This chapter has two objectives. The first and more general is to discuss the conceptual and technical problems of defining urban areas. (The term ‘urban areas’ is used in preference to particular words like ‘towns’ or ‘cities’, because it is the most general expression, without specific conventional connotations.) This is done in the first section. The second and more particular is to consider the applicability to Britain of one well-known definition of urban area: the Standard Metropolitan Statistical Area. This is done in the second section. The third section presents some results of the exercise; and a discussion of some policy conclusions rounds off the chapter.

The general problem

Definitions

That there is a problem and that it seems important, is evident from the literature. Statements are frequently made there about the percentage of the population of different nations that is urban, or about the growth of urban population in different nations as a percentage of their total population growth; but on inspection, such comparisons prove to depend on quite different national definitions of what is urban. In Denmark a place with 250 people is urban, in Korea a place with less than 40,000 is not (Hall, 1966, 19). So there is need for international standardisation. And even within any one nation, there is need for a closer definition of what is meant by an urban area.

In preparing this collection of essays, we have tried to ensure that a coherent theme runs through all the contributions, namely, the importance of the spatial dimension in the formulation of public policy. Though the volume deals explicitly only with Britain, we are confident that readers will recognise the general nature of the issues that are raised, which have not in the past received their due recognition.

Apart from the common interest of the authors in this particular field of enquiry, there is another bond that we should like to record. All the contributors pursued their undergraduate studies in geography at St Catharine's College, Cambridge, under the guidance of Mr. A. A. L. Caesar following his return to the College twenty years ago in 1951. It was under his guidance that all the contributors were first introduced to spatial policy problems of the British economy.

Over the last decade the primary fuel base of the British economy has been substantially widened. To the country's traditional and indigenous energy mainstay, coal, and its first modern partner, oil, has been added the power of the atom and of natural gas. In the process, and not withstanding the new-found hydrocarbon resources under the North Sea, the emotional reluctance of a time-honoured energy exporter for the first time to become dependent upon substantial energy imports has been to a large extent overcome. Meanwhile, the producer–supplier industries operating in the British energy market have steadily improved the efficiency of their production, processing, transport and distribution systems and have there-by come to enjoy falling real costs albeit in varying degrees. The market, as a consequence, has progressively developed a more competitive quality.

Setting the pace in these developments has been the oil industry. Taking advantage of the international industry's widening resource base – especially the development of low cost fields in North and West Africa – and the global persistence of excess production capacity relative to existing demands (Adelman, 1964 a, 1964b), it has been able to procure its crude-oil supplies in a market which has been characterised by a downward pressure upon field prices. Despite the rise in the general price level, the 1970 price for Persian Gulf crude oil f.o.b. stood at about one-third of what it was twenty-five years earlier.

In many ways, the most fundamental, striking and effective aspect of post-war government policies aimed at redressing inter-regional economic inequalities in Britain has been the promotion of industrial movement from the prosperous central regions to the lagging peripheral regions of the country. At the same time, what few coherent intra-regional planning policies are being implemented, notably for such areas as south east England, the Glasgow region, and the west Midlands, depend essentially upon the physical transfer of employment from congested central conurbations or cities to outlying settlements, including such famous new towns as Stevenage, Harlow, Cumbernauld and the incipient Milton Keynes. At both scales, national and regional, employment mobility is thus of crucial importance for successful implementation of government physical and economic planning policies, as well as for any prediction of the future configuration of the country's space-economy.

The validity of these assertions is evident not least from the spate of government legislation concerned with the promotion of employment mobility which has been enacted since 1945. Stimulated initially by the Barlow Commission Report on the Distribution of the Industrial Population (Royal Commission, 1940), and encouraged increasingly by political pressures from such peripheral areas as Scotland and Wales, successive post-war governments have placed on the statute book at least eight major Acts concerned wholly or in part with encouraging the movement of economic activity within Britain (Dowie, 1968; McCrone, 1969).

The last decade has witnessed a growing public interest in the geography of economic and social change. The emotions aroused by the prospective impact and cost of the London motorway box, the vigour with which Scottish and Welsh interests canvassed for an aluminium smelter and the conflicting interests exposed by the Roskill Commission's search for a site suitable as an international airport are common currency amongst informed opinion today. The issues may at first sight appear diverse. Yet underlying the continuing debate on what should be done to counter the persistently high levels of unemployment in the less prosperous parts of the country and the much publicised ‘drift to the south’, and surrounding the many arguments over shifts in local or sub-regional patterns of land use together with the frequently associated issues of environmental quality, there is a single unifying theme. This is the uncertainties, for individuals and corporate groups, inherent in making decisions concerning the spatial allocation of resources and activities. This new focus of interest and debate, which has given a renewed importance to the study and interpretation of geographical space, stems from the convergence of four developments in particular.

Social versus spatial inequalities

Great advances have been made in Britain during the present century in the attempt to equalise opportunities and conditions between different socio-economic classes. Poverty and inequality still exist, to a degree that varies according to the definitions employed.

The main purpose of this chapter is to explore one aspect of the relationships between certain long-term trends in the industrial and occupational structure of the population and the ecological structure of British cities. There are certain indications that the distinction between the skilled manual workers and the semi-skilled and unskilled manual workers is gaining a new significance in its urban spatial context. However, it is important to be clear at the outset that firm empirical evidence for the main outline of the argument is still lacking; a secondary objective of the chapter therefore is to argue for more research to be focussed upon this issue.

Changes in the industrial and occupational structure

It is initially worth adumbrating some of the well-established trends of changes in the country's occupational structure. Overall, it is evident that the total labourforce has been growing very slowly – at about 0.3 to 0.4 per cent per annum in England and Wales from 1931 to 1961. This rate of change masks more substantial changes at different levels, with more rapid rates of decline at the bottom of the occupational hierarchy and more rapid rates of increase at the top. The population as a whole is getting more skilled and the number of male semi-skilled and unskilled workers has declined absolutely in recent years (Knight, 1967). This decline in the proportion and number of less skilled workers is not taking place at the same pace between industries and occupational categories.

The residential differentiation of the urban population, a function of the axes of social differentiation extant in the society concerned, is reflected in a sifting and sorting of populations and locations. As the city develops typical patterns of differentiation become apparent. Different areas become associated with particular types of population and certain systematic relationships between geographical space and social space appear. The concern of the present chapter is with the spatial aspects of residential differentiation and, more particularly, with the validity of certain general models of this spatial structure which have appeared in the literature.

Discussions of the spatial aspects of urban structure generally concern themselves with three general models of urban form: the zonal, the sectoral, and the multiple nuclei. In the present case only the two former approaches will be examined. In contrast to the multiple nuclei model both the zonal and the sectoral analogies are concerned with the structural connotations of a particular set of differentiating processes, predict particular patterns of residential differentiation, and lend themselves readily to empirical test. The multiple nuclei ‘theory’ may be regarded as a caveat to the more general zonal and sectoral models.

THE ZONAL MODEL OF URBAN GROWTH AND STRUCTURE

The concern of the early Chicago ecologists with the differences in environment and in behaviour between different parts of the city led not only to descriptive studies of the ways of life to be found in particular natural areas, but also to a concern with the general features of urban structure.

The neighbourhoods which comprise the residential fabric of the city and form the framework for analyses of the relationship between locality and behaviour differ on innumerable grounds. In terms of physical structure they differ in density and age of development, in geographical position, and in types of dwellings. More importantly, from a behavioural perspective, their populations differ in age and sex composition, in occupations, incomes, and styles of life, in political, ethnic, and religious allegiances, and in a wide range of attitudes and behaviours. Each of these differences contributes its portion to the patterning of the urban residential system. In any attempt to understand this system it is necessary that each of these diverse differences should in some way be taken into account. The measurement problems which this involves form the major preoccupation of the present chapter.

UNIVERSES OF CONTENT, INDICANTS AND PROPERTIES

The conceptual tools which form the basic instruments of scientific endeavour may be classified under many rubrics. For our present purposes it is only necessary to distinguish between three such constructs: universes of content, indicants, and properties. In combination these three provide the basic tools for measurement and classification.

The universe of content

The universe of content forms the subject-matter of an analysis. It consists of the set of elements to which the procedures of measurement or classification are to be applied.

RESIDENTIAL DIFFERENTIATION

The urban community is neither an undifferentiated mass nor a haphazard collection of buildings and people. In the residential differentiation of the city the urban fabric comes to resemble a ‘mosaic of social worlds’. Similar populations cluster together and come to characterize their areas. As Park put it:

The residential differentiation of the urban population takes place in terms of many attributes and in many ways. Almost any criterion which can be used for differentiating between individuals and groups may become the basis for their physical separation. The process of separation may be accomplished through force, through a variety of sanctions, through a voluntary aggregation designed as a defence against unfamiliar ideas or customs or as an escape from persecution and discrimination, and through a selection of market forces. In much early town planning residential differentiation and segregation appear as prime characteristics.

It is characteristic of much work in urban ecology that while the fact of urban expansion is taken for granted, little attention is paid to other aspects of social or cultural change taking place in the wider society. All too frequently the urban community is taken as a phenomenon sui generis and considered in isolation. The hallmark of the macro-social approach to the study of residential differentiation within the city is that it attempts a rapprochement between urban ecology and the wider study of socio-cultural change. In the process an attempt is made to relate the bases of residential differentiation to the process of modernization.

Classical human ecology is a product of early twentieth-century Chicago. It is perhaps more than coincidental that the initial stimulus to the macrosocial approach to urban phenomena is a product of the burgeoning midtwentieth- century cities of the American West Coast, cities which have frequently been hailed as the harbingers of the ‘new urbanization’. There is an implicit assumption in much of the work produced under the aegis of classical human ecology that in some way the Chicago of the 1920s represents the typical case of the urban community. The experiences of the new communities of the West Coast and the growth of a comparative perspective in the social sciences, a development greatly speeded by the Second World War, have undermined the simplistic assumption that any single city can be typical of urban society and have caused a more searching inquiry to be made of the socio-cultural matrix in which urban communities are embedded.

Well-supported theories are still in short supply in the social sciences. Observations and hunches, on the other hand, are plentiful. The task of bringing theory and observation together is rendered difficult by the imprecision of many theoretical statements in the social scientific literature and by the dubious validity of many of the observations. It will be all too apparent that both faults are characteristic of much of the argument which forms the substance of the present work. Rather than an integrated theory of the residential differentiation of the urban population, complete with supporting evidence, what is presented should be regarded more as a series of essays towards such a goal. The generalizations offered frequently go beyond the evidence. This is particularly the case when attention is paid to the non-Western or pre-modern city. In the present state of the social sciences, however, it is felt that the dangers of over-generalization are less pronounced than are those of under-generalization. If nothing else, it is hoped that the essays will throw up a number of useful ‘Aunt Sallys’.

An overall summary of the book is provided in a brief last chapter. Chapters 1 and 2 are concerned with delimiting the nature and significance of residential differentiation. In the first chapter the notions of neighbourhood and of neighbourhood differences are explored in terms of a variety of human behaviours.

Like the society in which it exists, the modern city is highly differentiated. Different parts of the city are associated with different populations, with different opportunity structures, and with different reputations. The geographical framework of the city provides the basis for the emergence of a mosaic of social worlds. The increasing movement characteristic of modern society has almost certainly lessened the salience of location in the day-to-day lives of city-dwellers, but it remains the case that residence in one part of the city rather than in another has implication for a wide range of behaviours and biographies. The effects of location may be expected to be most pronounced on those whose daily movements are more or less confined to the bounds of their immediate neighbourhood – the young, the old, and the ‘care-takers’ – but the role of the local community in the initial socialization process and in the provision of a reference for social comparison purposes, ensures that the influence of the ‘neighbourhood effect’ is felt across a variety of activities and groups. Different populations have different relationships to their locale and areas which are suitable for one group may be quite unsuitable for another. The diversity of urban life and society both reflect and demand a diversified community.

The application of factor analytic techniques to urban residential differentiation has produced relatively consistent results. It seems rather well established that, at least in the Western world, much of the detailed variation in the characteristics of urban sub-communities may be interpreted in terms of three or four underlying constructs relating to differences in socio-economic status, in family composition, in ethnicity and in mobility. The demonstration of factorial invariance, however, is not to be equated with an explanation of factor structures. Having established the empirical validity of the major axes of residential differentiation the next step is to attempt the explanation of their significance and to examine the relationship they exhibit to other facets of human behaviour and social structure. It is to this task that the attention of this and of the next chapter are directed.

As is true in the case of many other sociological phenomena, the explanation of ecological structure may be attempted at either of two levels: a micro-social and a macro-social. In the micro-social approach attention is focused on the relationship between residential differentiation and patterns of individual decisions and behaviour. In the macro-social approach attention is focused on the relationship between residential differentiation and certain global characteristics of the encompassing society. A satisfactory theory of residential differentiation must include both approaches.