|
Housing
Design and the Virtual Community
Mark David Major, Tim Stonor,
Alan Penn and Bill Hillier
Published in the 19th International Making Cities Livable Conference
- Children and Youth in the City Proceedings, Charleston, SC USA, 9-13
March 1997
Keywords: Housing, Children, Configuration, Design, Space Syntax
Abstract
The paper presents 'intelligent'
new methods of graphically and statistically representing patterns of
pedestrian movement and space use in a modern housing estate in the
United Kingdom. Configurational modelling, or 'space syntax', is used
in conjunction with quantitative observational data on movement and
space use to demonstrate how the design of the estate has contributed
to the spatial separation of adults and children, namely through the
complexity and small-scale of routes in the estate; the high number
of spaces constituted by blank walls; and, the limited visibility and
access of 'public' spaces in the estate from dwelling entrances. Because
of this, the design of the estate itself can be implicated in the absence
of a 'natural', everyday surveillance of children by adults, which has
left several spaces on the estate dominated by teenagers and children.
The higher-level social outcome has been fear in the perception of this
estate as 'no-go' area, though actual rates of crime do not justify
this fear. The paper concludes with a discussion of design solutions
which can facilitate and strengthen the child-adult relationship in
housing.
Housing and the modern paradigm
There are few sectors of
architectural design in recent history which set the context for current
theoretical debate so clearly as housing. This is because housing is
preeminently social in its intent. It comes closest to public awareness,
not only in the provision of dwelling space in which we all live, but
also in the way it serves to construct the open space structure of the
city through which we all move, work, socialise and take leisure. Housing
is the main constituent building type of urban form.
However, over the last decade architectural practice has been faced
with a serious challenge. The housing estates of the 1960's and 1970's,
even some more recent, are now the subject of major redevelopment and
refurbishment programmes. As much architectural ingenuity is being expended
in correcting the errors of 'modernist' design as was spent on the original
designs themselves. The process appears to be one of simplification;
of replacing the highly contrived schemes with simpler prototypes; replacing
access ways with streets, scissor sections with simpler subdivision
and bringing doors to the ground level. The solutions are intensely
pragmatic, and on the face of it pose a problem for architectural theory.
They replace the products of apparently sophisticated theory with apparently
simple, often normative, rules of thumb. The rules have been developed
and applied not through rigorous analysis and logic, but often through
the intuition and experience of designers on the ground looking at specific
problems and trying to solve them.
In this paper we will argue that this process actually backs up the
proposition that architecture is a theoretical subject, that form and
function are related, and that these apparently normative solutions
to the problems of design can be demonstrated to have an analytical
and theoretical basis to them. This argument is based on the description
of a recent case in which new methods of 'configurational analysis of
space' or 'space syntax' have been used to analyse the everyday patterns
of space use and movement in a modern problem estate to directly feed
into the design solutions in its redevelopment. However, this case is
not only of interest in terms of the light it throws on how architectural
theory can feed into design, but also for fundamental theoretical notions
it gives rise to of how the design of space can either generate or hinder
the construction of a 'virtual community' in our cities. This community
is virtual because it exists before interaction but it is crucial to
the the everyday surveillance of children and policing of strangers
on our streets. It is suggested that this virtual community is a fundamental
aspect of a healthy, urban life.
Configurational analysis
of space
Configurational modelling,
or space syntax, is a set of techniques for representing space in a
building or city. For example, movement in a city tends to approximate
lines so one technique is to draw the set of longest and fewest lines
of sight and access, see Figure 1a. The stationary use of space in a
building or city tends toward convexity, i.e. the mathematical definition
of all points being visible to each other like a group of people gathered
in a circle, so another representation is the collection of all the
'fattest' two-dimensional 'lumps ' in a system, see Figure 1b. Finally,
the potential for seeing and moving can be represented as isovists,
or fields of vision, which are visible and accessible spaces which we
might move to from a point, or particular set of points, in space, see
Figure 1c. We can complicate the representation by using any combination
of these three depending on the problem to be researched, see Figure
1d and e.
 Fig. 1  Fig. 2
While these techniques are ways of representing space, some clarification
is needed for what we mean by 'configuration'. For example, if there
are two objects in a mathematical relationship to each other it can
said that 'a' is to 'b' as 'b' is to 'a', see Figure 2a. However, once
this relationship is established with regards to a third, in this case
the surface of the earth, there is a configurational relationship, see
Figure 2b. For example, 'a' is to 'c' as 'b' is to 'c' but in order
to reach 'b' from 'a' we have to pass through 'c'. This can be seen
more clearly in the corresponding graph. Next the idea of permeability
or connection can be introduced. If the objects are in a symmetrical
relationship where all spaces are maximally shallow from each other,
we can say that 'a' is to 'b' as 'b' is to 'c', see Figure 2c. Finally,
if ' b' is be placed on top of 'a', an asymmetrical relationship results.
This is because to reach 'b' from 'c' you have to pass through 'a' but
it is not necessary to move from 'c' to 'a' through 'b', see Figure
2d. This is what is meant by configuration; a relational system where
any local change in that system has global effects across the system.
In previous studies it has been found that axial lines are most revealing
representation of pedestrian and vehicular levels in urban systems.
In Figure 3, we have generated an axial model of Greater London, approximately
between the North and South Circular roads, and have asked the computer
to mathematically measure the relationship between every space in London
to every other space in the system, relativised for the size of the
system, a measure called 'global integration' because it measures relationships
globally across the system. The computer colours the map from dark through
to light with dark lines being the most integrated, or shallow, and
the light colours being the most segregated, or deep. This global integration
map of London shows that the most integrated space in the urban system
is Oxford Street. Let us be clear at this point that the model is a
purely mathematical representation of pattern - land use, density, and
other functional uses have not been taken into account though we can
later add these into the model. Despite this, the model provides us
with a very realistic picture of London.
Click here to see Fig. 3
We can also ask the computer to provide a more localised picture of
the city by examining only those space to only three steps away from
every space, a measure which we term 'local integration', see Figure
4. When we do this, the analysis picks out a set of lines corresponding
to London's high street structure.
Click here to see Fig. 4
Natural Movement
A common criticism of modern
housing schemes is that they lack 'vitality' and at worst turn into
'urban deserts' devoid of people even in the middle of the day. It is
a longstanding question whether the design of the schemes is in any
way at fault in this, or whether the blame rests with the way that the
schemes are managed, or with the socioeconomic status of the residents
and so on. The multivariate nature of this sort of socio-spatial question
has made it seem almost impossible to resolve one way or the other.
However, the new techniques for describing and quantifying the geometric
and topological form of urban space allow us to approach some of the
simpler questions at the heart of this issue.
Since we can use our spatial models to represent existing built conditions,
we can also amend them to describe and quantify radically different
spatial designs. In this way, we can begin to 'control' the design variable
in studies of other aspects of urban function. When we do thisit follows
that it is possible to detect effects of spatial design on patterns
of pedestrian movement simply by observing pedestrian flow rates at
a number of locations in the street grid and then using simple bivariate
statistics to look for the relationship between configurational measures
of those locations and flows. A large number of studies have now established
that integration is consistently the strongest predictor of pedestrian
flow rates (see Hillier et al, 1993, for a comprehensive review of the
evidence). Integrated spaces carry greater pedestrian flows than more
segregated ones. The effects are strong and consistent. For example,
we can see this if we examine a line map for an area in north London
called Barnsbury where detailed observations of pedestrian movement
patterns have been made over a period of years. Each street segment
has been observed for a total of over 50 minutes at different times
of day and on different days of the week. The all-day mean hourly pedestrian
flow is noted on each segment, see Figure 5.
 Click here to see
Fig. 5
The scattergram of the measure of 'local integration' in a much larger
model of the area against the logarithm of pedestrian flow rates shows
a strong correlation. In this case the model is much larger than that
shown, extending approximately two kilometres away from the observation
area in all directions. The correlation is remarkably strong at r=.735,
p<.0001, see Figure 6.
 Fig. 6
The key discovery is that the correlation is between pedestrian flows
and a purely spatial measure of the pattern of the street grid. No account
has been taken of the location of attractors or generators of movement
in constructing the measure of integration. It seems that movement patterns
result naturally from the way the spatial configuration of the street
grid organises simplest routes (involving fewest changes of direction)
to and from all locations in an area. Of course, this runs counter to
the premises of traffic modelling which holds that the key facts in
urban systems are the distributions of activities and land uses that
'generate' or 'attract' flows between different geographic locations.
Our results show that the primary fact is the pattern of space, and
that if there is a relationship between land uses and pedestrian flows
(which there certainly is - you find more people on streets with shops
than on streets without) this is most likely to be due to retailers
choosing their shop sites with care in order to take advantage of the
opportunities for passing trade provided by the natural movement pattern
resulting from the grid. These findings would suggest a marked shift
in emphasis from existing urban theory is required.
The probabilistic interface,
or virtual community
In the city which works well,
spatial configuration and its effects in generating the natural, everyday
policing and surveillance of space, goes almost unnoticed. Our research
seems to bring quantitative confirmation to the observation Jane Jacobs
made some twenty years ago that 'the public peace... is kept primarily
by an intricate, almost unconscious, network of voluntary controls and
standards among the people themselves and enforced by the people themselves'
(Jacobs, 1961). Hanson has termed this natural policing of the street
as a product of the virtual community.''A form of group awareness in
a collectivity, virtual because it has not yet been realised through
interaction among its members (and) people are members by virtue of
spatial presence' (Hanson et al. 1987). Hanson adds that 'spatial layout
in itself generates a field of probabilistic encounter with structural
properties that vary with... the layout...(and) that this encounter
field may be a more important social and psychological factor than has
been allowed' (Hanson, et al. 1987).
It is only where the problematic designs generated by the poor socio-spatial
theories of the recent past have created practically unworkable pieces
of urban space that the question of the relationship between design
and social function seems to come up. These 'problem estates' are now
helping us push forward a more robust set of theories of the relationship
between the city and society. One of the main ways we are doing this
at the Space Syntax Limited is by becoming actively involved in redevelopment
masterplans for 'failed' inner city housing estates. In these projects,
the new techniques are being used for three main purposes. First, as
diagnostic tools to help pinpoint the precise problems of a specific
estate, and so better to direct design effort and expenditure towards
those aspects of a scheme that really matter. Second, as models of pedestrian
movement to predict both rates and patterns of movement, to help eliminate
the under use and abuse of space that seem to characterise so many of
their problems and to try and reintegrate their communities into the
larger scale neighbourhood. Third, in order to achieve a natural relationship
between local and global spatial properties so that they become more
intelligible in their own right and as local areas with respect to their
context. In each of these we are gaining new data and discovering new
phenomena to theorise about. Live redevelopment projects of this sort
always demand methodological development and form the core of our fundamental
research activity. A recent project has proven to be a particularly
valuable 'learner' in this respect.
Marple Square Housing Estate,
Nottingham
The layout of dwellings in
the estate is characteristic of the contemporary shift from the monolithic
scale of high-rise, high density system-built housing estates to low-rise,
low density villages in the city, see Figure 7.
Click here to see Fig. 7
Although the underlying conception of this form of architecture is radically
different, representing a shift from the ideal of an urban utopia to
that of the romantic urban village, both architectural types can be
seen to use similar spatial means in generating their plans, the difference
laying in the scale. In Marple Square, the plan is generated by the
repetition and duplication of the simplest element of a single dwelling
which is then either rotated or shifted in a geometric relationship
with its neighbour, or a series of neighbours, in plan. In another well-known
'problem estate', Chalkhill Housing Estate in north London, the element
was the 'slab block', but the result in terms of enclosure of convex
spatial elements is similar. This becomes clear in the figure-ground
of the estate with the buildings in white and the spaces between these
buildings in black, see Figures 8 and 9.
  Click here to see
Figs. 8 and 9
For a 'problem estate' Marple Square would appear to be well connected
to its surrounding area, especially to the south. In all the estate
has 18 streets or pathways connecting outward from the estate with 14
of these connecting into Huntingdon Street and Woodborough Road, however,
the majority of these do no penetrate very deep into the estate. Indeed,
there is currently no single visual link or pathway which moves completely
across the estate without at least one change in direction.
One clear characteristic in the design is the use of the repeated single
dwelling elements to enclose interior spaces on the estate onto which
the entrances of dwellings open. In this way, dwellings within the estate
turn their backs to the primary traffic-articulated streets and tend
to conceal dwelling entrances from the larger scale streets. This is
a common trait in many 'problem estates' because of the belief that
it was desirable to separate entrances from movement for privacy and
security reasons.
Over a two day period, a total of 240 observations were carried out
within the estate and in its surrounding area. These observations were
broken down into weekday and weekend observations as well as into five
time periods from 8-10, 10-12, 12-2, 2-4 and 4-6. By analysing these
observations graphically and statistically we were able to investigate
how far the structure of space, and the pattern of space use within
this structure, could be held to be responsible for the problems within
the estate.
First, is a dotmap of all adults observed in movement over two entire
days (one weekday and one at a weekend), see Figure 10.
 Click here to see
Fig. 10
It shows the high concentrations of movement along the perimeter of
the estate along Woodborough Road, a high degree of movement bisecting
the estate along Alfred Street Central, as well as high levels of movement
along the southern section of St. Ann's Way. In addition to this we
can see that significantly lower levels of people were observed in the
north of the estate as opposed to the south and that the number of people
observed decreases the deeper into the internal pedestrian-only pathways
one moves. The picture that this map provides us suggests that movement
in the estate moves along the perimeter street of Woodborough Road,
enters the site along Alfred Street Central or St. Ann's Way, then disperses
among the complexity of routes available from these streets. Movement
out of the estate operate in reverse. Woodborough Road is clearly the
main sink and source for all larger scale movement.
Second, is a dotmap of all stationary children observed over two entire
days (one weekday and one at a weekend), see Figure 11.
 Click here to see
Fig. 11
It is clear that a high number of children are focused in the recreation
ground area of the estate suggesting that this is working as a focus
for children. In addition to this, some stationary children were observed
in the area of the estate west of Heskey Close and south of Alfred Street
Central around the internal links off Norman Close. Although the recreation
ground is successful in focusing children into a single space it does
so to such an extent that the ratio of adults-to-children in the recreation
ground during the weekdays is approximately 6 children to every 1 adult
and during the weekend this ratio increases to 10-to-1.
Upon closer analysis it can be seen that this segregation of children
and adults is pervasive. Next is a series of maps which allow us to
see the patterns of space use by categories of men, women and children,
see Figure 12a, b and c.
 Click here to see
Fig. 12
The maps highlight the lines on which the top 50% of movement is concentrated
for each category within the estate (excluding the perimeter streets).
This shows that men are using the primary vehicular streets (on average
one step from the perimeter), women also use these streets, but also
concentrate one step off these streets (and two steps from the perimeter),
whereas children can be found in spaces which are two steps from these
primary streets (thus three steps from the perimeter). This would seem
to indicate that the different categories of people are using space
quite differently from each other. We could infer from this that men
are moving from their dwellings (usually from the back door) to parked
vehicles and then out of the estate either on foot or by vehicle. Women
would appear to move on foot from dwelling entrances to primary roads
then out of the estate to the perimeter roads and children appear to
be either moving out of the estate along the same routes as women or
moving from dwelling entrances and back doors deeper into the more secluded
areas of the estate. We can interrogate our observations statistically
to clarify these differences. In Figure 13a, b and c, the correlations
between men and women, women and children and children and men reveal
a selective set of spaces which are being used primarily by one category
of people and not by the other.
  Click here to
see Figs. 13 and 14
By far the clearest example of this is shown when we correlate moving
adults with stationary children statistically, see Figure 14.
When we look at the spatial analysis of the estate bedded in a map of
the whole of Nottingham we can clearly see the global structure of Nottingham
is focused in a ring which passes through the central square in Nottingham
and to the north along the main east-west road into the town centre,
see Figure 15, Ilkeston Road, the primary west-east link of Gregory
Boulevard as well as Mansfield Road and Lenton-Radford Boulevard which
integrate north-to-south.
 Click here to see
Fig. 15
The estate itself is segregated and relatively complex in terms of the
degree of spatial break-up, see Figure 16a.
 Fig. 16a
Next is the correlation of depth from Woodborough Road against people
observed moving through the estate, see Figure 16b. We see that as depth
from Woodborough Road into the estate increases, the levels of movement
decrease. This bears out the observations of movement made on the ground
that suggest that Woodborough Road is the main sink and source of large
scale movement for the estate. The fact that it is also the most integrating
of the surrounding streets in the city as a whole leads to a relatively
good correlation between movement and local integration in the estate.
This correlation has an r^2 value of .583 indicating that we can predict
close to 60% of differences in movement in the estate, a value which
in our experience is high for a housing estate, see Figure 16c. This
gives us confidence that we can simulate proposed changes to Marple
Square with a degree of accuracy.
  Click
here to see Figs. 16b and c
We have in Figure 16d analysed a proposal made by the Joint Centre for
Urban Design at Oxford Brookes University. Several improvements have
been made to the spatial design of the area based upon the recommendations
of our initial report (Hillier & Major, 1994).
 Fig.16d
First, the scale of space is much larger and compatible to the more
traditional areas of Nottingham. Second, the patches of segregation
created by the internal pedestrian-only pathways and dead-end spaces
have been eliminated leaving the estate relatively free of segregated
areas. Third, the articulation of routes from the south (Huntingdon
Street) to north (Crammer Street) makes movement through the estate
simpler and complements the already strong west-east through-route of
Alfred Street Central. If we look at a scatter between global and local
integration and compare it to the scatter of the existing system we
can see that the high degree of layering typical of housing estates
and Marple Square has been eliminated and that the estate makes a very
tight scatter across the slope of correlation, see Figure 16e and f.
  Click here to
see Fig. 16e and f
This suggests that with this proposed model of space a highly intelligible
local housing area will have been created in the centre of Nottingham
with good levels of natural movement, in which the boundary that defines
it as an estate would have been mainly eliminated. If at the same time
internal remodelling to reverse the entrances to the dwelling units
is carried out as proposed it should be possible to reconstitute Marple
Square as an integral part of the city of Nottingham rather than as
an enclave.
Conclusions
The concept of the field
of co-encounter, or virtual community, between different social categories
in the everyday use of space, would seem to have far reaching social
implications. Our studies have shed light on the way spatial configuration
constructs patterns of movement and encounter and, we believe, on the
social pathologies that can ensue. There is remarkable consistency in
the way that post-war public housing in the UK has sought to segregate
its public open space from the surrounding street fabric. This is demonstrated
by substantially reduced integration values for housing estate spaces
compared to traditional street areas in configurational analyses. As
might be expected the presence of pedestrians also drops substantially
in estate interiors, to the point at which they are sometimes referred
to as 'deserts' or as being in a state of 'perpetual night'. It seems
that spatial segregation serves to isolate these estates from through
movement to the point at which you can be alone in most spaces for most
of the time. However, where we observe space use patterns by different
categories of people simultaneously we find still more suggestive results.
Patterns of space use by children and teenagers of school age differ
radically from those of adults. Children gather in groups, often not
moving much but using space to 'hang out' in locally strategic locations
which are cut off from the outside world in the estate interior.
These locations tend to remove them from informal surveillance by adults
moving into and out of the estate, and if we look at the correlation
between adult and child presence in estate interiors we find a characteristic
'L-shaped' distribution. Where there are high numbers of adults there
are low numbers of children, and where there are high numbers of children
there are low numbers of adults. In normal urban streets there is a
much stronger correlation between adults and children suggesting that
an informal interface is maintained. These findings appear to be general,
see Figure 17.
  Click
here to see Fig. 17
In recent doctoral research Jianming Xu has studied a sample of postwar
UK housing estates and detected a correlation between the mean integration
of the estate and the strength of the L-shaped effect, the more segregated
the estate the greater its effect on separating the different categories
of user, see Figure 18.
 Click here to see
Fig. 18
These findings are now being added to by more recent studies of crime
locations which suggest that specific locations in estate interiors
(when thery are emptied of normal adult levels of movement by being
segregated from through movement to and from surrounding areas) become
the favoured locations for specific types of crime and abuse.
It seems quite possible that the configuration of urban space through
its effects on patterns of movement may construct informal probabilistic
interfaces between different categories of people. The interface between
shop owners and buyers makes transaction possible, that between adults
and children may turn socialisation and control into natural and unforced
processes. Alternatively, where space structures lead to a polarisation
of space use by different social categories, we suspect that distrust,
stigmatisation and crime result. It seems possible given this view of
the relation between social processes and spatial configuration that
the theories which gave rise to zoning of 'communities' in their own
'territories' served to foster the social pathologies they were intended
to control. If this is so it is little wonder that 'theory' has gained
such a poor reputation amongst practitioners and the public alike.
We believe however, that the new form of analytic theory which is being
developed in studies such as these holds a great deal of hope for the
future. By applying the lessons learned from studies of successful and
failed urban areas we believe that it is possible to be both objective
in assessments of different design strategies, and to deal with individual
cases and their different contextual areas without recourse to the kinds
of normative design prescriptions about which designers and the public
alike are becoming so suspicious.
References
Coleman A. (1985) Utopia
on trial: vision and reality in planned housing. Hilary Shipman, London.
Hanson, Julienne and B. Hillier. (1987) 'The Architecture of Community:
Some New Proposals on the Social Consequences of Architectural and Planning
Decisions' from Architecture and Behaviour. Volume 3, Number 3, pp.
251-273.
Hillier, Bill, (1996). Chapter 4, 'Cities As Movement Economies' from
Space Is The Machine. Cambridge University Press, Cambridge, England.
Hillier B. & Major M.D., (1994,) The Urban Village Revisited: The
Spatial Design of Marple Square Housing Estate, Space Syntax Limited
report.
Hillier B., Penn A., Hanson J., Grajewski T., Xu J. (1993) Natural Movement:
or configuration and attraction in urban pedestrian movement, Planning
and Design: Environment and Planning B, Pion, London.
Hillier B. & Penn A., (1989) 'Dense Civilisations: or the shape
of cities in the 21st century' from Applied Energy. Elsevier, London.
Hillier B. (1986) 'City of Alice's Dreams' from Architects Journal.
London, 9th July 1986, 39-41.
Hillier B. & Hanson J. (1984) The Social Logic of Space, Cambridge
University Press, Cambridge, England.
Jacobs, Jane. (1961) The Death and Life of Great American Cities. Penguin
Books, Harmondsworth, England.
|
