Townscapes and landscapes by zhangyun


									                              Townscapes and landscapes

                                       Mike Savage

In this paper I draw on various recent research projects relating to issues of townscape
and landscape. My focus is on understanding how people construe spatial boundaries and
how these intersect with their social identities. Methodologically, I am interested in how
we use forms of mapping to develop our – still limited - understanding of these issues,
and I am also keen to understand these issues recursively, that is to say that our methods
themselves feed into, and are inextricably connected with, the very processes that they
seek to unravel. Rather than present a detailed account of any particular piece of research
I will seek to summarise and synthesise in an attempt to address the issues raised by the
conference organisers under this session.

I take the starting point of our discussions to be the recognition of the lack of clear
boundaries between town and country, or more generally between any kind of spatially
delimited territory. In a global, networked, mobile environment, attempts to of this kind
are bound to fail. However, it is also clear that older accounts of globalisation which
emphasise time-space compression, the rise of „non-places‟, the homogenisation of space,
etc, fail to recognise the growing importance of spatial particularities and what might be
called the „enchantment of the landscape‟ in the early 21st century. The multiple coding
of urban space and spatial activities, the proliferation of flows and mobilities, the
digitalisation of diverse modes of communication, etc all produce highly specific kinds of
spatial intensities, where aesthetics, ethics and attachments are mobilised in powerful and
pervasive ways. The challenge, as I see it is to build on the recognition of splintering,
fragmentation, proliferation etc, to explore contemporary dominant forms of spatial
ordering. In this process, old forms can be re-created amidst and alongside the flux of

Let me use a personal anecdote to explain this point further. With my partner and young
boy I recently moved to live in the city of York. The couple buying our house in
Manchester were moving from London and scrutinised potential properties on websites
before visiting Manchester only for one day to visit and check on the small number of
properties they had identified as potentially suitable. We were also able to visually
inspect and enjoy the immaculate presentations of various houses without leaving our
desks. We also used websites to look for a house in York, and – since Roger Burrows was
advising us - we even used <www.upmystreet> and similar web resources to look for
suitable postcodes to move to. The necessary property searches were done digitally
(which caused problems since it turned out that planning permission required for our
York house had been approved in 1973, and did not show in the planning records which
had only been digitalised since 1980, but that is a minor issue). The actual transfers of
moneys when the transaction took place was done entirely electronically (our solicitor
had a nice way of talking about sending the money „up‟ and „down the wire‟. Even
though these transfers were not exactly simultaneous (we had an anxious hour or two
when we might have been sleeping in a hotel), nonetheless with remarkable speed. So,

even the business of moving into a new house – necessarily spatially fixed as they are -
intensive of time, effort (physical and emotional), has been radically reformed through
technological, spatial and other forms of change.

And yet, a month after we moved, I bought a copy of Rowntree‟s famous poverty study
of York, written in 1903. It contains a very interesting map of the social geography of
York which differentiates the streets of the poor, the respectable, and the servant keeping
classes. I was struck by the fact that not only had we moved to a „servant keeping street‟,
but also, that without exception, all the areas we had looked in were also those of the
„servant keeping classes‟. 100 years on, the social geography of York seems to have been
reproduced almost entirely. York is of course, an unusual place: it has not been gentrified
(since the middle classes never really left the centre), it does not have an „inner city‟ etc1.
I was also struck by a second observation. It turned out that a significant number of
households in the street contained people actively involved in architecture, archaeology
or art. My partner is an architectural historian. We even returned home from a day‟s
outing once to find our next door neighbour sitting on his doorstep reading a copy of
Helen‟s book. We had, in a sense, and without knowing anyone in that street prior to our
move, ended up living alongside „people like us‟.

The challenge then is to recognise how forms of homogeneity, similarity and cohesion
are produced through processes of disorganisation, fragmentation and mobility. We do of
course have numerous theoretical resources for exploring this issue, ranging from Roland
Robertson‟s globalisation theory, through Arjun Appadurai‟s insistence on scapes,
Castells on the network society, and of course actor network theory, chaos theory, and so
on. All well and good. But these need to be set to work. I am increasingly interested in
exploring these issues through understanding the role of what might boringly be called
„research methods‟. Since it is precisely because of the methods that data is now
collected, organised, distributed and analysed that is central to the remaking of space, we
need to critically reflect on how methods themselves produce spatial orderings2. Taking
as point of departure Nigel Thrift‟s (2005) argument about who we live in an era of
„knowing capitalism‟, allied to Nik Rose‟s (1990; 2005) observations about how we need
to specifically scrutinise and critique the isolation and delineation of the empirical, we
will be able to develop a more powerful account of the dominant orderings of
contemporary socio-space.

In the next section I pursue this issue through exploring the features of what I call the
„descriptive turn‟. I explore the significance of the rise of „mapping‟ methods, and
consider the ability of the social sciences to engage with these. In the second part of the
paper I examine the remaking, and yet the reproduction, of spatial orderings in Britain,

  Yet the study of gentrification poses interesting issues. There is an argument that in some cases
gentrification is actually about the restoring of places that had lost status and standing in the post war years
back to the glory of their Victorian position. This would probably apply to the Islington phenomena, for
instance. Of course, as Butler and Robson (2005) has shown, gentrification is not a unitary phemomenon
and there are fundamental differences between the Islington and Docklands experience, for instance.
  This interest in the „politics of method‟ is a key theme of the ESRC Centre for Research on Socio-
Cultural Change (CRESC) of which I am Director and which seeks to nurture and develop methodologies
to address theoretical claims about sociio-cultural change.

drawing on data collected by the „Cultural Capital and Social Exclusion‟ project, where I
focus on the relationship between people‟s cultural positioning and their spatial
attachments, using both survey sources and in-depth interviews. I use this material to
argue for a „granular‟ conception of the landscape. People do not usually see places in
terms of their nested or relational qualities: town against country: region against nation,
etc. but compare different places with each other without a strong sense of any
hierarchical ordering. I further argue that the culturally privileged groups are highly
„vested‟ in place, able to articulate intense feelings of belonging to specific fixed
locations, in ways where abstract and specific renderings of place co-mingle. Less
powerful groups, by contrast, have a different cultural geography, which hives off fantasy
spaces from mundane spaces.

Spatial identifications and the „descriptive turn‟

There is a major challenge in thinking how we research local attachment and identity?
Whether we use local case studies or national surveys, the spatial scale is predefined
through our very research instruments. Research methods have certain renderings of
space and landscape built into them, which make it impossible to simply deploy them as
neutral research tools. Furthermore, dominant social science research methods are
attracted to an analytical model, in which they seek to explain „surface‟ phenomenon in
terms of underlying causes or processes. Positivists, critical realists, and interpretativists
find common cause in the idea that „underlying‟ forces produce social outcomes. John
Pickstone (2000) sees this analytical „way of knowing‟ as deriving from practices of early
19th century chemistry, and sets it against other ways of knowing, notably classification,
and experimentalism, both of which have little hold in the social sciences

Analytical methods treat the landscape in one of two ways: either as the contingent
outcome of underlying processes or as the „context‟ in which social relations are located.
Both these perspectives involve bracketing the landscape, seeing it as either above or
below the „real‟ stuff of social science. Timothy Mitchell (2002) has examined how the
practices of colonial social science relied on the formation of a „locationless logic‟,
dependent on abstraction. In this vein the rise of the sample survey as possibly the prime
instrument of social science from the early 20th century can be seen as the quintessential
mode by which social scientists seek abstracted knowledge, through the use of inferential
statistics which allow generalisation on the basis of a sampled population. The sample
survey is quintessentially dependent on the nation as imagined community, with the
boundaries of its sample being defined by national borders, and the space within these
borders treated as empty homogeneous spaces from which the national social structure
can be delineated. The abstracted variables and measures which can be manipulated
through survey analysis thus become the deep structuring forms that analytical social
science uses. Key governmental technologies, from the inflation rate to unemployment
rates depend on the deployment of this kind of information.

From within the perspective of the sample survey, the region is usually the smallest
viable spatial unit on which such data is amenable to analysis, and it is for this reason that

the region emerges becomes administratively important as a significant intra-national
boundary. In the British case, the administrative nature of these regional boundaries
explain why they do not necessarily map onto historic boundaries associated with
counties, and also explains the relatively muted symbolic power of the administrative
region in the British context. Yet Roger Burrows and I (2006) have argued that the
sample survey is now in decline as the predominant research instrument, and with this
decline, different ways of conceptualising and thinking about space and attachment
become possible.

We contend that the „glory years‟ of the national sample survey are in the past, for four

1. In an intensely researched environment, survey response rates have been steadily
falling, and it is proving more difficult to obtain response rates of circa. 70-80% which
were thought normal up to the 1980s. People no longer treat it as an honour to be asked
their opinion, but instead are more likely to see it as a nuisance, at worst an intrusion.
This problem now even extends to the Census itself, where since 1991 there have been
serious problems where a large number of households avoided filling in census forms
(Simpson and Dorling, 1994). This problem is instructive, but not overwhelming, because
survey statisticians have developed methods for estimating the attributes of „the missing‟,
and it still remains possible to generalise on the basis of smaller, biased, samples.

2. Sample survey face difficulties in measuring diasporic, migratory populations who
straddle national boundaries. Although many survey researchers are interested in
comparative and global processes, their approach depends on comparing different
national cases. No survey researcher interested, for instance, in the Jewish population, has
ever combined Jewish members of different national samples into one datafile and treated
this as their unit of analysis. This is for good reason, since the resulting population would
not be representative and would violate the assumptions of survey research. But this point
confirms the argument that the sample survey depends on, and constructs, the nation as
imagined community.

3. There is now a substantial fracture within the survey research community. Whereas
surveys historically were led by academics, and were strongly supported by public
funding, we now see the proliferation of survey research in private companies, notably in
market research. Such survey research now has very limited reference to academic
expertise. Unlike public surveys which are freely available, and form the bedrock of
academic statistical expertise, the fact that their data is commodified is central both to
their market and their purposes. One of the most important of such surveys is the British
Market Research Bureau (BMRB) panel survey, which has a very high sample size of
40,000 which parallels the largest public surveys and asks questions every month. Yet,
the BMRB has hardly been used by academic sociologists (for an exception, see Savage
et al (1992)). The resulting analyses are not subject to the kind of multivariate analysis
preferred by social scientists but are displayed visually through forms of cluster analysis,
so making the results accessible to a wide audience in corporate marketing departments.
Very few socio-demographic variables are used, and class is measured through the

largely discredited (by academics) market research categories, A, B, C, D and E. Even
though no self respecting academic sociologist would dream of using such measures, and
even though the official National Statistics Socio-Economic Classification has developed
a refined and elaborate means of classifying occupations, this message seems largely
irrelevant to the powerful bastions of market researchers.

Interestingly, the form of analysis used in this market research is predominantly not of the
analytical kind, but depends on forms of clustering, often using forms of visualisation.
Typically, a firm wanting to commission research on the market for its products will get
access to a series of figures in which their brand is located on a figure produced by some
kind of clustering, so that the firm can see which other brands it is located close to. This
allows the firm to think about how to position its brand vis-a-vis its competitors. This is a
very different way of thinking to that which is deployed in analytical social science, but is
rather akin to what Pickstone sees as forms of „classification‟, or what Latour (2005)
invokes in his call to treat the world as „flat‟. ,

4. This leads logically to my final and most important point. Most powerful institutional
agents now have more effective research tools than sample surveys: they can draw on the
digital data generated routinely as a by-product of their own transactions: sales data,
mailing lists, subscription data, and so forth. When sample surveys became popular, in
the 1950s, they met some resistance from those who insisted on the need to research
„whole populations‟, usually in the form of intensive case studies, or who insisted on the
impossibility of abstracting from „context‟ (see Savage forthcoming). As late as the early
1950s, the hybrid mix of anthropologists, surrealists and sociologists who ran Mass-
Observation insisted that their methods of eliciting narrative accounts of purchasing
decisions was more valuable than the abstracted accounts provided by survey researchers.
Anthropologists and psychologists, strongly influenced by Lewin‟s field theory (Lewin,
1951), thought it essential to explore the dynamics and relationships between all the
parties in specified socio-spatial settings. This was not a defence of ethnographic
methods, since quantitative researchers, especially those associated with early social
network methods, also championed such a view. Looking back, we can see how the
sample survey researchers won this battle because they were able to demonstrate the
effectiveness of their technology to powerful stake holders in the context of the 1950s.
And, although they claimed to win this battle in the name of science, we might also note
that cost effectiveness was on their side.

However, in the current situation, where data on whole populations are routinely gathered
as a by product of institutional transactions, the sample survey seems a very poor
instrument. To give a simple example of the power of routine transactional data over
survey data, does not need to market its books by (a) predicting on the basis
of inference from sample surveys the social position of someone who buys any given
book and then (b) offering them other books to buy which they know on the basis of
inference people from that social position also tend to buy. They have a much more
powerful tool. They know exactly what other books are bought by people making any
particular purchase, and hence they can immediately offer such books directly to other
consumers when they make the same purchase. Hence the (irritating, though often

tellingly useful) screens offering versions of „Other people who have bought x have also
bought y‟ which confronts the Amazon customer. Similar principles are used by
supermarkets through data gathered by their loyalty card schemes where they can identify
for any given customer – without knowing anything very much about their personal,
„social‟, characteristics – what other kind of goods they might be liable to buy if they
buy, for instance, organic bananas. They can hence bypass the principles of inference
altogether and work directly with the real, complete, data derived from all the
transactions within their system.

What is important in these methodological innovations is a new way of re-enchanting the
landscape. Rather than the abstract, empty homogeneous space of the sample survey, or
the intimate space of the local case study, these methods produce complete maps of
„neighbourhood types‟, and articulate a new geography of „granular‟ space. Here,
different locations are arrayed with other „like‟ places, regardless of their physical
proximity or whether they exist within the same region, metropolitan area, etc. Such
geodemographic systems „capture personal data triggered by human bodies and…use
these abstractions to place people in new social classes of income, attributes, preferences,
or offences, in order to influence, manage or control them‟ (Lyon, 2002: 3). For some
commentators such procedures are viewed as deeply emblematic of informational
capitalism (Elmer, 2004) where „[p]ersonal information is increasingly the basic fuel on
which economic activity runs. ..Companies offering geodemographic profiling data are
the 21st-century equivalents of the great energy companies of the 20th (Perry 6, 2005: 17).
This body of work seeks not to isolate any one „deep‟ variable - social class, ethnicity,
stage of the life course, gender, educational attainment and so on – but fuses these, (and
others) as manifested through residential location. From an analytic point of view this
represents a radical collapsing of these very common sociological variables onto small
neighbourhood areas (postcodes mainly). The exact mixture of the sociological
„elements‟ that become „fused‟ together in each category of these socio-spatial
„compounds‟ is empirically determined by the statistical purchase each gives in
explaining small-scale spatial variations in consumption patterns (Burrows and Gane,
2006). It turns out then that knowledge of the spatial location of someone is increasingly
an important proxy for all manner of sociological information: indeed to the extent that
there is no need for other social measures. Richard Webber, one of the most important
researchers here, has concluded that
        „the type of neighbourhood in which a consumer lives is a significantly more
        predictive piece of information than any person or household level
        discriminator…By implication therefore it is almost certain that significant
        neighbourhood effects must operate for many of the behaviours tested‟ (Webber,
        2004: 1)

Rather than „obsessing‟ on the best variable, or cluster of variables, that predict a range of
outcomes, we might more powerfully use much more variegated postcode information as
the most telling social indicator within a broader, descriptive, research strategy. One‟s
position in the landscape now becomes a central issue.

My point here is that contemporary research methods imply and construct different
notions of landscape than were enshrined in the sample survey, and we cannot deploy any
specific research method without recognising their complicity in specific kinds of spatial
construction. And through being implicated into the research circuits of „knowing
capitalism‟ they prove highly „productive‟. Postcode information can be sold so that it
allows institutions to target particular areas for mail shots, marketing and for planning
decisions as to where to locate new stores, services and so on. Such data is also
increasingly used by local authorities and the like. Given this wealth of data, inferential
statistics look pretty limited by comparison

What is implicated in the proliferation of these transactional classifications? Identity and
meaning is produced not through abstracting from, but by classifying, landscapes. Place
becomes granular and non hierarchical: post-codes are allocated to classification with no
concern about their region, or their closest city. They are not nested into larger spatial
units, but combined with others at possibly great spatial remove.

Elsewhere, I have examined historical patterns of local attachment in Britain since 1945
(Savage 2007), where I draw on archival records of in-depth interviews conducted by
social scientists in various periods of time. As late as the 1960s, these reveal hierarchical,
rather than granular conceptions of identity. To give one example, in the early 1960s, the
educational sociologist Brian Jackson, conducted a study of „working class community‟
is his home town of Huddersfield. Jackson had shot to fame with his book with Dennis
Marsden Education and the Working Class, a study of the attitudes of working class
youngsters to education in Huddersfield, Yorkshire. With his colleague Dennis Marsden,
Jackson was a pioneer in the use of in-depth interviews as a central device for eliciting
the view of those neglected by conventional social science research methods. Working
Class Community is in many respects a remarkable book, deliberately evoking the rich
cultural practices of its working class as a counter to the stereotypes of middle class
commentators. Jackson skillfully excavates the extent, yet also the limits of community
practices in the last period when Huddersfield could still represent the affluent working
class, from the bowling clubs, brass bands, working and men‟s clubs. The records of his
study, left in Qualidata, include a number of unstructured interviews he conducted with
young, working class, couples.

During his research he asked how some young couples felt about living in Huddersfield.
Here are some typical responses.

      „Well, I think I‟m proud to have been born here. I think I‟ve warm feelings about it.
      I‟m not ashamed to the place anyway. The place where you‟re born is different
      from anywhere else, isn‟t it? I think down south, I should be frightened to live
      down there (No. 1 husband)

      Whenever I‟ve been down south, I‟ve always felt inferior somehow. I‟ve always
      felt that everybody down there knew a lot more than we did. They‟re always much
      more up to date, aren‟t they (No. 1 wife).

      No, I don‟t think either of us likes the place very much. It‟s just that we live here,
      family ties I suppose. And it‟s so difficult making a move, I don‟t think we‟ll move
      out, but I wouldn‟t say we were proud to live here. Not a bit…. If anybody asks you
      where you come from and you have to say Huddersfield, well you feel a bit
      uncomfortable. Your heart sinks. I don‟t know what it is. I think perhaps it‟s the
      speech, they way you talk. (2)

      I‟ve never thought about it. It‟s alright I suppose. I‟m not bothered. Well, when you
      go away a lot of people have never heard of it (3)

      ….and then you see living in Huddersfield you are bound to feel a bit differently
      about it, aren‟t you. It‟s your home town, and you know so many people. It‟s not
      friends, no, it‟s not people you go out with, but people you meet in the street and
      knowing where to go, knowing where everything is. And having a house and a
      secure job. That‟s a lot really to give up (4 wife).

      well it‟s as good as the next place, isn‟t it. (The chief thing they valued the town
      for, was work. They felt that they never fear for a job living in a place like this.
      They weren‟t particularly proud of living here)‟ (6 husband)

      Well it‟s alright. It‟s alright for work, shouldn‟t say I was thrilled about it. Don‟t
      misunderstand me, I don‟t mean that we dislike the place (8)

      Well I‟m proud to be Yorkshire, not Huddersfield particularly, but I‟m proud to be
      a Yorkshireman. To me, Yorkshire is the county. I think anybody coming from
      abroad, when they think of a county they think of Yorkshire, don‟t they. You think
      of people as being tough (10)

      Well, we‟re not that proud Huddersfield, we‟re a bit more proud than we might
      have been a few years ago, now that Huddersfield‟s more on the map. One or two
      things have happened here. We‟ve the choirs and the choral society, people know
      where Huddersfield is, they don‟t say Huddersfield, where‟s that. I‟ll give you an
      instance (11).

Four points stand out here. Firstly, this is not, it would appear, an enchanted landscape. In
the quotes above, there is little evidence of a developed cultural geography where
respondents feel willing or able to compare specific features of Huddersfield with other
places salient to them. No other places (other than the county of Yorkshire and the
„south‟) are referred to. Huddersfield is, simply, where they live, the place that their life is
centred on. Secondly, and relatedly, the question is largely seen as bearing on pragmatic
issues regarding the quality of housing and the availability of jobs. The main criterion for
assessing a place is whether they offer people the chance to live a decent life. There is no
aesthetic sense regarding the quality or aura of place on display, no idea that location is
an aspect of consumer choice. However, the familiarity of living in Huddersfield gives
residents a sense that they belong in the place, albeit ambivalently. In the way
emphasized by many post war anthropologists, having kin in the area explains why they

are there and gives a certain kind of rationale. Thirdly, residents are aware of how
Huddersfield might be seen by outsiders, and feel that on the whole, it does not measure
up. Their articulation of who these outsiders are is unclear, though there is a sense that
they are southerners, and probably Londoners. For most respondents, it is not clear what
the actual location is of such outsiders, but there is simply an implicit sense that
Huddersfield is not the „centre‟, but part of a stigmatized periphery. This is a hierarchical
model in which a local inside is juxtaposed to a generalised but powerful „outside centre‟
from which these residents are excluded. Fourthly, the dominant moral claim which is
allowed within this way of thinking about place is one based on local patriotism or more
precisely „pride‟ (as with interview 10). One cannot just leave Huddersfield behind, and
therefore one has no option but to champion one‟s place in the scheme of things, however
limited one ultimately recognises your local belonging to be3.

I have dwelled on these examples because they seem so different to the kind of responses
I obtained when I conducted 182 interviews with Gaynor Bagnall and Brian Longhurst in
the later 1990s in the Manchester area, which we report in Globalisation and Belonging.
Admittedly, we interviewed predominantly middle rather than working class respondents,
mainly in a series of out of town or suburban locations (Ramsbottom, Wilmslow,
Cheadle, though we also interviewed in Chorlton). Nonetheless, on nearly every count,
the dominant mode of local attachment was radically different to those elicited by
Jackson. Firstly, many of our respondents had an elaborate cultural geography and a wide
range of referents by which they judged (morally, aesthetically and instrumentally) the
quality of their current abode. Secondly local economic factors are rarely cited as
significant in assessing the quality of place, if only since it was widely believed that one
could travel long distances to work so that one‟s appreciation of one‟s residential location
was essentially a separate issue to that of one‟s work. Thirdly, there is little sense of the
moral supremacy of „outsiders‟ in judging their place. London, and the south of England,
does not serve as the automatic reference point any more. It is for this reason that we saw
many of our respondents as forming a „Northern English middle class‟, characterised by
high mobility within Yorkshire, Cumbria, Lancashire, Northumbria and a broad
affiliation to the North. Finally, there was no sense of a primordial attachments based on
being „born and bred‟ in a specific location. Rather, people felt they belonged when their
residential location was aligned with their biography, when it was the kind of place for
„someone like me‟. This judgement was based not on the predominance of any one place
– for instance, London – but on the specific biographical referents of the individuals‟

In short, the people we wrote about in Globalisation and Belonging an
„enchanted landscape‟, one where places could be mapped through an extensive range of
spatial referents. Administrative boundaries, whether of region or county, had relatively
little purchase. People with similar cultural geographies often lived alongside each other,
in a process of socio-spatial sifting rather similar to my own experience moving to York

  In this paper I only have space to talk about this one example, but in Savage (2007) I use other sources to
argue that the Huddersfield cases are typical of the findings of numerous other studies.
  For the argument I am developing here, I am simplifying and omitting important qualifications,
differences between the four places, etc.

which I mentioned at the start of this paper. Places were not placed hierarchically, but
relationally to other kinds of granular places that were similar to those that they lived in
now: Cosmopolitan Chorltonians looked to London; country loving Ramsbottom livers
looked to the Yorkshire Dales, and so on. People saw this landscape as one premised on
the flow of „scapes and refused to abstract themselves from it.

Mapping the flat world

Many of the people we researched in Globalisation and Belonging are, in a sense great
„mappers‟. In this last section I want to explore – again methodologically – how we might
take further our understandings of this thoroughly „mapped‟ nature of contemporary
identities and attachments. The mappings I refer to include both those implicit in new
research methods, the proliferation of classifications, and the way that people themselves
map their identities. The challenge for social science is now to move beyond its rather
tired analytical strategies and meet the challenge of the map head on through a concern
with landscaped description. In this respect, we are undoubtedly aided by the recent
„descriptive turn‟, the insistence by different writers that we should seek not to abstract
but to describe, and by describe contest powerful contemporary classifications.

Latour‟s Reassembling the Social is one interesting manifesto to this effect. In refusing to
ontologise the social, he instead argues that we need to see socio-technical relations as
complexly networked, fluid and mobile. There is no social „stuff‟ other than the stuff
itself. There is no privileged arena of the social. Descriptive strategies, which trace
contingent and multiple ties are better able to unravel these complex connections and
disconnections than analytical models which reduce the world to „deep‟ structure.

This manifesto for a descriptive social science is in many ways an attractive and
compelling one, yet also frustrating in its lack of specificity. In calling for us to treat the
world as flat, Latour is not seeking to make the claim that we should really see the world
as flat. This would fly in the face of Latour‟s insistence that technologies are agents.
Given that most technologies – from wireless communication, through hydraulic systems,
and air travel - depend on the world not being flat, it would contradict his own argument
to insist on flatness as an ontological category. Rather, his appeal to flatness is a
rhetorical means of avoiding either „deep‟ structural variables, or „transcendent‟
categories, floating free of the landscape. But nonetheless, although this serves us well
enough for these purposes, it is frustrating in its lack of clarity. This relates to a second
methodological problem. It is akin to that which has been amply discussed by historians
over many years, namely, how do we know how to distinguish „historical facts‟ from
„facts‟ (see notably E.H Carr in his famous What is History). Given that there are so
many facts, how does one decide which ones connect, which ones to map? Latour‟s
problem, of course is even more intimidating than Carr‟s since he would admit natural
and technological „facts‟ into the picture and not restrict his focus to „human‟ facts alone.
Latour presumably refuses to offer much advice about how to select from the plethora of
facts available since to offer a „methodology‟ would itself be to define an external world
around which one can develop knowledge, hence contravening his own assumptions.

Hence he prefers broader invocations, such as we should „follow the actor‟. But how do
we know when we are, or are not, following the actor? Carr‟s response to this similar
kind of problem invokes narrative: when facts can be arrayed in narrative form, this
makes some facts more important than others, and hence provides a means by which
historical change can be meaningfully unravelled. The grain of historical facts can be
winnowed from the chaff of facts in terms of their centrality within a narrative. This, of
course, depends on prioritising time over space, in ways which involve abstracting from
the landscape5. Latour‟s advocacy of literary approaches implies a rather similar
resolution to this problem. Yet this advocacy of a narrative based approach then flies in
the face of actor network theory with its own anti-humanist insistence that objects are
agents – since it ends up privileging the author as narrator. Here there is a striking
divergence with writers such as Walter Benjamin or Gilles Deleuze who are much more
concerned to refuse this kind of narrative temptation. Given that object/agents rarely
work through narrative, but increasingly frequently deploy digital forms of reasoning and
ordering, Latour‟s ultimate reliance on narrative to demonstrate the connections between
numerous agents seems to rely on a humanism that he elsewhere repudiates.

In the light of this, we might be better served therefore by more systematically working
through what mappings might involve. This is a means of taking up Deleuze‟s challenge
to think „diagramatically‟ as a means of teasing out forms of intensity and their
localisations. And in thinking about how to develop this – and perhaps surprisingly -
Bourdieu‟s thinking proves to be of some value. Bourdieu‟s field analysis takes up issues
around descriptive and geometric approaches to analysis. Using his preferred sporting
metaphors, Bourdieu sees social life as organised around different „fields‟ in which
players struggle for position and advantage. This metaphor of sport is used to show why
there are no „neutral‟ positions within a field6, and that it is inevitable that all players are
required to contest against others – even from hopeless situations. It also emphasizes that
„players‟ are located on a landscape. We are inextricably „thrown‟ into a landscape from
which one has no choice but to battle for position, but our ability to command height is an
important feature of agency.

It is also not recognised enough in the UK that Bourdieu‟s emphasis on social space and
on his concept of the field is wedded to the use of Geometric Data Analysis (Leroux and
Rouanet 2005). Derived from the French mathematician Benzecri, correspondence
analysis seeks to render position as a series of co-ordinates in Euclidian space, and
through this means to portray maps of „social space‟7. Such maps can be composed in
several dimensions, and can be used to reveal social relations not in terms of deep
structuring properties, or in terms of reified variables (indeed his analysis in Distinction is
couched specifically as a critique of such variable-centred analysis), but in terms of their
„surface‟ relationship to each other. The ordering of these maps need to be explored
  Though see Moretti‟s brilliant exploration of ways that novels can be mapped to reveal features that are
obscured through orthodox narrative analysis.
  C.f. „it is not so easy to describe the “pure” gaze without also describing the naïve gaze which it defines
itself and vice versa: and that there is no neutral, impartial, „pure‟ description of either of either of these
opposing visions‟ (Bourdieu 1985: 32)
  One can thus trace Bourdieu‟s increasing deployment of the concept of field in his later work to his
interest in correspondence analysis which originated in the 1970s.

inductively, through trying to make sense of the spatial distribution of the variables
according to their position, rather than through imputing the power of „master‟ variables.

One advantage of this is that it can be put to work to explore issues around the
relationship between the social and spatial. Let me do this by deploying evidence from a
collaborative ESRC funded study „Cultural Capital and Social Exclusion‟ (conducted by
Tony Bennett, Elizabeth Silva, and Alan Warde as well as myself), which constructs a
systematic map of cultural taste, participation and knowledge in Britain using both survey
and qualitative evidence8. A correspondence analysis we have conducted on our survey
data is reported in Figure 1, on the basis of associations from 168 different variables
covering taste and participation for music, reading, art, eating, sport, TV, and film9.
Variables from different fields are colour coded to aid interpretation: blue variables
pertain to music; red to leisure; black to sport; green to film; purple to reading and brown
to eating out. There are no socio-demographic variables used to define this space: the
focus is entirely on the ordering of these cultural practices in their own terms. We can
therefore read Figure 1 as a map which needs to be read inductively10.

We can read Figure 1 according to the way that variables are distributed within it. Those
that are on edges are relatively exclusive compared to those in the opposite parts of the
Figure. When variables are located closer together, this means that they co-exist amongst
the respondents. Thus looking at the top left hand side of Figure 1 we can see that those
who like science fiction books also like heavy metal music, for instance. By inspecting
these visual array, we are able to see what variables they are incompatible with11.

Figure 1 reveals a „flat‟ cultural map. By comparing the variables on the left hand side
with those on the right, we can see that those who are active, and who like cultural
activities and are on the left, those who are mainly inactive on the right. Unlike
Bourdieu‟s own analysis, this map does not primarily distinguish „high‟ from „popular‟
culture, but the „engaged‟ from the „disengaged‟. If we look at the top and bottom of
Figure 1, we see that those at the bottom like the established, „legitimate‟ cultural forms
of classical music, opera, art galleries and museums, whilst those at the top like the
commercial culture of popular music. If we then super-impose socio-demographic
variables onto this space (Figure 2), we can see that the first axis is associated with class

  The survey was a national random survey of 1564, with a response rate of 53%, and an ethnic boost of
200, which took place in 2003-04. Further details can be found in the special issue of Cultural Trends,
issues 2/3, 2006.
  Details on how this correspondence analysis was conducted can be found in Gayo-Cal et al 2006. We
acknowledge the support of Johs Hjellbrekke (Bergen), Brigitte LeRoux and Henry Rouanet (Paris V) in
this work.
   Of course, like any map, this is a specific kind of construction which depends on the operationalisation of
variables, etc (for details of which see Gayo-Cal et al 2006). Although this is a descriptive method, we do
not claim that this is a „neutral‟ description of an exterior reality.
   In order to aid interpretation, only those variables which contribute disproportionately to the two axes are
shown here. Others variables, which tend to be clustered towards the centre of the Figure and which are not
well separated are omitted. We only show axes 1 and 2 here, since these are significantly more important
than axis 3. For further discussion of the use of correspondence analysis in our work see Savage et al 2005
and Gayo-Cal et al 2006.

and educational qualifications (those in higher social classes and are well educated are on
the left), whilst the second axis is organized by age (younger people are at the top).

How far is this cultural map also a spatial map? How can we relate the abstract forms of
cultural taste and participation depicted in Figure 1 to people‟s sense of spatial
attachment and identification? Bourdieu himself is hesitant about how he sees the overlap
between geographical and social space, and this hesitancy is itself instructive. His
numerous rhetorical flourishes about the importance of physical space in his work rarely
amount to more than asides (see for instance, his account of „site effects‟ in The Weight of
the World, or of his account of the phenomenological framework for his thinking in
Chapter 4 of Pascalian Meditations). There are seemingly frustrating formulations.
„Social space tends to be translated, with more or less distortion, into physical space….
The space is defined by the more or less close correspondence between a certain order of
coexistence ‟ (Bourdieu 2000: 134-5, my italics). Motifs recur, the most important of
which is the way that the physical separation between Paris and the provinces is an aspect
of the operation of cultural capital (with a pun on „capital‟) itself (see Bourdieu 1985:
124; 1999: 125). These never amount to a fully elaborated theoretical framework,
however, but this is itself instructive.

Bourdieu‟s conception of the relationship between social and physical space needs to be
unravelled here. At one level, he models social space on the properties of physical space.

       „Just as physical space, according to Strawson, is defined by the reciprocal
       externality of positions (…), the social space is defined by the mutual exclusion,
       or distinction, of the positions which constitute it….. Social agents, and also the
       things insofar as they are appropriated by them and therefore constituted as
       properties, are situated in a place in social space….‟ (Bourdieu 2000: 134, italics
       in the original).

This insistence on the role of fixed physical space may fail to grasp the increasingly
dynamic and mobile nature of contemporary social and cultural forms, and implies a
„friction of distance‟ as if physical space itself imparts certain causal powers. „Thus the
distance of farm workers from legitimate culture would not be so vast if the specifically
cultural distance implied by their low cultural capital were not compounded by their
spatial dispersion‟ (Bourdieu 1985: 124), somehow implying that spatial dispersion has
its own metric independent of „culture‟.

Yet, although apparently making this claim about the parallel character of physical and
social space, elsewhere, Bourdieu insists that the idea of social space is simply a
heuristic‟, and not to be reified so that it treated as „objective‟

       The mere fact that the social space… can be presented as a diagram indicates that
       it is an abstract representation, deliberately constructed, like a map, to give a
       bird‟s eye view….. Bringing together… positions which the agents can never
       apprehend in their totality and in their multiple relationships, social space is to the
       practical space of everyday life, with its distances which are kept or signalled, and

       neighbours who may be more remote than strangers, what geometrical space is to
       the “travelling space” of ordinary experience‟ (Bourdieu 1985: 169).

With this move, Bourdieu recognises the way that his own correspondence diagrams are
themselves particular kinds of objectifications, specific abstractions which are not to be
confused with social relations themselves. Within this heuristic, descriptive, spirit there
are some interesting observations to be gleaned from Figure 1. Looking at the left hand
side, along axis 1, most of the variables arrayed indicate participation in particular kinds
of leisure activities (museums, rock concerts, art galleries, French restaurants, opera, the
cinema). On the other hand, most of those activities which one normally does „at home‟
do not register on Figure 1. Only seven of the sixteen genres of TV programmes
contribute to axes 1 and 2, and even these seven do not contribute strongly: the most
important is that those who like soap operas who are situated on the middle right hand
side of Figure 1. Film genres, which can also be viewed at home, are also relatively
unimportant. Musical taste by contrast is highly discriminating, and although most music
is also listened to at home, we can also see that the taste for musical genres is close to
participation at musical events: thus those who go to the opera and orchestral concerts are
in the bottom left of Figure 1, close to those who like classical music. Those who
sometimes go to rock concerts are close to those who like rock music, in the top left of
Figure 1. We can go further than this: with few exceptions, indicators of participation,
rather than taste, are on the edges of Figure 1, which means they are more socially

We can see that cultural tastes and activities themselves vary in the extent to which they
demand spatial exclusiveness, and Bourdieu‟s theoretical apparatus is more salient for
those activities which occur in public rather than domestic settings. As IT and other
media allow an increasing access to a range of information, entertainment, and leisure to
be lifted out of the landscape, and transmitted remotely, so they are less distributed
spatially. The „suburbanisation‟ of cultural life (Silverstone 1996; 2000) is likely to entail
the limited significance of the kind of socio-spatial exclusiveness that form the basis of
Bourdieu‟s approach. However, Figure 1 also reminds us that considerable numbers of
activities, those that still depend on being conducted in fixed physical locations: going to
French restaurants, art galleries and the opera, for instance, remain highly exclusive.
Elsewhere (Savage 2006b) I have argued that this janus-faced role of the home is deeply
important to grapple with: the home is both an inward facing space whereby mobile
technologies can be mobilized to allow communication at a distance, and it is also an
outward looking object that cannot but mark one off – socially through spatially – from
others and define „where you are coming from‟

However, we can take our mapping further than Bourdieu was able to, through plotting
the location of every single individual in our survey according to their responses to the
168 questions on their cultural taste and participation. One of the interesting features of
correspondence analysis is that it allows us to map individuals, as well as abstracted
variables. Figure 3 shows the locations of all the individuals in the survey, and also the
post codes of all those that we interviewed in depth in a follow up to the survey. By
comparing the testimony from their in-depth interviews with their position in social

space, we can consider how social and physical spaces relate in their testimonies. This
approach, I contend allows us to combine the sample survey with the case study in an
unusual and revealing way: we are able both to position individuals within the „abstract
space‟ of the national survey, and in their own, „local‟, words. We are thus uniquely able
to reflect on how social identities are related to the physical landscape and embody
different conceptions of space according to one‟s position within our cultural map.

Our in-depth interviews focused on people‟s cultural tastes in music, reading, eating,
personal and domestic appearances, but we also asked them how they came to live in
their current house, and what their dream house was. These two questions in particular
often allowed respondents to talk about their spatial attachments, but these also occurred
from time to time in response to other questions. As Figure 3 shows, we were able to
interview people from all corners of our cultural map, though here I focus on seeing how
those on the left hand side differ from those on the right hand side.

Our in-depth interviews revealed some similarities in how people talked about their
physical location, but also some telling contrasts, especially between those on the left
hand side of Figure 1, and those in the centre or right hand side. Two extended case
studies usefully reveal important differences. The couple located in the bottom left
quadrant, YO31, the corner of „established culture‟, who we shall call Jane and David,
epitomize the key tenets of what I have called „elective belonging‟ (Savage et al 2005).
Their autobiographical account of how they came to live in York emphasizes contingency
and choice. David ruminates

       How did we come to York then? Jane and I first met in 1984 working up in Fort
       William in Scotland and we did a season there together, we started going out
       together and when we left at the end of the year, she went home to the west coast
       of Scotland and I came back down to Leeds because my mother lives in Leeds.
       And we decided we wanted to work together again the following season so the
       three places we‟d earmarked that we thought we would like to work was York,
       Chester or Channel Islands and York was the nearest one to Leeds, I hopped on a
       bus from Leeds one day and came up, did the trawl round the job centres and I got
       myself a job at a hotel in York and started in the February 1985 and Cherie came
       down about the June or something, a few months later wasn‟t it.

Even though Jane and David had no local „roots‟ (he being Welsh, and she Scottish), they
had both chosen to put down roots in York, a city which they now completely identified
with. Their work was bound up with the leisure and tourism industries, and they lived in a
„period‟ Georgian flat in a fashionable terrace. In talking about their dream home, they
sought only to put down even stronger roots through claims on historical features of

       Jane. We‟d not live in this barn! I would love to have a really, really
       old house, not out in the countryside, in the town, a town house, a really really
       old, old house you know, about 14th century, that‟s what I would like
       Low ceilings and beams?

       Jane: Yes, and all sorts of funny little bits that you walk up to and down to, and
       that‟s what I would love. Do you know York at all, you know the water tower at
       Lendal, the river, the old water works, you know where Lendal bridge is? ….
       well there‟s a tower there that was actually a water tower and it‟s about 15th
       century and they‟ve just announced that they‟re turning it into flats and I‟m not
       normally an envious person, but I‟d sell my granny to buy one of those flats
       they‟re actually advertising it internationally, it‟s gonna go for a fortune, it‟s
       gonna be really something I think.
       So that would be not for what they’re doing in terms of all the modern fittings but
       the history of the building
       Jane: It‟s Grade 1 listed, they‟re actually gonna have to do it with like the Civic
       Trust and things so it‟s gonna be done in keeping, a property like that would be
       fantastic…… We have a friend that actually rented an apartment, when you go up
       along the Stonegate there‟s an opening on the left hand, York is full of these little
       hidden properties, you don‟t know anything about and it‟s a lovely kind of arcade,
       it‟s in towards a lawyer‟s office and there‟s a wrought iron balustrade and you go
       down and round it, there‟s this fantastic huge Jacobean house building that‟s in
       flats and the back garden overlooks ..... doesn‟t it and when Anne had her
       apartment, we went to visit, when you went inside it there was an enormous
       staircase, it was as wide as this whole room with huge stone balustrades that went
       right up and up and then divided in two, fantastic place it is. To have something
       like that, I don‟t mean the whole house , just a little bit.

We see here a strong claim on place, seen as an intertwining of instrumental and aesthetic
orientations. The spatial scale of interest is remarkably minaturised, with specific streets
and buildings being identified as being of special significance. There was little sense of
regional attachment, or even to their own national origins in Scotland and Wales, which
were seen as being of background relevance.

Contrast a very different way of talking about place, from an elderly, retired, widower
living in Airdrie in Scotland. G34 is on the bottom right hand side of Figure 2, on the less
engaged side of our „cultural map‟. Unusually, the interviewer himself had noted the
respondents‟ strong local attachment in his notes

       He kept mementoes of Airdrie in an old bureau, and showed me a poem about a
       mining disaster of 1939."I love Airdrie" he proclaimed. And you could see why.
       His life had been spent there, his family had been part of the history. His
       pseudonym was borrowed from his grandfather, a footballer for Airdrie who had
       gone to play for Bolton Wanderers in England.

The sense that Robert‟s biography was historically anchored in Airdrie comes over very
strongly at numerous points in the interview.

       I was born in a local avenue, and I come up from a big family and no one has left
       in Airdrie. The only one that ever left was my brother, he was killed in Anzio in
       1944, he was killed in Anzio beach-head and that was only when, my other

         brother‟s Derek and Alan and Keith was that one that was killed but Arthur and
         Jack and they were all n the Armed Forces. And then there was Terence and
         myself who done our national service. But we always loved Airdrie and a lot of
         memories and a lot of things.

This is a very different kind of history to that which Jane describes of York, for although
the historical features of the city are clearly very important to her, they were objectified
in the city‟s physical infrastructure. York history is portrayed as static and unchanging, in
contrast to the more intimate rendering of the intersection between autobiography and the
physical landscape discussed by Robert.
For Robert, there was a dichotomy between Airdrie, his home town, the locus of his daily
life, and places outside which represented spaces of power, and spaces of fantasy. The
locus of abstract power was Glasgow, on which he spoke especially clearly when talking
about his hostility to its cultural prentensions.

         when you see them - putting on exhibitions, human dirt, and they put that up and
         they put anything up, they put - one time there was in Glasgow and it was a pile of
         bricks, I says „what the hell are they talking about?12‟

         Well, I think myself like, we‟ve got a big .wire horse going into Glasgow, and it
         costs thousands of pounds but they‟d have been better putting a merry go round
         and letting weans go round and round. They‟re talking about, we‟ve got two
         lovely parks, Queen Mary and George here and they‟re talking about doing away
         with them

Glasgow figures here as an abstract space of cultural power, compared to the intimate
local sites he knows well, and the two are identified in relational terms, with Glasgow
benefiting from Airdrie‟s loss. When asked about his dream house, Robert also had a lot
to say.
         I think it would be a country cottage. I think , because I love animals and I love
         plants and I think , I‟ve seen ...... country cottage that you could actually make
         into your own and - seclusion would be great, just hearing water running and - I
         think that‟s - we could have that, we could have that, but again, you need
         *CA You could have had that some years ago if you had wanted rather than buy
         this, you could have moved out.
         Me and Betty was always talking about going to St Andrews and we always liked
         the area, every year we went for a fortnight but we seen a house in Langlands
         Road, and it was just near the football pitch. There‟s a wee football pitch ....... she
         thought it would be ideal…... We liked St. Andrew‟s, it would have been ideal
         because I love it, (you can leave) your door and you can walk any road

  The fact that the „pile of bricks‟ story is also told about the London galleries is instructive of how
Glasgow figures as an abstract space of power.

We see here the spatial separation of ideal, fantasy, location from where Robert actually
lived. Interestingly, having played with the idea of „abstract‟ country space, this is then
revoked in favour of St Andrews, also talked about in highly specific ways. But it is not
that Robert favours specific, „actual‟ spaces rather than abstract spaces. He is also happy
to evoke ideas of abstract space, with respect to certain kinds of leisure activities. He
ruminated about .

           A drive in the country to lovely seaside, lovely seaside and a good day of, me and
           my wife used to go to Linlithgow, and we‟d make it a day out because we .... big
           supermarket there so we‟d just go there and down the supermarket, there then go
           off and we used to sit by a loch and we went to Saltcoats it was the same. Visit a
           country manor‟ .

The difference between Jane and David, and Robert lies not in their balance between
abstract and particular spatial referents, but in the extent to which these could be
reconciled. For Jane and David, York figures as an intense space, whereas for Robert
abstract space was separated from lived space, even though the boundaries between these
were always contested and fluid. For Jane and David, abstract evocations about „history‟
could be levered onto specific locations. David wondered about choosing into live in „that
little Dutch house up by Clifton‟13. Jane responded, „Yeah, that‟s a strange little house
isn‟t it‟, to which David tellingly replied „It‟s a strange house up there, yeah. Something
with history, something with character‟. Here notions of history and character, whilst
appearing to evoke particularity and specificity, actually operate as highly general and
abstract terms.

This contrast between these different evocations of place and identity on the right and left
hand sides of Figure 2 is found amongst other interviews. The „engaged‟ and „privileged‟
on the left hand side of Figure 2 do not evoke an attachment to the „space of flows‟ but
are actually highly vested in their current location. Most of them identify a revised,
renovated version of their current house as being their dream house. What matters in part
is having the right kind of home, seen an interior space, as a machine for living, and
partly living in the right kind of place for „someone like us‟, one which is in keeping with
the biographical characteristics of the respondent. Maria, from Leeds (LS26) had a dream
about reconciling the abstract home as machine with her current house.

           I‟d want a library because we‟re both avid readers and we‟ve nowhere to put the
           books. We‟ve got the study but it‟s teeming at the seams with books and they‟re
           spilling into the bedroom and I don‟t want books in the bedroom because I‟ve
           been told off for having them there by my doctor before. So I want a big library.
           I‟d want somewhere for the PCs to be put side by side because we both like being
           in our PC, he doesn‟t like mine and I don‟t like his. But it splits us up and neither
           of us are happy with that. I‟d like a huge kitchen, I love cooking and I‟m teaching
           Fruit Bat14. He was the microwave king before he moved up here. I‟d have a
           huge garden with, it‟s a mess but we both love gardening and I‟ve not done any

     Clifton is an area of York, a few hundred metres from their current residence
     Fruit bat is the name of her partner, who Maria met on the web.

       this year, he‟s not because he‟s been looking after me. But if we had the money
       we‟d employ a gardeners, with more than just apple trees, we‟d have loads and
       loads of fruit trees. The vegetable patch would be bigger because we do like
       growing our own fruit and veg, but it‟s healthy and it‟s more taste. So a bigger
       garden. There‟d be somewhere to entertain outside, normally in summer we just
       take out the plastic chairs and tables and whoever‟s coming brings, like my
       brother when he came and his family, all the times my mum and dad come, they
       bring their own as well so we can just have the whole family in the back garden.
       It can be a bit of a squeeze if the boys want to play, a bigger garden, designated
       eating area with proper furniture there, not plastic. Three or four bedrooms,
       because I want a bedroom always to be ready for Fruit Bat‟s family when they
       come up. At the moment they have to sleep on a mattress down here because the
       study‟s in one bedroom, the second bedroom is decorated at the moment but it‟s
       bare floorboards and the walls are being plastered and everything so no one can
       sleep there. Yeah, that‟s it.

Maria evokes a private suburban home, detached from its surrounds. Interestingly, this
vision goes alongside her sense of escape from Leeds, where she had been brought up in
an area of working class terraced housing. This is in contrast to Susan, located on the
right hand side of Figure 2 (ML6). .

       Jane I would have this house. But in a different location!
       But that location wouldn’t be a million miles away from what you said
       No, I would go back to East Kilbride or if I could survive June, July and August, I
       would buy a great big villa in Majorca.

Or a similar kind of account was revealed by OX44, also on the top right hand side of
Figure 2.

       *Edie What sort of house I would have, probably still a four bedroom house but
       bigger downstairs and a big tree house or something in the garden for Timmy and
       maybe a little pool. But yeah, I don‟t have anything elaborate, I just want but if
       money were no object, maybe a place in Spain, but I don‟t know. Just think
       maybe a nice holiday home somewhere, that‟s what I‟d like, no big expectations.
       No, just a pool
       *Edie Just a pool and a big tree house!

This was of hiving off spatial zones appears clearly different from those located on the
left hand side of Figure 1. CF14 (3), a household comprised of a University lecturer and a
doctor, clearly identify their chosen location, a desirable corner of Cardiff, as entirely at
one with where they want to be. Having identified themselves as „urban people‟, Susan
went on to note that

       If money wasn‟t, I, I can‟t, I can‟t think of a place any, I suppose I would get a
       house like this but if it wasn‟t on a main road. That‟s one thing that I‟d like but
       apart from that, I don‟t think…
       *SA I was wondering what you could possibly find what you could complain
       about with this house. But yes, you’re quite right, if you’ve got children I suppose
       the main road could be…
       Yes, I‟d like it with a quite road for the kids but yes, I mean, we‟re really happy
       here. We had a much smaller house before and we just, I think, bit by bit we‟d
       just like to decorate it. But apart from that, I can‟t imagine anything more even if
       there was no, it‟s in the, it‟s inner city and, and yes, people that are really nice and
       I can‟t imagine wanting anything more than this, really

Few of these respondents refer to „nested places‟, that is places seen as being in the
hinterland of other locations. For all these respondents, spaces operate as granules, which
can be connected with any other kind of space. Unlike Jackson‟s Huddersfield interviews,
there is no sense of being judged from outside. However, those on the right hand side of
Figure 3 do articulate a sense that they cannot readily detach themselves from where they
live, and often articulate modes of belonging which hive off different kinds of attachment
according to specific functions: leisure, fantasy, and so forth. By contrast, on the left hand
side, the more privileged respondents seek to reconcile desire with their actual location,
and are highly vested in a place which they chose, and which then becomes a badge of
identity. It is precisely the fact that this chosen place is isolated as a granular, and cannot
easily be placed in a spatial hierarchy that allows them to identify it as a place of their


I have discussed many issues in this paper, so let me try to bring some threads together by
way of conclusion. Firstly, I have argued for the increasing power of granular spatial
identities associated with forms of „elective belonging‟. Rather than identifying or
disidentifying with generic categories - regions, towns, or even the countryside, this
involves identifying the kind of area – regardless, to some extent of location – as
appropriate for „people like us‟.

In making this argument, I have also been concerned to explore the kind of methods that
we need to allow us to explore these issues. Both the local case study and the national
sample survey fail, in my view, to register the complexity of the „enchanted landscape‟.
Spatial identities are produced by predominant modes of spatial classification that are
now institutionalized in numerous forms and which allow places at a distance to be
compared and made apparently comparable with each other.

We can, in this sense see processes by which social and spatial maps are increasingly
likely to coincide, as people locate and are located through principles of small scale
classification where small scale locations are treated as equivalent and not nested
hierarchically (through being seen as located within regions, cities, metropolitan regions

or whatever). Of course we need to recognize complexity and variation in local and
spatial identities, but nonetheless, the extent of variation from those which are revealed in
(even quite recent) historical accounts are striking. As I have argued elsewhere (Savage et
al 2005) rather than kinds of tensions between „born and bred‟ and „migrant offcomers‟
which have been so amply reported in post war community studies, we increasingly see
the tone of places set by middle distance migrants, those who move to a place because it
is seen to suit „people like them‟ and in the process generate a self fulfilling prophecy.

In examining the nature of spatial identities revealed by the Cultural Capital and Social
Exclusion project, we see that claims on place are central for advantaged social groups. In
contrast to the literature which portrays elites as somehow rising above space, detached
from any particular location, it actually appears that the engaged are highly vested in their
location. This of course is amply demonstrated also by the literature on gentrification
(Butler 2003). It is those who are less culturally engaged who feel trapped in space and
seek, if only in fantasy, to escape fixed locations. The concerns with location evident
from both groups are less with generic kinds of spaces - whether cities or regions, but
more in terms of the specific connotations of particular places.

This is my final point. Rather than concentrate on processes of spatial abstraction, it is
more interesting to note how far people focus on specific, fixed, even „auratic‟ locations.
Thinking about how we conduct socio-cultural mapping should be a central concern for
those with interests in this issue.

References (to add)

Butler, T., and Robson, G., (2004), London Calling, London, Berg

Deleuze, G., (2000), A Thousand Plateaux

Gayo-Cal, M., Savage, M., Warde, A, (2006), „A Cultural Map of the United Kingdom,
2003‟, Cultural Trends

Latour, B., (2005), Reasembling the Social, Oxford, Clarendon.

Savage, M., (2006), „Housing and the Spatialisation of Class‟, paper prepared for
CRESC/Housing Studies Conference, Manchester, February

Savage, M., (2007), „Histories of Belonging‟, forthcoming in Social Research Methods

Savage, M, Bagnall, G., Longhurst, B.J., (2005), Globalisation and Belonging, London,

Savage, M., and Burrows, R., (2006), „The coming crisis of empirical sociology‟,
forthcoming in Sociology


To top