Contested  Terrains: Social  Relations
of  Production  and  the  Living  Place

K.  P.  Moseley

In petty commodity production, which flourishes in most pre-industrial cities,
residence  and  workplace  are often  joined,  and  a good  part  of  subsistence
may  still be produced  within  the household  itself.  Capitalist  development,
on the other hand, entails the progressive commoditization  of labour,  land,
and housing as well as other goods,  and a correlative tendency to  separate
production  from  the  home.  Residential  and  production  units  may  remain
intermingled  in spatial terms, with both workers and  owners living close to
work.  The  advent  of  large-scale  factory  production,  however,  is one  of  a
complex of factors promoting the concentration and segregation of economic
functions  and  social classes within a rapidly expanding urban  space.1 These
tendencies,  despite uneven  development,  local reverses, and  new locational
trends, are still gathering  force  at the level of the world-system  as a whole.

The separation  of  workplace  and  living  place has more  than  spatial  im-
plications.  With  the  splitting  of  the  "space-time  trajectories"  of  everyday
life2  a  social  distancing  also  occurs,  between  spheres  now  governed  by
distinct  sets of  social and  economic  relationships, oriented  to different  ob-
jectives, and  subject to different  forms of resistence and control. There has
been an analytic disjunction  as well, with such fields as Marxist economics,
class analysis, and industrial sociology, on the one hand, the study of  urban
social structures,  housing, and  spatial patterns,  on the other,  consigned  to
quite different  galaxies. With the exception of a few special areas — the study
of  occupational  or ethnic communities,  for  instance — the  conjunction  of
the  spheres  of  production  and  reproduction  was long  neglected.

The relationship between these two domains is now gradually coming into
focus,  with  developments in urban  theory playing a leading role. Here,  the
principal  approach  to  the  problem  is through  the  analysis  of  space  — the
struggle  between  rich  and  poor,  between  capital  and "communities,  to  ap-
propriate  space for  different  uses or  for  exchange. This theme,  of  course,
is not altogether new. Despite a veil of naturalistic metaphor,' 'class'' strug-
gles over land and housing in capitalist  societies were in fact  preoccupation

2  K.P.  MOSELEY

of the urban ecology of the Chicago School.3 It was in the  1960s, however,
that  urban  conflicts  came  into  their  own,  as  object  of  analysis  as  well  as
historical event,  aided by a revival of  Weberian  and  Marxian  ideas.  In  the
former perspective, deployed by such analysts as Rex and Pahl, distributional
struggles over  housing and urban  space were seen as another  form  of  class
struggle, separate from  the realm of work.4 Marxist theory has taken a  fur-
ther,  more different  step,  attempting  to  incorporate  urban  phenomena  to
the wider framework  of political economy, in which the antinomy of capital
and  labour  in  production  plays  a  dominant  role.

Having internalized, as it were, the duality of  production/accumulation,
on the one hand, and reproduction/consumption,  on the other, the pioneers
of  the  new  "urban  political  economy"  have  emphasized  different  aspects
of the urban link. For Harvey,5 the city was above all a site for the manage-
ment, circulation, and investment  of capital, its "built  form"  serving as  "a
resource  system  . ..  for  the  production  of  value  and  surplus  value."6
Castells,7 on  the  other  hand,  has  seen  the  urban  spatial  unit  as the  major
locus of  "collective consumption,"  the reproduction  of  labour  power  and
the socialization  of  costs of  reproduction  by the state. Cutting  across  these
differences  (in any case increasingly  synthesized  in recent  work),  is a com-
mon emphasis on the primacy of capital, generating and exploiting  housing
and  spatial  inequalities,  and  subordinating  consumption  to  accumulation
needs.

The degree of  determination  by the  accumulation  process,  however,  the
extent to which urban structures and  processes may escape or resist its im-
peratives, or even emerge on  different  bases altogether,  remain  matters  of
intense debate. One may distinguish  three  broad positions on these points.
One is an economistic Marxist-structuralist  tendency,  initially embraced  by
Castells,8 a number of other French scholars,9and,  for  a time, by Harvey10as
well.  Here, the theorization  of  not  only  the  urban  economy  but  of  urban
forms of consciousness and politics, tends to be closely subordinated  to the
central categories and tendencies of the capitalist mode.  Saunders,"  on  the
other  hand,  a trenchant  critic  of  this  school,  has  maintained  a  rear  guard
intellectual operation in which distributional strata and conflicts  are almost
completely divorced from  the production realm. A more dialectical style of
Marxist analysis,  finally,  has now taken  center  stage,  where Castells12  and
Harvey13  have  rejoined  more  inveterate  critics  of  the  structuralist
approach.14  Here  the  emphasis  is  on  the  relative  autonomy  of  the  ur-
ban/community  level, where specific  forms  of  culture,  consciousness,  and
conflict  with capital take shape; and the centrality  of processes at  this level
for  the political survival — and eventual demise — of  the capitalist  forma-
tion  as a whole.

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  3

If  a  dialectical  approach  proves  fruitful  in  advanced  capitalist  contexts,
this should be even more the case in areas such as West Africa,  where capital
is less deeply embedded, and confronted  by local structures and  orientations
of  very different  descent.  This paper15  will try to look  at  both  parts  of  the
world,  in  the  course  of  a  preliminary  reconnaissance  of  the  literature  on
workplace-living place links. The focus will be on wage labour, capitalist pro-
duction units, and the spatial context and patterns in which both factory and
workforce are involved. Remarkably  little work has been done on this com-
plex  of  variables.  The  urban  political  economy  literature  raises  interesting
questions,  discussed  in  the  first  section  of  this  essay,  concerning  the  rela-
tions  of  different  sorts  of  capitalists  and  consumers  in  land  and  housing
markets.  Despite  frequent  references  to  "labour,"  however,  the  urban
theorists pay little attention  to that other  "contested  terrain"16  — made up
of  real  production  units  and  the workers  and  work  relations  they  contain.

The rest  of  this  paper  will attempt  to  fill  in this gap, looking  first  at  the
impact of industry on worker housing, and then turning to the implications
of residential patterns for working-class consciousness and behaviour, within
the workplace  and  without.  The field  of  industrial  sociology thus  provides
a valuable complement to the analysis, although explicit connections to hous-
ing are  still few  and  far  between.  One area  of  convergence,  given  some at-
tention  by Mingione,17 involves the impact of the internal differentiation  of
the workforce;  another,  pioneered  by Burawoy,18  incorporates  the."mode
of reproduction of  labour"  (including company housing), to the analysis of
factory regimes. Industrial location, finally, which we often take as a  "given"
to which  workers simply respond,  may be powerfully  affected  by the class
and  community  structures  of  particular  locales.19  The  relevant  literature,
needless to  say,  has  only been  partially  scanned.  The outline of  important
questions, tentative connections, and  areas of  ignorance however,  begin  to
be  discerned.

Land,  Labour  and Capital
Many of the early factories  in Europe and America were dispersed in small
towns and villages, often finding their work-force already at home, as it were,
or sometimes erecting new housing  for  their  use. By the  late  19th  century,
however, industries  became increasingly concentrated  in large cities, where
masses of cheap labour had been summoned to struggle for  employment —
and shelter — as best they could. Here, companies rarely provided  housing
for their employees, and still less, before the 20th century, did the state. This
function,  rather,  was largely consigned to an expanding real estate  market,
where  an  array  of  profit-oriented  developers,  landlords,  builders,  and
speculators was coming to the fore.20 Thus the contradictory  vocation of  the
city — to serve the reproduction of both capital and labour power — entail-

4  K.P.  MOSELEY

ed  ambivalent  relations  of  dependency  and  antagonism  not  only  between
labour  and  its employers,  but  between  each  of  these  and  a  heterogeneous
property class.

The housing question is an aspect  of  a broader  struggle over the disposi-
tion and costs of space — a resource essential for all activities, yet  relatively
finite in supply. This is particularly the case in cities, where demand is inten-
sified  by the  density  and  purchasing  power  of  potential  users,  and  again,
by the economic advantages conferred  by the very agglomeration of  popula-
tion  and  activities in space.21  Housing,  of  course,  can  be  produced,  but  it
remains  relatively  expensive,  due in  part  to  constraints  on  productivity  in
the construction industry, and also to restrictions on "substandard" and self-
built  forms.  Real  estate  markets,  moreover,  are  notoriously  "imperfect",
marked by strong tendencies towards speculation, inflation  and waste. Prices
reflect  not only initial purchase plus improvement costs, but are  powerfully
affected  by the externalities  of sheer location,  including the  characteristics
of  adjoining  areas.

Additional  tolls  (the absolute  or  monopoly  forms  of  ground  rent  noted
by Marx) may be extracted for particularly unique or strategic locations which
may  otherwise  be simply  withheld  from  use.22

Under  market  conditions,  then,  the  distribution  of  urban  space  will  be
strongly skewed  by differences  in purchasing  power,  with  some  categories
of users perhaps excluded altogether. This is spatially expressed in the segrega-
tion  of  different  types  of  economic  activities,  on  the  one  hand,  different
classes of  residential consumers,  on  the  other.  Low-income categories  are
pushed towards undesirable, deteriorated,  or distant  locations, and  tend  to
suffer  from absolute shortages in housing supply: old housing stock may be
expropriated  for  purposes  of  redevelopment,  while  new  low-income  hous-
ing  is relatively  unprofitable  to  build.23  Despite  improvements  in  real  in-
comes  and  housing  standards,  and  constant  interventions  by  the  state,
distributional crises stubbornly reappear, in new conjunctures,  cities and parts
of  the  world.

With a few possible exceptions (e.g. Singapore?), housing inequalities in
the Third World, along with real estate markets, are still in a phase of  rapid
growth,  widely generalized  only since the  Second  World  War and  the  end
of  colonial  rule. West African  cities,  for  instance,  are still  marked  by  the
persistence of customary tenure and inheritance rules — particularly in slower-
growing cities and those with  strong indigenous cores — and  the  often  ex-
tensive tracts of land administered by the state.24 With rapid  urban  growth,
however,  critical supply problems emerge, due to extensive commercial  in-
vestment and speculation in real estate,25 and a polarized  social  structure in

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  5

which low local wage levels coexist with elite incomes, housing standards and
real estate prices which tend towards metropolitan norms. Thus, in addition
to the familiar  problems of congestion and deterioration of old housing stock
(insofar  as this exists), illegal subdivisions and constructions, squatting, and
self-housing  of  the most  minimal kind  have occurred  on an  unprecedented
scale.26

The precise spatial distribution of different  sectors and classes is  affected,
among other things, by a city's position in wider national and  international
divisions of labour. There have been dramatic changes in this respect in re-
cent  decades,  spurred  by  global  competition  and  increasing  technical
possibilities for the separation of central corporate administration from  pro-
duction  cities.  Investment  flows  have thus  become  much  more  sensitive  to
worldwide variations in labour markets, tax and tariff  regimes, and  general
operating costs.27 In the advanced  capitalist countries, many older  industrial
areas have declined, while a partially overlapping set of major  metropolitan
centers have been reorganized around corporate administration and services,
and the consumption  needs of  an increasingly whitecollar  and  professional
labour  force. These trends have supported  a process of  "gentrification"  of
the urban  core, and  a correlative displacement  of  productive  activities  and
the  working  class  towards  peripheral  zones.28

Such neo-imperial  "world cities"29 appear in certain  developing countries
as well: Hong Kong, Mexico City, or Singapore, for instance, perform  similar
economic functions  as a regional plane. Here, the problems associated with
the metropolitan  form,  such as congestion, infrastructural  deficiencies,  and
increased  commuting 
time  and  costs,  are  likely  to  be  even  more
pronounced.30  As in many lower-ranking Third  World  cities, moreover,  the
growth  and  urban  concentration  of  industry  and  wage labour  may  still  be
underway (sometimes as a direct result of relocation from  higher-cost  areas
overseas). The content of "tertiarization" also differs,  being weighted towards
the expansion  of  an  informal  sector  of  low-income,  small-scale  enterprise,
necessitated  by  inadequate  supply  of  waged  and  salarised  work.31

The role of  real  estate interests  in these distributional  and  spatial  shifts,
particularly in the advanced  countries, has been a major  concern  of the  ur-
ban  political  economists.  Mingione  has  noted  the  underlying  cleavages
involved:

the rentiers and  speculators want to impose higher costs to get an im-
portant share of the produced surplus value; the capitalists theoretically
oppose this tendency as they are obliged to pay for  it through  a direct
subtraction of surplus value or through higher wages; the workers strug-
gle for  better  housing and  living conditions  at  decreasing  costs  . .  ?2

6  K.P.  MOSELEY

Interpretation  is coloured  by the historical  role  of  property  capital,  which
antedates industrial capitalism, fails to play an equivalent role in raising pro-
ductivity or lowering costs and long remains confined  to a small-scale, petty-
bourgeois form. Thus for  Ricardo, Marx and Engels,33 and certain  theorists
today,34  ground  rent  may  erode  productive  capital  and  brake  long-term
capacity  for  growth. Similarly,  the proclivity  of Third-World  bourgeoisies
to invest in real estate and trade has been cited as proof  of their  incapacity
to  play  an  economically  progressive  role.35

In mature capitalist  formations,  however, as Harvey suggests,36 property
investment may simply figure as a "secondary circuit"  of capital, taking  up
the slack in periods (and places - KPM) marked by sluggish demand and pro-
fits in the goods-producing sectors. Particularly with the emergence of large-
scale property  development  corporations,  merged  with  financial  and  even
industrial capital,37 nothing prevents real estate proceeds  from  being recycl-
ed back to industry when market conditions change. Thus, if property capital
is an agent of the restructuring of major metropolitican centres, it is not likely
to be the cause — still less when we remember that  production  is not  being
abandoned,  but relocated,  in part  to less developed  areas  overseas.  There,
extensive concentrations  of  capital  in  real  estate and  construction  are  also
beginning to appear, in tandem with, not instead of,  industrial growth.38  In-
deed,  the extent  to which  real  estate  may  provide  a  strategic  niche  for  in-
dustrial capital accumulation, facilitating, at a later stage, entry into the more
competitive often  foreign-dominated,  industrial  sphere,  is a matter  deserv-
ing more  research.

Property  and  industrial  capital  are also  linked  more  indirectly,  through
housing costs and wage rates, as they come to bear on the worker  himself.
This is a key aspect of the workplace-living place connection that is our prin-
cipal concern,  and  thus will be alluded  to at  later  points  in the  text.  Here
we should note the basic variations in the wage-rent equation,  which struc-
turalist  analysis  has identified  but  not  yet  satisfactorily  explained.  On  the
other hand, capital may "externalize" housing and urban  costs by  shifting
them to labour, in which case property capital has most to gain. Living con-
ditions may then deteriorate, or wage pressures and adjustments  may occur
to cover the additional costs. In the latter case, there is an effective  subtrac-
tion from  industrial surplus value, which state rent controls or housing sub-
sidies may do something to restore.39 In these formulations,  it seems, labour
never wins; the empirical evidence, however, suggests that  the average  real
housing  conditions  of  labour  in  core  countries  have,  over  time,  clearly
improved.40

A  more differentiated  treatment  of  the  share-out  of  urbanization  costs
might be possible if we consider the distinction between large and  small-to-

CONTESTED  TERRAINS:  SOCIAL  RELATIONS  OF PRODUCTION  7

medium  scale  business  concerns.  While  the  former  may  thrive  on  spatial
mobility  and  long-distance  integration  of  their  activities,  small  ones  more
often rely on localized economic networks, and suffer disproportionately  from
urban redevelopment  and locational change.41 If ground rents or wages rise,
moreover,  larger, oligopolistic firms are more able to absorb this by raising
prices; their employees, in fact,  are likely to be well-organized  and  capable
of pressing home wage demands if housing or transport costs increase. Com-
petitive  sector  workers,  on  the  other  hand,  will tend  to  bear  the  brunt.42
Such contrasts  are even more extreme in the Third  World,  where  profits
and wages in the  "informal  sector"  are relatively depressed,  and where the
need  for  physical  proximity  of  small-scale  activities  to  each  other  and  to
customers is more acute.43 These variations, however, have been recuperated
by  yet  another  diffuse  functionalist  model  in  which  the  small-scale  sector
is seen to represent  less the inadequacies  or limits of  capitalist  growth  than
the very conditions  of  its success. The basic argument  here is that  various
forms of subsistence production and non-wage work keep general reproduc-
tion costs in check. Even wage labour, can be paid at less than "value" (i.e.
fully  monetarized consumption costs), providing  "excess" surplus-value  for
capital, and/or,  through  "unequal  exchange," effectively  subsidizing con-
sumption  costs  of  wage  labour,  even  overseas.** Such  formulations  have
also  been  extended  to  explicitly  refer  to  such  phenomena  as  self-housing,
squatting, flimsy construction, and sites-and-service schemes which, by lower-
ing  housing  costs,  ultimately  rebound  to  the  benefit  of  formal-sector
employers.45

Various aspects of the super exploitation/surplus-transfer  model have come
under  attack;46  as for  low-cost  housing,  one wonders why, if  it is so  func-
tional  for  capital,  it  is confronted  by such  incessant  efforts  at  regulation,
demolition,  and  renewal,  replicating  the  experience  of  the  developed
world.47  Our main concern here is to note that,  as with urban  structuralist
theory,  the  question  of  who  benefits  from  cheap  housing  is  actually  left
unresolved.  As  the  structuralists  would  point  out,  first  of  all,  the  gain  to
employers might only be property capital's loss; moreover,  petty  landlords
often do gain from letting "irregular" housing, sometimes at the same price
that would be paid for  more standard buildings and plots. Or windfall  dif-
ferentials  in real income may accrue to the formally  employed,  whose wage
levels may be predicated  on standard  vs. informal  housing costs, and  who
in any case will be enjoying  access to the same cheap housing as the  poorer
informal-sector  employees  and  the  self-employed.48

The wage-rent equation may also be affected  by capitalist  cycles of  con-
traction and growth. With respect to the advanced capitalist countries, it has

8  K.P.  MOSELEY

been suggested that in times of expansion and high labour demand,  industrial
capital may side with workers in favour of state intervention to reduce hous-
ing costs, while in periods of contraction, industrial and property capital are
more  closely  aligned.49  There  seem  to  be  many  exceptions  to  this  rule,
however,50  as  well  as  significant  international  variations  despite  similar
economic trends. The power of labour and other popular forces, along with
differences  in  inherited  political  culture,  need  obviously  to  be  taken  into
greater  account,  as  does  the  budgetary  and  fiscal  situation  of  the  state.
Through the political process, moreover, a variety of coalitions of business
and  labour  interests  may be  forged,  oriented  to  strategies  of  growth  and
redistribution in good times, or defense  of the local economy  in periods of
competitive  stress.51 The current  downturn  has been  generally  marked  by
declining or stagnant employment and real wages in industry, a polarization
of incomes within expanding sectors, and a shift in state spending from  social
welfare to infrastructural  costs. There has also been  extensive  competition
and variation between cities with respect to capital flows; particularly  where
restructuring has been most successful, the lower echelons of the labour  force
are likely to be caught in a scissors  movement  between  low wages  and in-
creasing  rents.52  Again,  it  is in the intersection  of  these  larger  shifts  with
those on local terrain that one is most likely to find answers to the question
of "who pays?" and to identify the conditions under which  "unproductive"
capital may thrive.

The variability of the role and content of state policy is even more  marked
if we include the Third World in our compass. Not only development  and
class formation, but the very capacity of the state to articulate class interests
and regulate urban processes is extremely uneven here. The dialectic of the
state and international  capital has historically  been  prominent,  while con-
flicts  over the structuring  of space at the municipal  level have  emerged  in
tandem with the establishment  of new urban  centers, European  settlement
and capitalist-industrial  penetration.  As Cooper53  suggests, the struggle  to
gain  control  over  urban  land  use  and  residence,  as  well  as  labour  and
workplace, has become increasingly acute in recent years. As elsewhere, the
net effect  of state welfare and planning measures seems to be to  reproduce
rather  than  overcome the inequalities  produced  by the market  itself.54

Marxists  of the structuralist  school,  as we have seen,  tend  to  emphasize
the strength of capitalist interests, and to down-play possibilities for  urban-
based redistribution and change. Recent work,  however — notably  that of
Castells55 — is less pessimistic on this score, while Mingione has noted  that
reforms  may be so extensive as to undermine  the solvency  and  legitimacy
of  the state, and even "endanger  the rhythm and the possibility  of the ac-
cumulation  process  itself."56  Distributional  struggles  are  given  greatest

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  9

weight,  finally,  in  non-Marxist  work  — but  here  they  are  often  viewed  in
isolation  from  the the  problem  of  the  reproduction  of  capital  or,  indeed,
from  the  capitalist  system  as a whole.  Saunders57  has  taken  a  particularly
strong position  on this point,  initially arguing that  the analysis  of  distribu-
tion should be delinked not only from  accumulation processes, but from  any
special  focus  on  housing  or  the  use  of  space.

Our  own  survey  of  the  field  suggests  contrary  conclusions.  In  capitalist
settings, at least, the "built environment" clearly occupies a privileged posi-
tion  among  both  consumption  and  investment  goods,  and  can  only  be il-
luminated by a political economy approach. As suggested at several points,
however,  elegant models of the accumulation process must be tempered  by
a primary fidelity to historically and spatially-specific  groups and structures,
themselves conditioned by broader networks of relations created by markets,
states, and the world system as a whole. Moreover, the influence  of  capital
or  of  productive  organization  on  the  economics  of  consumption  cannot
resolve in advance the political issues relating to lines of cleavage, forms of
consciousness  or  the effectiveness  of  struggles  in the distributional  sphere.

Housing  Strata  and  the  Industrial  Division  of  Labour
The spheres  of. production  and  consumption,  each  of  which involves  both
social and  spatial dimensions,  can be approached  through  various  units of
analysis. The city, as we have seen, has often  been taken as the spatial unit
of reproduction or distribution,  frequently  counterposed to the organization
of  production  in the factory,  workplace,  or  firm.  Consumption  is not  con-
fined  to  cities,  of  course,  but  commoditization,  markets,  and  state  provi-
sion are there most  concentrated  and advanced.  For purposes  of  more  dif-
ferentiated  analysis, 
the  progressively  smaller  units  of  community,
neighbourhood,  living place, and home or residence are variously employed.
The literature on urban land and housing makes remarkably little  reference
to that basic social, budgetary, and residential unit — the household — now
a  focus  of  growing attention  in both  community  and  labour  research.  The
household  continues to be a site for  work  of  various  kinds,  including  con-
tractual outwork for capitalist concerns; the dominant secular trend, however,
is for  the major  source of livelihood to be put  outside the domestic  sphere.
Those who work  for  wages outside the home will be our  major  focus  here.

The relations of social inequality and conflict  characterizing each of these
two terrains, and above all the connections between them, have been a ma-
jor preoccupation of the urban political economy school. As Harvey observes:

the split between the place of  work  and  the place of  residence  means
that  the struggle of  labour  to control the social conditions  of  its own
existence  splits  into  two  seemingly  independent  struggles.  The  first,

10 K.P.  MOSELEY

located in the workplace, is over the wage rate, which provides the pur-
chasing  power  for  consumption  goods,  and  the  conditions  of  work.
The second, fought in the place of residence, is against secondary  forms
of exploitation and appropriation represented by merchant capital, land-
ed property, and the like. This is a fight  over the costs and  conditions
of  existence  in  the  living  place  .  .  .5^ 

•

As suggested, however, by the reference to "seemingly independent"  strug-
gles, the degree of  congruence or autonomy  of  the two spheres is a  matter
of considerable debate. These discussions, which have been building up over
the past decade, have largely referred  to advanced-capitalist  contexts;  their
significance, however, escapes these confines, spilling over, in fact,  to touch
on our most general conceptions of class and social change. A  first step, in
any case, might be to identify  the significant  social divisions and  categories
in the housing sphere.

Theories  of  Residential  Inequality
If all big capitalists were also landlords, all small capitalists,  homeowners,
and all labour,  renters or sans domiciles, matters  would be simple. In  fact,
housing, perhaps more than other forms of consumption, gives rise to  signifi-
cant  divisions  within  both  proletariat  and  bourgeoisie.  One  approach,
pioneered  by  Rex and  Moore,  has  been  to  treat  the  "distribution  of  life-
chances" in housing as the basis of a distinct class hierarchy, giving rise  to
struggles over the use, not  of the means of production,  but  of  the  "means
of  residence" and  "domestic property."59  Later  work  spun  out  numerous
categories based on differences in access, tenure, and the social status of users;
the systematic  analysis  of  property  classes,  however,  was lost  from  sight.
Saunders,  following  Pahl, has noted that  a more rigorous  application  of  a
Weberian  approach  might  yield three  main  groups:

those who live off  the economic returns from  house ownership  (such
as landlords and private developers), those who use housing purely as
a  means  of  consumption  (tenants in  the private  and  public  sectors),
and  those who, in the process of  consuming  housing,  typically  enjoy
a return  on capital  (most  owner-occupiers,  who achieve  considerable
capital  gains  from  their  ownership  of  housing).60

Further  distinctions  might  be  made  among  large  and  small  property
capitals, public and private provision,  and  particular  modalities  of  pricing
and distribution. This sort of analysis is not incompatible with early Marxist
approaches to the subject,61  in which the linkages of heteroclite urban  strug-
gles to  the global  properties  of  capital  began  to  be  explored.

The enlarged,  synthetic  framework  of  class analysis  implicitly  promised
here  has  not,  unfortunately,  been  entirely  realized  in  recent  work.

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  11

Saunders,62 at  one extreme,  has opted  for  a view of the political process in
which  "consumption  locations,"  now  seem to  be  significantly  determined
by housing tenure, take pride of place. These strata, however, are dissociated
from  capital,  occupation,  and (production-based)  class in not  only  analytic
but  very real terms: the generation  of income and consumption  differences
by  the  larger  economic  structure,  along  with  the  functions  of  property  in-
vestment  for  large-scale  accumulation,  are  almost  entirely  lost  from  view.
This stance  is in  part  a  reaction  to  the  structuralists,  particularly  Harvey,
who took an opposite tack: arguing that  "the dichotomy between living and
working  is an  artificial  division,"  Harvey63  attempted  to  reduce  consump-
tion categories to the dichotomy of capital and labour  based in the produc-
tive  sphere.

As suggested by the preceding section of this paper,  this strategy is  fairly
successful  at the level of  capital, which, in taking various fractional  forms,
does indeed dominate production  and accumulation in the housing field.  It
creates serious problems, however, when subordinate housing categories are
addressed.

Initially,  Harvey64  had  simply  referred  to  these  as  "occupiers"  or  con-
sumers,  divided  by  income,  tenure,  neighbourhood,  and  the  like.  These
diverse groupings were subsequently hypostatized as mere manifestations  of
"labour"-in-general,  which,  through  a  series  of  elisions,  was  attributed  a
privileged  role with  respect  to  consumption  and  orientation  to  use.65  And
consumption,  being  tied  to  the  reproduction  of  labour  power,  could  then
become,  by  implication,  a  moment  of  the  production  process  itself.

This  was not  to  deny  the phenomenal  reality  of  consumption  cleavages;
these were highlighted, in fact,  but  in quite invidious terms. As Cox observ-
ed, with  the commodification  of  the living place, the working  class and  its
everyday  community  dissolve  into

distributional groupings according to race, income, location or any other
social cleavage created by the necessarily uneven development of capital
.  .  .  Fractions  of  labour  are  willing  to  identify  with  those  sharing  a
similar  consumption  status and  develop consciousness opposing  their
claims  to  those  of  other  groupings.66

It is homeownership,  among other  housing  inequalities, that  is the  leading
villain of the piece, dividing workers by neighbourhood, placing some of them
"on  the side of  the principle of private proverty,"  and generally serving as
an  instrument  of  "divide  and  rule."67  Beneath  the  "surface  appearance"
presented  by such cleavages, the  "hidden  essence  . ..  the struggle  between
capital  and  labour,"  lies  concealed.  6S

12 K.P.  MOSELEY

Such  reductions,  no  less  than  Saunders,  neo-dualist  approach,  seem
altogether  premature.  Other  and  more  recent  Marxist  work,  however,  in-
cluding that of Harvey himself, is more cautious on these points. Mingione,
for instance, notes that reproduction includes "various social groups, which
cannot be considered productive workers,69 while Castells had pointed to the
emergence of  disparities related not to "class relationships but to the posi-
tion  in  the  consumption  process  itself  .  .  ."70  In  The  City  and  the
Grassroots,11  and  Consciousness and  the  Urban  Experience,12  he  and
Harvey, respectively, have gone much further  along these lines. Taking up
a theme introduced by Lefebvre years ago, both books emphasize the non-
class, space-based, and cultural character of urban groupings; capital remains
the main antagonist, however, along with the primacy of exchange-value and
accumulation, and the incessant urban contradictions, which it entails. For
Harvey,  however,  the connections between such groups and capital at  the
level of economic structure is still a major  concern. The key linkage is now
located  not in production as such, but at the moment of exchange — par-
ticularly through the circulation of revenues, which structures all other social
solidarities  and  distinctions  in their  turn.  Urban  social relations, then, re-
main subject to the class antimonies of capitalism, although, once "institu-
tionalized  and  reified,"73  they take on an  apparent  life  of  their  own.

This  perspective  helps  open  the  way  to  a  more  empirical  analysis  of
workplace-living links, something which the urban political economists con-
tinue to neglect. Such an investigation demands that we move from  an ex-
clusive focus  on urban  consumption  and the general impact  of  capital,  to
bring real production units — factories, corporations, and so on — into view.
This is all the more necessary if we are to study labour in the living place
(real, existing labour, that is, and not a metaphor), for it is at the point of
production  that  labour  can be clearly  identified  as such. Of  course,  there
is a workforce  internal to the housing field,  in construction,  maintenance,
and  so on.  But  even there,  it will tend  to dissolve into  "consumers"  and
"citizens" outside the hours and place of work — unless specific mechanisms
intervene to carry over workplace identities into the residential sphere. The
critical task is to ask to what extent and through what processes this might
occur,  and  how it might react back  on workplace relations in turn.

The  Impact  of  Industry
This down-to-earth strategy yields immediate results. With the recent excep-
tion  of  Harvey,  who, as just  noted,  has acknowledged  the importance of
money incomes, urban theorists have tended to imply that consumption  dif-
ferences  are  somehow  engendered  in  the  consumption  process  itself.  But
labour is in fact divided within the workplace as well as in everyday life out-
side. Looking at even a single firm,  let alone an industrial sector or a coun-

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  13

try, three major  sorts of differentiation  can be seen.74 The first,  which seems
intrinsic to  the industrial  labour  process (at  least  under  competitive  condi-
tions),  involves the division  of  labour  in production,  and  the  arrangement
of tasks in a hierarchy of wage and skills. The second is the overlapping but
broader  segmentation  of  workers  into  primary  and  secondary,  core  and
periphery, or formal and informal  sectors, based on additional factors  such
as unionization, job  security,  and labour  market  structures.  (The  differen-
tiation of industries themselves with respect to scale, organization, and market
position  is an  important  underlying  variable  here).  Finally,  and  with  con-
siderable  local  variation  in  import  and  form,  is  the  fragmentation  of  the
workforce  along  ethnic,  cultural,  and  sexual lines,  insofar  as these  ascrip-
tive  criteria  are  interwoven  with  the  structural  divisions  just  described.

There is remarkably little in either the urban or industrial literature which
simultaneously  brings  workplace  status  and  housing  situation  into  view.
Everything suggests, however, that these work-based divisions may account
for  a large part  of the housing variations that  cut across the working class.
Insofar  as this is the case, then conflicts  around the living place are, indeed,
"displaced"  from  the  "work  process,"  as  Harvey  once  put  it,75  but  in  a
much more direct  way than he had in mind.  As he now suggests, given  the
importance of work status for income, and of income (and, perhaps, ethnicity)
in housing markets,  it is hard to  see how it could  be otherwise. And  if  this
is correct, then Saunders' notion, that consumption-based  divisions "are no
less  'basic'  or  'fundamental'  "  than  those  derived  from  work  and  class,76
is certainly  placed  in  doubt.  Given  the variable  development  and  scope of
housing and labour  markets, between countries and over time, the  strength
of correlation  between the two spheres will also vary, of course. We cannot
attempt a systematic comparative treatment here, but will try to draw atten-
tion  to  some  preliminary  evidence  which  may  stimulate  further  work.

One of the contexts in which such correlations clearly appear is the plann-
ed housing  estate designed  entirely  or  in part  for  members  of  the  working
class. Here, workplace  divisions may not  only carry  over into  a.segregated
community  life  — as  with  migrant  hostels,77  for  instance  — but  may  be
reflected  quite exactly in differences  in the location  and  quality of  dwelling
units.  Parkin  provides  striking  details  of  this  sort  for  two estates  in  Kam-
pala,  Uganda,78  and  in  India  and  in  socialist  countries,  public  housing  is
even more commonly  tied to gradations  of  status  at work.79  In such cases,
of course, deliberate policies of a redistributive or egalitarian kind may also
constrain  the  expression  of  income  differences  in  the  living  place;  racial
segregation or egalitarian social norms may have similar effects  in unplann-
ed market  settings.80 Here,  however,  marked  housing differences  are  more
common,  especially insofar  as minimal housing standards are absent or not

14 K.P.  MOSELEY

enforced.  Engels'81  pioneer  study  of  mid-nineteenth  century  Manchester,^
with its constrasts of  respectable working-men's  homes with  the sties and
cellars of Irish migrants and other subproletarians, is a classic contribution
in this respect (and one cannot imagine him arguing that these housing  dif-
ferences  are the cause of  the cleavages within  the working  class!). Recent
British data reveals that a strong linear correlation between income/occupa-
tion  and  housing  situation  still  stands,  extending  to  wide  variation  in
homeownership  between  skill  categories  within  the  working  class.82  The
duration and stability of employment, along with age and household com-
position,  are  additional  factors  here.83

Other research focusses on neighbourhood variation as opposed to dwell-
ing units as such. Harvey,84 for instance, has traced the association between
occupational categories and spatial differentiation  of 19th-century Paris, while
Kornblum's85  study of  South Chicago steel workers describes the intricate
interrelations of labour recruitment and work status, ethnicity and community
life. Others have noted the impact of industrial relations on inner-city Black
American neighbourhoods, increasing either unemployment or housing costs,
and reinforcing  housing abandonment and decline.86 Ethnic segregation, at
least among wage workers, is relatively unimportant  in much of the Third-
World;  distinctions  between  'strangers'  and  indigenes  in  African  cities,
however,  may  have  clear  occupational  and  residential  links.87  Although
neighbourhood variations in concentrations of wage workers, and of skilled
vs*  unskilled,  can  certainly  be  found,88  spatial  differentiation  among
workers tends to be restricted by the compression of wages and the absolute
scarcity  of  housing.  Thus,  even  in shanty  towns and  squatter  areas,  there
tends to be extensive occupational  mix.89 Such housing is generally  linked
to low incomes and purchasing power, but the coexistence of homeowners,
landlords and renters, and of different  types of housing, suggests considerable
internal  variation  in  this  respect  as well.

Central location, as noted earlier, may be particularly important for small-
scale firms  and the  self-employed.  Economic restructuring,  however, may
both  multiply  such  enterprises  and  threaten  their  hold  on  central  space.
Mingione,  looking  at  Italy,  has noted  the way in which  stability  and  cen-
trality  of  residence is connected  to the core-periphery  segmentation  of  the
labour force, arguing that the unemployed are virtually barred from  access
to the cities where they might  find  jobs.90 Apart  from  some limiting cases
(e.g. the pass law regimes in settler territories), such exclusion seems more
a wish than a reality. Mingione's treatment is interesting, however, in that
it treats housing differences  as a symptom, not a cause, of the disaggrega-
tion of the working class. Marginal and unemployed workers, he suggests,
"have no chance of gaining anything through factory struggles,"91 and thus

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  15

direct  their  welfare  demands to the state; core workers, on  the other  hand,
act first, and more effectively,  in the workplace, where they attempt  "to com-
pel  individual  capitalists  to  pay  directly"  for  increases  in  reproduction
costs.92  Not  only  wages, but  free  transportation,  for  instance,  or  subsidiz-
ed housing,  may figure among their  demands. This directs our attention  to
another  possible workplace  link — the extent to which  non-wage consump-
tion  needs  of  workers  may  be  provided  for  directly  by  the  firm.

The work of Michael Burawoy offers  a valuable starting point for analysis
here.  One  of  the  determinants  of  different  production  or  factory  regimes,
he argues, is "the mode of  reproduction  of  labour  power"  — which  varies
as to whether provisioning is assured by auto-subsistence, the wage, the fac-
tory,  or  the  state.93  Capitalist  development  tends  to  separate  the  worker
from  the means of both subsistence and production: his "appearance at the
factory  gates  has  to  be  renewed  each  day  if  he  or  she  is  to  survive."94

Pure cases of such "market despotism," however, are actually quite rare,
and  one of the interesting variations that  occurs is the provision  of  worker
housing by the employer,  which,  we should  note,  could be a public agency
as well as a private  firm.  This sort  of  housing,  which  has been  of  episodic
importance in the West, is even more significant when other parts of the world
are  taken  into  account.

In the  early  stages  of  capitalist  development,  wage workers  often  retain
access to a more or less autonomous subsistence base. One way of recruiting
and stabilizing labour under such conditions is to supply and oversee workers'
consumption, in some cases combining this with restrictions on mobility and
alternative sources of supply. This gives rise to what Burawoy calls the  "com-
pany  state,"  which  "intervene(s)  coercively  in  the  reproduction  of  labour
power, binding community to factory through non-market as well as market
ties."95  Such patterns appeared  in post-Emancipation  Russia,96  in some of
the early Lancashire cotton mills,97 and in many ventures in rural,  frontier,
and colonial areas where shortages not  only of  labour but  of private hous-
ing stock may have played a role. Major 20th-century examples are the min-
ing compounds  of  central and southern Africa98  designed  in the latter  case
less to  stabilize  migrant  labour  than  to  prevent  it  from  settling  down.

A  second,  more  liberal  type  of  company  housing  scheme appeared  in  a
number of core industrial areas during the century prior to the Great Depres-
sion. A forerunner  was the model worker community established by philan-
thropist  mill  owner,  Robert  Owen,  partially  replicated  by  other  well-
established  firms  in  England,  New  England,  and  on  the  Ruhr.99  Labour
shortage was still a factor in some cases, as suggested by the dormitory system
devised for  the young farm  girls brought to work in the early New England

16 K.P.  MOSELEY

cotton  mills.100 By the late  19th century, the socialization  of immigrants and
the containment of radicals seem to have been more prevalent concerns and
company  towns  reappeared  as one of  several possible  components  of  the
"welfare capitalism" adopted by many major American corporations at the
time.101  In the  big cities,  however,  rentier  capital  could  now more conve-
niently take  care of  housing  supply.

A third variant  — which no one, to our knowledge, has written about as
a type — is the housing routinely  provided  by many large scale companies
and  public services in colonial,  ex-colonial,  and  other  less-developed  areas
outside the world system's metropoles. Here, a history of housing and labour
shortages is combined with a relatively non-interventionist  labour  relations
regime: living quarters  figure  more as a perquisite of  office  than as an in-
strument  of control. Thus, it is the highest-paid  professional  and executive
cadres  who  receive  priority  here.  Consider  the vast  stock  of  fine  housing
established  for  administrators  in the colonies,102 or built  by companies  for
their senior staff  (e.g. the old UAC compounds in many West African cities,
'Shell Camp' in Port Harcourt, or Delta Steel Township in Warn, Nigeria).
For  lower  cadres, the most  comprehensive  schemes outside the  uniformed
services seem to have been offered  by the railways — often the largest single
employer apart  from  ports and mines. A well-documented  case is the East
African  Railway  Estate  in  Kampala,  which  housed  all  railway  workers,
dividing them  into seven  "house classes" according to their  administrative
grade.103  Company  housing  seems to  have been  given  special emphasis in
settler areas: it still constituted  17 per cent of the total Nairobi housing stock
in  1968.104

In  interwar  England  and  America,  in  the  meantime,  company  welfare
schemes were being whittled down in the face of labour militancy and business
decline. These problems also revealed particular drawbacks which company
housing entailed. On the one hand, workers found  they had to vacate com-
pany  housing  if  they  went  on  strike.105  On  the  other,  companies  such  as
Pullman and Bethleham Steel found themselves torn between retaining their
worker-tenants and the occasional need to shed labour in line with business
conditions  or  rationalization  schemes.106 As Harvey  notes.

The solution for the individual capitalist is to withdraw from  the pro-
duction of consumption fund items for the workers he or she employs.
But  the  problem  remains  for  the  capitalist  system  as a  whole. 07

For these and other reasons, company provision of housing has become very
uncommon in core countries since the Second World War. On the other hand,
housing construction  by various cooperative  and  non-profit  organizations

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  17

including trade unions assume some importance particularly in the Scandina-
vian case,108 and state intervention,  through a varied patchwork  of  measures
to moderate prices and increase supply, accelerated throughout the capitalist
world. Generally, there has been an expansion of  public housing (especially
in Britain and Scandinavia) and private homeowning (especially in America,
but  now a widespread  trend); private  rental  stock  is now in  decline in  core
countries, but is growing in much of the developing world. The specific con-
nections of these developments to industry  and industrial workers have still
to be unravelled,  although a few broad  points can  be made. First,  employ-
ment status can still condition access to state-provided goods, and we should
investigate further  the extent  to which  this is so.109 Second, we suspect  that
there is still significant  variation  among  firms  as concerns the provision  of
consumption  goods  or  subsidies,  and  that  this  is shaped  by  further  varia-
tions  along  inter-sectoral  and  international  lines.

A good  case in  point  is provided  by  Japan,  where comprehensive  pater-
nalistic welfare  regimes, often  including significant  housing benefits,  remain
typical  of large corporations.110  Four  features  of  this situation  deserve par-
ticular  mention.  Japan's  industrial  structure  is extremely  segmented;  com-
pany  benefits  are  largely restricted  to  the  large-scale sector  and,  within  the
sector, to the permanent as opposed to the temporary or contract labour  force.
At the same time, housing is very scarce and costly, and state-provided social
insurance  and  consumption  provision  is meager  indeed.  Finally,  Japanese
industry  has enjoyed  a uniquely  favoured  position  relative to the competi-
tion  and  contraction  that  has  afflicted  the  world  economy  over  the  past
decade. Elsewhere, the re-privatization  of consumption  costs by both  states
and  companies  has  been  the  general  trend.

In light of this analysis, it seems safe to conclude that the situation of labour
in the living place may reflect,  rather  than  deflect  from,  differences  emerg-
ing within the world  of work.  In neither case is it useful  to attempt  a treat-
ment of "labour"  in undifferentiated  terms (still less to identify labour with
"consumption,"  as if other  strata do not  exist). The complexity of  factors
and variations affecting  the housing sphere suggests further  that the autonomy
of the distributional order vis a vis production should be conceded for analytic
purposes  —  but  for  not  much  more  than  that.  Substantively,  pace
Saunders,111  housing  and  other  kinds  of  consumption  are  secondary  —
ultimately shaped by the production  mode that dominates the social system
as a whole, and immediately affected  by the sorts of links to production and
employment traced above. Distributional  cleavages may be "secondary" in
another  sense,  of  course,  in  that  they  distract  the  workers  from  the  main
issues, particularly  that  of  ruling class control  of  production  and the state.
This  brings  us  to  the  final  topic of  this  essay  — the  way  in  which  various

18 K.P.  MOSELEY

features  of  the  living  place  in  fact  condition  the  political  attitudes  and
behaviour  of  the  working  class.

Living-Place  Patterns  and  the  Labour  Struggle
The analysis  of  cleavages  in  the residential  domain  has  been  strongly  col-
oured by apprehensions concerning their possible political effects.  The  Left,
traditionally,  has  been  indifferent  to  "urban"  issues;  marked  by  Engels'
negative evaluation  of the housing question,"2 its occasional  alliances with
urban reform  movements have been accompanied by deep reserve. Since the
protests of the  1960s, as we have seen,  these attitudes  have begun  to  shift.
The economistic strand of structuralism,  however,  marked  a reversion  to a
more orthodox political stance, resting on three main points. One has already
been noted — the idea that apparent working class gains in the housing  field
actually redound to the benefit  of  one or another  fraction  of  the  capitalist
class. The rise of  homeownership  in the United  States,  for  instance,  or  of
self-help housing in the Third World, has been interpreted as a more or less
deliberate counters to economic crisis and social unrest.113  Secondly,  hous-
ing differences promote "parochialist forms of community action that typical-
ly  lead  one  fraction  of  labour  to  benefit  at  the  expense  of  another,"114
deflecting attention from  class struggle, and reinforcing  divisions within the
working class.

Most serious, perhaps, is the way in which the commercialization  of home
and  community  may  promote  the  domination  of  capital  at  the  level  of
ideology, values, and the everyday, intimate spheres of life. As Harvey puts it,

it is only in terms of  an all-embracing  domination  of  labour  in  every
facet  of  its  life  that  the  "work  ethic"  and  the  "bourgeois  values"
necessarily demanded by the capitalist work process can be created and
secured.

Particularly with the spread of homeownership, labour is won over the prin-
ciple of private property and to the pursuit of status and autonomy  through
the individualistic and competitive acquisition  of goods; the relationship of
classes is correspondingly  concealed.116

These are interesting hypotheses which, however,  are weakened  by being
stated in overly schematic terms. Several preliminary  objections,  relating to
the general significance of distributional cleavages, can be made. If two-thirds
of housing units in the United States are occupied  by owners, for  instance,
and one-third of those in Britain publicly provided on the basis of need, are
the functions  of these phenomena for  capital — here, through  profits  from
real estate, there, through lowering the necessary wage — really the key ex-
planatory  point?  As  Saunders  once  noted,  such  arguments  may  entail  a

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  19

"systematic  devaluing  of 
the  casual  significance  of  working-class
demands.""7  Even  in  the  American  case,  the  "suburban  compromise"
represented  concessions  to  labour,  not  merely  its  cooptation.118  Granted,
this takes us far from  revolution — but should we regret a scenario predicated
on the immiseration of the working class? Or combat housing reform in order
to hasten  the capitalist  demise?  Engels may  have thought  so,  but  few  con-
temporary analysts are so blunt. Rather, housing gains in the advanced coun-
tries are implicitly accepted as one of several indices of labour strength, while
expectations  that  Third World  shanty towns might  provide new centers  for
revolt  have  been  quietly  put  aside.

The  contrast  between  living-place  and  workplace  demands,  more-over,
needs  to  be  more  carefully  thought  through.  Rather  than  rigidly  counter-
posing  the  two,  we  might  observe  that  the  class  struggle  itself  is  typically
expressed in distributional terms; even in the factory,  conflict  is more centred
on  the wage bargain  than  on  issues of  property  or control.  This,  of  course,
can  be  dismissed  as  "mystification"  as well.119  In  any  case,  the  derogation
of real day-to-day struggles as "mere trade unionism" or "false consciousness"
does not seem to get us far.  It should prove more fruitful  to admit that  "con-
sumption interests are real and vital," not to be dismissed as "mere ideological
barriers" to an ideal polarity of  class.120 Nor are such considerations  limited
to high-consumption areas of the world: current assessments of African  labour
studies  also  suggest  revisions  along  these  lines.  121

The  recent  work  of  Harvey  and  Castells,  finally,  goes  far  in  answering
these complaints: the  importance  of  spatial  and  distributional  issues,  their
inescapable imbrication  in wider projects  for  change, are  no longer  left  in
doubt.  Castells,  of  course, was already  quite  sympathetic  to  urban  protest
groups. Many of these, he had argued, may go no further  than  ameliorative
reforms  — but  some,  in certain  conjunctures,  and  if linked  to the  broader
struggle of  the working  class, might  constitute  "urban  social  movements"
of  revolutionary  thrust.122 While his immediate expectations remain  modest,
Castells has now taken a further  step: it is the multi-class urban  movements
that  are  the primary  potential  source  of  change  and  they  should  distance
themselves  from  the class-based  movements and  parties,  which  may  other-
wise tend to divide, distract,  or coopt.123  Harvey,  similarly, now  recognizes
"different  loci  of  consciousness  formation"  and  resistence,  of  which  pro-
duction  is only one; with a  "material  base in  daily urban  life,"  they  "can-
not be dismissed as false."124 At the same time, however,  they breed  various
forms  of  fetishism' and  fragmentation  which only a broader  political  move-
ment  can  ultimately  resolve.

In both  formulations,  the problem  of reduction  is avoided,  with  the link
between  "urban"  and  production-based  cleavages  consigned  to  the  outer

20  K.P.  MOSELEY

reaches of the infrastructure  and the state. One might object,  in fact,  that
they have gone too far in this respect; more proximate, day-to-day connec-
tions are empirically noted, but not sytematically explored. The remainder
of  this paper  will be devoted  to this problem,  focussing  on the impact of
residential variations on class practices of various kinds. Two concrete con-
texts will be involved  — not only the local community, but the workplace,
that  "hidden  abode of production"125  where the urban theorists  apparent-
ly still  fear  to  tread.

Spatial  Matrices  of  Class
Lojkine  notes  that  the  subordination  of  labour  to  capital  may occur

at  the level of  space, itself  .  .  .either  directly  (when  the town is the
appendage of the factory, e.g. as in a mining community) or indirectly
. ..  when the firm is not involved in either the production or the manage-
ment  of  housing  .  . .126

The differences  between these two situations, however,  are equally worthy
of note. Burawoy has noted how managerial control is strengthened by join-
ing reproduction  directly  to  the  firm;127  the  separation  of  the two,  which
structuralists  consider  a sort  of  original sin,128 thus  appears  in a very  dif-
ferent  light.  Rather  than  merely deflecting  energy from  the class struggle,
it gives it space to mature — space in a literal as well as metaphorical sense.

The significance  of spatial appropriation and control is illustrated by the
short-lived cites ouvrieres, launched in Paris in 1849: unpopular with workers
due to their strict controls, they were even more opposed by conservatives,
"as breeding grounds for  socialist consiousness and as potential hearths of
revolution."129 At the same time, some businesses in the United States were
seeking to insulate their workers from  "urban agitation" by building isolated
company towns.130 Even the residential clustering of workers around a single
firm  would  have different  implications,  if  the housing  was not  company-
provided but privately acquired. And with increasing real wages and market
choice, workers could "move their community life out from  under the eyes
of  their  employers,"131 while various fractions  of capital developed  vested
interests  in workers'  consumption  and  purchasing  power.

The variegated spatial patterning of working-class settlements is also not
without  redeeming  effects.  Expectations  of  the mobilisation  of  labour-in-
general are at least premature; outside the workplace, rather, the real work-
ing class coalesces on local, regional, and national terrains. As Engels132 was
particularly  aware,  it was the concentration  of  workers  not  only in large-
scale units of production, but in homogeneous residential areas, that would
promote a wider consciousness of class. In such communities, distinctive sub-

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  21

cultures can emerge or be preserved, providing some insulation  from  the in-
dividualism,  mobility  aspirations,  and  consumption  styles  of  the  middle
class."3  Apart  from  community  itself,  three  major  sources  for  such  alter-
native  meanings  might  be noted  here. One is the  family,  in which  effective
relationships,  reciprocity,  and  use-values  prevail,  all  potentially  subversive
of  the  bourgeois  order  if  projected  too  far  outside  the  home.134  A  second
is the pre-capitalist  cultural elements often  associated with ethnicity, nationali-
ty, or  religion.  Lubeck,  for  instance,  has insisted on the role of  Islamic net-
works in  Kano in the constitution  of the Kano working class.135 Finally,  of
course,  is  the  fund  of  shared  experience  and  class  perceptions  which  is
engendered  in  the  process  of  work  itself.

A  large body  of  research  has shown  that  attachment  to workbased  iden-
tities  tends  to  be  maximized  in  more  or  less  isolated  "occupational  com-
munities," where the boundaries of work and community relationships closely
coincide.136 African  data also illustrate this point: the relative cohesion  and
militancy displayed  by miners,  railwaymen,  and  dockers has been traced  to
their  residential  concentration  as  well  as  distinctive  characteristics  of  the
work.137  Factory  workers  still  tend  to  be  relatively  recent  recruits,  few  in
number and spatially dispersed, but the direction of the relationship between
residence and  class-formation  seems to be the same.138 Generally  speaking,
ecological segregation  and residential stability can reinforce  communal-class
identity even where workers move between communal-sector  firms and jobs;
informal  recruitment  networks,  moreover,  may tend  to cluster  co-residents
in  similar  kinds  of  work.139

The local community  not  only provides a milieu  (indeed,  a chef  lieu)  for
class-formation,  but  interacts with class mobilization in synergistic ways. As
Harvey  has noted,  "the  'principle  of community'  is not a bourgeois  inven-
tion,"  and  may  serve as  "a  defensive  and  even offensive  weapon"  for  the
working  class.140 Castells goes further:  people require spaces as a  "material
basis"  of  opposition,  multiple  space  outside  work  where  they  may:

find  values, ideas, projects and, finally,  demands that  do not  conform
to the dominant social interests. The control over space is a major  bat-
tle  in  the  historic  war  between  people  and  the  state.141

We should  recall  here the strategic  role of  the popular  quarters  in the mili-
tant  urban  protests  of pre-industrial  Europe,142 the now universal  combina-
tion  of local traditions with the new oppositional tendencies of the industrial
working class, and  the segregated  Black communities  of America,  England,
and  South  Africa  which  have been the site of  rebellions  in recent  times. As
such cases suggest, the residential community may unite fractions of the work-
ing class divided  between different  kinds of  work,  and  bring together  those

22  K.P.  MOSELEY

with work and those without.  Here, too, relations of trust, permitting sus-
tained  cooperation,  effective  organization,  and  social control  may emerge
from  personal  interaction  and  informal  ties.

The interaction of neighbourhood with workplace and trade union, along
with the territorial hierarchy that frames the political process, provide futher
mechanisms of aggregation,  often  encapsulating local ethnic solidarities as
well.143 Consciousness and loyalties may remain fragmented,  of course, and
the politics of  ethnic or localistic arithmetic provides an unromantic view,
at  best,  of  the mobilization  of  the working class. Without  a  combination
of the respective issues, discontents, and personal attachments  of both the
workplace and living place, however, the strength and committments of the
larger  class  movement  may  be  undermined.  Although  the  fate  of  such
movements is not Castells' concern,  his accounts  of urban  protests in e.g.
Glasgow, Chile, and Madrid indicate the extent to which labour unions and
parties may draw on residential issues and bases of support.144  Conversely,
the decline of the American  Socialist Party has been linked to the breakup
of  old  inner-city  neighbourhoods,145  and  the  relative  lack  of  living-place
militancy  in  much  of  subsaharan  Africa  may be a  factor  in the weak  ties
between radicals and the grassroots there (South Africa  being an exception
which  supports  the rule).

It is thus no accident that policies of urban renewal and residential disper-
sal in the West have often  been aimed at  long-established  and  chronically
troublesome working-class residential zones.146 In the Third World,  on the
other  hand,  the main  target has been  "spontaneous"  settlements, and  the
affront  to market  discipline and  tp the legal and  symbolic order  that  they
represent.147  In  all  cases, as  Harvey  notes,

Control over spatial organization  and authority over the use of space
becomes a crucial means for the reproduction of social power relations.
The state, or some other social grouping such as financiers, developers,
or  landlords,  can thus  hide their power to shape social  reproduction
behind  the seeming  neutrality  of  their  power  to  organize space.148
Where labour, in any particular spatial context, cannot be brought under
control, the response of capital is often  to move away. While the cost and
power of labour  may be the most important  factors  here, the character of
local subcultures and  communities is often  involved as well. Too much or
too little community cohesion, informal social control, or attachment to cer-
tain production  methods, divisions of  labour,  or different  kinds of  work,
may all be factors prompting industrial relocation — and the spread of class
and  community  formation  to  new terrains.149

Looked  at  in  the  context  of  community,  the  meaning  and  impact  of
homeownership among workers also appears in a much more variable light.

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  23

Intraclass  conflict  over  the  neighbourhood  "turf,"  first  of  all,  may  occur
even  where  owner-occupancy  is  nil.  And  insofar  as  a  neighbourhood  is
working-class in composition, and particularly if residential turnover is low,
there seems little reason for homeowning to have particularly divisive effects.
Commercial considerations often  remain secondary; indeed, there is substan-
tial  "irrationality"  in  the  disposition  of  real  estate  by  owner-occupiers  in
general,  not to speak  of the relatively modest  houses likely to be owned  by
workers. The predominance of orientations to use, needs for  self-expression
and  security,  and  attachments  to particular  socio-spatial  contexts  of  every-
day  life,  may play  a  part  in  this  respect,  along  with  market  disadvantages
that  tend  to limit capital gains. The subjective  significance  of  homeowning
varies further  along neighbourhood,  subcultural and national lines.150 In the
Third World, meanwhile,  we are discovering the presence of  landlords  and
speculators even in traditional indigenous settlements, shanty-towns and low-
income  housing  projects.151  The significance  of  this,  however,  in  terms  of
individual  accumulation  or  differentiation  within the working class, is still
not  altogether  clear.  The  African  data  suggests  that  in  the  popular
neighbourhoods, profits and accumulation are generally slight. Most houses
are owner-occupied  and,  unlike land,  are  rarely  sold.  A  few  cheap  rooms
may be let out, if not appropriated by kin, yielding a somewhat higher stan-
dard  of  living but  rarely  a  promotion  to  the  rentier  class.152

A nucleus of homeowners, in fact, may strengthen working-class  defenses
against various encroachments on local terrain. In advanced capitalist coun-
tries,  many  instances  have  been  noted  in  which  small-scale  capitalist  and
homeowner interests have opposed property speculation and redevelopment
schemes.153  That  fractions  of  labour  may  be  allied,  not  with  capital  in
general, but  with  particular  fractions  of  the bourgeoisie,  is worth  pointing
out. Small-scale capital, though accorded short shrift  by Marxism-Leninism
(if  not by the socialist tradition as a whole), has certain anti-capitalist  pro-
clivities, in part because of concrete attachments to craft, vocation, or locality
that may run counter to unalloyed commitments to profit  and exchange. This
is even more the case in less-developed  economies where pre-capitalist  sen-
timents and relations are still particularly strong. Here, petty-bourgeois status
may rest on a fragile material and even legal base (particularly, perhaps, where
real estate is involved), requiring the maintenance of extensive kinship, ethnic
and  patron-client  ties.  At  the  community  level,  then,  landlords  and
homeowners tend to play pivotal social and political roles, joining local and
external networks in ways often  critical for  bringing in amenities and  war-
ding  off  renewal  and  demolition  schemes.154

The social heterogeneity which such coalitions imply, joining  formal-sector
workers with various echelons of the self-employed,  is commonly associated

24 K.P.  MOSELEY

with a diffuse  "urban populism" as opposed to radicalism of a class-conscious
kind.155 These contrasts, however, should not be overstated: they rest on an
idealised proletariat  that  only rarely anywhere can be found  and they may
mask  considerable  homogeneity  at the level of  culture and life  styles rein-
forced by the mixture of different  kinds of work both within households and
in the course of individual lives. It seems more interesting to note the way
in which  populism  itself,  as Castells puts  it,  "walks  on the thin  edge bet-
ween clientelism and the triggering of urban social movements,"156 and how
complex social networks may promote "a dangerous autonomy"157  among
the poor inimical to ruling class control. Community issues, may stimulate
consciousness of the shortcomings of the larger system, and the  formation
of coalitions favourable to change; such local alliances may lend support to
workplace  protests,  in  turn.158

In sum, either the intra-class differences  or the motley alliances emerging
in the living space seem to have necessarily negative net results. It is the disper-
sion and breakup of the local community,  rather,  that is the greater  threat
to  the unity  and  consciousness  of  the  working  class, and  it is in  such  cir-
cumstances,  perhaps, that  homeownership  may show a more sinister  face.
Study after  study, in different  parts of the world, shows that relocation  from
old  popular  and  working-class  neighbourhoods  to  more  recently  settled,
heterogenous and suburban areas entails a decline in social cohesion in both
class and  community  terms.  The new residential  areas

bring together a population  of strangers  . . . Unrelated  by the ascrip-
tive ties of kinship, long-standing neighbourliness and shared work ex-
periences  .  .  . workers  on  the  estates  tend  to  live a socially  isolated,
home-centered  existence.159

It is here that  "commodity  consciousness"160 and pursuit of the  "personal
life" can prevail, while attitudes to work, workmates, and the labour move-
ment take on a more instrumental cast. Somewhat ironically, state-provided
benefits  may  reinforce  this  state of  affairs,  reducing  dependence  not  only
on  the  employer,  but  on  kinship  and  neighbourhood  links  as well.161

These  tendencies  should  be  reinforced  in  mix-class  neighourhoods
dominated by the life-styles of the middle class. This point applies particularly
in Western countries where the incomes of  the stable working class permit
substantial emulation along these lines. Such residential areas "rule out the
convergence of workplace and neighbourhood loyalties" that earlier genera-
tions  experienced,162  and  may  reinforce  estrangement  from  the  urban
newcomers who have succeeded to the old central slums. Even in these cir-
cumstances, however, and where homeownership prevails, work-based iden-
tities and grievances may remain intense.163 Where working-class  residents

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  25

predominate,  moreover,  and with the passage of time, strong local primary
groups,  selective settlement  and job  recruitment  patterns  — in short  a neo-
traditional  form  of  community  — may well re-emerge.  Here, Western  and
Third-World  experience  is again  analogous  — the  old  homeowning  Black
suburbs of Detroit or Los Angeles, for instance, or the new low-income hous-
ing areas  that  ring  Rio,  Lagos,  or  Singapore,  could  all  be  looked  at  more
closely  in  this  regard.

Resistence  and  Control  at  Work
The  final  issue to be considered  here is the impact  of  the housing  situation
of workers on workplace behaviour. This is critical for all broader generaliza-
tions concerning housing and the working class, although actual data is still
remarkably scarce. Despite frequent  reference to relations of production, rela-
tions  in  production  are  rarely  considered  by  urban  theorists;  industrial
sociology, on the other hand, may refer to community structure, but seldom
to housing  as such.  We hasten  to  add  that  both  these generalizations  hold
less true  for  the  literature  on  the  Third  World  where various  connections
between work  and  housing are often  more clearcut.  (One wonders  if it was
not  Burawoy's research in Zambia, in fact, that  first  alerted him to the wider
significance  of  this  theme).

As we have seen,164  Burawoy  points to the association  between  company
provision of  worker  housing and  the reinforcement  of  managerial  control:
in some cases, "through their control of housing, provisions, company stores,
education  and  religion,  masters  were  able  to  consolidate  their  rule  in  all
spheres  of  life."165  This sort  of  formulation  points the way to more detail-
ed investigation  of the links between particular  kinds of provisions and  par-
ticular kinds of behavioural effects.  The compound  system (and  "company
states"  in  general)  seems  to  have  been  particularly  geared  to  simply  keep
workers at work, workers whose merely  "formal"  subordination  to  capital
had  to  be  backed  up,  not  only  by  a  particular  housing  regime,  but  by
manipulation  of  credit  and  debts,  provision  of  recreation  and  other  in-
ducements, and by a panoply of extra-economic controls. As for  performance
in the labour process itself, the compounds, in South Africa,  were designed
to  prevent  diamond  thefts;166  otherwise,  improvements  in  productivity
depended  on  a  further  set  of  punishments,  bonuses  and  fines.167

Burawoy's treatment  suggests that  state intervention in housing will have
a countervailing effect: along with othet types of collective provision, it creates
guarantees  of  subsistence  independent  of  the wage and,  especially,  of  par-
ticular firm. This, in combination with the regulation of industrial  relations
by the state, signals the transition  from  a "despotic" to a "hegemonic"  fac-
tory  regime,  characterized  by  formalized  work  rules,  negotiation  and

26  K.P.  MOSELEY

consent.168 Two variant  forms  of  the housing/workplace  nexus might also
be mentioned  here. For  one,  as  Burawoy  is aware,  where  the  state  is still
directly involved  in labour  control,  even  state-provided  housing  may have
coercive effects.  Apart from  Stalinist labour regimes, this can be seen in con-
temporary  South  Africa,  where  migrant  workers  lose  access  to  the  state-
provided hostels (and any other residence) if they leave their jobs. This assures
a  stable  labour  for  unpleasant  and  poorly  paid jobs  they  might  otherwise
be loath to take.169 Secondly, as we have pointed  out, company housing may
also coexist with liberal or hegemonic labour systems, with effects  that need
to be further explored. If such housing has not prevented strikes, for instance,
it may have made them less frequent,  and  there are indications  that  labour
turnover  might  be  reduced.170

Housing shortages and related  problems, conversely,  may be reflected in
labour demands and various forms  of discontent. In contemporary capitalist
countries, where markets  for jobs  and  housing are generally  quite  distinct,
it is hard to find  detail on these points. But Jeffries  notes, for  instance, that
housing problems were among the issues underlying the 1961 Sekondi railway
strike,171 and one suspects that more intensive research would  unearth many
cases of this kind. Interesting comparative data exists for  the Soviet Union,
where it is housing,  not jobs,  that  is in scarce supply,  and  where  both  are
ultimately provided  by the state.  While  the housing-enterprise  linkage has
been  attenuated  in  recent  years,  it  remains  a  potential  instrument  of
managerial control: workers are loath  to go so  far  as to  lose their  housing
privileges,  or even their  places  on  the  waiting  lists.172 Younger  and  lower-
echelon workers have less to lose in  this respect,  especially  those  employed
in less well-endowed regions or firms. For these categories of workers, then,
labour turnover is now extremely high, while inadequate housing, particularly
in  peripheral  regions,  is  a  major  cause  of  such  open  labour  protests  as
occur.173

What  of  the  relationship  of  housing  status  to  variations  in  workplace
behaviour of more routine kinds?  Harvey once suggested  that  homeowner-
ship  for  workers  in  Western  countries  was designed  to  promote

a  sense of satisfaction  and  contentment  that  will lead to  spontaneous
co-operation  and  efficiency  in  the  work  place.174

From the actor's point of view, however, the private home may be a symbol
of  autonomy  and  self-sufficiency  relative  to  forces  emanating  from  both
capital  and the state.175 Is such consiousness  necessarily  false?  Wright and
Singlemann have noted the existence of "semi-autonomous employees," who
retain substantial control in the labour process,176 while Burawoy links loss
of autonomy in subsistence with  submission to workplace controls.177  pUt-

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  27

ting  these  ideas  together,  might  there  be  a  category  or  "semi-authonous
reproduction"  in  which  homeownership,  inter  alia,  might  play  a  contem-
porary role? This raises interesting questions, in turn.  For instance, is it the
"semi-autonomous employees," historically a particularly resistent  fraction
of  the  labour  force,  that  are  most  likely  to  own  homes?  Are  local  labour
markets such that  being tied to a place also ties one to a certain job?178 Does
ownership entail greater expense than would renting, and can costs be shared
out by letting rooms or by increased household  size? Most important  of all,
perhaps, is whether the house is still being paid for or if it is owned  outright.

In  capitalist  settings,  owner-occupancy  now  typically  involves  purchase
rather than inheritance, and substantial long-term debt. And a worker  "mor-
tgaged  up to  the  hilt,"  Harvey  notes,  is "a  pillar  of  social stability,"179  as
homeownership schemes have long assumed. The evidence seems to support
this  point.  A  mortgage  creates  a  double  vulnerability:  heavy  installment
payments and the threat of loss of equity if payments fall behind. Thus, Cana-
dian  data  suggest  that  would-be  owners  are likely to be  extremely  faithful
to the job,180 and  a New Jersey study found  that workers on the  "mortgage
treadmill" tended to accept harder, longer and more closely supervised work
in order  to maximize their wage.181 Similar  findings  emerge from  socialist
Hungary,  where both  private home construction  and small-scale  enterprise
have burgeoned in recent years. As in the Western cases, however,  workers'
positions  in  the  life  cycle  strongly  pattern  the  results.  Young  Hungarian
workers,  for  instance,

face  an  acute  shortage  of  housing  .  .  . They have no  alternative  but
to  begin  construction  of  their  own  homes,  which  may  take ten  years
or  more.  During  this time they will also  face  the costs  of  bringing  up
their  families  .  .  .182

Older  workers,  on  the other  hand,  are more likely to  have completed  this
process  and,  with  other  household  members,  to be earning  income  on  the
side. With their consumption relatively independent from the job, "they were
in a stronger position to withhold effort  than were younger workers in need
of  ready  cash  .  .  ,"183

This need not suggest a conscious radicalism, of course. Canadian owner-
workers,  for  instance,  feel  that  homeowning  has conservative effects;  only
a  minority,  on  the  other  hand,  felt  that  they  were  less  likely  to  agitate  at
work  or go on strike.184 It is indicative, perhaps, that the unusually prolong-
ed and militant Hormel strike in the United States occurred in a town whose
owner-occupancy  rate of  80 percent  is one of  the highest  in the  country.185
Just as company housing failed to prevent worker militancy 50 or  100 years
ago,  homeownership  too  may  prove to  be  less reactionary  than  originally

28  K.P.  MOSELEY

supposed.  If  workers seek market  freedoms  and  increased  control  of  per-
sonal space,186 if worldwide structural tendencies now favour  privatization,
small enterprise, and even work at home, what is important is to not simply
reject but to recuperate these trends, grasping the whole cluster of variables
that condition their effects  and manipulating them insofar  as possible so as
to  shift  the balance of  advantages  towards  the poor  and  working class.

This brings us back to the Third World,  where both  "semi-autonomous
reproduction" and semi-proletarianization are relatively widespread. Here,
the  contradictions  between  residential  conditions  and  the  imperatives  of
formal-sector employment are greater than in the West, and, given low wages
and inadequate public resources, much more difficult  to resolve. As Cooper
has eloquently observed,  the anarchy which prevails with respect  to much
' urban housing in Africa has been felt an indirect but insidious threat to con-
trol over the labour force.187 Apart from  the symbolic respects of the'spatial
order,  there are two main  aspects of  the housing  situation  that  should  be
noted.  For  one, various  forms  of  irregular  housing can  be very cheap in-
deed. Thus, Turner contrasts the "supportive shack," which is virtually free,
and the  "oppressive house"  which, even if  subsidized,  might  absorb over
half  a Mexican  worker's  pay.188 A modest  but  permanent  self-built  house
might cost only about three month's wages — while it might take three years
of a worker's income to buy a house in the USA.189 The "mortgage trap,"
like open unemployment, is a rare luxury, virtually unavailable to the work-
ing poor,  and  home-ownership  is often  postponed  to  the  end  of  a wage-
worker's career.  It is relatively unlikely, then, to be a significant  source of
discipline  on  the job.

The structural assymetry between work and residence is re-inforced  insofar
as workers retain control of rural land — a common situation in the African
case, at least.190 Whether these rural connections entail net expenditures or
gains in income is not at all clear, but they certainly offer  a cheap subsistence
redoubt  in  case  of  urgent  need.  Historically,  low  costs  of  reproduction,
generally, and access to rural land, particularly, have been associated with
a "backward bending" supply curve and high turnover in wage labour, with
resistence expressed through sheer withdrawal from work. Rising living costs
and  scarcity of jobs, however,  have dramatically  reversed this situation in
recent  years.191  As  far  as  private  housing  is concerned,  we  unfortunately
have little other information on how residence affects  behaviour on the job.
One  possibility  that  comes  to  mind  is  that  chain  migrations  and  chain-
employment may have a residential complement, with "land-lords" also oc-
cupying or being linked to positions of authority at work.192 Another, given
the inadequacies of public transport in many parts of the world, is the con-
nection between distance of home from  work, on the one hand, and, on the

CONTESTED  TERRAINS:  SOCIAL  RELATIONS  OF PRODUCTION  29

other,  lateness, absenteeism  and  pressures  on employers to cover  transport
costs  in  wages  or  in  kind.193

The  question  of  location  reminds  us,  finally,  of  the spatial  dimension  of
capital  itself,  which  moves  production  facilities  between  countries,  coun-
trysides  and towns, finding  new reservoirs of labour at destination, or  drag-
ging their workforce  in their wake. In older industrial areas, threats of  plant
closure  or  relocation  have been  leading  to  "give-back"  contracts,  reducing
fringe  benefits  and/or  wages, and  to  a general  enhancement  of  managerial
control.  "Hegemonic  despotism"  is the rather  fanciful  name that  Burawoy
has given to this new production regime, in which the "reproduction  of  labour
power  is bound  anew  to  the production  process,  but  . ..  at  the level of  the
firm,  region  or  even  nation-state."194  The new destination  areas are also  not
secure, and rising wages, union militancy, or political tensions may put capital
on the move again.  Resistance outside the workplace, however, may increase
—  "resistance  from  neighbourhoods  that  do  not  want  to  disappear,  from
regional  cultures  that  want  to  cluster  together,  and  from  people,  that,
previously  uprooted,  want  to  create  new  roots."195  Such  movements  are
fragmented,  however, both  from  each other, and with respect to more deter-
minate  levels of  control:  "relationships  of  production  are integrated  at  the
world-wide level, while experience is culturally specific and power is still con-
centrated  in  the  nation-states  .  .  ."196

fet  local  spatially-based  conflicts  also  express  the  contradictions  of  the
wider  system,  and  are  an  element  of  its  prolonged  "crisis-transition,"197
hopefully  towards  a  system  based  not  on  indefinite  accumulation  but  the
satisfaction  of more authentic human needs. The programmes of the working-
PIJK-C parties of Europe have begun to incorporate  urban  and territorial  con-
cerns, placing workplace and living-place issues under one political roof  and
providing  an essential  opening  to a broader  popular  base.198 For  reasons  of
praxis as well as theory,  then,  the  connections  between  the  two terrains,  as
mediated  by both  capital  and labour,  should  remain  a key area of  reflection
and  research.

Footnotes

1.  See Harvey  (1982),  Yancey  (1976),  Gordon  (1978).
2.  Giddens (1983:135  ff.).
3.  See Saunders  (1981:  Ch.  2 - 3 );  Rex  (1971).
4.  Saunders  (1981:  Ch. 4).
5.  (1973,  1981).
6.  Harvey (1981:103).
7.  Castells  (1977). 
8.  Ibid.
9.  See Harloe  (1977);  Pickvance (1976a,  1976b).
10.  (1982).

'

30 K.P. MOSELEY

11.  (1981,  1984).
12.  (1983,  1984).
13.  (1985).
14.  See,  e.g.,  Lefebvre  (1974);  Gottdiener  (1984).
15.  An initial version, devoted  to theoretical issues and advanced-capitalist  case materials, was
presented to the Faculty Seminar, Faculty of Social Sciences, University of Port  Harcourt,
June  1986.

16.  This  term  is  borrowed,  of  course,  from  Edwards  (1979).
17.  (1981).
18.  (1985).
19.  See Massey  (1984); Gordon  (1978).
20.  See,  e.g.,  Walker  (1978);  Harvey  (1985: Ch.  3).
21.  See  Lojkine  (1976).
22.  Harvey  (1973:  Ch.  5);  Lamarche  (1976);  Saunders  (1981:  232-35).
23.  See,  e.g.,  Harvey  (1973:  esp.  Ch.  2); Castells  (1977: Part  V).
24.  O'Connor  (1983:172,  181, 224-27);  Cooper  (1983b:31 -34).
25.  See esp.  Evers  (1976,  1984);  Jaramillo  & Schteingart  (1985).
26.  Pacione  (1981);  Cornelius  1976; Gilbert  & Gugler  (1982: Ch.  5).
27.  Massey  (1984);  Scott  & Storper  (1986).
28.  Cohen  (1981);  Fainstein  & Fainstein  (1982); Zukin  (1982).
29.  Friedmann  & Wolff  (1982).
30.  For a general treatment  emphasizing these trends, see Mingione (1981). On  Third-World
commuting problems, see the chapters on Singapore, Lagos, and Bombay in Pacione (1981).

31.  Gilbert  & Gugler  (1982: Ch.  2);  Portes  & Walton  (1981:  Ch.  3).
32.  Mingione  (1981:36-37).
33.  See  Saunders  (1981:234 -  36).
34.  Mingione  (1977,  1981);  Lamarche  (1976).  See  also  Saunders  (1981:233  ff.);  Pickvance

(1976a:272-78).

35.  Baran  (1960)  was  a  major  progenitor  of  this  view.
36.  Harvey  (1981).
37.  Lojkine  (1976:135-38).
38.  Evers  (1984:  esp.  491-92);  Jaramillo  & Schteingart  (1985:186-90).
39.  See Cox  (1981:440-42);  Pickvance  (1976a:272-73,  277-78);  Mingione  (1981:101).
40.  Saunders  (1981:234  ff.);  Edel  (1984:  Ch.  10).
41.  Lojkine (1977); Massey (1984:150 ff.). For details on New York, see Zukin (1982); on  19th-

century  Paris,  Harvey  (1985:119  ff.).

42.  Mingione  (1981:56-58);  Edwards  (1979:  Ch.  9).
43.  See,  e.g.,  Marris  (1962).
44.  See Portes  & Walton  (1981:  Ch.  3).
45.  Ibid.,  pp.  94  ff.
46.  E.g.  Connoly  (1985).
47.  Cf.  Cooper  (1983b);  Gilbert  & Gulger  (1982:113).
48.  On  such windfall  differentials  (and  the obverse: the excessive cost  of  conventional  hous-

ing  for  manual  workers),  see Turner  (1977),  esp.  pp.  5 8 - 6 3,  71 - 7 6,  8 0 - 8 5.

49.  Harvey  (1981:116;  1982:552-53);  Mingione  (1981:28,  37).
50.  It  was  during  the First  World  War,  for  instance,  that  workers  — with  business  support
—  won  rent  controls  in  Glascow  (Castells  1983:34-7).  For  the  interwar  period,  see
Pickvance  (1976a:282)  and  Edel  (1984:280).

51.  See Mingione (1981:101-102); Harvey (1985:266 ff.); Fainstein & Fainstein (1984:179 ff.).
52.  Mingione (1981:128 ff.); Fainstein & Fainstein  (1984:167 -  70); Saunders (1984:210 -11).
53.  (1983b).
54.  See e.g.  Gilbert  & Gugler  (1982:99  ff.);  Stren  (1982).
55.  (1983,  1984).
56.  Mingione  (1981:24).
57.  (1981:Ch.4,8).

CONTESTED  TERRAINS:  SOCIAL  RELATIONS  OF PRODUCTION  31

58.  Harvey  (1982:547).
59.  Rex  (1971);  see Saunders  (1981:115,  137).
60.  Saunders (1981:63).
61.  E.g.  Harvey  (1973); Bookchin  (1974); Castells  (1977).
62.  (1981:269 ff.;  1984: esp.  203-10).
63.  (1982:547).
64.  (1973).
65.  Harvey  (1981:103;  1982:550).
66.  Cox  (1981:444).
67.  Harvey  (1982:552).
68.  Ibid.  p.  557.
69.  Mingione (1981:31).
70.  Cited  in  Saunders (1981:198).
71.  Castells (1983).
72.  Harvey (1985).
73.  Ibid.,  p.  265.
74.  For the following, see esp. Edwards (1979:163 ff.); Bonacich (1979); Rubery (1984); Bromley

& Gerry (1979).

75.  Harvey  (1981:115).
76.  Saunders (1984:207).
77.  E.g.,  Sitas  (1985).
78.  Parkin  (1969:13  ff.).  For  a European  case,  see Castells  (1977:110).
79.  See Burawoy  (1985:197).
80.  See,  e.g.  Gans  (1962:21,  82,  85),  or Jeffries  (1978:22).
81.  (1958).
82.  Agnew  (1981:462-63).
83.  For West Africa,  see Peil & Sada (1984:297); Barnes (1979:63-64;  1982:13, 21).  Here,
homeowning is strongly associated with self-employment, though both may be made possible
by previous savings  from  wage work.  Canadian data (Pratt  1986: 379 ff.)  suggests that
it  is the sale  of  a house  that  may  provide  capital  for  starting a business  on  one's  own.

84.(1985:103-105).
85.  (1974).
86.  E.g.  Harvey  1973:60-64);  Castells  (1977:411-12).
87.  Jeffries  (1978:14-15,  35); Parkin  (1969:1 - 4,  1 4- 15).
88.  E.g.,  Sandbrook  & Am  (1977:29-30);  Peil  (1981:75).
89.  Portes (1985:11 -14);  Connoly  (1985:70 -  72).
90.  Mingione  (1981:53,  147  ff.).
91.  Ibid.,  p.  38.
92.  Ibid.,  pp.  2 3 - 2 4.
93.  Burawoy  (1985:12 ff.,  126).
94.  Ibid.,  p.  31.
95.  Ibid.,  p.  92.
96.  Bendix  (1974:169  ff.).
97.  Burawoy  (1985:91-92).
98.  Ibid.,  Ch.  5; van Onselen  (1976).  The latter,  pp.  3 4 - 8,  133-36,  also notes the impor-
tance of  self-housing  within the Rhodesian  compound areas,  and its relationship to the
stratification  of  the labour  force.

99.  Bendix  (1974:49-52);  Bookchin  (1974:111 — 15);  Edel  (1984:276-79);  Walker

(1978:179-81).

100.  Burawoy  (1985:100-101).
101.  Edel  (1984:279-80); Edwards (1979:91  ff.).
102.  The classic study is King's (1976) study of colonial India, where minute gradations in oc-
cupational status were mirrored on the residential plane. See also O'Connor (1983:175 ff.).

103.  Grillo  (1973:23-24,  87); also,  Jeffries  (1978:17-18).
104.  O'Connor  (1983:171).

32  K.P.  MOSBLEY

105. Edwards (1979:91). This may have posed less of a problem to African  workers, who often

had  rural  homes to which  they could  return  (cf.  Grillo  1973:44-48).

106.  Stark  (1982:312-13);  Harvey  (1982:557-58).
107. Ibid., p. 558. The sheer cost of employee housing for companies should also be taken into
greater account, as well as the availability of  alternative sources of  housing supply.  Cf.
van Onselen  (1976:34-38,  129-32);  Walker  (1978:189-90).

108.  Cronberg  (1986). Union  housing  in New York  is mentioned  by Edel (1984:310),
109.  The importance of formal sector employment for access to public housing estates is men-

tioned  for  Mexico (Arias & Roberts  1985:163-64)  and  Ghana  (Peil  1981:73).

110.  See Burawoy  (1985:145:45); Anon.  (1983); Garnett  (1985).
111.  See n.  79, above.
112.  Engels (1979). Also noted in Burgess (1978:1125 -26); Edel (1984:173, 189-92); Bookchin

(1974:114).

113.  Harvey  (1981:116);  Burgess  (1978), esp. p.  1120.
114.  Harvey (1982:560); see also Lloyd (1979). Quite unintentionally, the neo-Weberian spec-
tre that  Saunders  (1984:213-15)  has summoned  up  — in which  antagonistic and  "ex-
ploitative" relations between homeowners and public sector renters displace the traditional
class divide •— might  be cited  in confirmation  of  such  fears.

115.  Harvey  (1982:556-57).
116.  Agnew  (1981:458-60,  466-68);  Cox  (1981:447-48);  Harvey  1982:550-51).
117.  Saunders (1981:241).
118.  Edel (1984), esp. Ch.  10. >
119.  Cox (1981:276).
120.  Sauttders (1981:276).
121.  See Freund  (1984), esp. p.  13; Cooper  (1983b); Sandbrook  (1981).
122.  Castells  (1977:263, 324 ff.).
123.  Castells (1983), esp. Ch.  32; (1984:249-50).
124.  Harvey  (1985:262, 251).
125.  Marx, cited  in Burawoy (1985:210).
126.  Lojkine  (1977:143).
127.  Supra, pp.  26-27.
128.  See esp. Harvey  (1982). Cf.  Henderson  & Cohen  (1979:10), who link  the separation of

work  and  residence with  problems  of  workplace controls.

129.  Harvey  (1985:145).
130.  Edel  (1984:276-77).  See also, Massey (1984, esp. pp.  54, 61-62).
131.  Gans (1984:284).
132. (1958)
133.  See Gans (1962); Yancey (1976); Castells (1977), Ch. 7; Harvey (1985:18-19, 256-57).
134.  Cox (1981:433-34);  Harvey  (1985:258-59).
135.  Lubeck  (1981), esp. pp. 47, 60-66.
136.  See Blauner  (1968): Lockwood  (1982:361-62,; Kornblum (1974).
137.  Sandbrook  (1981:16-20,  25); Parkin  (1969:53).
138.  See Sandbrook  & Am  (1977: 52-53);  cf.  Peil(1981:82  ff.).
139.  Yancey  (1976); Grillo (1973:55  ff.).
140.  Harvey (1981:117).
141.  Castells (1983:70). For the importance of places and social relations outside the point of
production for the crystallization of class identity, see also Massey (1984:43, 58); Cooper
(1983b:36), and Katznelson (1981). We were unfortunately  unable to fully consult the lat-
ter  source by the time of  writing.

142.  Hobsbawn  (1959:108  ff.).
143.  See, e.g.  Kornblum  (1974);  Remy (1975).
144. Castells (1983:31-32,  36,  199 ff.,  24$-49).
145. Edel (1984:307 -10).
146.  See, e.g.  Castells (1977:296-97,  319-20);  Harvey  (1985:€h.  3), esp. pp. 1O2-1O3.
147.  Cooper (1983b:32-34); Castells (1983:189); Perlman (1976:Ch. 7-8); Marris (1962:Ch. 9).

CONTESTED  TERRAINS:  SOCIAL  RELATIONS  OF PRODUCTION 33

148.  Harvey  (1985:23).
149.  Massey (1984); Gordon (1978). On the impact of ethnic sub-cultures, see Fainstein& Fains-

tein  (1982:171 -72);  Henderson  & Cohen  (1979:24-26).

150.  See Agnew  (1981:468  ff.);  Saunders  (1984),  esp. pp.  204-205,  221 - 2 3;  Edel (1984).
151.  Aronson  (1978);  Stren  (1982),  esp. pp.  94-95;  Smart  (1986).
152.  Barnes (1982), esp. pp. 17-22, and (1978); White (1983); Marris (1962), esp. pp.  19-22,

26-27,  79.

153.  Lojkine  (1977:147-50);  Zukin  (1982);  cf.  Cox  (1981:436-38;  442-43).
154.  Chege (1981); Barnes (1979; 1982:22-24); Sandbrook  (1982: 147-56,  195-98); Castells
(1983:Ch.  19,20). Conversely,  where owners  are weak  or  indifferent,  communities  may
be destroyed:  see White  (1983);  Marris  (1962:Ch.  7); Gans  (1972).

155.  Sandbrook  & Arn  (1977);  Bujra  (1978-79);  Nelson  (1979);  cf.  Cornelius  (1971).
156.  (1983:179).
157.  Cooper  (1983b:38).
158.  Peace  (1975);  Gordon  (1978:36,  40-42).
159.  Lockwood (1984:367). See also Castells (1977:98 ff.); Marris (1962:Ch. 8, 123 ff.);  Hender-

son  & Cohen  (1979:20-22).

160.  Lockwood  (1984:371).
161.  Montgomery (1985:4). For other second thoughts on state welfare provisions, see Castells

(1984:245); Saunders  (1984:211).

162.  Montgomery  (1985:3-4).
163.  Ibid.,  p.  4;  Tyler  (1986).
164.  Supra,  pp.  26  ff.
165.  Burawoy  (1985:92).
166.  Van  Onselen  (1976:128-32).
167.  Ibid.,  p.  169; Burawoy  (1985:226-28,  231 -34).  Cf.  Crisp  (1983).
168.  Ibid.,  pp.  125  ff.
169.  Sitas  (1985).
170.  The railway workers of  Kampala and Sekondi-Takoradi  show remarkable stability, with
most  workers having served  for  10 years or more, and with  turnover  rates of  only a few
percent a year (Grillo  1973:39-40; Jeffries  1978:16). As only one-third of the Ghanaian
workers were company-housed, however, other factors must also be involved — probably
the  railway's  very  superior  wages  and  over-all  conditions  of  work  (cf.  Grillo  1973:62).

171.  Jeffries  (1978:76,  95).
172.  Jones (1981:267). Workers in self-managed  enterprises in Yugoslavia face similar constraints

(Sharon  Zukin,  personal  communication).

173.  Jones  (1981:267,  273);  Burawoy  (1985:198).
174.  Harvey  (1982:557).
175.  Gans  (1984:307);  Saunders  (1984:219-23).
176.  Wright  & Singelmann  (1982:182-83).
177.  Burawoy  (1985:89).
178.  Edel  (1984:192-93)  notes  recent  developments  which  belie Engels'  fears  on  this  point.
179.  Harvey  (1982:551).
180.  Pratt (1986:366 -  89). They would be more likely, too, to send their wives to work (ibid.,

392-95).

181.  Montgomery  (1985:4).
182.  Burawoy  (1985:194).
183.  Ibid.,  p.  189.
184.  Pratt  (1986:390).
185.  Tyler  (1986).
186.  Harvey  (1985:254).
187.  Cooper  (1983b:22-23, 30).
188.  Turner (1977:58 -  62). In much of West Africa,  rents for a self-contained  flat  would rare-

ly be  less than  a  month's  minimum  wage.

189.  Turner  (1977:80);  Edel (1984:306).

34  K.P.  MOSELEY

190.  Studies in East Africa suggest that about three-quarters of urban workers have land (Sand-
brook  1981:3; Bujra  1978 -  79:56). Of the Kampala railwaymen, over 70 percent had rural
houses as well (Grillo  1973: 44-45).

191.  Sandbrook  (1982:126-27).
192.  Cf.  Remy (1975); Cohen  (1969); Dahya (1974).
193.  In Nigeria, transport as well as housing allowances are standard benefits in large-scale/formal
firms, and staff buses are also a common sight. For women, especially, proximity to work
to home may be essential to regular appearance at work, or even to entering the labour
market  at  all (di Domenico  1983:261, 263-4;  Prakash  1983:84-87).

194.  Burawoy (1985:150).
195.  Castells  (1983:241-42).
196. Castells (1983:310).
197.  Mingione (1981:39).
198.  See Adams (1986:19-20); Lojkine (1977:154); Mingione (1977:105). For a historical over-

view, see Edel  (1984), Ch. 6 and 9.

References 
Agnew, J.A.  1981. "Houseownership and the Capitalist Social Order," in Dear and Scott 1981:

^

Anon.  1983. "Japanese  Women and  Work," Newsletter of  International Labour Studies 19

457-80.

* (Oct.): 16.

Arias, Patricia and Bryan Roberts 1985. "The City in Permanent Transition: The Consequences

of  a National  System of  Industrial  Specialization,"  in Walton  1985:149-75.

Aronson, Dan R. 1978. "Capitalism and Culture in Ibadan Urban Development,"  Urban An-

thropology 7,3:253-69.

Baran,  Paul A.  1960.  The Political Economy of  Growth. New York: Prometheus.
Barnes, Sandra T. 1979. "Migration and Land Acquisition: The New Landowners of Lagos,"

African  Urban Studies 4 (Spring):59- 70.

1982. "Public and Private Housing in Urban West Africa:  The Social Implications," in Mor-

rison and Gutkind  1982:5-32. 

,

Bendix, Reinhard 1956.  Work and Authority in Industry. Berkeley: University Of California Press.
Bohaciah, Edna 1979. "The Past, Present, and Future of Split Labor Market Theory," Research

in Race and Ethnic Relations, I.  Greenwich: JAI Press.

Blauner, Robert  1968. "Work Satisfaction and Industrial Trends in Modern Society," in Wm.

J.  Goode,  ed.,  The Dynamics of  Modern Society, New York: Atherton:  110-31.

Bookchin,  Murray  1974.  The Limits  of  the City. New York:  Harper  and  Row.
Bromley, Ray and Chris Gerry, eds.  1979. Casual Work and Poverty in Third World Cities.

New York: Wylie.

Bujra,  Janet  M.  1978-79.  "Proletarianization  and  the 'Informal  Economy'^  A Case Study

From  Nairobi," African  Urban Studies 3 (Winter): 47-66.

Burawoy, Michael  1985. The Politics of  Production. London: Verso.
Burgess, Rod 1978. "Petty Commodity Housing or Dweller Control? A Critique of John Turner's

Views on  Housing Policy,"  World Development VI, 9/10:  1105-33.

Castells, Manuel  1977.  The Urban Question. London: Edward  Arnorld.
1983.  "The  City and the Grassroots. Berkeley: University of California  Press.
1984. "Space and Society: Managing the New Historical Relationships" in Smith 1984:235 - 60.
Chege, Michael  1981. "A Tale of Two Slums: Electoral Politics in Mathare and Dagoretti,"

Review of African  Political Economy 20 (January-April):  74-88.

Cohen, Abner 1969. Custom and Politics in Urban Africa: A Study ofHausa Migrants in Yoruba

Towns. Berkeley: University  of  California.

Cohen, R.B. 1981. "The New International Division of  Labour, Multinational  Corporations

and  Urban  Hierarchy,"  in  Dear anq" Scott  1981:281-315.

Connolly, Priscilla 1985. "The Politics of the Informal Sector: A Critique," in Nanneke Redcliff

CONTESTED  TERRAINS.- SOCIAL  RELATIONS  OF PRODUCTION  35

and Enzo Mingione, eds., Beyond  Employment:  Household,  Gender and Subsistence. Ox-
ford:  Blackwell, pp.  55-91.

Cooper, Frederick,  ed.  1983a. Struggle for  the City: Migrant Labor,  Capital, and  the State in

Urban  Africa.  Beverly  Hills: Sage.

1983b. "Urban  Space,  Industrial  Time, and Wage Labor  in Africa,"  in Cooper  1983a:7-50.
Cornelius, Wayne A.  1976. "The  Impact of Cityward Migration on Urban  Land and  Housing
Markets:  Problems  and  Policy  Alternatives  in  Mexico  City,"  in  Walton  and  Masotti
1976:249-70.

Cox, Kevin R.  1981. "Capitalism  and Conflict  Around the Communal  Living Space," in Dear

and  Scott  1981:431-55.

Crisp, Jeff  1983. "Productivity and Protest: Scientific Management in the Ghanaian Gold Mines,

1947-1956,"  in  Cooper  1983: 91-130.

Cronberg.Jarja  1986.  "Tenants'  Involvement  in  the  Management  of  Social  Housing  in  the
Nordic Countries," Scandinavian Journal of Housing and Planning Research 3, 2:65 -  87.
Dahya, Badr 1974. "The Nature of Pakistani Ethnicity in Industrial Cities in.Britain," in Abner

Cohen,  ed.,  Urban  Ethnicity.  London:  Tavistock.

Dear,  Michael and  Allen  J.  Scott,  eds.  1981. Urbanization  and  Urban Planning  in Capitalist

Society.  London:  Methuen.

di Domenico, Catherine  1983. "Male  and  Female  Factory Workers in  Ibadan,"  in  Christine

Oppong,  ed.,  Female  and  Male  in  West Africa.  London: Allen  & Unwin.

Edel, Matthew, PUiott D. Sclar, and Daniel Luria, Shaky  Palaces: Homeownership  and Social

Mobility  in  Boston's  Suburbanization.  New  York:  Columbia University  Press.

Edwards, Richard 1979. Contested  Terrain: The Transformation of the  Workplace in the Twen-

tieth  Century.  New  York:  Basic  Books.

Emmanuel,  Arghiri  1972.  Unequal Exchange.  London:  New Left  Books.
Engels,  Friedrich  1958.  The  Conditions  of  the  Working  Class in England  in  1844.  Stanford:

Stanford  University  Press.

1979.  The  Housing  Question.  Moscow:  Progress  Publishers.
Evers, Hans-Dieter  1976. "Urban Expansion and Landownership in Underdeveloped Societies,"

in Walton  and  Masotti  1976:  67-79.

1984. "Urban  Landonwership, Ethnicity and Class in Southeast Asian Cities," International

Journal  of  Urban  and  Regional  Research VHI, 4:481 -96.

Fainstein,,'fcprman  I. and Susan S. Frainstein  1982. "Restructuring the American City: A Com-
parative Perspective," in their (eds.), Urban Policy  Under Capitalism. Beverly Hills: Sage,
pp.  161-89.

Freund, Bin  1984. "Labor and Labor History in Africa: A Review of the Literature,"  African

Studies Review 27,  2:1-58.

Friedmann,  John and  Goetz Wolff,  1982. "World City  Formation: An  Agenda  for  Research

and  Action,"  International  Journal  of  Urban and  Regional  Research  VI,  3:309-44.

Gans,  Herbert  J.  1962.  The  Urban  Villagers. New  York:  Free Press.
1984.  "American  Urban  Theories  and  Urban  Areas:  Some  Observations  on  Contemporary
Ecological  and  Marxist  Paradigms,"  in Ivan  Szelenyi,  ed.,  Cities  in Recession.  Beverly
Hills:  Sage,  278-308.

Garnett,  Nick  1985. "A  Tough  Life  on the Line at  Nissan,"  Financial  Times  (18 Sept.): 14.
Giddens, Anthony 1983. A  Contemporary Critique of Historical Materialism. Berkeley: University

of  California  Press.

and David Held, eds. 1981. Classes, Power and Conflict.  Berkeley: California  University Press.
Gilbert,  Alan and Joself  Gugler  1982. Cities, Poverty,  and Development:  Urbanization  in the

Third  World.  Oxford:  University  Press.

Gordon,  David  M.  1978.  "Capitalist  Development  and  the  History  of  American  Cities,"  in
William K. Tabb and  Larry Sawyers, eds., Marxism  and the Metropolis.  New York: Ox-
ford,  pp.  2 5 - 6 3.

Gottdiener,  M.  1984.  "Debate  on the Theory of  Space: Toward  an Urban  Praxis," irt Smith

1984:199-218.

Grillo,  R.D.  1973. African  Railwaymen.  Cambridge: Cambridge  University  Press.

36  K.P.  MOSELEY

Harloe,  Michael,  ed.  1977.  Captive Cities.  New York:  Wiley.
Harvey,  David  1973. Social Justice and  the  City.  Baltimore: Johns  Hopkins.
1981. "The Urban Process Under Capitalism: A Framework  for  Analysis," in Dear and Scott

.  1981:91-121.

1982, "Labour, Capital, and Class Struggle Around the Built Environment in Advanced Capitalist

Societies,"  in Giddens  and  Held  (1982) 545-61.

1985.  Consciousness and  the  Urban Experience.  Baltimore:  Johns  Hopkins.
Henderson,  Jeff  and  Robin  Cohen  1979.  "Work,  Culture  and  the  Dialectics of  Proletarian

Habituation,"  Papers in  Urban and  Regional  Studies  (Birmingham)  3:3-34.

Hobsbawn,  E.J.  1959. Primitive  Rebels.  New  York:  Norton.
Jarafnillo,  Samuel  and  Martha  Schteingart  1985.  "Capital  Accumulation  and  Housing  Pro-

duction  in  Latin  America  1960-80,"  in  Walton  1985:  176-94.

Jeffries,  Richard  1978. Class, Power and Ideology in Ghana: The Railwaymen ofSekondi.  Cam-

bridge:  Cambridge  University  Press.

Jones, T. Anthony  1981. "Work,  Workers and Modernization in the USSR," Research in the

Sociology  of  Work  I.  Greenwich:  JAI  Press.

Katznelson,  Ira  1981. City  Trenches. Chicago:  University  of  Chicago  Press.
King,  A.P.  1976.  Colonial  Urban Development,  London:  Routledge  and  Kegan  Paul.
Konrad, Gyorgy and  Ivan Szelenyi  1977. "Social Conflicts  of Underurbanization,"  in Harloe

1977.  157-73.

Kornblum,  William  1974. Blue  Collar Community.  Chicago:  University  of  Chicago  Press.
Lamarche,  Francois  1976.  "Property  Development  and  the  Economic  Foundations  of  the

Urban  Question,"  in  Pickvance  1976: 85-118.

Lefebvre,  Henri  1976.  The Survival  of  Capitalism.  London:  Allison  and  Busby.
Light,  Ivan  H.  1972. Ethnic Enterprise in America.  Berkeley: University  of California  Press.
Lloyd,  John  1979.  "Neighbourhood,  Community  and  Class: The  Role of  Place  m the

Formation  of Class Consciousness," Papers in Urban and Regional Studies (Birmingham)
3:63-76.

Lockwood,  David  1984.  "Sources  of  Variation  in  Working-Class  Images  of  Society,"  in

Giddens  and  Held  1984:359-72.

Pickvance  1976:119-145.

1977:141-56.

Lojkine,  Jean  1976.  "Contribution  to  a  Marxist  Theory  of  Capitalist  Urbanization,"  in

1977.  "Big  Firms'  Strategies,  Urban  Policy  and  Urban  Social  Movements,"  in  Harloe

Lubeck,  Paul M.  1981. "Class  Formation  at the Periphery:  Class Consciousness and  Islamic
Nationalism among Nigerian Workers," Research in the Sociology of  Work I. Greenwich:
JAI  Press:  37-70.

Marris,  Peter  1962. Family and Social Change in an African  City.  Evanston:

Northwestern  University  Press.

Massey, Doreen  1984. Spatial Divisions of Labour.  London: Macmillan. Mingione, Enzo 1981*.
Social Conflict  and  the City.  Oxford:  Blackwell  Montgomery,  David  1985.  "America's
Working  Man,"  Monthly  Review  $1, 6 (November):  1-8.

Morrison, Minion K.C, and Peter C.W. Gutkind, eds., 1982. Housing the Urban Poor in Africa.

Syracuse:  Maxwell School of  Citizenship and  Public  Affairs,  Syracuse  University.

Nelson, Joan  M.  1979. Access to Power; Politics and the Urban Poor in Developing  Nations.

Princeton:  Princeton  University  Press.

O'Connor,  Anthony  1983.  The African  City.  London:  Hutchinson
Pacione, Michael, ed. 1981. Problems and Planning in Third WorldCities. London: Croom Helm.
Parkin,  David  1969. Neighbours  and Nationals  in an African  City  Ward'. London:

Routledge  and  Kegan  Paul.

Peace,  Adrian  1975.  "The  Lagos Proletariat:  Labour  Aristocrats  or  Populist  Militants,"  in

Sandbrook  and  Cohen  1975: 281-302.

Peil,  Margaret  1981. "Workmates,  Kin  and  Friends:  The  Social  Contacts  of  West  African
Urban Workers," Research in the Sociology of  Work I. Greenwich: JAI Press} 71 -104.

with  Pius  O.  Sada  1981. African  Urban  Society.  New York:  Wiley.

CONTESTED  TERRAINS:  SOCIAL  RELATIONS  OF PRODUCTION  37

Perlman, Janice E.  1976. The Myth  of Marginality: Urban Poverty and Politics in Rio de Janeiro.

Berkeley:  University  of  California  Press.

Pickvance, C.G.  1976a. "Housing: Reproduction of Capital and  Reproduction of Labour  Power:

Some  Recent  French  Work,"  in  Walton  and  Masotti  1976:271-89.

pd.  1976b.  Urban Sociology:  Critical Essays.  New York:  St.  Martin's,  85-118.
fortes,  Alejandro  1985.  "Urbanization,  Migration  and  Models  of  Development  in  Latin

America,"  in  Walton  1985:  109-251.

and  John  Walton  1981. Labour,  Class, and  the  International  System.  New  York:

Academic  Press.

Prakash,  Brahm  1983. The  Urban  Dead-End?  Bombay:  Samaiya  Publications  for

ILO/International  Council  on  Social  Welfare.

Pratt,  Geraldine  1986. "Against  Reductionism: The  Relations of  Consumption  as a Mode of
Social  Structuration,"  International  Journal  of  Urban  and  Regional  Research  X,
3:377-400.

Remy, Dorothy  1985. "Economic Security and  Industrial Unionism: A Nigerian Case Study,"

in  Sandbrook  and  Cohen  1975:161-77.

Rex, John  1971. "The Concept  of  Housing Class and the Sociology of  Race Relations," Race

Rubery,  J.U.  1981. "Structured  Labour  Markets,  Worker  Organization,  and  Low  Pay,"  in

XII,  3:293-301.

Giddens  and  Held  1981:330-48.

Sandbrook,  Richard  1981. "Worker  Consciousness and  Populist  Protest  in Tropical  Africa,"
Research in  the  Sociology  of  Work  I.  Greenwich:  JAI  Press:  1 -36.
1982.  The  Politics  of  Basic Needs.  London:  Heinemann.
ind  Jack  Arn  1977.  The  Labouring  Poor  and  Urban Class Formation:  The Case of  Greater
4ccra.  Montreal:  McGill  University,  Centre  for  Developing-Area  Studies.
ind  Robin  Cohen,  eds.  1975.  The  Development  of  an African  Working  Class.  Toronto:
University  of  Toronto  Press.
Saunders,  Peter  1981. Social  Theory  and  the  Urban Question.  Holmes  and  Meier.
1984.  "Beyond  Housing Classes: The Sociological  Significance  of  Private Property  Rights in
Means of Consumption," International Journal of  Urban and Regional Research 8, 2: 202 — 27.
Scott, Allen J. and Michael Storper 1986. Production, Work, Territory. Boston: Allen and Unwin.
Sitas, Ari  1985. "Moral Formations and Struggles Amongst Migrant Workers on the East Rand,"
Labour,  Capital  and  Society  18, 2:372-401.
Smart, Alan  1986. "Invisible  Real  Estate: Investigations into the Squatter  Property  Market,"
International  Journal  of  Urban and  Regional  Research X,  1:29-45.
Smith,  Michael  Peter,  ed.  1984.  Cities in  Transformation:  Class,  Capital,  and  the  State.
Beverly  Hills:  Sage.
Stark,  David  1982.  "Class  Struggle  and  the Transformation  of  the  Labour  Process:
A  Relational  Approach,"  in  Giddens  and  Held  1982:310-29.
Stren,  Richard  1982.  "Housing  Policy  and  the  State  in  East  Africa,"  in  Morrison  and
putkind  1982:81 -  104.
Turner,  John  F.C.  1977. Housing  by People:  Towards Autonomy  in Building  Environments.
New York:  Pantheon.
Jyler,  Judith  1986.  "There's  More  Happening  in  Austin  Than  Meets  the  Eye,"  Frontline
(26 May)  1, 4.
van Onselen, Charles  1980. Chibaro: African  Mine Labour in Southern  Rhodesia  1900 -1933.
London:  Pluto  Press.
Walker,  Richard A.  1978. "The Transformation  of Urban Structure in the Nineteenth Century
and the Beginnings of  Suburbanization,"  in Kevin R. Cox,  ed.,  Urbanization and  Conflict  in

Market  Societies.  Chicago:  Maaroufa,  165-212.

Walton,  John,  ed.  1985.  Capital  and  Labour  in  the  Urbanized  World. Bervely  Hills: Sage,
and  Louis  H.  Masotti,  eds.  1976,  The City  in  Comparative  Perspective.  New  York:  Wiley.
White,  Luise  1983. "A  Colonial  State  and  an  African  Petty  Bourgeoisie:  Prostitution,
Property,  and  Class  Struggle  in  Nairobi,  1936-1940,"  in  Cooper  1983:  167-94.
Wright, Erik Olin and Joachim Singelmann  1982. "Proletarianization in the Changing American

38  K.P.  MOSELEY

Class Structure,"  American  Journal of  Sociology  88, Supplement:  S176-209.
Yancev  William  L., Eugene P. Ericksen and  Richard  N. Juliani  1976. "Emergent  Ethnicity:
A  Review  and  Reformulation,"  American  Sociological  Review  41, 3:391 - 4 0 3.
Zukin,  Sharon  1982. Loft  Living: Culture and  Capital  in  Urban  Change.  Baltimore:
JohnsHopkins-