within which infrastructure networks
are being ‘unbundled’ in ways that help sustain the fragmentation of
the social and material fabric of cities. (Graham & Marvin, 2001, p. 33)
We argue here that this view of a ‘splintering urbanism’ is a useful lens to view
the changes engendered by an increasing number of gated communities in the
UK.
Gated communities suggest that residential segregation does more than sim-
ply lead to a withdrawal of certain groups into spatially fixed enclaves. The
paper examines the impacts for public service providers and policy-makers
related to gated developments. We also extend our analysis to consider the
potential for a dynamic, overlapping and fluid sense of segregation and its
impacts. In essence we argue that gated communities provide a refuge that is
attached to social networks, leisure, schooling and the workplace via paths
which are used to avoid unwanted social contact. Our argument is that each of
these spaces more or less segregates its occupants from social contact with
different social groups, leading us to suggest that the impact of such residential
division resembles a seam of partition running spatially and temporally through
cities, what we term time-space trajectories of segregation.
Auge (1995) has argued that non-places, including airports, supermarkets,
hotel lounges and shopping malls, provide a sense of travelling without moving
by virtue of the serial similarity of such spaces. Auge considers the example of
an anonymous businessman moving from his home to the airport, a large simple
building, from which he travels to another similar airport and stays in a hotel
much like those in his home town. Such non-places can also be seen as the
protected and detached world of international and local elites moving through
space without contact with other social groups.
Other conceptualisations may usefully be attached to those of Auge’s thesis.
Hannigan’s (1998) ‘fantasy city’ suggests that at the heart of current imperatives
of city development is the desire for experience without danger leading to the
creation of what Goldberger (1996) has termed ‘urbanoid’ spaces—spaces which
look ‘real’ but are devoid of the diversity that they used to support. It is now
possible for social factions to exercise unprecedented control over their experi-
ence of the city in terms of to whom, how and when social encounters are made.
Such trajectories of segregation extend patterns of residential separation so that
socialisation, work and leisure may be viewed as bounded activities that may
have negative impacts on wider social cohesion and our experience of diversity.
In short, how is empathy for inequality and social problems engendered if it is
never or rarely experienced?
Sennett (1970) and Lasch (1996) have both suggested that the city is the site of
a retreat by political and social elites away from the city into their own enclaves
and out of the social life of the city. Richard Sennett argues that:
the realm of the city where … dissonances and conflicts are played out
among strangers has been ‘abandoned’ to the middle and lower classes.
I use the word ‘abandoned’ because the signal feature of the new elite
878 Rowland Atkinson & John Flint
in these cities is that it has withdrawn from the public realm. (Sennett,
2001, p. 181)
Sennett argues that the ‘new’ elites do not seek to engage in the political and
social life of the city in ways that those of earlier cohorts did. The street and the
city, as an essential aspect of identity, he contends, are being eroded. If we
unpack this line of thinking further we may argue that while political and social
elites often still work and ‘play’ in the city their presence is less discernible and,
at times, almost invisible or highly secluded.
An important element of this withdrawal is a wider search for security (Low,
2003). As safety has increasingly shifted from being a social good to a com-
modity (Hope, 1995), differential access to resources contributes to security
differentials which manifest themselves spatially as elite groups seek to inhabit
‘security enclaves’. Such ‘enclaves’ offer security through ‘bubbles of gover-
nance’, in which security technologies are purchased and deployed around a
‘defensive exclusivity’ aimed at insulating communities from dangerous out-
siders (Reiner, 1992; Shearing, 1995; Crawford, 1997; Girling et al., 2000). Gated
communities appear as the crystallisation of such processes.
The paper is presented in three sections. The first details what we see as the
key theoretical hallmarks by which residential segregation and dynamic flows of
separation can be understood in the contemporary city. The second draws on the
empirical results of recent research on the numerical significance and location of
gated communities in England, supplemented by a discussion of their local
impacts. Finally this material is drawn out in its implications for understanding
a dynamic conception of segregation extending beyond residential location. We
conclude by considering the significance of these developments for those left
outside the gates and for policy-makers.
The Key Features of Gated Communities in England
Our operational definition of gated communities was as follows: Gated com-
munities are residential areas or a development that is fenced or walled-off from
its surroundings, either prohibiting or controlling access to these areas by means
of gates or booms. The concept can refer to a residential area with restricted
public access (other terms include—security villages, fortress neighbourhoods,
exclusive leisure developments and so on). The research comprised a postal and
telephone survey of all of the local planning authorities in England (383), which
achieved a 93 per cent response rate.
A series of key actor interviews were conducted with national housing,
builders, estate agents and developers’ representative groups, architecture and
planning bodies, local government organisations and other interested organisa-
tions totalling nine interviews. Finally, case studies were undertaken at ten gated
sites within four authorities identified through the postal survey returns. These
involved structured observation during visits and interviews with the following
in each area:
• Gated community tenant/resident association chair
• Representative of tenant or resident association (i.e. non-gated and adjacent
area)
• Representatives from the local authority:
Fortress UK? 879
—Planning
—Housing
—Environmental services
• Community police
• Developers of gated communities in the locality
• ‘Mystery shopping’ calls to sales offices (to establish envisaged community
profile and criteria for residency)
An analysis of the legal documentation for GCs from these developers was also
carried out. The case study areas were as follows:
1. Metropolitan borough—London borough of Lambeth
2. A large ‘provincial’ town—Oxford City
3. Commuter belt London—Runnymede, Surrey
4. A northern town—Chester le Street, County Durham
It should be noted that all of our case studies had gates and walls with punch
code and/or key entry systems. Only in the Surrey GCs did we find additional
guards while one of the Lambeth GCs had CCTV. Built form varied consider-
ably, three of the GCs were large converted homes while five consisted of
new-build development that had put gates in place. In Lambeth and Chester le
Street the GCs were not in suburban locations with some interviewees suggest-
ing that the security of gating helped to sell the developments in what were
perceived to be unsafe areas.
Through the survey research we identified around 1,000 gated communities in
England. These developments are generally small (most less than 50 units) but
are spread well across England, though they are clustered and more extensive in
the South East. A large number of factors were posited as explanations of gated
communities as places of residence amongst those seeking to purchase or rent
property. Case study interviews suggested that GCs are appealing to a diverse
spectrum of households. While a common characteristic of these developments
is the relative expense of acquiring a property within them, thereby resulting in
a predominance of wealthy individuals, the communities appear to have both
young professional and retired elderly residents and also a mix of single, couple
or family households.
Motivations to Move to Gated Communities
The perceptions of local authority officers, national housing organisations and,
to an extent, developers, attributed security and exclusivity as the two most
important aspects driving demand for gated developments. A number of plan-
ning officers identified a growing demand for what one termed ‘total and
absolute security’. This was regarded as being influenced by the US housing
market, and in response UK developments were being deliberately designed as
‘safe and predictable islands of safety’. One community police officer reported
that residents of the gated communities in his operational area regarded the
surrounding neighbourhoods as crime-prone localities, despite the fact that they
had very low crime rates. Significantly none of the national bodies we inter-
viewed had any clear policy position on GCs and were generally ambivalent
about their positive or negative qualities, beyond operational difficulties arising
from their lack of permeability.