What is a city?

Pile, S., 1998. What is a city? In D. Massey, J. Allen, & S. Pile, eds. City Worlds London:
Routledge. (page 1-19)
The defining of a city is subjective—could be attracted to its offers or wanting to get away from it e.g.
how holidaymakers choose to advertise cities.
Features of the city can also be found outside of the city—but smaller in scale; is size distinctive to
what cities are?
A particular combination of elements to be a city?— very few cities shave the same mixture of
elements/features; something can be a city feature yet not all cities will have it and it doesn’t define
exactly what a city is.
“We tend to conflate the physical and the human aspects together and, while we can say with some
confidence what we mean by ‘urban’ in physical terms, it is much more difficult to spell out its social
significance.” – Robson, 1975, p.184
City has social significance:
Tangible aspects vs. intangible aspects—
Intangible qualities of city life that makes it distinctive e.g. luxury and poverty, amenity and pollution,
tradition and innovation, order and disorder, volatility and conflict, difference and indifference, public
services and welfare provisions, freedom and dependency (tensions).
Experiences of the city will differ—perhaps due to location.
Cities exaggerate relationships—more intense activities.
Cities contain different physical features and experiences—stories to tell.
Howard’s plan of the city separated different aspects of the city rather than bring them together.
Mumford; city as a social institution—forms of social interaction that occur in cities cannot be found
anywhere else.
Mumford; city is a “geography plexus”—networks within and outside of the city.
City as a personal and social drama—like nowhere else can bring people together. But as strangers
accumulated, cities became areas of dissipation.
1. Cities have something to do with the kinds of association that form between individuals and
groups within the city. What is distinctive about urban life is its vitality, vibrancy, creativity,
novelty and intensity.
2. The form of urban association change with ‘nearness’— the ways in which social relations
are made through proximity and distance, closeness and remoteness.
3. Individuals, groups and ways in which the urban institutions operate are all significant.

The role of race and class in urban marginality

In the early 1990s, when ‘quartiers sensibles’ (at-risk neighbourhoods) came to the forefront
of public debates, sociologist Pierre Bourdieu made the following observation about the
problem of scale in social science research:
‘The perfectly commendable wish to go see things in person, close up, sometimes leads
people to search for the explanatory principles of observed realities where they are not to be
found (not all of them, in any case), namely at the site of observation itself. The truth about
what happens in the “problem suburbs” certainly does not lie in these usually forgotten sites
that leap into the headlines from time to time.’ (Bourdieu, 1999, p. 181)

But even though Wacquant acknowledges the reality of discrimination in France,

discrimination almost completely disappears in his description and analysis of the
marginalization of French banlieues. Implicitly, the logic seems to be that if ethnic lines did
not create a ghettoization comparable to that of the USA, then ethnicity and ethnic
discrimination played no structural role at all. In fact, a major part of Wacquant’s
comparative analysis is focused on geographic stigmatization, ‘the single most protrusive
feature of the lived experience of those entrapped in these sulfurous zones’ (p. 169), rather
than on ethnic stigma. Likewise, he suggests that ethnic discrimination plays a minor role
compared to age (specifically, youth): ‘age tends to trumps ethnicity in the Red Belt’ (p. 193).
Wacquant’s claim that stigma is attached to age and place of residence, and to a much lesser
extent to ethnic characteristics, results in a dramatic understatement of the experience of
racial discrimination in France.

Heterogeneity measured by nationality is an indubitable fact (far less indubitable than the
supposedly ‘well-documented frictions’ (p. 152) Wacquant asserts are created by this
heterogeneity). But this assertion is highly dependent on the categories used to measure the
population. If we speak about ‘non-whites’, homogeneity becomes a striking feature of the
banlieues. In France, no ethnic statistics exist to measure a ‘non-white’ population, and the
debates about the relevance and dangers of such statistics are ongoing. This review is not the
appropriate venue for entering the discussion; suffice it to mention the marked and increasing
interest in France in the emergence of an identity based on the perception of being rejected as
‘non-white’, that is, a ‘racialized’ population predominant in the banlieues.

Describing this process, Wacquant gives a rather rosy gloss to the history of public housing
agencies; in his view, the officials ‘have worked diligently […] to prevent the constitution of
these notorious “immigrant ghettos” […] thorough a meticulous management, […] aiming to
scatter foreign tenants and so-called problem families’ (p. 154). In fact, this concern for
‘scattering’ was also driven by colonialist ethnic categories. Moreover, public housing
agencies were willing to revise their view of ethnic concentration as a problem if such
populations could be collected in the most dilapidated buildings of the least prosperous
neighbourhoods.

Because of the way new spatial categories were introduced, immigrants and residents of the
suburbs were never described in terms of the racism and discrimination they faced, excluding
them from participation in the economy; instead, they were considered in terms of the
problems they supposedly posed to the entire society, because of their insurmountable
‘difference’. For Wacquant, ethnicization is the erroneous importation of foreign categories,
those of race.
He often speaks of the ‘Red Belt’ and ‘working-class territories’ or ‘(post industrial)
working-class territories’. ‘Red Belt’ is the translation of ‘banlieues rouges’ or ‘ceinture
rouge’, expressions that designate cities run by local Communists, with a high percentage of
public housing and a large influx of immigrants after World War II, as well as a strong
workingclass culture and a high degree of classconsciousness. The term ‘Red Belt’ nicely
contrasts with ‘Black Belt’, underscoring the differences between the two countries.

Urban Marginality in the Coming Millennium

Along with the accelerating economic modernisation caused by the global restructuring of
capitalism, the crystallisation of a new international division of labour (fostered by the frantic
velocity of financial flows and workers across porous national boundaries) and the growth of
novel knowledge-intensive industries based on revolutionary information technologies and
spawning a dual occupational structure, has come the modernisation of misery— the rise of a
new regime of urban inequality and marginality. (For a fuller argument, see Wacquant,
1996a.)

Four Structural Logics Fuel the New Marginality:

1. The Macrosocial Dynamic: The Resurgence of Social Inequality

Post-industrial modernisation translates, on the one hand, into the multiplication of highly
skilled positions for universitytrained professional and technical staff and, on the other, into
the deskilling and outright elimination of millions of jobs for uneducated workers (Sassen,
1991; Carnoy et al., 1993). What is more, today, jobless production and growth in many
economic sectors are not a utopian possibility but a bittersweet reality. Witness the virtual
emptying of the harbour of Rotterdam, perhaps the most modern in the world and a major
contributor to the rise of unemployment in this Dutch city to above the 20 per cent mark. The
more the revamped capitalist economy advances, the wider and deeper the reach of the new
marginality, and the more plentiful the ranks of those thrown into the throes of misery with
neither respite nor recourse, even as of®cial unemployment drops and income rises in the
country.
Put differently, advanced marginality appears to have been `decoupled’ from cyclical ¯
uctuations in the national economy. The consequence is that upswings in aggregate income
and employment have little bene® cial effect upon life-chances in the neighbourhoods of
relegation in Europe and the US, while downswings cause further deterioration and distress
within them. Unless this disconnection is somehow remedied, further economic growth
promises to produce more urban dislocation and depression among those thrust and trapped at
the bottom of the emerging urban order.

2. The Economic Dynamic: The Mutation of Wage Labour

The new urban marginality is the by-product of a double transformation of the sphere of
work. The one is quantitative and entails the elimination of millions of low-skilled jobs under
the combined press of automation and foreign labour competition. The other is qualitative,
involving the degradation and dispersion of basic conditions of employ- ment, remuneration
and social insurance for virtually all but the most protected workers.
First, a signi®cant fraction of the working class has been rendered redundant and com- poses
an `absolute surplus population’ that will probably never ®nd regular work again.
Secondly, and more importantly, the character of the wage±labour relation itself has changed
over the past two decades in a manner such that it no longer grants foolproof protection
against the menace of poverty even to those who enter it.
In short, where economic growth and the correlative expansion of the wage sector used to
provide the universal cure against poverty, today they are part of the malady.

3. The Political Dynamic: The Reconstruction of Welfare States

For, alongside market forces, welfare states are major producers and shapers of urban
inequality and marginality. States not only deploy programmes and policies designed to `mop
up’ the most glaring consequences of poverty and to cushion (or not) its social and spatial
impact. They also help to determine who gets relegated, how, where and for how long.
States are major engines of stratification in their own right and nowhere more so than at the
bottom of the socio-spatial order they provide or preclude access to adequate schooling and
job training; they set conditions for labour market entry and exit via administrative rules for
hiring, ®ring and retirement; they distribute (or fail to distribute) basic subsistence goods,
such as housing and supplementary income; they actively support or hinder certain family
and household arrangements; and they codetermine both the material intensity and the
geographical exclusivity and density of misery through a welter of administrative and fiscal
schemes.

4. The Spatial Dynamic: Concentration and Stigmatisation

In the post-war decades of industrial expansion, poverty in the metropolis was broadly
distributed throughout working-class districts and tended to affect a cross-section of manual
and unskilled labourers. By contrast, the new marginality displays a distinct tendency to
conglomerate in and coalesce around `hard core’ , `no-go’ areas that are clearly identified—
by their own residents, no less than by outsiders—as urban hellholes rife with deprivation,
immorality and violence where only the outcasts of society would consider living.
Brixton in London, these entrenched quarters of misery have `made a name’ for themselves
as repositories for all the urban ills of the age, places to be shunned, feared and deprecated. It
matters little that the discourses of demonisation that have mushroomed about them often
have only tenuous connections to the reality of everyday life in them. A pervading territorial
stigma is ®rmly af®xed upon the residents of such neighbourhoods of socioeconomic exile
that adds its burden to the disrepute of poverty and the resurging prejudice against ethnic
minorities and immigrants (an excellent analysis of this process of public stigmatisation is
offered by Damer, 1989, in the case of Glasgow).
Now the neighbourhood no longer offers a shield against the insecurities and pressures of the
outside world; it is no longer a familiar and reaf®rming landscape suffused with collective
meanings and forms of mutuality. It turns into an empty space of competition and con¯ ict, a
danger-®lled battleground for the daily contest of survival and escape. This weakening of
territorially based communal bonds, in turn, fuels a retreat into the sphere of privatised
consumption and strategies of distancing (`I am not one of them’ ) that further undermine
local solidarities and con® rm deprecatory perceptions of the neighbourhood.
 There is no Turkish ghetto in Berlin, no Arab ghetto in Marseilles, no Surinamese
ghetto in Rotterdam and no Caribbean ghetto in Liverpool. Residential or commercial
clusters fuelled by ethnic af®nity do exist in all these cities. Discrimination and violence
against immigrants (or putative immigrants) are also brutal facts of life in all major urban
centres of Europe (Wrench and Solomos, 1993; BjoÈrgo and White, 1993). Combined with
their typically lower-class distribution and higher rates of joblessness, this explains the
disproportionate representation of foreign-origin popula tions in urban territories of exile. But
discrimination and even segregation are not ghettoisation. Such immigrant concentrations as
exist are not the product of the institutional encasement of the group premised on rigid spatial
con® nementÐ as evidenced by rising rates of intermarriage and spatial diffusion when
education and class position improve (Tribalat, 1995). Indeed, if anything characterises the
neighbourhoods of relegation that have sprouted across the continent as mechanisms of
working-class reproduction have ¯ oundered, it is their extreme ethnic heterogeneity as well
as their incapacity to supply the basic needs and encompass the daily round of their
inhabitantsÐ two properties that make them anti-ghettos

Harding, A. & Blokland, T. (2014) Urban Theory London: Sage

The Oxford English Dictionary defines urban as ‘of, living in, or situated in, city or town’.
A city, meanwhile, is either an important town or a town awarded privileged status by charter,
especially if it also contains a cathedral (although, just to confuse mat- ters, not all cities have
cathedrals, nor have all towns with cathedrals become cities!).

Due to technological developments, suburbanisation has enabled the affluent to move to the out-
skirts and find convenient ways to commute to places of work in town. Household staff have been
replaced by electrical equipment, from washing machines to security systems, and private daycare
centres have replaced indi- vidual nannies, segregating the lives of rich and poor, and those serving
the rich no longer live in the attics and basements of their houses.

Since roughly the 1980s, a revival of inner-city living can be seen. This process of hollowing out inner
cities has been reversed somewhat, but not with the result, over time, of more social mixing.
Upgraded neighbourhoods in those inner cities, products of what urban theorists call gentrification,
tend to be homogeneous again. Cities thus show intra-urban inequalities just as much as we see
inequalities between cities.

We see a spiky urban world within cities, and the differences are big. As has been pointed out by
various urban anthropologists (Gupta & Ferguson, 1997), the differentiations between social
groups in cities wide apart may actually be smaller than those with their immediate city fellows.
Whether or not cities are increasingly polarised – for example, have a sharper contrast between the
wealthy and the poor with the middle disappearing, as has been suggested in connection to
globalisation (Sassen, 1991) – is subject to discus- sion (Mollenkopf & Castells, 1992) and varies
between cities, depending on their position in the global network of finance, capital and industrial
produc- tion and on national contexts of the severity of wealth inequality which dif- fers greatly even
between OECD countries. While cities are becoming more unequal (Dreier, Mollenkopf, &
Swanstrom, 2013), similar to inter-urban inequalities, then, some cities may be spikier than others.

Inequalities versus differentiations: vertical and horizontal paradigms

Inequality is what sociologists like to call a relational concept: someone or some social group is
always in an unequal position in relation to some indi- vidual or group who has more: more money,
more access to resources such as good education or health care, and so on. Traditionally,
sociologists have viewed inequality in terms of social class: the position of people in relation to the
means of production, with an antagonism between the working class and the owners of the means
of production. Such a dichotomy is no longer accu- rate, if indeed it ever was—it is no longer
possible to ascribe people class positions based on the occupation and income of the head of
household. Moreover, the idea of a clear class structure itself has been more and more intensively
debated (for an overview, see Devine, 1997; Crompton, 1998). So social scientists have discussed
other ways to talk about a stratified society and generally agree that more subtle, less bounded
notions of class help make sense of stratification. Perspectives vary widely on what combination of
assets or forms of capital should be decisive in ordering social positions in a society, but for urban
theory, the importance of this discussion lies in the underlying definition of inequality as unequal life
chances based on possession of assets. Scholars who study social stratification and class are said to
work with a vertical paradigm: higher and lower social positions are relatively easily differentiated,
as economics is the prime determinant of the social structure in this paradigm (Grusky, 1994).
 Urban theory is primarily interested in spatial expressions of such inequalities and in the possible
ways in which space may organise or even contribute to them. While social positions explain, to a
large extent, where one lives, urban theory is not just about describing social stratification expressed
spatially; urban theo- rists are also interested in how inequalities are reproduced through the
spikiness of the urban landscape of unequal resources. To give a brief example, economic
geographers have asked how locations of jobs and residential locations of job seekers are related,
and developed the idea of a spatial mismatch (Preston & McLafferty, 1999): changes in both
where people live and where low-entry jobs are located make it increasingly hard to find jobs nearby
for those who need it most, or getting to work comes with very high trans- action costs, so that
ethnic segregation can result in occupational segmentation ‘as the geographical scope of possible
jobs is limited by the need to minimize com- muting time on public transport’ (Coe, Kelly, &
Yeung, 2007: 385). Here, the spatial structure of the city – though by no means natural or given but
the effect of state regulations – organises and contributes to income inequality. Does it also
reproduce inequality? As we will see, this is one of the hot discussions in the field. Sharkey (2013)
provides compelling evidence that neighbourhoods are implicated in the severe stratifications across
generations along lines of class and race in American big cities, but, as we will see, the ways in which
neighbourhoods func- tion as perpetuating machines of poverty or wealth are not that clear.

In general sociology it has also been acknowledged that inequalities are not just about having money
or not. Money alone is an important but not sufficient resource to develop capabilities: to live a life
as one desires for oneself (Sen, 1999). It has hence been argued that other factors point to complex
dimensions of stratification, such as one’s education or the network on which one can draw for
support or for information on new opportunities, or the ways in which one manages to earn social
status in the eyes of others, build a reputation and achieve or maintain a position. This is why the
sociological literature likes to turn to Bourdieu’s (1984) notions of positions defined not just by
money, but by economic, social, cultural (and, in some corners of the debate, political) capital. Once
we start to do this, however, the clear picture of social classes within a vertical social structure
becomes blurred.

Beyond either horizontal or vertical

That social stratification models cannot adequately be described by the tradi- tional Marxist model
of social classes or Weberian model of status groups has of course gained ground over the years in
the sociology of stratification. Stratification theories understood a society’s social structure as
layered accord- ing to a hierarchical principle of vertical rather than horizontal differences. Anthias
(2005) discusses the alternatives to this model.

Anthias sees problems with this perspec- tive too (Anthias, 2005: 36–37) and proposes to speak of
narratives of location and positionality as concepts that may help us tell the story of – for us here
urban – inequalities: a story about how people place themselves in terms of gender, ethnicity and
class at a specific point in time and place (location) and a story about ‘being placed’. This reminds us
of the notions of location and posi- tion that Bourdieu had used. We will thus take away from this
discussion, which tends to become highly abstract, that given the problems with seeing stratifica-
tion as a rather fixed system of classes or status groups, urban inequalities can be understood as the
ways in which positionality formed by class, race, ethnicity and gender influences the durable access
to resources and assets (or capabilities: Sen, 1999). For urban theory, the two questions then are
what mechanisms and processes affect the durable inequalities as they are spatially expressed in the
city and how the spikiness of the city affects the durability of these inequalities.

Much of urban studies that engages with divisions, however, does not link these necessarily to
inequalities and their reproduction but primarily to living the city and questions of identity. The
approach to the city as a site of differ- ence rather than focusing on inequalities in life chances and
outcomes, or resources, can be traced back to the very origins of urban sociology. Notwithstanding
the neo-Marxist perspectives that looked at the production of space through a Marxist lens and
would definitely not prioritise questions of difference and how difference is lived as such, a rather
separate strand of urban sociology and urban anthropology has done precisely that. The city, these
scholars maintain, is ‘more than a backdrop’ and ‘actively intervenes in the social process. It is
created by our collective attitudes and actions and in turn impacts who we are by becoming part of
our consciousness and influenc- ing our behaviour’ (Flanagan, 2010: 15).

Living in a world of difference

More than in our earlier chapters we see here the connection between city, col- lective attitudes,
consciousness: in short, the city can structure personal experiences, influence personal styles and
tastes or make them possible, be a site of fun, desire and entertainment – as in Hannigan’s Fantasy
City (1998) – or be the site of identity politics and other forms of political expression (see Flanagan,
2010: 9–30). With the idea of the city as anonymous, as the expression and site of individualisation
and rationalisation, came the concern of urban sociologists with community. For Simmel, the
urbanite lived in a world of difference, differ- ence as ‘a profound social reality, lived over and over
in glancing encounters of the street’ (Tonkiss, 2005: 11). For the Chicago School sociologists, too,
differ- ence of class and of ethnicity and race structured the city. As Tonkiss (2005: 15) notes,
‘identity and community invariably were tied to spatial separation and cultural differentiation.
Community gave a social form to these patterns of dif- ference.’ Through a natural process of sifting
and sorting, groups different in class, ethnicity and race formed a mosaic of social worlds, as Park
used to say, where each group found its source for identity, sociability and institutions in particular
neighbourhoods in the city. McKenzie assumed that neighbourhoods produced by sifting and sorting
were defined by ‘peculiar selective and cultural characteristics’ (McKenzie, quoted in Tonkiss, 2005:
42), discussions that, similarly to Wirth’s work, move to urban culture and to ‘an understanding of
“personality” as a collective quality’ (Tonkiss, 2005: 43).

Cities as sites of resources: space and inequalities

Once urban sociologists started to be critical of the neo-classical economic per- spective that had
long informed the understanding of cities as the product of market forces that sifted people into
places as if they were all equal parties with given preferences in the market, a ‘new urban sociology’
(Gottdiener & Hutchison, 2011) developed based on political economy perspectives, including
the insight that power interests were unequal and that the role of politics and government in
urbanisation was a central one. For the spatial expression of inequalities within cities, we note that
scholars since the late 1960s have started to see these no longer as the result of a natural process of
sifting and sorting, as was common in the ecological tradition, but instead as the prod- uct of
advanced capitalism, including its supportive patriarchy, and government as supporting the capitalist
system. Moreover, they have started to think about city neighbourhoods and urban institutions –
schools, health care centres and the like – as sites of reproduction of labour.
The production of space

The socio-spatial perspective (Gottdiener & Hutchison, 2011: 2) that developed from here asks
what role economic, political and social institutions play in creat- ing and changing settlements, and
through what processes they are given mean- ing by local residents. It includes three elements that
Gottdiener and Hutchison (2011: 13) consider distinctive for this new urban sociology: its shift to a
global perspective; its attention to political-economic pull factors in urban and subur- ban
development; and its appreciation of the role of culture in urban life and in the construction of the
built environment.

Seeing the city as both the product and reproducer of inequalities, then, rests on an understanding
of space that leaves behind the idea that space is a container in which stuff happens. Instead, space
is thought to constitute social relations and meanwhile produced through social relations, and
constantly re-created (Gottdiener & Hutchison, 2011: 19). Lefebvre (1991) therefore argued
that space was a social product, produced on three interconnected levels. First, space was produced
by a spatial practice of production and repro- duction. Second, representations of space were
produced by government sur- veyors and planners, by architects and scientists, reflecting their
power positions, and creating the rationalised city of maps and plans. Third, repre- sentational space
included space as it is produced through the way it is lived and experienced: space as produced
through experiences, practices of the everyday, and imagined. As Tonkiss (2005: 3) has argued,
Lefebvre helped us see that the urban is not just material, but made of ‘meanings, language and
symbols’: ‘there is no such a thing as empty space. Space is always and only produced as a complex
of relationships and separations, presences and absences.’

While the production of space helps us understand the urban form, the repro- duction of inequalities
and the role of this urban form in this process cannot be thought through with this perspective
alone. Whereas sifting and sorting may not be natural but indeed the product of the mode of
production, neither is individual agency irrelevant, as Duncan shows. Duncan (2013: 231) writes:

There is a belief in our society that the social value of an individual must be roughly commensurate
with the social value of the place he frequents. In a society which is socially mobile the objects and
settings that one sur- rounds oneself with are an inseparable part of one’s self. This argument
logically leads to the notion that if one shares a setting with another then, to a degree, he shares his
identity thus allowing stigma to spread by spatial association. A tramp does not have a ‘normal’, that
is propertied, identity but, to put it in Goffman’s terms, one that is ‘spoiled’. Such people must be
separated out for their spoiled identities can spoil the setting and by extension the ‘normal’ people
who are a part of it. Hence the hierarchical division of space according to social value is of critical
importance to the host population.

How, then, can agency be connected to the theory of production of space? Prime and marginal
space, the concepts that Duncan uses as a scale of relative terms for this hierarchy, imply no
inherent value of space itself, but value is assigned to space by how it is viewed and used:

Thus the tramp must learn the social value that the host group assigns to different types of urban
space and to the regular temporal variations in these values: he must then resign himself to
spending as much time as possible in marginal space. Marginal space includes alleys, dumps, space
under bridges, behind hedgerows, on the roofs of buildings, and in other no man’s lands such as
around railroad yards which are not considered worth the cost of patrolling.’ (Duncan, 2013: 229)

Segregation

To say that social space is inscribed in spatial structures means to see physical space as one of the
sites were power is asserted and exercised. But inequalities in positions, as we have seen in our
discussion of intersectionality above, are being produced at the intersections of people’s lives in
terms of different posi- tions they hold in relation to class, gender, sexuality and race and ethnicity.
Segregation is the term that urban scholars use to refer to the positionality of groups in the city and
the extent to which they occupy separate sites in the city. Social theorists like Bourdieu and Anthias
have not been extensively used in the debate on segregation. As we will see, the focus of this strand
of urban studies, which has produced a huge number of empirical studies, particularly in the United
States, is primarily on the measurement of segregation and on theorising and measuring the effects
of segregation, a ‘cottage industry’ (Slater, 2013) that we will discuss in more detail later in our
section on neighbourhood effects.

Segregation can refer to ethnic or racial segregation and to social segregation, that is, the
segregation by social class. As socio-economic status differences manifest themselves spatially, with
the affluent living in neighbourhoods with high-status reputation and the poor living in undesirable
areas, the city reflects the social stratification of the society at large, another way of saying that
social space is inscribed in physical space. Such segregation is not voluntary: contrary to what new
eco- nomics often assumes, people are not equal market competitors on a free hous- ing market.
Not only wealth limits people’s housing choices; race, gender and sexuality also affect people’s
opportunities of where to live. Browne, Lim, and Brown (2009: 6–7), for example, discuss the
concentration of gays in American cities as the appropriation of territory as a defensive act in a
heterosexist society, out of a need for ‘safe’ spaces in a society that oppresses gays as men in
relation to heterosexual men.

This, then, is linked to a discussion, important to housing studies, of how much choice and constraint
there is in the housing choices people make (for an overview, see Clark & Dieleman, 2012). Do
people act as free agents on the market, finding ways to consume the type of fixed housing
preference that they have, given the economic resources at their disposal? Neo-economists would
certainly assume that the market simply does its work. Forced or voluntary seg- regation has been
theorised mostly by saying that segregation either is the result of structure (e.g. agents having
different economic capital gives or limits their options to act as free agents on the market), or, given
a specific volume of eco- nomic capital, is an expression of lifestyle preferences. This discussion has
gained particular relevance in recent decades in attempts to explain ethnic con- centration, or the
phenomenon that people who share ethnicity tend to live in neighbourhoods with others from their
own group. Noting that financially they may not need to do so, scholars have looked for other than
structural explana- tions. It is clear from this discussion that the simple division between choice
constraints and preference is not very useful, not least because the stories people tell of why they
live where they live contain their own rationalisations.

Moreover, where people move to is not a free choice either: neighbour- hoods have symbolic
meaning or aggressively exclude, affecting where people live (Pattillo-McCoy, 1999), and a gap
between one’s habitus and habitat may affect this choice, as DuBois (2013: 176) signalled more than
a hundred years ago:

The Negro who ventures away from the mass of his people and their organ- ized life, finds himself
alone, shunned and taunted, stared at and made uncomfortable; he can make few new friends, for
his neighbours however well-disposed would shrink to add a Negro to their list of acquaintances.
Thus he … feels in all its bitterness what it means to be a social outcast.

Social segregation

Ethnic or racial segregation is the form of segregation we may be most familiar with. The confusing
thing here is that social segregation is also used in contrast to residential or spatial segregation:
where the latter refers to the question whether groups different in ethnicity/race or class are living
separately from each other in the city, here social segregation refers to the ways in which people
who may reside in mixed neighbourhoods have separate social worlds of institu- tions, networks and
the like. Atkinson (2006) has contributed to this line of thought when he discussed how middle-class
and upper middle-class residents live in their own enclaves and travel the city through tunnels on
their way to socially segregated places of work and leisure, like philharmonic orchestras or expensive
restaurants. Butler and colleagues have shown how middle-class resi- dents living in socially and
ethnically mixed neighbourhoods in London make school choices that ensure that their children are
surrounded by other children with the same habitus, out of fear of social contamination (Butler
& Hamnett, 2007). Watt (2009) studied the ways in which middle-class residents in suburban
London used their socially mixed neighbourhoods and discusses the social seg- regation in the
facilities they use and avoid in terms of selective belonging.

Such social segregation is more difficult to measure systematically than residen- tial social
segregation, as it requires the study of what people do rather than where they live. Unsurprisingly,
segregation studies have primarily focused on the meas- urement of various forms of residential
segregation, assuming that the geographical neighbourhood of where we sleep matters most to
whom we engage with, and that spatial distance not only expresses social distance but also acts as a
basis for dis- tinction in practices that reinforce the habitat, in the formation of social networks and
in the generation of forms of capital. The study of segregation has been heav- ily dominated by
American scholarship, where scholars have traditionally been interested in measuring racial
segregation and particularly black/white segregation.

* Pakistanis and Indians in Brick Lane, London, where the ethnic character of the neighbourhood is
the basis for the commercialisation of segregation.

Suburbanisation

Suburbanisation is generally understood residentially, as the process whereby an increasing number

of people move out of the city to its outskirts, often to single family houses rather than apartments.
In Europe, suburbs include working-class families and do not mean single family housing at all.
Similarly, large council estates in British cities were often built on the edges. While these can be seen
as sub-urban in the sense that they are subordinate to the central city in many ways, it may make
more sense to refer to these areas as periphery in relation to the centre, a phrasing common in Latin
American urban studies (see Ingram & Carroll, 1981; Gilbert, 1996; Marques et al., 2008) where
favelas can be found on the edges of the central city. As in European cities, affluent enclaves of
suburban character can be found beyond those areas again.

Ghettoisation as a spatial process of marginalisation

Poor neighbourhoods, or even ‘ghettos’ where pov- erty and ethnic minority or racial status
coincide, are the spatial expressions of processes in which local labour market situations, the state
and other ‘institu- tional assemblages’ entrench poverty (Theodore, 2010: 170). They are spatial
expressions of processes of marginalisation. Marginalisation is the opposite process of the
advancement of the clustering of forms of capital elsewhere in the city, and the processes that make
locations where strong positions can be main- tained take place in a double bind with processes in
the other direction. As we will see below, these include economic restructuring under conditions of
techno- logical development and globalisation, the articulation of racism and discrimina- tion, spatial
stereotyping, the ideological construction of ‘blaming the poor’, if not punishing them, based on
behavioural understandings of poverty, and neo- liberal policy-making in times of austerity.

Ever since the onset of industrialisation and the intense urbanisation that came with it, social critics
have been concerned with the living conditions of the lower and working classes, who are the
majority in the city: ‘temporarily endur- able existence for hard work’ is how Friedrich Engels
described their lives after visiting London and Manchester at the end of the nineteenth century, and
their segregated lives seemed to matter to no one: ‘society, composed wholly of atoms, does not
trouble itself about them; leaves them to take care of themselves and their families, yet supplies
them no means of doing this in an efficient and per- manent matter’ (Engels, 2011: 54. Moreover,
neighbour- hoods may be homogeneous in class and race because society involuntarily segregates
some of its members (Gans, 2008: 356) based on racism and discrimination.

‘Ghetto’ as a term in urban studies

The use of the term ‘ghetto’ in urban sociology goes back to Wirth’s (1928) early work with the same
title. Wirth aimed to show how the Jewish European ghetto was reproduced in Chicago at the time,
tried to classify the Jews living there according to their lifestyle characteristics and noted that the
ghetto was the product of the historical exclusion of Jews from mainstream society. It was strictly
used to refer to Jewish neighbourhoods, and Wirth did not see them as much different from other
ethnic neighbourhoods where first-generation migrants embarked on their way to assimilation
(Haynes & Hutchison, 2008: 349–50). Robert Weaver’s (1940) first study of racial segregation
then linked the term to black residents, whilst Kenneth Clark discussed in Dark Ghetto (1965) the
problems that came with racial segregation, referring to both its ‘pervasive pathology’ and its
‘surprising human resilience’ (quoted in Haynes & Hutchison, 2008: 351).

Public opinion and discourse has come to refer to such areas as ghettos, increas- ingly also in
European media. The tendency is a ‘dilution of the notion of ghetto simply to designate an urban
area of widespread and intense poverty, which obfuscates the racial basis and character of this
poverty and divests the term of both historical meaning and sociological content’ (Wacquant, 1997:
342)

While in the US such neighbourhoods have long been called ‘ghettos’ and European media have
hyped this moral panic (Slater, 2010), European scholars have been hesitating to use the term, or
have argued that it analytically makes no sense to do so, often claiming that Europe does not have
‘American condi- tions’ (Wacquant, 1997, 2008; Agnew, 2010; Nobles, 2010; Slater, 2010; Peach,
1996; Simpson, 2007). More than in the US, it seemed in welfare states such as France, the
Netherlands, the Nordic countries, Germany and even Great Britain as if poverty, unemployment
and their political and social consequences had disappeared in the booming years of welfare growth
and welfare state develop- ment after the Second World War. From the 1970s, however, new forms
of polarisation began to characterise European cities (Häussermann, Kronauer, & Siebel, 2004:
7) so that European scholars started to note the emergence of new poverty, often linked to
migration (‘ethnicising’ poverty – see Häussermann et al., 2004: 9) and to find spatial expression in
neighbourhoods where poor, migrants and the unemployed concentrated, although such
concentration could by no means be compared to the concentration known from the US (Musterd
& Ostendorf, 1998b). Madanipour (2011: 188–91) prefers to speak of social exclusion rather
than poverty or marginalisation, as this allows him to stress the cultural, economic and political
dimensions rather than just whether one is poor or not. In European countries such as Germany,
Sweden, the Netherlands and France, social exclusion has also been the core concept in policies
aimed at com- bating disadvantage, whereby the ‘spatiality of exclusion’ has often, as in Britain,
resulted in ‘attempts to dismantle pockets of deprivation without necessarily dismantling the causes
of deprivation or the forces bringing them together in particular enclaves’ (2011: 191). Madanipour
does see a relevance of spatial practices for exclusion. Exclusion, he writes (2011: 191), is a matter of
resource access of various sorts; and such resources have ‘clear spatial manifestations’.

Apart from poverty not being as extreme or as extremely concentrated, it has been pointed out that
the scale, the exclusive black demography of the impov- erished US tracts, and the ubiquity and
lethalness of crime in the American ghetto are all unlike the characteristics of very poor European
areas (Western, 2010; see also Wacquant 1993: 366–7). Ghettos as neighbourhoods of the
unemployed have been analysed by Americans, as O’Connor (2004: 69–70) has warned her
European audience, within the tradition of ‘analyzing the deviance of the poor in the ghetto without
questioning the role of the free economic mar- ket as such’. While this may be a little too strong and
not do full justice to Wilson and other American scholars, O’Connor of course correctly points out
that taking American frames of references as ‘theories’ to understand empirical patterns of
concentrated poverty and the processes of marginalisation that bring them about and make them
durable must be done with caution (Wacquant, 1996). Various scholars have pointed to the
metaphorical meaning of ‘ghetto’ in international scholarship nowadays (Monteiro, 2008: 378), with
its strong behavioural connotations (Blokland, 2008b). We return to this below, as this ideology of
blaming the poor for their own poverty is one of the important pro- cesses of marginalisation, a
process of symbolic violence.