See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/46541370

A Study of Space in Mumbai's Slums

Article  in  Tijdschrift voor Economische en Sociale Geografie · February 2010

DOI: 10.1111/j.1467-9663.2009.00576.x · Source: RePEc

CITATIONS READS

62 2,144

1 author:

Jan Nijman
Georgia State University / University of Amsterdam
63 PUBLICATIONS   1,608 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Jan Nijman on 07 April 2021.

The user has requested enhancement of the downloaded file.

A STUDY OF SPACE IN MUMBAI’S SLUMS tesg_576 1..14

JAN NIJMAN
Department of Geography & Regional Studies, Urban Studies Program, University of Miami, USA.
E-mail: Nijman@miami.edu

Received: March 2009; revised June 2009

ABSTRACT
The urban slum in the less developed world has an overwhelming significance of place for its
dwellers: it determines who they are, what they do, where they go, and whom they know. Unlike
most Western cities where the different realms of life (residential, work, religious, public, etc.) are
spatially segregated, here they are all functionally and spatially integrated. A close examination of
slum spaces in Dharavi, Mumbai, reveals such overlapping spatial patterns and raises some
fundamental questions. Is there a proper definition of the slum? How should we conceive of the
slum community and its spatial features? How useful or problematic are Western concepts of
residential segregation, ghettos and enclaves? It is argued that the historical persistence of urban
slums points to their indispensability, with the tacit (if inconsistent) approval of the state. Slums
not only provide shelter to a large urban labour force but also a milieu that is conducive to intense
social organisation and economic production.

Key words: Slum, space, urban, India, Mumbai, Dharavi

URBAN SLUMS AND THE CONGRUENCE
OF ECONOMIC AND SOCIAL
GEOGRAPHY
A close examination of the nature of space
in urban slums reveals the congruence of econ-
omic and social geography. Generally speak-
ing, the urban slum has an overwhelming sig-
nificance of place for its dwellers: it determines
who they are, what they do, where they go,
and whom they know. Unlike most West-
ern cities where the different realms of life
(residential, work, religious, public, etc.) are
spatially segregated, here they all seem func-
tionally and spatially integrated. To be sure,
this is the case in Dharavi, a large contiguous
slum area in the centre of Mumbai, India.
It is the geographical perspective, therefore,
that has particular potential in understanding
how slums work, whence they come, and why
they persist. And, I will argue, it is this spatial
perspective that can guide us to a better theo-
retical understanding of urbanity in the global
‘South’. Western concepts of the modern city
and of modern urban spaces have been difficult
to reconcile with empirical observations of the
South Asian city, in particular. As Rao (2006,
p. 225) puts it: ‘As a social and cultural form,
the modern South Asian city has been a site of
theoretical anxiety and ambiguity’.
Nonetheless, geographical studies of urban
slums are quite sparse.1 In 1985, Gilbert and
Ward (1985) lamented that ‘no writer has
traced the geography of low-income settle-
ments in any third world city over the whole
postwar period’. And in a review essay earlier
this year, Grant (2009, p. 218) noted that the
Annals of the AAG ‘published only two articles
containing the words slums, squatters, or infor-
mal settlements in their title since 1976’.
One of the reasons for this scarcity lies in
the unfortunate balkanisation of the discipline.
Our understanding of slums is hampered by
the separations between economic geography
and cultural geography on the one hand, and
between urban geography and development

Tijdschrift voor Economische en Sociale Geografie – 2009, Vol. ••, No. ••, pp. ••–••.
© 2009 by the Royal Dutch Geographical Society KNAG
Published by Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA
2 JAN NIJMAN
geography on the other. ‘Hampered’ may be
an understatement since the separation of sub-
disciplines has basically inhibited geographers
from studying slums altogether: urban geogra-
phers ‘don’t do’ slums because they pertain to
questions of development while development
geographers ‘don’t do’ slums because they
have traditionally maintained a rural focus.2
There is no question that the topic of slums
has recently come to the forefront in the news
media, popular books and academic writings
(e.g. Davis 2004, 2006; Mitlin & Satterthwaite
2004; Neuwirth 2004). The path-breaking UN
Habitat report of 2003 was one of numerous
publications in the past few years that alert us to
the gravity of the challenge. With rapid and
continued urbanisation in the less-developed
world and in the absence, thus far, of effective
policies, it is likely that precarious living condi-
tions in cities will become the main challenge
to human development in future decades.
Slums represent a spatiality that is often at
once cultural and economic. Certainly, this is
so in a number of areas in Mumbai, India, and
Dharavi is a particularly striking example.
Mumbai is a city with many different faces that
is, in fact, hard to know, not least because of
the enormous diversity of its populations and
the extremely wide gap between rich and poor
(Grant & Nijman 2004), It is useful, therefore,
to start with a brief overview of the historical
geography of Mumbai’s slums.
MUMBAI’S SLUMS AND THE BIRTH
OF DHARAVI
Greater Mumbai comprises the peninsula
bound by the Arabian Sea to the West, Thane
Creek to the East, and Vasai Creek and Ulhas
River to the north. It is connected to the main-
land in the north and northwest. The area’s
dimensions are about fifty kilometres from
North to South and an average of ten kilo-
metres from West to East. The current popu-
lation is around twelve million people and
the average population density is 24,000
people per square kilometre. The city grew
spectacularly in the decades following Inde-
pendence and its geography became increas-
ingly dense. Recent decades witnessed a shift
of population from the South to the Northern
suburbs.
© 2009 by the Royal Dutch Geographical Society KNAG
The geographical constraints of the Island
City put a premium on space and have histori-
cally influenced land values and land use in
the city. There has been a steep gradient
in land values from the South to the North.
In the mid-1990s, the unprecedented influx
of foreign corporations contributed to an
extreme escalation of land values, making
Mumbai for some time the most expensive city
in the world (Nijman 2000).
The city’s housing problems have been on
record for over 150 years, ever since it turned
into an urban centre of significance. At the
middle of the nineteenth century, after a
couple of decades of rapid growth, the city’s
population reached half a million. The urban
area then was, of course, much smaller than it
is now and covered only the southern extension
of the peninsula. Most parts of Mumbai had a
population density thirteen times higher than
London at the time. The city’s waxing popula-
tion led to an acute shortage of housing and
serious problems with the provision of water,
sanitation and drainage.
There was a big difference between Euro-
pean and ‘native’ residential quarters with the
slums heavily concentrated in the latter. The
Native Town grew virtually unplanned and
without any consideration of the quality of
life of its inhabitants. Despite the fact that
the housing problems in the Indian quarter
were common knowledge, no substantial
measures were taken by colonial govern-
ments and ‘nothing of consequence was ever
achieved’ (Dossal 1991, pp. 196–197), The
1872 Census reported that: ‘The houses of
Bombay [Mumbai] are far too few in number
to afford proper accommodation for its inhab-
itants’ (Sundaram 1989, p. 56) and this became
a mantra to be repeated in every Census to
follow to this day.3
Until about 1900, Mumbai did not extend
beyond Mahim Creek and was confined to what
is known as the ‘Island City’ (the southern part
of the peninsula), Dharavi was then referred to
as one of the Koliwadas, an area belonging to
the tribal Koli fishermen who are Mumbai’s
original inhabitants. In 1864 the estimated
population in the area was 992 (Dossal 1991,
p. 197).
Around the turn of the twentieth cen-
tury, unsanitary conditions plagued the city,
A STUDY OF SPACE IN MUMBAI’S SLUMS 3

Figure 1. Overview of Greater Mumbai.

© 2009 by the Royal Dutch Geographical Society KNAG

4 JAN NIJMAN

Source : Perry-Castaneda Library Map Collection of the University of Texas at Austin.

Figure 2. The geographic extent of Bombay in the early twentieth century.

© 2009 by the Royal Dutch Geographical Society KNAG

A STUDY OF SPACE IN MUMBAI’S SLUMS
especially the so-called Native Town, the segre-
gated section where Indians lived in very high
population densities. It was from there, from
Mandvi to be precise, in 1869, that the bubonic
plague spread outwards to the rest of the city
and then across most of the subcontinent. The
epidemic killed nearly 200,000 in Mumbai and
eight million in all of India. It was decided that
the gradual expansion of polluting industries
in Native Town had to be halted and moved
to a distant tannery town in the north (Dossal
1991, p. 202). That town was Dharavi.
Dharavi was born as a result of expulsion of
noxious activities from the city to the periphery.
The first tannery arrived in 1887 and by the
end of 1890 a number of Muslim tanners from
Tamil had settled in the area as well. While
there was considerable resistance by many
firms (including foreign multinationals who for
some time continued to operate in the south)
eventually the result was a concentration of
these polluting activities in what was then the
urban fringe.
Other early settlers included the Kumbars,
a large Gujarati community of potters (another
heavily polluting industry), who were given a
99-year land-lease from the colonial government
in 1895. In the following years they gradually left
south Mumbai to settle in the southwest corner
of Dharavi, where they are still today. From the
beginning, then, Dharavi was a place that com-
bined living and working. The first mosque, Badi
Masjid, dates from 1887 and the oldest Hindu
temple, Ganesh Mandir, was built in 1913.
At Independence, on the eve of rapid
increases in urbanisation across India, most of
Dharavi was built up but it also still counted
various empty spaces and it was a favourite
dumping ground for companies operating
across the city. As rural-urban migration esca-
lated and as Dharavi continued to be a major
destination for overflow migration from the
Island City, the area very quickly became
extremely dense during the 1950s and after.
By the 1970s, as Dharavi had become increas-
ingly central to the growing metropolis, more
and more industries considered unfit for a
central urban location were driven further out
of the city. The abattoir (in Bandra) was folded
in 1980 and the large tanneries were ordered to
leave the area – today, only the smaller ones are
left. In recent years, the manufacturing of
5
finished leather products took over from first
stage processing. Some of the areas vacated by
the big tanneries became the sites of (residen-
tial) redevelopment projects, mostly along the
south-eastern edge of Dharavi.
Despite a range of successive policies (since
Independence) to ‘fight’ the slums, they have
steadily grown and now house at least half of
the total population of Greater Mumbai (Das
2003, p. 210). Slums are scattered across the
city and have very high densities – yet they
occupy only about 12 per cent of the land. The
most recent Slum Redevelopment Scheme
dates from 1995 and is closely related to the
liberalisation of the urban land market and
rapidly increasing real estate values. But while
private developers have brought large numbers
of new expensive homes on the market that are
out of reach of most Mumbaikars, slum rede-
velopment has stalled (Nijman 2006, 2008).
Due to the city’s expansion to the north and
northeast, Dharavi now finds itself right in the
centre of Greater Mumbai, strategically located
with connections to all three of the city’s com-
muter rail corridors (Western, Central and
Harbour lines). It has fast connections to
the main business districts in the south; it is
adjacent to the new Bandra-Kurla commer-
cial complex (across Mahim Creek); it is very
close to the new industrial/back-office centre
around Andhere-East, and it is two miles from
the airport.
INSIDE DHARAVI
Dharavi occupies about 432 acres and has an
astounding population density of more than
1,200/acre. The area is clearly bounded by the
railroads to the southwest and southeast and by
Mahim Creek to the north. There is some rede-
velopment (high rise) along the edges, espe-
cially in the south, but most of it is untouched
by direct government intervention, especially
the inner parts of the slum.
Dharavi houses an estimated 67 slum com-
munities, though much of this depends on
definitions (Graber et al. 2005). About one-
third were (descendents of) immigrants from
Tamil Nadu (mostly Muslims from Tirunelvelli
district), another third were Maharashtrians,
and the rest came (originally) from Gujarat,
Karnataka, Kerala and Uttar Pradesh. The
© 2009 by the Royal Dutch Geographical Society KNAG
6 JAN NIJMAN

Source : Nijman (2008).

Figure 3. Mumbai’s slums.

© 2009 by the Royal Dutch Geographical Society KNAG

A STUDY OF SPACE IN MUMBAI’S SLUMS
Source : photograph by Zach Woodward.
Figure 4. Dharavi: a view from above.
survey also counted 27 temples, 11 mosques,
6 churches and 1,044 manufacturing units
(Sharma 2000, p. 80). As far as population
and manufacturing is concerned, estimates
vary notably. A recent count arrived at 550,000
people and 1,230 manufacturing units (MM
Consultants 2006). It is estimated that about 70
per cent of Dharavi’s residents work there
(Graber et al. 2005, p. 16).
The overwhelming majority of Dharavi resi-
dents are Dalits (formerly known as Untouch-
ables) who combine material poverty with
social stigma as soon as they move outside of
their circles. They reside in tight community
clusters within the slum generally based on
regional origin and professional status – living
anywhere else is virtually unthinkable. The
result is a social and cultural residential mosaic
in which people are very much identified in
terms of where they belong. Venturing out-
side of that designated territory can be accom-
panied with apprehension, stress and feelings
of insecurity (Pendse 1995).
The internal spatial patterns of Dharavi are
mainly a function of immigration dynamics
(cultural origins), industrial/commercial clus-
7
tering and external access. The main arteries
cut through the area and provide access to
buses, trucks and lots of other traffic. These are
the arteries that keep Dharavi connected with
the rest of the city. These streets are quite wide
(60 ft and 90 ft, supposedly, but not really) and
they are lined with retailing, food stands,
kiosks, taxis, small restaurants, some hotels,
etc. Lots of people are moving through these
streets and many are actually not from Dharavi.
Indeed, a visitor might not realise he is actually
inside a slum area.
Off the main streets, all that changes, espe-
cially if one enters into the core areas that are
(part) residential: the streets are narrower and
do not allow heavy traffic. There is shade, less
exhaust fumes, usually less dirt, less noise, and
the environs are much more intimate. It is like
stepping inside, and in many ways it a relief
from the urban overdrive that characterised the
main roads. But social control is immediately
apparent, behaviours are routinised and codi-
fied, and outsiders are immediately noticed.
Moving into the core of Dharavi, it gets harder
to distinguish exterior from interior, public
from private, and accessible from inaccessible.
Getting off the main arteries one may also be
entering a part of the slum that is more exclu-
sively industrial, especially if it is highly pollut-
ing industry. These areas are almost all pushed
to the edges of Dharavi and to enter them
from the main arteries one would be moving
away from the core. For example, the southwest
corner of Dharavi has a major cluster of plastic
recycling factories, with some estimates of well
over 500 units. Other major clusters include
small tanneries further to the north and the
potteries in the southeast corner of Dharavi.
The main streets are wide enough to allow for
trucks bringing and getting materials, but often
they are not paved. These areas are intimate in
the sense that nobody is passing through unless
they have business there. These areas are also
extremely dirty, reeking, and one has to be
careful where to step and to avoid collisions
with vehicles and human carriers.
Most of Dharavi combines a whole range
of functions of living, retailing, wholesaling,
manufacturing, consumer services, producer
services, public functions like schools, houses
of worship, civic organisations and so on.
Economic activities are often inseparable from
© 2009 by the Royal Dutch Geographical Society KNAG
8 JAN NIJMAN
ethnic identities and from the highly localised
cultural milieux of the slum. The tanners
are Tamils and often Muslims, the embroi-
dery workers hail from UP, the potters from
Gujarat.4 The smaller firms in a particular
industry are often subcontractors to the bigger
firms in that industry so there is a powerful
overlap of ethnic ties, kinship ties and econ-
omic interdependence. Civil society operates at
a fine scale in Dharavi.
Seen from the sky above, it may be hard to see
any clear spatial organisation, apart from the
main arteries, but on the ground the lines are
very clear and very well known. The use of space
here is more deliberate and intense than in any
other urban environment imaginable. Every bit
of space is allocated and this is known exactly by
the locals: they know who belong where, what
belongs to whom, what is private and who has
the rights to it, and what is public (and for
which public). There is a great deal of tolerance
in terms of human density and movement, but
at the same time a powerful realisation that
territorial control is fundamental to long-term
survival and identity.
In Dharavi, the use value of land and space is
extremely high while exchange values are often
hard to determine because much of the occu-
pation is extra-legal. But the potential market
values are tremendous if Dharavi were re-
developed and became part of the urban land
market. It explains the keen interest of many
developers. This matter will be discussed later.
First, we must scrutinise and discuss the notion
of slum itself.
THE CONCEPT OF THE SLUM
What is a slum? Even if we concentrate ex-
clusively on the quality of housing, it is by no
means easy to identify slum dwellings or slum
areas on the ground. UN Habitat’s (2003, p.
12) official operational definition of a slum is:
an area that combines, to varying extents,
the following characteristics (restricted to
the physical and legal characteristics of the
settlement): inadequate access to safe water,
inadequate access to sanitation and other
infrastructures; poor structural quality of
housing, over-crowding and insecure resi-
dential status.
© 2009 by the Royal Dutch Geographical Society KNAG
It is a broad ranging definition yet the focus
is strictly on housing and living conditions –
namely, it does not consider non-residential
functions of the slum. This is the common
view of slums. Similarly, the Government of
India (1988, p. 5) defines slums as ‘housing
that is unfit for human habitation or detri-
mental to safety, health and morals of the
inhabitants’.
Slums also vary in degrees of legality (formal
vs informal settlements), size, area, configura-
tion and perceptions of boundaries. This is
not just academic: where government does not
recognise the legality of slums, there are no
government services (water, rubbish collec-
tion, etc.) and so slum dwellers must organise
themselves; where government does recognise
slums, they may qualify for redevelopment
schemes. On the other hand, the designation of
‘slum’ can foreshadow eradication.
Appadurai suggested a full spectrum of
‘housing conditions’ from the multi-million
dollar apartments on Malabar Hill down to the
homeless: ‘Public sleeping is a technique of
necessity for those who can be at home only
in their bodies’. (Appadurai 2000, p. 638). In
between are, for example, the chawls (single-
room dwellings originally built for the textile
workers, the more permanent slums such as
Dharavi, and the highly transient make-shift
‘homes’ along main streets, rail roads and
deserted back alleys known as jopad-pattis.
Should a slum and its territorial extent be
defined on the basis of community? To outsid-
ers, slums tend to appear as more or less con-
tiguous areas of decrepit housing, without
much consideration for possible internal differ-
ences. But to those inside the slums, territo-
riality is often hugely important in terms of
belonging, identity, safety, community, status
and political organisation (Pendse 1995; Vora
& Palshikar 2003). But the definition of such
communities, at sometimes a very fine scale, is a
momentous task and can be highly contested.
When Davis (2004, p. 14) postulates that ‘the
five great metropolises of South Asia alone
contain about 15,000 distinct slum communi-
ties’ he must resort to the crudest of guessing
games, or repeat the guesses of others. The fact
is that nobody knows and that most estimates
are based on a jumble of conceptions, defini-
tions and counts. Comprehensive and accurate
A STUDY OF SPACE IN MUMBAI’S SLUMS
maps of many slum areas and slum communi-
ties often do not exist.
What is and what is not considered a slum in
local parlance is by definition contextual and
can be deeply political.5 Most Mumbaikars (and
they rarely if ever visit Dharavi) routinely refer to
all of Dharavi as a slum. Some who reside there
actually object to the term, keenly aware as they
are of the political ramifications: the Kumbars
refuse to accept the term slum for their neigh-
bourhoods and they are strongly opposed to
any notion of redevelopment of the area.
When trying to understand the spatiality of
urban slums such as Dharavi, we find that exist-
ing conceptual frameworks, derived from urban
studies in the West, are not much of a help – but
their consideration is useful for our purpose as
it suggests the need for a new way of ‘seeing’
slums and urbanity altogether. There are two
matters, in particular, that need some discus-
sion. The first concerns the relevance or irrel-
evance of the notions of ‘segregation’, ‘ghetto’
and ‘enclave’. The second pertains to the perva-
sive separation, in western urban contexts, of
residential and economic urban functions.
In his discussion of segregation in American
cities, Marcuse (2001, pp. 3–4) uses the follow-
ing definitions of ghettos, enclaves, and exclu-
sionary enclaves:
A ghetto is an area of spatial concentration
used by forces within the dominant society
to separate and to limit a particular popula-
tion group, externally defined as racial or
ethnic or foreign, held to be, and treated as,
inferior by the dominant society.
An enclave is an area of spatial concen-
tration in which members of a particular
population group, self-defined by ethnicity
or religion or otherwise, congregate as a
means of protecting and enhancing their
economic, social, political and/or cultural
development.
An exclusionary enclave is an area of spatial
concentration in which members of a par-
ticular population group, defined by its posi-
tion of superiority in power, wealth, or status
in relation to its neighbors, cluster as a
means of protecting that position.
Marcuse’s aim is really to distinguish between
voluntary and involuntary segregation and to
determine whether ‘segregation that is socially
9
acceptable may be differentiated from that
which is undesirable’. In short, voluntary segre-
gation that does not exclude ‘others’ is accept-
able and of a different nature than forced
segregation (which typically involves people
deemed ‘inferior’),
Even in the US context, these distinctions
are not always straightforward.6 In Mumbai’s
slums, the entire conceptual scheme is out of
place. First, while arguably slums at the level of
communities are formed voluntarily because
they provide comfort, security and econ-
omic networks (enclaves), more generally
slums emerge because the poor are for all
practical purposes excluded from the formal
housing market. In other words, they are often
compelled to erect slums (ghettos),
Second, if slum communities are best viewed
as enclaves, then they are very often exclusion-
ary, namely, they do not normally include
people of other religions, other castes or other
regions. But this is hardly a function of their
‘superiority’ – it is a necessary and basic survival
strategy in what is otherwise a hostile urban
jungle.
Third, segregation processes in the West
tend to be about the desirability or undesira-
bility of congregation of groups with different
identities. In India’s slums, on the other hand,
the formation of spatial communities is a func-
tion of space itself: those 67 communities or so
inside Dharavi have formed not because people
decided they did not want to live among others
but because it is the only way to secure and
maintain a space! This is illustrated with the
slum version of the ‘gated community:’ smaller
communities within large, dense, slum areas
are sometimes gated in the sense that their
territory is clearly marked and that traffic
is controlled with one or more designated
entrances. This reflects intense competition for
space as well as high ethnic concentration and
segregation. Social control within these com-
munities tends to be very strong. Clearly, these
are quite different from the affluent gated com-
munities in American cities – but they are gated
nonetheless.
The second problem with Western concepts
of urban studies concerns the spatial separa-
tion, in the West, of different urban functions.
Marcuse makes the point that segregation of
urban functions in American cities follows
© 2009 by the Royal Dutch Geographical Society KNAG
10
a logic that is either cultural, economic, or
(power) political: ‘they fall into three quite
separate and distinguishable . . . divisions by
culture, by functional economic role, and by
position in the hierarchy of power’. This
perspective does not apply in Mumbai. First,
India’s caste system is far from dissolved in the
lower echelons of society, even in urban areas.
In as far as it is based on jati, positions in the
caste hierarchy are closely associated with
professional status. For example, the Kumbars
(potters), are a ‘caste’, a neatly defined cultural
ethnic group, and an industry all at once. So are
the Charmakars (leather makers), the Tamil
Muslims (tanners), or the Kolis (fishermen),
Hence, we see a complex blending of culture,
economics and hierarchical position. Mar-
cuse’s argument that economic functional
segregation is ‘essentially independent of cul-
tural differences’ simply does not apply in this
context.
Without an economic function, the slum
would lose an essential part of its rationale. This
is why a slum is emphatically not an urban
ghetto in the modern American sense.7 In
the US, urban ghettos are typically inner-city
neighbourhoods where as a result of economic
restructuring employment opportunities have
disappeared and where, precisely because of
the lack of jobs and economic activity, decay
prevails. Dharavi, in contrast, is buzzing with
economic activity. Seventy per cent of its resi-
dents are said to be working inside Dharavi
(Graber et al. 2005, p. 31), and entrepre-
neurship is all around. Dharavi is better char-
acterised as a densely packed working class
residential area mixed with (light) manufactur-
ing, retailing and a whole range of other func-
tions. In 2002, Dharavi’s ‘GDP’ was estimated
at 1500–2000 crores or about US$ 360
million (Sharma 2000, p. 79). An American-
style ghetto, it surely is not.
This also means that the urban slum is more
than a ‘reserve army of labor’ (Breman 1996;
2006) or ‘a stealth workforce for the formal
economy’ (Davis 2006), Certainly, slums do
house such labour but it is entrepreneur-
ship and production in the slums themselves
that have been systematically overlooked. Many
slum dwellers are employed in firms that
operate from within the slum, whether their
work takes place from home or on the company
© 2009 by the Royal Dutch Geographical Society KNAG
JAN NIJMAN
work floor. Home-based entrepreneurship is
an essential element of the informal economy
and confirms the importance of housing (in
the slum) as a productive asset and its role
in promoting economic activity (Gordon &
Nell 2006).
The significance of social networks seems
hard to overstate. This dimension of the local
cultural milieu can be thought of in terms of
social capital. In Bourdieu’s words, ‘social
capital is formed in the context of a “durable
network” and provides each of its members
with the backing of the collectivity-owned
capital, a “credential” that entitles them to
credit in the various senses of the word’ (Bour-
dieu 1985, p. 55), The vital importance of this
social capital is not lost on slum entrepreneurs
and recent immigrants: one of the reasons
for the steady population growth of Dharavi in
past decades is that workers were consistently
recruited from the same ethnic/religious/
geographic/caste origins, as these communi-
ties provide themselves with a self-created
‘protective coat’ (Graber et al. 2005, p. 34),
THE UNINTENDED CITY
Generally, cities in the less-developed world
have a long history of spatial fragmentation in
terms of planned and unplanned areas, illegal
settlement, slum developments, etc. (Harris
1978; Balbo 1993; Balbo & Navez-Bouchanine
1995). In the case of Mumbai, from the late
nineteenth to the early twenty-first century the
city’s economy changed profoundly but slums
remained a constant feature of the landscape.
In Mehrotra’s words, it:
was never conceived or built in a singular
image. Instead, the city’s evolution con-
sistently makes evident a series of duali-
ties [such as] lifestyles, cultural attitudes,
planned intervention versus kinetic or incre-
mental growth, big moves versus small ges-
tures, passive versus active interventions,
governmental action versus private initia-
tive, the pukka versus the kutcha city, etc.
(Mehrotra 1991, p. 12)
Today, Mumbai is still two cities, one deemed
modern and desirable, and the other the city
of the slums, unintended and undesirable.
The undesirable city in colonial times was
A STUDY OF SPACE IN MUMBAI’S SLUMS
Native Town, today it is Dharavi. It is a mistake,
however, to think that they are truly separate.
In fact, the two cities are economically con-
nected and in some ways they exist by virtue
of each other, as in a symbiotic relationship.
The growth of slums was undesirable
because it crossed ideals of what the ‘urban’
was supposed to represent. Slums in urban
India have always exhibited a hybrid social
structure both urban and rural, allowing the
slum dwellers to be part of the city, in terms
of work, while continuing social and cultural
affiliations with the village (Sen 1975). Indeed,
the maintenance of ancestral traditions, family
ties, and cultural identities, including caste,
was a condition for successful adaption in the
city. So, for example, in the urban slums the
extended family did not break down as mod-
ernisation theories would have it but they
rather flourished.
It is hard to imagine this persistence of slums
without effective facilitation by the state. It has,
after all, the power in principle to eradicate, to
legalise, to prohibit, to build roads and provide
services, or to do nothing at all. ‘Dharavi
illustrates how the state in fact endorses and
encourages illegality with one hand, while
trying to curb it with the other’ (Sharma 2000,
p. xix). For instance, the state funded the build-
ing of major access roads and internal arteries
that are essential to its economic functioning,
but it does not provide sufficient sanitation
(McFarlane 2008); it rules that certain noxious
activities are illegal but it does not enforce
their removal; and, most fundamentally, for all
practical purposes, the state has for 150 years
neither eradicated the slums nor redeveloped
them – it has, in effect, let them be.
At the same time, the urban elites and policy-
makers never accepted the slum as part of
the formal urban development process.8 The
imaginary of the modern city is closely related
to our views of London, Paris or New York, with
the emphasis on their global grandeur and
achievements while downplaying their own
local problems.9 In the mid-1990s, the Bom-
bay Chamber of Commerce (2003) initiated
a development programme with the title
‘Bombay First’ and it was clearly modelled after
the earlier ‘London First’. Presently, most of
such rhetoric has shifted to successful Asian
‘world cities’ such as Shanghai and Singapore.
11
The Bombay First programme and its suc-
cessors included vigorous efforts to maintain
and improve the Western appearance of the
southern part of the city with public-private
programmes involving road and pavement
maintenance, the removal of hawkers, ‘anti-
spitting drives’, and general clean up. In the
process, this part of the city witnessed the kind
of sterilisation and regulation that was long
typical of cities in the West, what Edensor
(1998, p. 213) calls ‘the erasure of much social,
sensual and rhythmic diversity in urban space’.
It only increased the contrasts with the world
of the bazaars and the slums (Grant & Nijman
2002; Nijman 2007a),
Prakash (2008, p. 2) argues that we should
question ‘the idea of the European metropolis,
defined as a bounded unit by modernist theory,
as the paradigmatic modern city’. We need to
rethink the history of urban modernity and
urban change and the first step, says Prakash,
involves:
expanding the focus beyond Europe and
North America to include the experiences of
urban modernity in Asia, Africa, and Latin
America. It entails approaching the histor-
ical experiences of modern urban forms and
transformations as ineluctably global, spe-
cific, diverse, and divergent.
The ‘promise of the modern city’, then, seems a
false promise. The city of the slums may not be
intended, but it is likely to be indispensable.
Acknowledgement
This paper is in part based on research funded by
the US National Science Foundation (BCS-0721025),
a collaborative project (with Richard Grant) on
the economic geography of slums in Mumbai and
Johannesburg. This paper is also based on the text of
a keynote address in a special session commemorat-
ing the 100th anniversary of TESG at the Annual
Meeting of the Association of American Geographers
in March 2009.
Notes
1. See Gilbert & Ward (1985); Eyre (1990); Oldfield
(2002); Olds et al. (2002); Nijman (2008). Much
has been written, especially since the latter part of
the 1960s, outside the discipline of geography, in
the fields of anthropology, developments studies,
© 2009 by the Royal Dutch Geographical Society KNAG
12
and urban planning (e.g., Abrams 1964; Clinard
1966; Mangin 1967; Turner 1972, 1991)
2. The so-called ‘cultural turn’ in economic
geography reflects recognition of this artificial
separation of social and economic realms. Theo-
retically, at least, there is a growing consensus
that the ‘stubborn locational rootedness’ of pro-
duction owes much socio-cultural context (Scott
2000, p. 31). Thrift referred to the cultural turn
as ‘the most important event to impact on the
sub-discipline in the last ten or fifteen years’. So
important, he went on, that ‘economic geogra-
phers have become some of the leading expo-
nents of cultural geography’ (Thrift 2000, pp.
689, 692). Still, this theoretical awareness has
not translated in more integrated empirical
approaches.
3. Useful appraisals of government policies regard-
ing slum rehabilitation can be found in Ruiter
(1999) and Mukhija (2003).
4. The communities in Dharavi have shown a great
deal of stability over time. There is some turnover
as young and better educated (second or third
generation immigrants) leave for higher educa-
tion or jobs elsewhere. This is the reason that, over
time, there had to be a steady replacement flow
of workers from the region of origin.
5. Gilbert (2007, p. 697) speaks of ‘the return of the
word “slum” with all of its inglorious associations’
and he warns against its colloquial usages and
political abuses.
6. Marcuse acknowledges as much. I am using his
concepts here not so much to critique the original
argument but rather to show that some key
notions in the Anglo-Saxon literature on urban
studies do not apply to the context of India’s
urban slums.
7. Some have argued that Dharavi, like Mumbai
at large, has lost some its cosmopolitan flair after
the riots following Babri Masjid in 1992 led to
more segregation: identity used to be mainly
region-based but now religion became more
important, setting in motion ‘a process of ghettoi-
sation.’ (Sharma 2000, p. xxxi).
8. See D’Monte (2002, p. 213) for a description of
the wide disparities of perceptions between the
elites and slum dwellers of Mumbai and its desired
future. The assertion of a new middle class with
strong negative views of slum dwellers (Sharma
2004) has contributed to Mumbai’s becoming a
‘revanchist city’ (Smith 1996), The prevalence of
this modern urban imaginary is also reflected in
© 2009 by the Royal Dutch Geographical Society KNAG
JAN NIJMAN
the title of a well-known volume: Bombay: Metaphor
for Modern India (Patel & Thorner 1996).
9. One of the key features of the revival in ‘com-
parative urbanism’ in the field of urban studies
is the avoidance of normative (modernist) com-
parative schemes (Nijman 2007b; Ward 2008;
Smith 2009).
REFERENCES
Abrams, C. (1964), Man’s struggle for shelter in an
urbanising world. Cambridge MA: MIT Press.
Appadurai, A. (2000), Spectral housing and urban
cleansing: Notes on millennial Mumbai. Public
Culture 12, pp. 627–651.
Balbo M. (1993), Urban planning and the frag-
mented city of developing countries. Third World
Planning Review 15, pp. 23–34.
Balbo, M. & F. Navez-Bouchanine (1995), Urban
fragmentation as a research hypothesis: Rabat-sale
case study. Habitat International 19, pp. 571–582.
Bombay Chamber of Commerce (2003), Bombay
First: Transforming Mumbai into a world-class city.
Mumbai: McKinsey & Co, Inc.
Bourdieu, P. (1985), The forms of capital. In: J.G.
Richardson, ed., Handbook of theory and research
for the sociology of education, pp. 241–260. Westport,
CT: Greenwood Press.
Breman, J. (1996), Footloose labour. Working in India’s
informal economy. New York: Cambridge University
Press.
Breman, J. (2006), Slumlands. New Left Review 40,
pp. 141–148.
Clinard, M.B. (1966), Slums and community develop-
ment: Experiments in self-help. New York: The Free
Press.
Das, P.K. (2003), Slums: The continuing struggle for
housing. In: S. Patel & J. Masselos, eds., Bombay
and Mumbai: The city in transition, pp. 207–234. New
Delhi: Oxford University Press,.
Davis, M. (2004), Planet of slums: Urban involution
and the informal proletariat. New Left Review 26,
pp. 5–34.
Davis, M. (2006), Planet of Slums. New York: Verso.
Dossal, M. (1991), Imperial designs and Indian reali-
ties. The planning of Bombay City 1845–1875. Delhi:
Oxford University Press.
D’Monte, D. (2002), Ripping the fabric: The decline of
Mumbai and its mills. New Delhi: Oxford University
Press.
Edensor, T. (1998), The culture of the Indian
street. In: N.R. Fyfe, ed., Images of the street: Plan-
A STUDY OF SPACE IN MUMBAI’S SLUMS
ning, identity and control in public space, pp. 205–221.
London: Routledge.
Eyre, L. (1990), The shanty towns of central
Bombay. Urban Geography 11, pp. 130–152.
Gilbert, A. (2007), The return of the slum: Does
language matter? IJURR 31, pp. 697–713.
Gilbert, A. & P. Ward (1985), Housing, the state and
the poor: Policy and practice in three Latin American
cities. New York: Cambridge University Press.
Gordon, R. & M. Nell (2006), Home-based entre-
preneurs: Research findings and recommenda-
tions. Available at <www.finmarktrust.co.za>.
Government of India (1988), Report of the National
Commission on Urbanisation, vol. II. New Delhi:
Ministry of Urban Development.
Graber, D., A. Kirschner, S. Mill, M. Schach, S.
Schmekel & H. Seligman (2005), Living and
working in slums of Mumbai. Working paper 36.
Magdeburg: Otto-von-Guericke Universitat.
Grant, R. (2009), A review of Planet of Slums. Annals
of the Association of American Geographers 99, pp.
216–219.
Grant, R. & J. Nijman (2002), Globalisation and
the corporate geography of cities in the less devel-
oped world. Annals of the Association of American
Geographers 92, pp. 320–340.
Grant, R. & J. Nijman (2004), The rescaling
of uneven development in Ghana and India.
Tijdschrift voor Economische en Sociale Geografie 95,
pp. 461–481.
Harris, N. (1978), Economic development, cities and
planning: The case of Bombay. Bombay: Oxford
University Press.
MM Consultants (2006), Dharavi Redevelopment
Project. Powerpoint presentation, Mumbai 2006.
Mangin, W. (1967), Latin American squatter settle-
ments: A problem and a solution. Latin American
Research Review 2, pp. 65–98.
Marcuse, P. (2001), Enclaves yes, ghettoes, no:
Segregation and the state. Lincoln Institute of
Land Policy Conference Paper.
McFarlane, C. (2008), Sanitation in Mumbai’s
informal settlements: State, slum, and infrastruc-
ture. Environment and Planning A 40, pp. 88–107.
Mehrotra, R. (1991), Bombay City: One space,
two worlds. Architecture and Design, November/
December, pp. 12–19.
Mitlin, D. & D. Satterthwaite (2004), Empowering
squatter citizens: Local government, civil society and
poverty reduction. London: Earthscan.
Mukhija, V. (2003), Squatters as developers? Slum
redevelopment in Mumbai. London: Ashgate.
13
Neuwirth, R. (2004), Shadow cities: A billion squatters.
A new urban world. New York: Routledge.
Nijman, J. (2000), Mumbai’s real estate market in the
1990s: Deregulation, global money, and casino
capitalism. Economic and Political Weekly 35/7,
pp. 575–582.
Nijman, J. (2006), Mumbai’s mysterious middle
class. IJURR 30, pp. 758–775.
Nijman, J. (2007a), Mumbai since liberalisation:
The space-economy of India’s gateway city. In: A.
Shaw, ed., Indian cities in transition, pp. 238–259.
New Delhi: Orient Longman.
Nijman, J. (2007b), Comparative urbanism. Urban
Geography 28, pp. 1–6.
Nijman, J. (2008), Against the odds: Slum rehabi-
litation in neoliberal Mumbai. Cities 25, pp. 75–
87.
Oldfield, S. (2002), Partial formalisation and its
implications for community governance in an
informal settlement. Urban Forum 13, pp. 102–115.
Olds, K., Bunnell, T. & S. Leckie (2002), Forced
evictions in tropical cities: An introduction. Sin-
gapore Journal of Tropical Geography 23, pp. 247–251.
Patel, S. & A. Thorner, eds. (1996), Bombay: Meta-
phor for modern India. Bombay: Oxford University
Press.
Pendse, S. (1995), Toil, sweat, and the city. In
S. Patel & A. Thorner, eds., Bombay: Metaphor for
modern India, pp. 3–25. Bombay: Oxford University
Press.
Prakash, G. (2008), Introduction. In: G. Prakash
& K.M. Kruse, eds., The spaces of the modern city.
Princeton, NJ: Princeton University Press.
Rao, V. (2006), Slum as theory. International Journal
of Urban and Regional Research 30, pp. 225–232.
Ruiter, W. (1999), Equity and slum redevelopment
in Mumbai. In: G. Chapman, A. K. Dutt & R.
Bradnock, eds., Urban growth and development in
Asia. Volume II: Living in the cities. Aldershot:
Ashgate.
Scott, A.J. (2000), Economic geography: The great
half-century. In: G.L. Clark, M.S. Gertler & M.P.
Feldman, eds., The Oxford handbook of economic geog-
raphy, pp. 18–44. New York: Oxford University
Press.
Sen, J. (1975), The unintended city: An essay on the city
of the poor. Calcutta: CRS.
Sharma, K. (2000), Rediscovering Dharavi. Stories from
Asia’s largest slum. New Delhi: Penguin.
Sharma, K. (2004), Mumbai’s other half. The
Hindu, 17 October 2004. Retrieved at <www.
countercurrents.org/gender-sharma171004.htm>.
© 2009 by the Royal Dutch Geographical Society KNAG
14
Smith, M. (2009), Editoral: Just how comparative is
comparative urban geography? A perspective from
archaeology. Urban Geography 30, pp. 113–117.
Smith, N. (1996), The new urban frontier. Gentrification
and the revanchist city. London: Routledge.
Sundaram, P. (1989), Bombay: Can it house its millions?
New Delhi: Clarion Books.
Thrift, N. (2000), Pandora’s box? Cultural geogra-
phies of economies. In: G.L. Clark, M.S. Gertler
& M.P. Feldman, eds., The Oxford handbook of
economic geography, pp. 689–704. New York: Oxford
University Press.
Turner, J.F.C. (1972), Housing as a verb. In: J.F.C.
Turner & R. Fichter, eds., Freedom to build,
pp. 148–175. New York: MacMillan.
© 2009 by the Royal Dutch Geographical Society KNAG
View publication stats
JAN NIJMAN
Turner, J.F.C. (1991), Housing by people: Towards
autonomy in building environments. London: Marion
Boyars.
UN Habitat (2003), The challenge of the slums: Global
report on human settlements 2003. London: Earthscan
Publications.
Vora, R. & S. Palshikar (2003), Politics of locality,
community, and marginalisation. In: S. Patel &
J. Masselos, eds., Bombay and Mumbai: The city
in transition, pp. 161–182. New Delhi: Oxford
University Press.
Ward, K. (2008), Editorial: Towards a comparative
(re)turn in urban studies? Some reflections. Urban
Geography 29, pp. 405–410.