Townscape: when new housing is being planned, or old housing being redeveloped, weaving it into the townscape is as important - socially, visually and functionally - as the design of the houses themselves. This article discusses Britain’s recent failure to achieve this and the opportunities for doing so that still exist. It goes on to analyse, by comparing old and modem examples, the proper relationships between housing and the fabric of the town
Originally published in AR November 1967, this piece was republished online in September 2011
Those who care about townscape have tended to direct their attentions inwards concentrating more on the city centre than its edges. Meanwhile, the planners (and it is at the planning stage that things start to go wrong) have been allowing housing to spread outwards or upwards with little idea of strengthening or adding to the visual form of towns, with no attempt to create or maintain local identity, 1, 2, and with no thought for what people really want.
Some of these planners are also architects, so it is no good shifting the blame on to that race of multipurpose Jacks-of-All-Trades. The architectural profession must take its share of the blame. The trouble is that most architects, instead of using their eyes, collect file upon file of information (A4 sheets, of course) and worry too much about the use of computers.
You cannot compute aesthetics. Housing uses up more land than anything else in our towns. It therefore makes a major contribution to their visual image. Its layout and design also affects the lives of people more than any other aspect of town design.
Fortunately, there is an increasing number of well-designed housing layouts which break with the tradition of sprawl which we inherited from the ‘twenties and ‘thirties; but the problem is far bigger than any particular site; it involves the form of the whole town.
The visual impact of high or low rise housing in the expansion of existing towns or in the development of new towns is seldom considered in relation to the function and appearance of the town or city as a whole.
Some new towns, notably Cumbernauld, 3, and the abortive Hook, 4, attempted to create a total image which should be easily legible and reflect its social organization. Unlike many large-scale attempts at housing, they tried to create a local identity without distorting the pattern of family and community life.
As far as one can judge, in visual terms Hook, had it been built, would not yet completed - about 1,000 of its projected population are there and its centre, slowly extruding from one end of the hilltop site to the other, has only a fraction of the quota of shops planned.
There are a few corner shops but these make little impact and mostly have the feeling of mini NAAFI’s. As it is incomplete, any attempt to assess its total identity
It may be argued that identity does not matter; that the need for towns to have a visual character distinguishing them from their neighbours is no longer architecturally important or socially necessary, given the personal mobility of today.
Indeed one may ask whether towns are valid units at all at a time when whole regions can be travelled quickly and comfortably? Has the basis of their existence disappeared with the methods of travel that formed them?
Looking ahead it is more than just conceivable that large or small housing estates (with the minimum of convenient local shopping), somewhat like detached neighbourhood areas, could be built in the countryside with road or monorail connections to shopping centres and places of recreation or industry.
As the recent range of consultants’ reports on expansion makes clear, somebody is soon going to build a linear town or a string of beads, or a directional grid, in which case the identity of the whole will be quite different from the concentric town layouts of the past as we know them, 5, 6, 7.
But these are not arguments against the desirability of achieving identity: they simply reinforce the need to produce positive visual identities for these new urban forms. The social and physical mobility, with which increasingly we find ourselves endowed, is likely to make us search for more rather than less fixed points against which we can enjoy our new found freedom.
Though we may live in a standard house, drive in a standard car, we are all the more certain to demand some difference, some personality, some identity in our homes and in our towns. Present urban layout tends to be dictated solely by the needs of traffic generation and bus stop or shopping catchments.
These are essential considerations in the planning of communities, but there are also architectural considerations which must both influence and freely interpret those functions. As long as we go on building new towns with town centres surrounded by areas of housing, it will still be socially important, for the reasons Kevin Lynch has given, to create in them a sense of visual cohesion.
Similarly, if we are to make the best use of visual capital, it is vital that the hundreds of existing towns, which are with us for a long time, should maintain their identities and their visual cohesion. Where they have lost those qualities we must recreate them as we modernize them.
If there is a social case for maintaining a recognizable and compact visual image in our existing settlements, there are many factors working against that aim-leaving aside the question of our skill to do it. These factors derive principally from growing affluence and our needs for space, privacy and mobility. But these new influences on our towns need not be destructive; our task is to see that they are used to create new forms and identities.
Space and privacy
Whilst frighteningly large areas of poverty still exist - one twelfth of the population below subsistence level and one quarter in houses without essential amenities - environmental deprivation is becoming as real as lack of purchasing power in wages.
The majority now have the right to expect that the environment of the community should improve to match the improving standards of their homes. We now want both private and communal space for relaxation outside as well as our Parker Morris interiors.
Most of us like at least a patch of outside space at ground level that is private. And as a nation one of the strongest desires that we seem to have is that houses should feel separate individual castles - even though the isolation may be symbolized only by a bit of chestnut paling and a windy passage.
In practice this innate desire is the root cause of the dissipation of town identity. But to rant about it because it uses up too much countryside or because (and this is debatable) it sprawls in a formless, artless way, fails to take into account that people seem generally to like and prefer (however small) a private piece of open space, 8, to a communal one.
If that is so, then that must be the architect/planner’s brief. He must learn to be less arrogant about what he thinks people ought to want and make full use of sociological information rather than select what suits his visually predetermined schemes.
From experiments under way, this sort of brief could result in a maximum density of probably around 140-160 persons per acre in densely urban situations. A scheme by Richard Sheppard, Robson and Partners for the GLC at Southwark would provide gardens and garages at this kind of density.
And in addition to these personal wants there are needs for open space for football or cricket or golf, in other words for formal sport, on top of which there are demands for space for informal activity (Gordon Cullen’s ‘wenching areas’?). In combination these inevitably tend to reduce the overall density and the compactness.
If we are stuck with space standards in terms of density, we are also stuck with the motorcar. If Buchanan’s environmental standards are applied, as surely they must be, assuming maximum car ownership and use, residential areas will be governed in their design and form by a further set of criteria.
The actual layout of roads in individual areas of course affects their environmental capacity. The greater the flow of traffic the less can roadside activity be serviced and the more we need to protect individual properties against noise and fumes.
Similarly, the need for physical and visual pedestrian segregation will be greater. Thus by a combination of these factors, the number of people in a residential area will be limited. The significance of this in visual terms is that the fabric of the whole town will increasingly become split into units separated by a network of roads carrying traffic about the town.
This could be a positive force working on the side of area and town identity. In town centres a new complex will often contain roads, car parks, offices and shops; even some residential units-all designed to work as a unit. Cumbernauld is a case in point.
But designing all these functions to be contained within a single structure will be of limited application only. The increased costs involved in providing the right environmental standards for residential units will prohibit extensive constructions of this type.
So the danger is that both existing towns and new towns will progressively be carved up into areas containing single uses - segregated housing areas with networks creating division between them, separated from main shops, comprehensive schools, offices and industry.
The South Hampshire Study represents a significant step in this direction, 9. Disregarding the obiter dicta of Jane Jacobs the report makes proposals as if the underlying land form did not exist. The first generation new towns began this trend, 10, and the concepts contained in Traffic in Towns may prove to be even more divisive in this respect, 11, 12.
Size of community and town form
The need for privacy, private and communal outdoor space and the attractions of personal mobility - seemingly unalterable historic symptoms of affluence - are now affecting and will increasingly dominate the remodeling and planning of our towns.
These growing needs are the principal causes of the changes in urban identity. Underlying these, however, is the fundamental problem of the deployment of population growth, which causes the continual expansion of towns.
The larger the settlement grows, the more difficult the task of maintaining or creating local identity. Large cities such as Liverpool, Birmingham or Sheffield still retain their visual, social and functional differences in their central cores, but as whole units of settlement they lost their identities long ago.
And the process of urban renewal is now rapidly reducing even those city centres to bleak conformity, broken only by the occasional gross vulgarity that becomes acceptable if only because it provides relief. (Walk down Victoria Street, SW 1, and see how that extraordinary building, Windsor House, scaled to apparently twice its proper size, has become a positive jewel.)
In the large cities and conurbations, where fast population growth is a continuing problem, an identity can and should be retrieved by breaking down the suburban sprawl into recognizable units. And here network roads could have a positive role. Once constructed, these networks would effectively cut the suburbs into sectors, giving each the chance to find its own identity, 13.
If this process can become a powerful tool reforming the suburbs, a growing awareness of the character of many of our city centres may help us to refashion them as the core of a cluster of suburbs, each with their own subordinate identities. In the smaller, free-standing, towns there is still time to set limits to their growth.
These limits would be drawn where it became necessary to siphon off growth into new settlements to prevent the town’s image from being destroyed. New housing constantly added at the perimeter, however well designed at the detailed level, must inevitably erode the legibility of the total image.
Decisions on the extent of the residential areas (the number of people living in the town) in relation to the town centre are crucial to the whole question of local identity. Of course some towns will have to expand; growth cannot be confined entirely to new towns, but those that are chosen must be capable visually of absorbing change. And the opportunity to recreate anew the identity of such towns must be firmly grasped.
Apart from visual criteria, there are other advantages in setting limits to growth in certain existing settlements. In traffic terms the greater the number of people the more vehicular trips that will be made, and the larger and wider the roads to carry them will have to become.
Consequently the economic costs of expanding such towns will be heavy. If the aim is to maintain the identity and cohesive unity of individual settlements, the starting point must be an assessment of the visual effect of housing expansion at the regional level.
The ability of towns to accept growth in relation to their visual and social identities is seldom, if ever, considered at any stage. Change and conservation can only go hand in hand if the process is conscious and planned. Thus, a visual survey is essential, at regional level, to discover the aesthetic constraints and opportunities of existing settlements and to assess their capabilities for housing expansion.