Social Geography, Spatial Structure and Social Structure

Author(s): B. S. Morgan
Source: GeoJournal , November 1984, Vol. 9, No. 3, Social Geography (November 1984),
pp. 301-310
Published by: Springer

Stable URL: https://www.jstor.org/stable/41143395

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Springer is collaborating with JSTOR to digitize, preserve and extend access to GeoJournal

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
 GeoJournal 9.3 301-310 301
© 1984 by D. Reidei Publishing Com
0343-2521/84/0093-0301 £1.50

Social Geography, Spatial Structure and Social Structure

Morgan, B.S., B.Sc, Ph.D., Department of Geography, King's College London,

Strand, London WC2R 2LS, UK

Abstract: This paper adopts a positivist approach to social geography, which is viewed as a
study of the two-way relationship between spatial structure and social structure. This rela-
tionship is examined in cities in contemporary Britain and the United States. The concept of
social structure is discussed in the first section. It is suggested, after Blau, that social structure
may be delineated by parameters which demarcate the lines of differentiation among people,
created in their social interaction. Area of residence may be considered such a parameter.
Social structure comprises a complex configuration of these parameters. In general, coinciding
parameters limit social mobility and lead to an atomised society. The processes by which
social structure takes on spatial expression are discussed in the second section. The salience of
various parameters of social structure in a spatial setting is assessed first. Behaviouralist,
managerialist and structuralist approaches to residential differentiation are reviewed. It is
argued that most insight will come from studies of the interaction between groups of house-
holds and the major institutions of the housing market in the context of characteristics of
the housing supply. The impact of neighbourhood on social structure is examined in the third
section. The role of propinquity in social interaction is discussed along with the role of neigh-
bourhood as a status symbol. Individuals may change their position along structural para-
meters as a consequence of their residential location. Residential segregation is also a means,
through its role in circumscribing contacts, by which the continuation of the stratification
system is ensured in the next generation. In conclusion, it is noted that the isolation of racial
minorities in the disadvantaged parts of the city poses a danger to social stability.

Social geography has neither a unified conceptual frame- This positivist social geographer, who was trained in
work nor an agreed content. There is even disagreement the mid-1960s and was heavily influenced by the spatial
on the order of sub-division it represents in a classification analytic paradigm, is primarily concerned in his teaching
of geography. Buttimer's (1968, p. 134) description of and research to elucidate the relationship between spatial
social geography as "a field created and cultivated by a structure (the distribution of social groups) and social
number of scholars rather than an academic discipline" is structure. This is a two-way relationship. First, spatial struc-
as accurate today as when it was written fifteen years ago. ture is largely a reflection of aspects of social structure, and
This paper presents the views of one cultivator. No attempt it is incumbent on the social geographer to investigate the
is made either to argue this view of the subject against those processes whereby social structure takes on spatial expres-
of other scholars, or to erect a field boundary between the sion. Second, spatial structure acts to reinforce, and to
proper concerns of the sociologist and of the geographer in influence the evolution of social structure. The paper is
the manner of Schnore (1961). comprised of three parts, with consideration of these two

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
 302 GeoJournal 9.3/1 984

key issues entiate peopleprefaced

being in terms of a continuous status scale,by a
social engender inequality.
structure. Income, wealth, education and power
Discussion fo
where spatial are examples of graduated parameters. Socialare
structures interaction is mor
expected to be greater within groups than between them in
the case of nominal parameters; it is expected to be inverse-
Social Structure ly related to the status distance between persons in the case
of graduated parameters. Substantial correlation of nominal
More print has probably been devoted to discussion of with
the graduated parameters forms the basis of what Blau
concept of social structure in the sociological literature
terms ordinal parameters, which divide people into groups
than to the definition of social geography in the geo-with distinct boundaries that are ordered in a hierarchy of
graphical literature. An uncontested definition of social
ranks. Thus, for example, Duncan (1961) has created an
index of occupational status by ranking occupational
structure would be so general as to be almost meaningless.
groups by the education and income of their members.
Udy (1968) suggests that one or more of five components
may be recognized - individual, group, morphological, When the hierarchically ranked groups exhibit little overlap
systemic and cultural - which are given different emphases
in status and there is minimal ¡ntergroup interaction, an
according to the philosophical and theoretical viewpoint of
ordinal parameter may delineate quasi-castes. Blau suggests
the scholar. However, there are two conceptions of social
blacks and whites are quasi-castes because racial differences
in the United States and United Kingdom are closely
structure that are generically different. The first, associated
with Radcliffe-Brown (1940), holds that social structure is
associated with differences in power, prestige, income and
a system of social relations among differentiated parts of a
education.
collectivity. The second perspective sees social structure as Individuals become integrated in groups through
a system of logical relationships among general principlesprocesses of recurrent social interaction and communica-
tion. Blau maintains that these groups and their members
which is designed as a theoretical interpretation of social
life. In this view, "the term social structure has nothingbecome
to integrated in the larger social structure because,
do with empirical reality but with models built afterprovided
it" the parameters intersect, every person belongs to
(Levi-Strauss, 1963, p. 322). a variety of groups and has multiple roles. Ingroup relations
Social structure is used here in the first sense as along some lines are necessarily accompanied by ¡ntergroup
referring to the differentiated, inter-related parts relationships
of a along other lines. For example, a manager and
an unskilled worker may be members of the same residents'
society or group. It may be defined as a set of persistent,
association
patterned social relationships between groups of people in but face one another across the negotiating
different social positions. It must be emphasised thattable
this representing management and union. In contrast,
coinciding parameters lead to an atomised society because
is a restrictive definition, which emphasises the group
ingroup preferences become cumulative, and contact is
element to the exclusion of relationships between institu-
restricted within groups that share the same affiliations.
tions. Social structure, as defined, describes observable
The inter-relation of parameters also has significance
conditions. Most soch.' theories seek to explain this struc-
ture by reference to tc^ '¡ological, economic, culturalfor
andsocial change. Social mobility between groups has
psychological faeton. Tr.us, for example, Marx explainsprofound implications for social structure since it changes
the class structure ana class conflict on the basis of the the distribution of people among social positions. For
dialectical interplay of productive forces with productive example, it is generally agreed that there are three essential
conditions for the maintenance of a class system in society
relations. Parsons explains social relations in terms of value
orientations which he acknowledges makes him a "cul- - class cohesiveness, the individuals' awareness of belonging
tural determinist"(Parsons 1966, p. 113). to a class, and the self-consciousness of classes. These will
This essay draws heavily on Blau's (1974, 1977) con- be undermined if there is substantial intergenerational
social mobility and marriage between classes. Blau argues
ceptualization. He argues social structure may be delineated
that in general social mobility between positions is greatest
by parameters which demarcate the lines of differentiation
among people created in their social interaction. Socialwhen structural parameters are weakly correlated since this
structure comprises a complex configuration of these para-attenuates ingroup relations and intensifies ¡ntergroup inter-
meters, which he suggests should be kept analytically dis-actions. Conflict ¡s also dissipated since most individuals are
members of many different groups and have conflicting
tinct prior to an attempt at synthesis. He distinguishes two
types of parameter. Nominal parameters, which engender loyalties. This in turn opens channels of communication, so
that compromises are possible when conflict threatens. In
heterogeneity in a community, divide the population into
sub-groups with distinct boundaries but without contrast,
an structural consolidation, particularly of graduated
parameters such as power, prestige, training and skill, rein-
inherent rank-order. Sex, race, religion and, critically from
the present perspective, area of residence are examples of forces the social barriers between groups, and forces
nominal parameters. Graduated parameters, which differ- individuals to turn within their own groups for social sup-

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
 GeoJournal 9.3/1 984 303

port. This analysis is a summarizing

makes the techniquesocial
which repartitions thesy
to change, total varianceultimately
and of the distribution of a set of variables over fo
since social sub-areas into sets of factors
change is representing groups of variab-
represse
At the most les with similar patterns of variation. Several
general level, reviews have th
cerned with attempted to generalise about the spatial structure of wes- the
investigating
social structure
tern cities on the
and basis of this literature
spatial (Rees 1972, Timms st
this scale, 1971). To quote the structure
spatial author of an important recent text- is
dent variable. There is a voluminous literature which relates
book, "by far the major finding is that residential differen-
tiation in the great majority of cities is dominated by a
the effects of the widespread industrialization and urbaniza-
tion of western society in the last two hundred years onsocio-economic dimension, with a second dimension charac-
interpersonal relations (for example, Wirth 1938, Tonnies terised by family status/life cycle characteristics and a third
1957), on class structure (for example, Marx 1967 edition) dimension relating to segregation along ethnic divisions"
(Knox1981,p.81).
and on the spatial distribution of social groups in the city
(for example, Warnes 1973, Goheen 1970). However, This is a common, but rather misleading, conclusion.
economic and political conditions and cultural values are Socio-economic status is generally the pre-eminent factor,
also influential, particularly insofar as they interact which
to frequently accounts for over a third of the total
determine how a given economic system places people with variance. However, this reflects the spatial congruence of
regard to the allocation of scarce resources. For example,an important group of structural parameters - occupation,
Musil (1968) has shown how the ecological structure of
income, number of years schooling, employment status
Prague changed in the twenty years after the Second Worldamong them - rather than the degree to which any one
War as a consequence of the adoption of socialist housingparameter which is indicative of socio-economic status is
policies. Wissink (1962) interprets the structure correlated
of with area of residence. Other important struc-
American cities in terms ofthat societies 'cultural system'
tural parameters, such as racial and ethnic status, do not
- in particular the Anglo-Saxon penchant for the country- co-vary spatially and are consequently represented by
side and low density single family homes, the effect of the separate, relatively unimportant factors (for example, Rees
1970).
frontier on attitudes to land, and a firm belief in the rights
of the individual and the regulatory power of the market. The salience of particular parameters of social structure
However, I home in here on a more detailed considera- in a spatial setting are best summarised by segregation
tion of the interrelationship between social structure andindices. The index of residential differentiation (RDI) sum-
marises
the spatial structure in cities in contemporary Britain and the segregation of n social groups in a single
the United States, the environments in which I have most measure (Morgan 1975, Morgan and Norbury 1981). The
longer
research experience. What follows is essentially a personal standing, more frequently used, index of dissimilarity
commentary on the major strands of the relevant literature; (D) is a special case of RDI, where n=2. Both indices are
simple
particular attention is paid to methodological issues and the measures of the unevenness of the distributions of
development of an agenda for future research. social groups. An index of 100 is indicative of total segrega-
tion where no sub-areas are socially mixed (or, from the
perspective being developed here, a perfect correlation
Spatial Structure as a Reflection of Social Structure
between a structural parameter and area of residence), an
index of 0 represents an even distribution.
The accepted model of the dimensions of residential struc- Race is the structural parameter with the most clear-
ture in British and American cities reflects three of the cut spatial expression in urban America. The average
most important groups of parameters of social structure -
segregation between blacks and whites over census tracts
social status, family status and ethnic status. The pre-
in central cities in 1960 was 75.1, but over half the cities
eminence of these dimensions was first postulated by had scores above 87.8 (Taeuber and Taeuber 1965). Segrega-
Shevky and Williams (1949). Shevky and Bell (1955) tion decreased in the sixties, so that the mean index was
accounted for them in terms of the changing scale 68.3of by 1970 (Van Valey, Roof and Wilcox 1977).
society associated with modernization influencing changes
National origin is also a more salient dimension of residen-
in the range and intensity of relations, differentiation
tialof
differentiation than socio-economic status. I have cal-
function and complexity of organization. Modernization culated, for example, an RDI on a tract basis of 53.66 for
undoubtedly brings in its train changes in the parameters
the of
segregation of seven ethnic minorities (excluding blacks
social structure, but Shevky and Bell offer no explanation
but including Puerto Ricans) and native whites of native
of why these should take on spatial expression. parents (NWNP) in the city of Philadelphia in 1970. On a
However, their hypothesis about the major dimensions
pairwise basis, the highest index was 84.23 between Puerto
of residential differentiation have been confirmedRicans
by and persons of USSR stock (over 90 % of whom
numerous 'factorial ecologies' (Sweetser 1965). Factor
were Jews); the highest for European stock was 70.93

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
 304 GeoJournal 9.3/1 984

between particularly Marxists, argue that

latter the this only leads to super-
group and
21 .22 betweenficial understanding
NWNP since the scarce resources whichperso
and
1983). These managers and institutions allocate
results are are socially broadl
and
Guest and economically determined.
Weed (1976)They suggest real understanding
for et
Boston and Seattle. of British and American urban areas will only be achieved
In contrast, the average RDI for eight occupational by a consideration of the role of cities within the structure
groups in 50 randomly selected SMSAs in 1970 was 18.59, of advanced capitalist societies. Bassett and Short (1980)
with a range from 9.75 to 25.29. Simkus (1978) reportshave a recently provided a detailed description of these two
mean D of 46.00 between the two groups at opposite endsschools of thought.
of the status spectrum, professional workers and labourers, Much of the argument between proponents of the
in ten urbanised areas. As would be expected, educational choice and constraint perspectives has focused on whether
and income groups show similar degrees of segregation asindividual households or the institutions of the housing
occupational groups (Farley 1977). market are the more appropriate units of analysis for the
Cross-societal comparisons of residential differentia- social geographer. In the words of one influential writer
tions between the United States and United Kingdom are "individual households are actors in a play designed and
difficult because census data in the two countries is produced by capitalism whose actions are constrained by
tabulated for sub-areas with different population a stage managed by the 'urban gatekeepers'. Therefore, to
sizes.
However, the relative degree to which structural parametersexplain why cities are as they are, it is necessary to change
take on spatial expression appears similar. Ethnic segrega-our explanatory framework and adopt a new mode of
tion, although pre-eminent, is probably less acute analysis
in the. . . (which) stress(es) that institutions rather than
United Kingdom. For example, D, calculated on household a ward units are the key to understanding urban pro-
basis (with an average population of 10,000) in London cesses but
inalso that institutions are most fruitfully analysed
1970 for the segregation between the indigenous as reflections of capitalist society" (Gray 1975, p. 231-2).
popula-
tion and, first, persons born in the Caribbean and, second, Part of this disagreement concerning the fundamental
persons born in Pakistan were 50.92 and 48.96 respectively unit of analysis reflects differences in aims. We are largely
(Peach 1975). It is also possible to compare the degree concerned
tohere with understanding the bases of residential
which the occupational and life cycle parameters take on Managerial ists are generally more concerned
segregation.
spatial expression from published work on British with the allocation of scarce resources; Marxists with under-
cities
(Morgan 1975, 1976). standing the underlying mechanisms that generate the social
Three major approaches have been employed to structures associated with different modes of production,
account for residential differentiation. The first focuses on particularly the capitalist mode. Nonetheless, along with
the actions, motivations and choices of individual house- Johnston (1982) and Herbert (1979) but contrary to Dun-
holds. It emphasises the interaction of the needs and aspira- can (1980) and Eyles and Lee (1982), I see these as poten-
tions of households with the morphological and locational tially complementary routes to understanding segregation
characteristics of dwellings. This viewpoint interprets social not as alternatives to be used in an eclectic way. While the
status segregation, for example, in terms of households' structure of residential space cannot be understood with-
attitudes to residential areas as status symbols (Wirth 1938, out an appreciation of the supply of, and differential access
Warner and Lunt 1941); their desire to use spatial distance to, housing, this can only provide a partial explanation.
to minimise the social distance between themselves and Household decisions about where to live are variously con-
higher status groups (Oyen 1969); the wish of high status strained: at one extreme high income households with sub-
groups to have good access to scarce urban resources such stantial reserves of capital have a virtually free choice; at
as good schools (Cox 1973) and high quality medical carethe other, homeless families being rehoused by local
(de Vise 1973). authorities have little option but to accept the first acom-
An alternative explanatory framework which has modation they are offered (Smith and Whalley 1975).
recently gained ground emphasises the overriding impor- Given my definition of social structures, I would argue that
tance of constraints emanating from the operation of the most insight comes from studies of the interaction between
housing market over which the individual has little control. groups of households and the major institutions of the
There is a substantial difference of emphasis between two housing market in the context of the characteristics of
groups of scholars who give pre-eminence to the role of housing supply.
contraints. One group, who may be broadly defined as Unfortunately there are relatively few studies of this
'managerialists' (Pahl 1975) focus attention on the actions, type, although there is a substantial body of literature on
ideologies and affects of various individuals (for example, the position of blacks in the housing market in the United
landlords and estate agents) and institutions (for example, States. In this country, Rex' and Moore's (1967) study of
building societies and local authorities) involved in the Sparkbrook, Birmingham was an important milestone.
supply and allocation of housing. In contrast, structuralists, They developed the concept of housing classes to describe

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
GeoJournal 9.3/1 984 305

groups of households with common housing opportunities Rose (1981 ) in four cities in the United States has explored
structured by economic, bureaucratic and political forces. the preferred residential environments of black profes-
This notion of class is explicitly based on Weber "who sawsionals who have the greatest opportunity to escape the
that a class struggle was apt to emerge whenever people
ghetto. In this country, Phillips (1981) has investigated the
in a market situation enjoyed differential access to housing experience of high status Asians who are beginning
property" (Rex and Moore 1967, p. 273). Two major
to settle in the suburbs of Leicester.
criteria were used to define housing opportunities - the This is in stark contrast with the failure of most ques-
size and stability of family income which reflected access tionnaire surveys of individual households to yield insights
to mortgage finance, and the ability to meet the 'need' and about family status segregation, which has been interpreted
length of residence qualifications to enter public sectoras the outcome of families changing their residence in res-
housing. ponse to changes in their housing and locational require-
Rex and Moore pointed out that housing classes took ments as they progress through the family cycle (Abu-
on territorial expression, but their primary object was to Lughod and Foley 1960). The evidence for this is tenuous:
understand the housing situation of low income West stage in the life-cycle (Speare 1970), marital status (Speare
Indian and Asian immigrants in the inner city. For our et al. 1974) and age of head of household (Long 1972) and
purpose, the relevant groups are defined by the parameters increase in family size (Brown 1975) are significant cor-
of social structure which take on spatial expression, that is relates of household mobility; housing space complaints are
occupational groups, racial groups, age groups and so on. the most frequently cited reasons for moving (Clark 1970,
These groups do not constitute housing classes because Herbert 1973). In many behavioural analyses no attempt
they are not all in competition for the same housing. For has been made to aggregate households into social groups,
example, a retired couple will not seek out the same type and in even fewer has there been emphasis on the impor-
of housing as a large family, an Italian family is unlikely tance of the housing market and housing costs in constrain-
to be in competition for housing in a Jewish area of an ing households' freedom of action (although Rossi's (1955)
American city. pioneering study of residential mobility in Philadelphia is a
The eventual aim must be to develop generalizations noteable exception). Murie (1974) was the first to dissent
about the housing aspirations and the constraints acting from the received wisdom. He concluded, from a survey of
upon the sub-groups delineated by cross-cutting structural movers in West Yorkshire, that "there is a wide and contra-
parameters: that is, for example, for high income high dictory variation in patterns of housing use among house-
status whites in the post-child phase of the family cycle, holds at similar stages of the family cycle" (Murie 1974, p.
for low status unemployed blacks in the child-bearing stage 114). This is scarcely surprising since the family cycle is a
and so on. We are a long way from realising this aim. nominal parameter which is cross-cut by such graduated
Choices may only be meaningfully considered when all the parameters as race, income and occupation, but we still
relevant constraints facing groups have been understood. have gained little insight into the extent to which various
Ideally a hierarchical sequence of research is envisaged in groups of households are constrained from satisfying their
which generalizations are developed first for quasi-castes changing needs. It even remains to ascertain the relative
such as racial groups, next for groups formed by ordinal importance of 'mover' and 'stayer' households on family
parameters and such as occupational groups, and finally status segregation. I have recently suggested (Morgan 1976)
for nominal groups such as family cycle stages. That is, on the basis of a study of Exeter, that the latter may exert
for example, we should only attempt to generalise about the greater influence on residential patterns.
white professional workers and black professional workers,
not professional workers as a whole, and about households
of a particular race, occupational group and family cycle
The Impact of Spatial Structure on Social Structure
stage rather than about all households at one stage of the
cycle. A parameter of social structure demarcates the lines of
It is instructive to review briefly the existing literature differentiation among people created in their social inter-
from this perspective. It has been noted that quasi-castes action. Blau (1974) considers area of residence to be a
are more segregated than groups defined by ordinal para- parameter. He suggests that if the various groups in a city
meters, which in turn are more residentially differentiated are located in different neighbourhoods, most encounters
than those defined by nominal parameters. The extent to will involve members of the same group, and there will be
which valid generalizations have been developed for groups little opportunity for ¡ntergroup contact. This strengthens
defined by a single structural parameter also declines from group barriers and increases the social distance between
quasi-castes to nominal groups. For example, substantial groups. Although this is a special case of the general prin-
understanding has been achieved of the constraints imposed ciple that correlated structural parameters counteract the
on racial minorities as a consequence of the operation of impact of heterogeneity rather than a distinctive feature of
the housing market. From this sound base, recent work by location, it is critically important to the social geographer.

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
306 GeoJournal 9.3/1984

However, there is some difference of view about the whereby the city is partitioned into several village-like areas
role of propinquity in social interaction which is reviewed where the groupings of people are of manageable propor-
by Irving (1977). At the simplest level, there is abundant tions is a means of social control to cope with the
evidence that interaction rates fall off with increasing dis- heterogeneity and potential for conflict and exploitation
tance (Morrill and Pitts 1967). Much of the disagreement that exist in a city. The residents of these distinctive areas
stems from the varying value and meaning of different are "distinctive earmarks which the residents, outsiders and
types of contact, and the role of propinquity as against external organizations seize upon to sort one another out
other structural parameters. For example, Gans (1961, for selective treatment" (Suttles 1972, p. 15/16). He dif-
1967) argues that shared interests and norms are more ferentiates between the responses of the rich and the poor:
potent factors in friendship formation than distance, the rich can select a residential area where the character of
although he accepts propinquity may be a necessary factor neighbours is assured by the costs of housing excluding the
because it influences the likelihood of meeting. He points less respectable members of society; the poor, in contrast,
out that much of the detailed research on the role of strive to attain the same result by personal covenant among
themselves
propinquity on social interaction has been carried out in to produce what Suttles terms the 'defended'
homogeneous communities (for example, Festinger,
neighbourhood.
Schacter and Back 1950), and that instances of conflict
While some of Suttles' views may be rather overstated
and lack generality, there is little doubt that some people
are as probable as those of cooperation in an heterogenous
do use area of residence as a status symbol, and that neigh-
neighbourhood. However, he does concede that homogen-
eity in one or two critical respects may be sufficient
bourhoodtoidentity is a reference point against which people
promote interaction. 'The fact that most people (in In some instances, neighbourhood names are
are assessed.
Levittown, New Jersey) were similar enough in even used as terms of status and rank. Warner and Lunt
age and,
(1941) note in their classic study of Yankee city that
to a lesser extent, income, enabled them to become friendly
with people of different occupations, religions,certain
ethnicgeographical terms were used to describe people's
place
backgrounds, or regional origins for the first time in in the rank-order. For example, the term 'Hill
their
lives" (Gans 1961, p. 177). Streeter' designated the upper class group, whereas 'River-
brooker' connoted the lowest status group. This suggests
It is difficult to accept Gans' views in their entirety.
that area of residence should possibly be considered an
Allport (1954) has specified four conditions which are
ordinal parameter.
necessary if contact is to lead to greater respect and accept-
We have considered so far the effect of spatial structure
ance rather than conflict and rejection. Prejudice is lessened
when two groups possess equal status, seek common ongoals,
social structure through its influence on patterns of
social interaction. However, individuals may also change
are cooperatively dependent on one another, and interact
their position along certain structural parameters as a
with the positive support of authorities, laws or custom.
consequence
There is substantial evidence to support this view with of their residential location. It is useful to
respect to the integration of blacks and whites differentiate
in stable between intra- and inter-generational changes
in this respect.
interracial neighbourhoods in the United States. Repeated
studies have found that integrated living in public housing
The best documented examples of intra-generational
developments which meet Allporťs criteria results in ex-
changes are to be found in the literature on the assimilation
change and accommodation of views and sharplyof reduces
ethnic groups in the United States. At the time of their
racial prejudice on both sides. Many whites reevaluate the these groups were forced into the areas of worst
arrival
stereotypes about blacks as a group that they have housing,
learnt but within these areas ethnic enclaves developed
through their previous socialization experience, and as react
immigrants sought out their fellow-countrymen as neigh-
to them as individuals rather than as an homogeneous bours. Group structures and norms were preserved within
category (Deutsch and Collins 1951, Jeffries and Ransford these enclaves while adjustments to living in an urban
1969). Rokeach (Rokeach et al. 1960, Rokeach and Mezei environment in a strange country were taking place. As
1966) goes so far as to suggest that white American rejec- immigrants became more assimilated into American society,
tion of blacks is motivated by assumed belief and valueand particularly as they became upwardly socially mobile,
differences. Seen in this light, residential integration offersthey began to disperse through the residential areas of cities
the hope that neighbours will learn that they do share some(for example, Myers 1950). However, there is evidence that
beliefs and values, while residential isolation may lead inthe degree to which an immigrant group is segregated may
time to the development of genuine differences, making independently influence ethnic behaviour and assimilation.
interracial contact in the future less likely. For example, Lieberson (1963) has assessed with reference
Some scholars argue that residential areas perform a ten cities in 1930 the role of residential mixing in the
to
more vital function than influencing the pattern of social ability of an immigrant group to speak English. It would
relations. Suttles (1972) suggests residential segregationbe expected that isolated foreign groups would have less

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
GeoJournal 9.3/1 984 307

reason and opportunity to learn English than groups morenalised values and norms are deeply embedded in neighbour-
widely dispersed among the native white population. hood experiences, settings and social cues, so that local chan-
Holding the median year of arrival of a group constant, he nels of communication support intergenerational communi-
reports partial tau correlation coefficients as high as .79 in cation of common values, beliefs and aspirations. Working
Philadelphia, .71 in Boston between the ability of a group class communities, therefore 'produce' individuals with
to speak English and its segregation from native whites. Hevalues conducive to being in the working class, for example,
also demonstrates that the likelihood of a group taking out and these values, which are deeply embedded in the cog-
citizenship is related to its residential dispersal, although innitive, linguistic and moral codes of the community,
both cases it is impossible to demonstrate the direction ofbecame an integral part of the conceptual equipment
causality. However, it is not clear how typical the case of individuals use to deal with the world. Finally, one of the
assimilation of ethnic groups is as an example of the impactmajor determinants of an individual's life chances, the
of residential propinquity on social structure. For example, school that he attends, is in many cases determined by the
Goldthorpe and Lockwood (1967) were able to find littleneighbourhood in which he lives.
evidence that affluent car workers in Luton were adopting
One does not have to be a marxist to accept much of
middle class ways of life despite their relative affluence and
Harvey's argument. Sociologists agree that the neighbour-
suburban residence among white collar workers.
hood is one of the most important sources of socialization
A group's residential location may cause some of its
after the home. There is substantial evidence that educa-
members to change their position along one or more
tional spending and attainment differ between neighbour-
structural parameters other than as a consequence of social
hoods. For example, Cox (1973) reports that the City of
interaction. The most startling example is the disad-
Los Angeles in 1967 had a pupil/teacher ratio of 27 and
vantaged position of blacks in American and, to a lesser
spent $610 per pupil, whereas Beverly Hills, one of its
extent, British cities in terms of access to jobs. The distance
suburbs, had a ratio of 17 and spent $1,1192 on each
between the place of residence of black workers in the
child. Furthermore, evidence is accumulating (Armor
inner city and employment opportunities have increased
1980) that a considerable number of white parents have
substantially in the post-war period with increasing indus-
moved from central cities which have a substantial black
trial and office developments in the suburbs. For example,
minority to white suburbs to frustrate court imposed
Christian (1975) has charted the loss of jobs from the black
school desegregation plans. There is less variation in educa-
ghetto in Chicago between 1965 and 1971: 22,000 jobs
tional provision between London boroughs, but there are
were lost to the ghetto as a consequence of over 1 ,000
substantial differences in attainment. For example, the
firms closing or moving away. The suburban locations
mean literacy score of children of non-manual parents in a
which many of the firms opted for, some as far as twenty-
Greenwich ward in 1968 indicated that they were two years
five miles away, are only accessible by car but under 20 %
ahead of children with a similar background in a ward in
of the black population owned cars. The inevitable con-
Tower Hamlets (Panton 1980). Panton goes on to demons-
sequence is that blacks have greater difficulty finding
trate that differences in the mean score in a sample of inner
employment than they otherwise would, and those that do
may be forced to accept low status jobs which are notLondon wards are related to their social composition - the
more middle class the ward the higher the score. Coleman
commensurate with their level of skill and experience.
(1966) found a similar neighbourhood effect in the United
Although Marx did not concern himself with residen-
States.
tial structure and latter-day marxists differ in the
importance they attach to place and spatial relations, the Beshers (1962) argues from a different perspective that
most coherent statement concerning the impact of spatialresidential differentiation is the means by which the con-
structure on intergenerational changes in social structure tinuation of the stratification system is ensured in the next
has been made by the marxist geographer, David Harvey generation through its role in circumscribing the choice of
(1975). He argues that while mobility chances must always marriage partners. For a stratification system to survive not
be open to the population, capitalist society must findonly must there be a limited amount of intergenerational
some way to organise and structure mobility chances insocial mobility but there must also not be substantial inter-
order to ensure social stability. Once inter-generationalmarriage between strata. It is generally agreed that higher
mobility is limited, social distinctions become relativelystatus parents seek to manipulate as far as possible the
fixed. He suggests that residential differentiation should besocial setting options that are available to their offsprings,
interpreted in terms of the social relations within capitalistand consequently their opportunities to interact with other
society, since segregation brings about differential access toindividuals on an equal footing. Besher argues the choice of
the scarce resources required to acquire 'market capacity' residential location is influential in this respect because an
(Giddens 1973). Market capacity from this viewpoint com-individual's choice of marriage partner will be constrained
prises a whole set of attitudes, values and aspirations as well
by whom she meets, and as she will more acquaintances
as distinctive skills. Socialization, early images and inter- locally than on the other side of the city, she will be more

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
308 GeoJournal 9.3/1 984

likely to marry the 'right' man if her parents are living in a form of collective behaviour that usually takes place in
the 'right' neighbourhood. It should also be noted, although response to perceived disadvantages; they may be seen as
Beshers does not emphasise it, that the effect of the local part of a continuum of protest which may range from
milieu on value systems is also likely to be influential. isolated actions undertaken by individuals to full-scale
There is evidence to support the view that segregation insurrection. It will be recalled that Blau suggests that the
underpins homogamy. Many studies (for example, Jones structural consolidation of graduated parameters reinforces
1967, Peach 1980) have illustrated that there is a close the social barriers between groups and forces individuals
association between the degree of homogamy and the to turn within their own groups for social support; at the
degree to which a group is residentially isolated, although extreme a revolutionary situation may be fostered- because
this may merely reflect the fact that perceived social social change is repressed until it erupts violently.
distance influences the choice of marriage partner and It has been shown that blacks in the United States are
residential location. A second type of analysis focuses on almost totally residentially isolated, and that West Indians
individuals. For example, I have demonstrated that the and Asians are highly segregated in British cities. While area
contextual position of the bride's parents in the social of residence is but one of a large set of correlated para-
environment of the local neighbourhood influences, albeit meters, it is one of the more critical since, as has been illus-
weakly, the chances of their daughter making a good trated, it influences others such as employment status and
marriage (Morgan 1978), and that there are more marriages educational attainment. The social and economic condi-
between persons from similar social environments even tions in ghettoes and some ethnic enclaves are partly
after controlling for the effects of homogamy and pro- influential in maintaining the self-perpetuating cycle of
pinquity (Morgan 1981b). poverty in which many racial minorities find themselves,
The critical unresolved question in studies of both while residential propinquity forms the basis of com-
marriage and other types of social interaction is the relative munities which may eventually explode in the search for
importance of distance and territorially on patterns of social and economic justice. However, rioting is not evenly
contact. On the one hand, as has been noted, there is distributed. Warren (1969) argues that the local neighbour-
abundant evidence that in general contact rates decline hood is more of a key social unit in black than in white
regularly with distance; on the other hand, some socio- communities because the ghetto compresses their arena of
economic divides operate as barriers to interaction. The social interaction and enforces their isolation from the
boundary of the black ghetto in North American cities is
wider society. He goes on to show, with reference to
one example of such a barrier, while Boal (1969) has participation in the 1967 Detroit riots, how neighbour-
demonstrated that the Shankill-Falls divide between
hoods with particular types of social structure were more
Protestants and Catholics in Belfast is another. More sur-
prone to rioting.
prisingly, he has also shown that two adjacent post-war
The economic and educational status of blacks has
neighbourhoods in south-west Belfast which differ only in
improved substantially in the last two decades in the United
their social status also comprise almost totally separate
States, while many West Indian and Asian immigrants have
territories. Residents of the high status neighbourhood of
become structurally assimilated into British society. Un-
Upper Malone had more contact with people from a high
fortunately, at least in America, this economic advance has
status area some miles distant than with residents of neigh-
not been matched by progress in developing an open
bouring Taughmonagh. There is an urgent need for more
housing market. The continuing isolation of racial
studies to focus on the separate and joints effects of dis-
minorities in the most disadvantaged parts of the inner city
tance and the social environment on patterns of social inter-
must pose a threat to the social stability in both Britain and
action between like and unlike individuals to improve
the United States.
understanding of the impact of spatial structure on social
structure. The paper has discussed social geography as the study
of the relationship between spatial structure (the traditional
concern of geography) and social structure (the 'subject
matter' of sociology). No attempt has been made to differ-
Some Concluding Comments entiate between the proper concern of the geographer and
sociologist: the paper has drawn heavily on the work of
This essay has attempted to illustrate the ways in which both. Both urban sociology and urban social geography
have
social structure takes on spatial expression, and the role of common roots in the Chicago school of human
spatial structure in turn in influencing the evolution ofecology.
social Although Park (1952) noted that spatial distance
and social distance are mutually reinforcing, the classical
structure. In conclusion, in the light of the widespread
noting in British cities in the summer of 1981 it is useful ecological
to position holds that spatial distance is not impor-
consider the role of residential segregation of racial tant in and of itself, but only insofar as it reflects social
relationships. This remains an influential view in some
minorities as a contextual factor in riot behaviour. Riots are

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
 GeoJournal 9.3/1 984 309

sociological ecology
circles. For andmost
the taug
in the path of an
sociologists
importantuntil
area
been to the fore in
with recent
their work
deeply
of place and distance
and onrelatio
spatial socia
matter dealt with
tanthere,
role whether
to play i
geography and taught in a geogr

References

Allport, G.W.: The Nature of Prejudice. Addison Wesley 1954. Gans, H.J.: The balanced community: homogeneity or heterogeneity
Abu-Lughod, J.L., Foley, M.M.: Consumer differences. In: Foote, in residential areas? Journal of the American Institute of
N.N., Abu-Lughod, J.L., Foley, M.M., Winnick, L. (eds.), Planners 27 (1961)
Housing Choices and Constraints. McGraw Hill 1960. Gans, H.J.: The Levittowners. Allen Lane 1967.
Armor, D.: White flight and the future of school desegragation. In: Giddens, A.: The Class Structure of the Advanced Societies. Hut-
Stephan, W.G., Feagin, J.R. (eds.), School Desegregation: Past, chinson 1973.
Present and Future. Plenum 1980.
Goheen, P.: Victorian Toronto, 1850-1900. Research Paper 127,
Bassett, K.A., Short, J.R.: Housing and Residential Structure: Department of Geography, University of Chicago ( 1 970)
Alternative Approaches. Routledge and Kegan Paul 1980. Goldthorpe, J.H., Lockwood, D.: Affluence and the British class
Beshers, J.M.: Urban Social Structure. Free Press 1962. structure. Sociological Review 11 (1963)
Blau, P.M.: Parameters of social structure. American Sociological Gray, F.: Non-explanation in urban geography. Area 7 (1975)
Review 39(1974) Guest, A.M., Weed, J.A.: Ethnic residential segregation: patterns of
Blau, P.M.:: Inequality and Heterogeneity. Free Press 1977. change. American Journal of Sociology 81 (1976)
Blau, P.M. A macrosociological theory of social structure. American Harvey, D.: Class structure in a capitalist society and the theory of
Journal of Sociology 83 (1978) residential differentiation. In: Peel, R., Haggett, P., Chisholm,
Boal, F.W.: Territoriality on the Shankill-Falls divide, Belfast. Irish M. (eds.), Processes in Physical and Human Geography: Bristol
Geography 6 (1969) Essays. Heinemann 1975.
Boal, F.W.: Territoriality and class: a study of two residential areas
Herbert, D.T.: Residential mobility and preference: a study in
in Belfast. Irish Geography 6 (1971)
Swansea. In: Clark, B.D., Gleave, M.B. (eds.), Social Patterns
Brown, H.J.: Changes in workplace and residential locations. Journal
in Cities. Institute of British Geographers 1973.
of American Institute of Planners 41 (1 975)
Herbert, D.T.: Introduction: geographical perspectives and urban
Buttimer, A.: Social geography. In: Sills, D.L. (ed.), International
problems. In: Herbert, D.T., Smith, D.M. (eds.), Social Prob-
Encyclopedia of the Social Sciences. Macmillan and Free Press
1968.
lems in the City. Oxford University Press 1979.
Irving, H.W.: Space and environment in interpersonal relations. In:
Christian, CM.: Emerging patterns of industrial activity within large
metropolitan areas and their impact on the central city work Johnston, R.J., Herbert, D.T. (eds.), Geography in the Urban
force. In: Gappart, G., Rose, H.M. (eds.), The Social Economy Environment: Progress in Research and Applications 1 (1978)
of Cities. Urban Affairs Annual Review 9 (1975) Jeffries, V., Ransford, H.E.: Interracial social contact and middle
Clark, W.A.V.: Measurement and explanation in intra-urban residen- class white reactions to the Watts riot. Social Problems 16
tial mobility. Tijdschrift voor Economische en Sociale Geo- (1969)
grafie 61 (1970) Jones, F.L.: Ethnic concentration and assimilation: an Australian
Coleman, J.S.: Equality of Educational Opportunity. United States case study. Social Forces 45 (1967)
Department for Education and Welfare 1966. Johnston, R.J.: On the nature of human geography. Transactions,
Cox, K.R.: Conflict, Power and Politics in the City: a geographic Institute of British Geographers NS 7 (1982)
view. McGraw Hill 1973. Knox, P.: Urban Social Geography: An Introduction. Longman
Deutsch, M., Collins, M.: Interracial Housing: A Psychological 1982.

Evaluation of a Social Experiment. University of Minnesota Levi-Strauss, C: Structural Anthropology. Basic Books 1963.
Press 1951. Lieberson, S.: Ethnic Patterns in American Cities. Free Press 1963.
Duncan, O.D.: A socio-economic index for all occupations. In: Long, L.: The influence of number and ages of children on residen-
Reiss, A.J. et al. (eds.), Occupations and Social Status. Free tial mobility. Demography 9 (1972)
Press 1961. Marx, K.: Capital. International Publishers 1967.
Duncan, S.S.: Book review. Environment and Planning A 12 (1980) Morgan, B.S.: The segregation of socio-economic groups in urban
Eyles, J., Lee, R.: Human geography in explanation. Transactionsareas: a comparative analysis. Urban Studies 12 (1975)
of the Institute of British Geographers NS 7 (1982) Morgan, B.S.: The bases of family status segregation: a case study
Farley, R.: Residential segregation in urbanised areas of the United in Exeter. Transactions, Institute of British Geographers NS 1
States in 1970: an analysis of social class and racial differences. (1976)
Demography 14 (1977) Morgan, B.S.: Residential segregation, marriage and the evolution
Festinger, L., Schacter, S., Back, K.: Social Pressures in Informal of the stratification system: a case study in Christchurch, New
Groups. Harper 1950. Zealand. Environment and Planning 11 (1978)

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms
 310 GeoJournal 9.3/1 984

Morgan B.S.: A Rose, H.M.:: The black professional and residential segregation in the to
contribution t
quity and segregation. American city. In: Peach, C, Robinson,Journal
V., Smith, S.J. (eds.),
(1981) Ethnic Segregation in Cities. Croom Helm 1981 .
Morgan, B.S.: A Rossi, P.H.: Why Families Move.
case A Study in the Social Psychology
study of et
methodology. of Urban Residential Movement. Free Press 1955.
Unpublished pape
King's College Schnore, London
L.: Geography and Human Ecology.(1983) Economic Geography
Morgan, B.S., 37 (1961)
Norbury, J.: Some
of residential Shevky, E., Williams, M.: The Social Areas of Los Angeles. Univer-
differentiation. D
Morrill, R.L., Pitts,sity of California PressF.R.:
1 949. Marria
mation field. Annals of the Association of American Geo- Shevky, E., Bell, W.: Social Area Analysis. Stanford University Press
graphers 57 (1967) 1955.

Mune, A.: Household movement and housing choice. Occasional Simkus, A.A.: Residential segregation by occupation and race in ten
Paper 28, Centre of Urban and Regional Studies, University urbanised areas, 1950-1970. American Sociological Review 43
of Birmingham (1974) (1978)
Musil, J.: The development of Prague's ecological structure. In: Smith, D., Whalley, A.: Racial Minorities and Public Housing.
Pahl, R.E. (ed.), Readings in Urban Sociology. Pergamon 1968. Political and Economic Planning Broadsheet 556 (1975)
Myers, J.K.: Assimilation to the ecological and social systems of a Speare, A.J.: Home ownership, life cycle stage and residential
community. American Sociological Review 15 (1950) mobility. Demography 7 (1970)
Oyen, D.: A search for predictors of residential mobility: a follow- Speare, A.J., Goldstein, S., Frey, H.W.: Residential Mobility, Migra-
up study of contextual position. Acta Sociologica 12 (1969) tion and Metropolitan Change. Ballinger 1974.
Pahl, R.E.: Whose City. Penguin 1975. Suttles, G.T.: The Social Construction of Communities. University
Panton, K.J.: Literacy in London. Geography 65 (1980) of Chicago Press 1972.
Park, R.E.: The Human Community. Free Press 1952. Sweetser, F.L.: Factorial Ecology: Helsinki 1960. Demography 2
Parsons, T.: Societies: Evolutionary and Comparative Perspectives. (1965)
Harvard University Press 1966. Taeuber, K.E., Taeuber, A.F.: Negroes in Cities: Residential Segrega-
Peach, G.C.K.: Immigrants in the inner city. Geographical Journal tion and Neighbourhood Change. Aldine 1965.
141 (1975) Timms, D.W.G.: The Urban Mosaic: Towards a Theory of Residen-
Peach, G.C.K.: Ethnic segregation and intermarriage. Annals of the tial Differentiation. Cambridge University Press 1971.
Association of American Geographers 70 ( 1 980) Tonnies, F.: Community and Society (translated by C.P. Loomis).
Pettigrew, T.F.: Attitudes on race and housing: a social-psychol- Harper 1957.
ogical view. In: Hawley, A.H., Rock, V.P. (eds.), Segregation Udy, S.H.: Social structural analysis. In: Sills, D.L. (ed.), Interna-
in Residential Areas. National Academy of Sciences, Washing- tional Encyclopedia of the Social Sciences. Macmillan and
ton, DC 1972. Free Press 1968.
Phillips, D.: The social and spatial segregation of Asians in Leicester.
Van Valey, T.L., Roof, W.C., Wilcox, J.E.: Trends in residential
In: Jackson, P., Smith, S.J. (eds.), Social Interaction and Ethnic
segregation. American Journal of Sociology 82 ( 1 977)
Segregation. Academic Press 1981 .
Radcliffe-Brown, A.R.: On social structure. Journal of the Royal de Vise, P.: Misused and misplaced hospitals and doctors. Resource
Anthropological Institute 70 (1940) Paper 22, Commission on College Geography, Association of
Rees, P.: Concepts of social space: toward an urban social geo- American Geographers (1973)
graphy. In: Berry, B.J.L., Horton, F.E. (eds.), GeographicWarnes, A.M.: Residential patterns in an emerging industrial town.
Perspectives on Urban Systems. Prentice Hall 1970. In: Clark, B.D., Gleave, M.B. (eds.), Social Patterns in Cities.
Special Publication 5, Institute of British Geographers (1973)
Rees, P.: Problems of classifying sub-areas within cities. In;: Berry,
B.J.L. (ed.), City Classification Handbook: Methods Warren,
and D.I.: Neighbourhood structure and riot behaviour in
Applications. Wiley 1972. Detroit: some exploratory findings. Social Problems 16 (1969)
Rex, J., Moore, R.: Race, Community and Conflict. Oxford Uni-
versity Press 1967. Wilson, A.B.: Residential segregation of social classes and aspiration
of highschool boys. American Sociological Review 24 (1959)
Rokeach, M., Smith, P., Evans, R.: Two kinds of prejudice or one?
In: Rokeach, M. (ed.), The Open and Closed Mind. Basic Books Wirth, L.: Urbanism as a way of life. American Journal of Sociology
1960. 44(1938)
Rokeach, M., Mezei, L.: Race and shared beliefs as factors in social Wissink, G.N.: American Cities in Perspective. Royal VanGorcum
choice. Science 151 (1966) 1962.

This content downloaded from

152.118.24.31 on Tue, 20 Sep 2022 21:39:11 UTC
All use subject to https://about.jstor.org/terms