total cost per house.
In working out these options, it is essen-
0.1 Compared to the scale of the housing tial that a number of crucial elements are
demand in most third world urban areas, taken into account. For instance:
the resources nationally available are tragi-
(a) The cost of service infrastructure,
cally minuscule. In India for instance, i.e. roads, water supply, sewer lines, etc.
the income of 25% of the families will buy
(b) The cost of providing mass trans-
them about 2 M2 of built up construction; port, which would of course vary with the
the next 50% can buy upto 5 M2 nor is densities, but also depends on the urban
it merely a matter of finance; steel, cement, structure (e.g. whether or not it is linear).
and other material resources are also crip-
(c) The construction cost of the units.
pling constraints.
(d) Some weightage against using
0.2 Yet space itself is a resource. nationally scarce materials.
Housing, after all, is much more than
(e) Weightage in favour of such con-
houses; it is really a system, a hierarchy figurations and densities wherein it is pos-
of activities and spaces, ranging all the sible to re-cycle human and other wastes
way from private family functions (like (gobar gas plants, bio-gas plants, etc.).
cooking and sleeping) to the principal com-
1.1 Providing housing involves much
munity meeting place.
more than just providing houses. The room,
0.3 There are two vitally significant the cell, is just one element in a whole
characteristics of the elements in this hier-
hierarchy of spaces human beings need in
archy; firstly, within each activity, there order to exist in a city.
are trade-offs, (especially in warm 1.2 This hierarchy is determined by
climates) between spaces which are cover-
many factors, such as climate, culturally-
ed and those open-to-sky; and secondly, defined life-styles, and so forth. For in-
these activities are mutually inter-depen-
stance, under certain Indian conditions, it
dent within the hierarchy, and there can be appears to have four major elements:
spatial trade-offs between them.
(a) The space needed by the family for
0.4 Since residential building sites are exclusively private use, such as cooking,
usually only one element in total land use, sleeping, storage, etc.
these trade-offs can best be perceived in the
(b) The areas of intimate fine-scale con-
context of the entire land-use allocation tact i.e. the door-step where children play,
pattern. To begin with what we need is: you chat with your neighbour, etc.
(i) A quantitative and qualitative ap-
(c) The neighbourhood meeting places
praisal of the hierarchy of spaces (both (e.g. in our villages, the village well) where
public and private) used by people in select-
you become part of your community.
ed third world human settlements, in order
(d) Finally, the principal urban area—
to understand the trade-offs which occur e.g. the maidan—used by the whole city.
within existing systems.
A space to kick a football, fly a kite, hear
(ii) A model which investigates various a political speech.
land-use patterns and which, for different
1.3 Each element in this hierarchy can
development and land values, optimises the consist of covered spaces and/or open-to-sky

spaces. For example, many of the private within the totally inadequate context pro-
activities at the micro end of the scale, such vided for them. Recently, at the environ-
as cooking and sleeping need not be ex-
mental conference in Stockholm, our Prime
clusively indoor but can — and do — take
Minister said that in countries like India
place in an open courtyard (provided of the most serious pollution of all was really
course that the family's privacy is reason-
poverty. And that statement can be ex-
ably assured). In fact, depending on the tended — for it is not just poverty, but the
cost of building construction this trade-off particular forms which poverty takes in
is automatically adjusted — each society our urban areas. Rural poverty in India
(and each family within it) finding its own is a different thing. The people are as
balance. This adjustment is of the utmost poor — often in fact poorer — but they are
importance, particularly in developing not as de-humanised. Obviously there is
countries, because they are usually in warm very little relation between the way our
tropical climates where a number of acti-
cities are built and the way people use
vities can indeed take place in the open. them. Not having the proper range of
1.4 The second important fact about spaces they need to live, people merely
the elements in this hierarchy is that they mis-use what they do have access to —
are mutually inter-dependent. That is to hence the thousands of families cooking on
say, less space in one area can be pavements, squatting along railways tracks,
adjusted by the provision of more in one and so forth.
of the other three. (For example, smaller
2.2 And of course the tragedy is that
dwelling units might be compensated by piling people one over another does not
larger neighbourhood community spaces, in fact 'save' much land for the city. For
in most urban areas around the world, only
1.5 To re-cap briefly, we perceive hous-
about 15% of land-use is devoted to residen-
ing not as cells in isolation but as a hier-
tial building sites. The rest is in other
archy of activities and spaces; secondly, space-extensive uses such as industry, ware-
these activities are mutually inter-dependent housing, and so forth. For instance, trans-
and there can be a spatial trade-off be-
port is usually between 25% to 35% of
tween them; thirdly, within each activity, land-use (higher in Los Angeles!). Even
there is a similar trade-off between spaces with the inclusion of local distributary
which are covered and those open-to-sky. roads, tot-lots, and so forth, housing occu-
2.1 To identify the hierarchy and to pies about a third of most cities. Thus
understand the nature of these trade-offs we see that doubling the number of dwel-
is of course the essence of the task of pro-
ling units on each site does not 'save' much
viding housing. Without this, one is in land for the whole city (though it could
grave danger of formulating the wrong ques-
mean much higher profits for the indivi-
tions. For instance, most attempts at low-
dual developer — which is of course the
cost housing perceive it only as a simpli-
reason it gets done). On the other hand,
stic question of trying to pile up as many halving the density on residential plots
dwelling units, as many cells, as possible could mean only a marginally larger city.
on a given site; with the result that in
2.3 Similarly, there are options involving
much of the Indian urban scene today, trade-offs between different land-uses in the
what we are observing is the desperate at-
city. For instance, New Delhi provides
tempt of people—especially the poor—to try about 1.5 hectares of open space per thou-
and somehow work out a pattern of living sand people. (The figure is considerably

higher in London and most new towns viable in the context of a land-use plan
around the world). Now 1.5 hectares per which develops corridors of high-density
thousand works out to 15 M2 per person demand. The distance that a person is wil-
—i.e.. Over 75 M2 per family. Ironic in-
ling to walk to the transit station depends
deed to think that families in Delhi —
on the mode of transport and differs from
crowded into little hovels — each have country to country, indeed very often from
over 75 M3 awaiting them along some city to city. Thus the cost and convenience
monumental vista —while the planners of the mass transport system is not merely
hesitate to provide them with 10 M2 in a function of overall densities, but depends
a private courtyard which they could really also on the structure of the city — namely
use. ...
that it be a linear system, or a combina-
2.4 None of the fore-going is meant to tion of linear sub-systems — with each sta-
imply that there are no high density resi-
tion having a sufficient hinterland.
dential areas. On the contrary, in certain
3.3 Let us take an example. In the case
sections of many third world cities, e.g. of Bombay, this hinterland per train stop
Bombay and Calcutta, the densities are averages about 8 to 10 minutes walk, i.e.
extraordinarily high. But these densities a little under one Km. Within such a radius,
are not achieved through high-rise build-
and assuming a dwelling unit of 32 M2
ings; no, they primarily result from the (cri-
carpet area and a communal space at 30
minal, omission of play spaces, hospitals, M2 per family (for tot-lots, health centres,
schools and other social infrastructure. For etc. — the school play-fields and other
instance, in Bombay city, open space is space-extensive uses are just outside this
about 0.10 hectares per 1000 persons — area), we find that with five-storey walk-up
and this includes the 'green' of the traffic tenements we can accommodate about
islands! Then again, road coverage is about 40,000 people on each side of the station.
8% of land useless than a quarter of Ten-storey buildings (using elevators) would
what it is in New Delhi (this goes a long send up this figure to 55,000. On the other
way towards explaining the great crowds hand providing ground floor houses on in-
we see on Bombay's streets — as compar-
dividual sites (each 4 m x 11 m) would
ed to the empty boulevards of Lutyen's accommodate 25,000 people on each side
of the station. (Density is not a direct func-
3.1 But then the question arises: will tion of the height of the buildings. This
lowering the densities increase dispropor-
is partly due to the fact that the taller the
tionately the cost of the service infra-struc-
buildings, the further apart they must be;
ture? In particular of the travel time and and partly due to the social infrastructure
travel costs, involved in the public trans-
area per family being constant at 30 M2).
port systems. This is indeed an important
3.4 These variations in net residential
question. Without transport there can be density will not, as was pointed out be-
no mobility, therefore no job choice — in
fore, make a great deal of difference to
- fact often no job at all — and for the urban the overall size of the city; but they will
poor, mass transport becomes as crucial make crucial mutations in the living pat-
a prerogative as housing.
terns — really the life-styles — of the
3.2 Now a mass transport system — people. Furthermore, they can also make
whether a tram, or a train, a bus in mixed a decisive difference to the cost of con-
traffic or on a reserved track — is essen-
structing the dwelling units themselves. For
tially a linear function. It only becomes again, in developing countries, there is a

great variety of simple materials and exist-
struction materials. Multi-storeyed build-
ing vernacular technology in which the ings must of necessity use steel and ce-
ground floor house can be built — as for ment — commodities which are in excru-
instance, brick-in-lime mortar walls with ciatingly short supply, in developing coun-
country-tile roofs, or even stabilised earth tries. On the other hand, the individual row
with a jhopra (palm thatch) roof. Further-
house can be constructed out of anything,
more, any open-to-sky space left on the from mud on up. The recent advances in
site is really an extra room, obtained at paper technology open up a world of new
no cost, usable at least three quarters of possibilities.
the year for essential family purposes. For
(vi) Of course, if the house in its early
instance, they might use 5 M2 of this open-
stages is constructed of brick, mud and
to-sky area for grinding masala, or sleep-
country tile, then it will not have a life
ing at night; thus saving on the cost of span of more than 10 or 15 years — as
constructing an extra room. And this of compared to an R.C.C. structure which will
course, makes operative exactly the kind have a life span of, say, 70 years.
of trade-offs discussed earlier.
4.2 Of course it is one thing to be able
4.1 From the work we have done so to identify optimal residential patterns and
far, it appears that in developing coun-
densities; it is quite another matter for the
tries, these trade-offs decisively favour a authorities to be able to stabilise densities
pattern of low-rise medium-density hous-
at these levels. This is of course the cru-
ing. For there are a number of additional cial question, and it is here that strategies
advantages; to wit:
must be developed, strategies which would,
(i) A low-rise building has a much in all probability, involve the mass trans-
shorter construction period. Thus the in-
port system. For instance, if we could use
terest cost of capital tied up during con-
the transport system to open up new areas
struction is considerably less.
for residential use, we would in effect be
(ii) It is incremental, i.e. it can grow subsidising low-cost housing indirectly
with the owners requirements and his earn-
through a subsidy on the transport system.
ing capacity. Eventually, the owner may This might well be preferable to a direct
want to add an additional floor or two, subsidy on housing, as that contradicts the
either for rental or for his grown-up chil-
actual value of the housing, and leads — a t
dren's families. (This would have the ad-
least under Indian conditions — to illegal
ditional advantage of increasing the hous-
transfers of the tenements.
ing densities; though it would entail a
4.3 In any case, one wonders whether
certain flexibility in the pattern of infra-
the critical issue in third world cities is
structure provided).
not so much a question of increasing den-
(iii) It has great variety, since the indi-
sities, but rather one of lowering them.
vidual owner can design and build it ac-
For instance if we can bring down the
cording to his own needs.
density in the residential areas to 80 to
(iv) An individual building his own 100 persons per hectare, it may become
house is a highly motivated person; this feasible to dispense with a central sew-
motivation might possibly engender an age system and instead recycle waste mat-
increase in per capita savings — so that ter (both human and animal) to consider-
housing is built without sacrificing other able advantage (cooking gas, fertilizer etc.)
national investment targets.
under Indian conditions this would have
(v) It need not use high-priority con-
the additional advantage of allowing the

people a pattern of life they are accustom-
between Rs. 200 and Rs. 500. Even if we
ed to. As though Mahatma Gandhi's vision assume a rent-paying capacity of a quarter
of a rural India had an almost exact urban of this income (high by Indian standards
for this income level), then, using brick and
5.1 In conclusion, it must be empha-
concrete, very little can be constructed for
sised once again that the problem of low-
money — somewhere between two and five
cost housing is not one of finding new square metres per family — and this for
"miracle" building materials or construc-
75% of the population! Furthermore, it is
tion technologies, but primarily a question not merely a question of financial budget-
of land-use trade-offs and allocations. We ing; there is not enough cement and enough
have squandered far too much time in a steel in the country to deal with our mil-
fruitless quest for architectural and engi-
lions of urban homeless in this way.
neering panaceas, when all along the plan-
5.2 Yet, if we look around the country
ners have stated the problem wrongly to we find that indigenous towns and villages,
begin with. By Government count, a few people — without benefit of planners, or
years ago, there was a back-log of almost architects, or engineers — have always made
twelve million housing units required in marvellous and ingenious trade-offs be-
our urban areas alone. To this must be tween open-to-sky space and built-up con-
added the four-fold increase in urban popu-
struction, indicating to us a lesson of de-
lation expected over the next three decades. cisive importance: namely, that in a warm
In contrast, our resources — both financial and tropical climate, space itself is a re-
and material — are minimal. About 25% source. It is imperative that urban plan-
of our urban households earn less than ners in the third world begin to use it that
Rs. 200 per month. The next 50% earn way.