You are on page 1of 12

Interdependent Urbanization in an Urban World: An Historical Overview Author(s): David Clark Source: The Geographical Journal, Vol.

164, No. 1 (Mar., 1998), pp. 85-95 Published by: Blackwell Publishing on behalf of The Royal Geographical Society (with the Institute of British Geographers) Stable URL: . Accessed: 08/06/2011 12:31
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . . Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact

Blackwell Publishing and The Royal Geographical Society (with the Institute of British Geographers) are collaborating with JSTOR to digitize, preserve and extend access to The Geographical Journal.

Vol. 164, No. 1, March 1998, pp. 85-95 TheGeographical Journal,

Interdependent Urbanization in an Urban World: an Historical Overview

DAVID CLARK Coventry Priory CoventyCV1 5FB. University, St, SubjectArea, Geography Thispaperwas acceptedfor in publication April1997 The distribution of the world's population is now more urban than rural. Contemporary and historical urban patterns are identified and their causes are evaluated. Urban development was largely confined to developed countries before mid-century but has spread to developing countries since. Both outcomes are seen as interdependent consequences of the growth and geographical extension of capitalism. The merits of the interdependency theory are assessed. Recent urbanization in Africa and Asia is a locational response to the new global economic order. Cities have grown because of the influx of manufacturing and service jobs from the developed economies, and the in-migration of workers displaced by agricultural adjustment. The prospects for further urbanization are considered.
KEY WORDS: world urban development,urbanization,capitalism,globalism,interdependencytheory.

HE LAST DECADE OF the twentieth century marks a major watershed in the evolution of human settlement,for it encompassesthe period during which the location of the world's people became more urban than rural (Clark, 1996). Variations among countries in the quality of their census data and in the ways in which urban areas are defined mean that it is not possible to be exact, but it is likely that 1996 was the year in which the figure of 50 per cent urban was achieved. Despite its symbolic significance, this historical event went largely unrecognized and unreported. More of the 5.4 billion inhabitants of the globe now live in urban settlements than in villages and hamlets.1 No longer are towns and cities exceptional settlement forms in predominantly rural societies. The world is an urban place. Urban development on this scale is a remarkable geographical phenomenon. Instead of being spread widely and thinly across the surface of the habitable earth, a population that is urban is one in which vast numbers of people are clustered together in very small areas. Levels of urbanization, however, are far from uniform (United Nations, 1991). They are high across the Americas, most of Europe, parts of western Asia and Australia (Fig. 1). South America is the most urban continent with the population in all but one of its countries (Guyana) being more urban than rural. More than 80 per cent of the population live in towns and cities in Venezuela, Uruguay, Chile and Argentina. Levels of urban development are low throughout most of Africa, South and East Asia (Brunn and Williams, 1993). Fewer than one person

in three in sub-Saharan Africa is an urban dweller. The figure is below 20 per cent in Ethiopia, Malawi, Uganda, Burkina Faso, Rwanda and Burundi. Despite the presence of some large cities, levels of urban development throughout South and South East Asia are low (Dogan and Kasarda, 1989; 1990; Chen and Heligman, 1994). An estimated 41 per cent of China's 1.2 billion people and 29 per cent of India's 0.96 billion lived in towns and cities in 1995. The Himalayan kingdom of Bhutan is reckoned to be the world's most rural sovereign state, with only six per cent of its population living in towns and cities. The urban world has emerged only very recently. Towns and cities have existed for over eight millennia, but fewer than three per cent of the world's population lived in urban places in 1800. According to Davis (1965; 1969) it was around 27 per cent in 1950, by which time most of the countries in what is now regarded as the developed world were predominantly urban (Fig. 2). Urbanization as a phenomenon that encompasses the majority of the world's population is a consequence of a massive rise in the percentage of the population that is urban in the developing world, especially in Africa and Asia (Gugler, 1988; Gilbert and Gugler, 1992). This shift in the locus of urban development raises far-reaching questions concerning the causes of recent and current urbanization in developing countries and its links with that in developed areas. This paper explores and attempts to explain such patterns and relationships in an historical context. It overviews the principal stages in the
? 1998 The Royal GeographicalSociety

Percentageof population urbanin 1995 ' O 19.9 20-39.9 40 -59.9" 60 - 79.9 80-100



--il . .. . . . . . .. .

~ ~ ~ ~
. ..

..... . .. . ... ... .. .. . . . . . . . . . .

. ... ... .. . . . . ... . . . ...... . . . . . . . .. . . . .. .. . . .. .. ... . . . . .... ..... . ... ..... . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . .

Percentageof population urbanin 1950

0- 19.9


20 - 39.9 60- 79.9

40-59.9 Ir
in that Fig.2. The of population wasurban 1950 percentageworld's



evolution of the urban world and suggests that they can be explained by an interdependency theory of global urban development. theoy Interdependency Urban development is the consequence of deepseated and persistent processes that enable and encourage people to amass in geographical space. Historically, two separate prerequisites were necessary: the generation of surplus products that sustain people in non-agricultural activities (Childe, 1950; Harvey, 1973), and the achievement of a level of social development that allows large communities to be socially viable and stable (Lampard, 1965). Urban historians suggest that these changes took place simultaneously in the Neolithic period when the first cities emerged in the Middle East (Wheatley, 1971). It is further thought that the volume of surplus product imposed a ceiling upon urban development in the pre-industrial society (Sjoberg, 1960). A second coincidence of economic and social change was associated with the rise of industrial capitalism in the late eighteenth century. This initiated powerful processes of urban growth and urbanization that led to the emergence of urban societies in Great Britain, North West Europe and North America (Pred, 1977). Although they may explain the principal historical turning points, theories of self-generated urbanization do not and cannot account for the recent urbanization of developing countries. This occurred in a world that was already partly urbanized, and the sheer scale and pace of the changes involved point to the operation of widespread and powerful non-local forces. Structuralist interpretations advanced by Wallerstein (1979) and Goldfrank (1979), and elaborated by Chase-Dunn (1989), Dicken (1992) and Taylor (1993), link recent changes in the roles and organization of the economies of developing countries to the growth and extension of capitalism in an emerging world system of nations. Urbanization can similarly be seen as an internal locational response to the absorption of such areas within an integrated global economy (King, 1990; Timberlake, 1984; 1987). Capitalism produces urbanization by concentrating production and consumption in locations that afford the greatest economies of scale, agglomeration and linkage, and where control over sources and supply can be exercised with maximum effectiveness, at least cost (Johnston, 1980). An important feature of this structuralistinterpretation is the emphasis that is placed on historical continuity. The urbanization of the developing world since 1950, and of the developed world before this date, have the same basic causes. They are interdependent consequences of the growth and expansion of capitalism. Structuralists see the spread of capitalism to the developing world as the most recent stage in the development of capitalism as an economic system

(Chase-Dunn, 1989). It is a result of changes in the ways in which wealth is accumulated, and the evolution of the world-system of nations (Table I). The former is a product of the sequential evolution of the prevailing economic formation from mercantilism, through industrial and monopoly capitalism, to transnational corporate capitalism (Castells, 1977; Goldfrank, 1979; Chase-Dunn, 1989). It has its own momentum in the form of the drive for ever-higher levels of output and profit through the development of new sources of wealth and units of production. It is characterized, according to proponents of the 'regulation school' (Boyer, 1990), by the periodic emergence of powerful social and cultural norms, such as Fordism and post-Fordism, which serve to regulate the inherently unstable course of accumulation (Jessop, 1990; 1992; Lipietz, 1992). The latter structural development is concerned with geopolitics and involves the division of the world into progressively larger spheres of economic association and exchange based upon changing space relations and systems of supply (Taylor, 1993). It is associated with the rise to economic and political dominance of a small group of core nations led by the USA as the foremost hegemonic power. Interdependency theory proposes a single explanation or interpretation for urbanization, whether in developed or in developing economies. It has echoes in dependency theory which explores and attempts to account for the links between development in core regions and underdevelopment in the periphery (Frank, 1967; 1969; Hette, 1990). Dependency theory suggests that underdevelopment is a result of the plunder and exploitation of peripheral economies by economic and political groups in core areas. Interdependency theory argues that urban development, wherever it occurs, is one of the spatial outcomes of capitalism. When seen from the developing world, most recent urbanization appears to be 'dependent', in the sense that it is introduced or imposed by the developed world. From a global perspective, however, all urbanization can be held to be interdependent in that it stems centrally from capitalism and its spatial relations. This is not to say that all urbanization has arisen in an identical way and is, therefore, the same in all countries. Capitalism has adopted different forms at different times, and is regulated in different ways, so producing spatially differentiated patterns of urban development at the global scale. The interdependency theory of global urban development can be criticized on four principal grounds. The first, in common with structuralist interpretations generally, is that it is stronger on suggested associations than on causal linkages. This is especially important given the debate between structuralists who see such links as arising directly from the mode of accumulation, and regulationists who




in Principal stages globalurban development 1780-1880 Modeof accumulation Economicformation Sourceof wealth unit of production Representative characteristics World-system Space relations Systemof supply Hegemonic powers Industrial capitalism Manufacturing Factory 1880-1950 Monopoly capitalism Manufacturing Multi-national corporation 1950Corporatecapitalism and Manufacturing services Transnational corporation, globalfactory Global Corporateimperialism USA 27 Africaand Asia New York,London,Tokyo

Atlanticbasin Interational Colonialism/imperialism State imperialism Britain Britain,USA 5 North-western Europe,the Americas,coastsof Empires London,New York

Urban consequences Level of urbanization startof period (%) 3 at Areasof urbanization Britain duringperiod Dominant cities London

trace their origins to social and cultural norms through which accumulation is regulated (Roberts, 1995; Painter, 1995). The fact that capitalism changed at a time of massive urbanization does not necessarilyimply a functional connection. Coincidence is not the same as causation and the mechanisms involved, which may vary over time and space, are matters for detailed empirical investigation and elaboration. A second reservation is that urbanization in the developing world lagged so far behind that in the developed world that it cannot be regarded as part of the same process. Britain was an urban industrial society for three-quarters of a century before any territory in what is now the developing world passed the 50 per cent urban threshold, and the urbanization of most of the developing world did not gather real momentum until after 1950. It is important, however, to place urbanization in its context of space and time. Global urbanization involves massive shifts in the distribution of population over a wide area and is inherently a slow process. It is perhaps no accident that self-sustaining urban development first occurred in Great Britain; a very small country where forces of urban growth were concentrated (Carter and Lewis, 1991). A sense of perspective is also important. When looking back over the last two centuries from the present, lags of a few decades appear to be of major significance. In the context of eight millennia of urban history they are trivial. A third criticism is that interdependency theory undervalues the rich traditions of urban development, supported by non-capitalist economic systems that existed in many developing countries. Highly successful urban civilizations existed in ancient Egypt, India, China, Cambodia, Peru, Mexico and Nigeria in states and economic systems that were religious, military or feudalistic in formation. Independency theory, however, recognizes the achievements of

non-capitalist economies, although it is argued that they were incidental to global urban development. Levels of productivity and surplus in early urban hearthlands were never high enough to facilitate selfsustaining urban development, and so their importance was localized. Rather than denying and devaluing their contribution, interdependency theory provides a powerful explanation as to why nonindustrial urban economies were not more successful. The final criticism is that capitalist theories do little more than state the obvious and often in a language that serves to obscure rather than to clarify. Capitalism is the prevailing economic formation in most countries. To say that it causes urbanization is to advance explanation and understanding very little as all social outcomes, both structural and spatial, are the products of capitalism. Such arguments have some validity at the most general level but they fail to distinguish between capitalism as an underlying principle and capitalism as a specific and evolving economic formation. The value of interdependency theory lies not in its foundations in capitalism per se, but in the links that it proposes between successive stages in the evolution of capitalism and urban development across the world. Theurbanization thedeveloped world of The extent to which urbanization in the developed and developing worlds is an interdependent consequence of the evolution of capitalism and its changing space relations becomes clear with historical analysis. A useful starting point is Weber's classic work on The growth of cities in the nineteenth century (1899). Mapping data of questionable quality, relating in some cases to long-forgotten countries, provides only the crudest of indications, but the limited extent of urbanization at the global scale is clear (Fig. 3). Only three areas in Great Britain, North-West Europe and the USA were more than 20 per cent



c 0 3
L o0 0O o s 0 c 0 cn0
) 0

co . cn T?0 0
0) CO 0

o o





urban in 1890. With less than three per cent of the world's population living in towns and cities, there was little or no urban development in most other territories. Urbanization in the developing world was restricted to concentrations of population around points of supply. The industries of the developed economies used domestic coal and iron ore to build and power machines to process cotton, sugar, jute, rubber, tobacco, wheat, tea and rice imported from imperial territories. The accumulation of these agricultural commodities led to limited urban development in developing countries that was not detectable at the global scale. 'Sao Paulo grew on the basis of coffee, Accra on cocoa, Calcutta on jute, cotton and textiles, and Buenos Aires on mutton, wool and cereals' (Gilbert and Gugler, 1992: 47). Urban development in association with agricultural supply similarly took place in the West Indies and Indonesia, Malaysia and the Far East (King, 1990). Although cities were established along the coasts of empire, these developments did little to change the overwhelmingly rural distribution of the local population. Industrial capitalism was succeeded by monopoly capitalism towards the end of the nineteenth century (Wallerstein, 1979). It was distinguished by a vastly increased scale of economic activity and the domination of newly-created international markets, within state-controlled empires, by a small number of producers in each sector. Monopoly capitalism emerged in response to the demand for products that was generated by the rapidly growing population of the industrial nations. This stimulated manufacturers to diversify from making heavy, crude products into the mass production of a wide range of consumer goods and services. Increased output occurred both because the core economies in Europe became more productive, and because the manufacturing belt of the USA attained core status alongside Britain, France, Germany and the Low Countries during the 1880s (Chase-Dunn, 1989). It was achieved through the consolidation of many factory enterprises into multinational corporations that typically engaged in many functions in many areas, both at home and in the periphery. Monopoly capitalism involved the ruthless exploitation of peripheral areas. The larger scale of industrial activity required the international sourcing of raw materials and marketing of manufactured products, so the success of the core regions became dependent on their ability to dominate and control overseas territories. This was either through formal imperialism, or else through corporate power and influence. Britain established itself as the leading imperial power after about 1880, when it increasingly drew its industrial raw materials including ores, oil and rubber from around the world and in return supplied its overseas possessions (in India, Africa, the

Far East and other territories) with railways, ships, machinery, arms and motor vehicles. Similarly, the USA rapidly became a major international player after 1909 when, symbolically, Selfridges store was opened in Oxford Street, London, at the very centre of the dominant power in the world economy (King, 1990). Thereafter, many major US corporations developed international spheres of operation. Monopoly capitalism produced further urban growth and urbanization in an expanded core, although urban development in the periphery remained limited. Precise comparison of the urban world in 1890 (Fig. 3) with that in 1950 (Fig. 2) is inappropriate because of the quality of the data, but the broad pattern of change is clear. Urbanization in the first half of the twentieth century occurred most rapidly and extensively in Europe, the Americas and Australasia. Most of the rest of the world was unaffected. Urban development between 1890 and 1950 is explained by processes of population concentration that were associated with the economic and political imperialism of the United States, Russia, the United Kingdom and France. High levels of urban development in Canada, South and Central America were a legacy of British trade and, more recently, corporate links with the USA. Limited urban development existed across the Russian empire in Asia, Central and Eastern Europe. Urbanization elsewhere in the periphery was largely a localized product of British and French imperialism. Although only a quarter of the population lived in urban places, the principal feature of the urban world, in 1950, was that the cycle of urbanization in the developed countries was, or was very nearly, complete (Davis, 1965). In most developing countries it had hardly begun. Theneweconomic order The developing world has urbanized since 1950 as a consequence of a new economic order resulting from the reorganization of production, labour, finance, service provision and competition, on a transnational basis. Over the past half century, an increasing share of production has been organized globally rather than within the narrow confines of nation-states or empires (UNCTC, 1993). Much production has shifted to the developing world both as a means of penetrating local markets and in order to use cheap labour to make goods for sale in the core economies and elsewhere (Frobelet al., 1980; Sit, 1993). Examples include electronic goods, drugs, motor vehicles, clothing, machine tools and domestic appliances. At the same time, several countries in the developing world have expanded their manufacturing capabilities, and the firms in these newly-industrialized economies have captured markets for their products in the developed world (Lo, 1994). The production of some foodstuffs has also been reorganized on a



commercial basis so that it can be exchanged globally. Domestic agricultural production in many developing countries has been replaced by production for export, a beneficial consequence, as far as global capitalists are concerned, being that it generates currency that can be used to purchase more imports and so increase external dependency. The transnationalization of production involves the manufacture of global products, with global brand names, which are assembled across the world from components made in a number of countries (Dicken, 1992; Dunning, 1992). It is achieved by direct investment by firms from the core economies in developing countries where they can take advantage of large pools of very cheap labour. A new pattern of specialization has emerged that owes less to traditional distinctions between core and periphery and more to the jobs that workers perform within transnational corporate empires. The basis of the new international division of labour is the direct employment of large numbers of workers in low-cost overseas territories to perform standard production tasks (Cohen, 1981; Feagin and Smith, 1987). Rather than peripheral supply and core area processing, which was the pattern under industrial and monopoly capitalism, the new economic order is one of peripheral production and manufacturing, and core area research, development, design, administration and control (Castells, 1992). This pattern was made possible by, and in turn gave rise to, a new pattern of international finance. A global system of supply and circulation has emerged in recent years in place of the bilateral funding arrangements, tied to trading blocs and dominated by governments, that existed at mid-century. The new system is directed and controlled by the economies of the developed world through a small number of powerful banks, finance houses and exchanges which rank alongside transnational corporations as global institutions (Thrift, 1987;1989). The world cities in which they are located are the command and control points of the global economy (Sassen, 1994; Knox and Taylor, 1994). Developments in production and finance are supported by the growth of the international service economy (Daniels, 1991; 1993). Service activities that were once domestically bound have reorganized on an international basis in order to serve the needs of businesses operating across the globe (Warf, 1989). This trend is reflected in the rise of the advanced producer services sector, which includes insurance, accountancy, real estate, legal, advertising, research and development, public relations and management consultancy firms. Global business is further facilitated by means of the organization of employee services, including hotel accommodation, car hire and personal finance, on an international basis. The new economic order emerged alongside, as

part-cause and part-consequence, of a new political geography (Taylor, 1993). By far the most important feature was the ending of imperialism by Britain, France, Belgium and the Netherlands and the attainment of political independence by many colonial territoriesin Africa and Asia between 1950 and 1980 (Corbridge, 1993). This added further changes to the political map, which had been transformed during the 1940s by the post-war redrawing of boundaries in Europe and by the withdrawal of the British from the Indian sub-continent. Together these developments produced a large number of new nation-states that were keen to participate in the world economy in order to enjoy the benefits of trade and Aid. The new pattern was created in conditions of relative peace and prosperity, certainly in comparison with those that prevailed in the previous half century with its two world wars and numerous regional conflicts. Theurbanization thedeveloping world of The new economic order is principally responsible for the recent rapid urbanization of the periphery (Timberlake, 1984; 1987). Transnational corporate capitalism produced and is producing urbanization in the periphery both directly, as a consequence of urban growth in response to localized investment, and indirectly, through its impact on traditional patterns of production and employment. The former arises because economic exchanges between core and periphery are spatially focused and so lead to a concentration of globally-related economic activity in urban places. Cities, especially national capitals and those with major ports or international airports, offer overwhelming advantages for profitable investment, affording wide access to cheap labour and to domestic markets. Such places are typically the major and in some cases the only centres in the country for large-scale industry, hospitals, universities, media services and facilities for sport and the arts. As cosmopolitan centres with good external connections they are attractive to corporate managers and specialist workers on overseas postings. They are likely to be the home base of local elites that shape behaviour and consumption patterns towards which others in the country aspire. The urban concentration of foreign investment-led economic activity is high across much of the periphery. In Indonesia, Forbes and Thrift (1987) found that overseas investment was largely restricted to the area aroundJakarta where all major foreign corporations had their headquarters. Abidjan, the capital of the Ivory Coast, has 15 per cent of the national population but accounts for more than 70 per cent of all economic and commercial transactions in the country. Bangkok accounts for 86 per cent of Gross National Product in banking, insurance and real estate, and 74 per cent of manufacturing, but has only 13 per cent of Thailand's population. Lagos,



with 5 per cent of Nigeria's population, accounts for commodities and the import of consumer goods. 57 per cent of total Value Added in manufacturing Cities were dominated by the White minority in the and has 40 per cent of the nation's highly skilled country, and other than those employed in domestic labour (Kasarda and Parnell, 1993). service and a very small number in industry and serUrbanization is also taking place as an indirect vice activities, Blacks were prohibited unless they had consequence of the impact of transnational corporate a job and accommodation. In the countryside, some capitalism upon the economies of developing coun- Blacks worked for White farmers but most were tries. The central argument here is that adjustments engaged in subsistence agriculture. The population in economic structure are enforced as the price, or was 17 per cent urban in 1970. The favouring of the penalty, for incorporation within the world economy. White colonialists, however, meant that social and These lead to the release of large numbers of workers health care services were city-based, and significant from traditional occupations, who flock into the differences in standards of provision existed between towns and cities and so contribute to urban growth urban and rural areas. This basic pattern was transformed during the and urbanization. Peasant farmers are foremost amongst those whose livelihoods are undermined by 1970s as a consequence of increased foreign investthe drive for production of goods that will generate ment and the opening up of external markets for the foreign currency, both to help reduce national products of Zimbabwe's farms and factories. indebtedness and to enable governments to acquire Urbanization occurred through net in-migration to the symbols of statehood such as grand presidential jobs in cities, as the manufacturing sector increased palaces and national airlines. Many have been its contribution to the Gross National Product from displaced from their traditional lands and means of 10 per cent in 1965 to 24 per cent in 1980 subsistence by the introduction of commercial agri- (Stoneman, 1979). At the same time, the mechanizaculture that is geared to the production of exotic tion of many of the larger commercial farms, and fruits, flowers and out-of-season vegetables for devel- their increase in size, generated a surplus of Black oped world consumers (Susman, 1989). They include labour in rural areas. Movement into the cities large numbers of the very poor who have no alterna- increased significantly after 1980 when the legislation tive sources of employment and must look to the city permitting ownership and residence in cities was for survival. Droughts and civil wars, especially in relaxed and removed. Many traditionally White parts of Africa have further undermined the viability areas of Zimbabwe's cities rapidly became Black of traditional farming, leading to increased rural (Cumming, 1990). Urban growth was compounded urban migration. when families were reunited and birth rates rose. The policies of post-colonial governments stimu- Some 31 per cent of the population was thought to late urban growth by further enhancing the attrac- live in urban places in 1995 and the population of tiveness of towns and cities at the expense of rural Greater Harare was in excess of 1.5 million. The areas (Auty, 1995). One way is through the exagger- recent rapid urbanization in Zimbabwe, in common ated bias of government expenditures on infrastruc- with many African and Asian countries, is a conseture and services in favour of urban areas. Another is quence of structural and associated spatial changes the higher wage rates and better employment protec- that are associated with the transformation of a rural tion that exist in cities because urban workers are subsistence into an urban-based and politically indeorganized into trade unions. A third is the decline in pendent commercial economy, which is incorporated the demand for locally-produced staples as urban within the global economic system. Detailed evidence on the links between global consumers develop a taste for imported food items. Such policies are creating 'backwash urbanization' production and urbanization in developing countries by destroying the vigour of rural areas and suffocat- is presently fragmentary. The research that has been ing the cities with the burden of the human casualties undertaken points to the existence of a general this process creates. The implications are seen in the relationship between incorporation within a global rapid growth and dire social and environmental corporate capitalist economy and urban growth, but conditions of many cities and others in the develo- with wide variations from country to country. ping world that are swamped by large numbers of Taiwan, Singapore and Korea, where there is a clear in-migrants looking for work and welfare (Berry, connection, and China, where urban development is 1973). largely a consequence of rural changes associated Many of the urban consequences of the absorption with economic liberalization, perhaps represent the into the global economy are exemplified by extremes. It is important also to distinguish within Zimbabwe, a country that attained formal sover- the periphery between experiences in South eignty in 1980 after 15 years of unilaterally declared America, where levels of urban development are independence (Drakakis-Smith, 1992). The modern historically high, and Africa and Asia, where they are urban system in Zimbabwe emerged under settler low. Both regions have been affected by the same colonialism to facilitate the export of various adjustments associated with the emergence of the



new economic order, but with different consequences. The effects in the former have largely been to consolidate existing urban patterns by compounding growth in existing centres. In Africa and Asia they have created and accelerated urban development where little existed before. Conclusions This paper has identified and attempted to account for the processes responsible for the creation of the contemporary urban world. The urbanization of developed countries took place before mid-century, and in the developing world has occurred since, but it is argued that the causes are similar and related. Both stem interdependently from the advance of capitalism and its spatial relations. The settlement patterns in most developing countries have been transformed in recent years as external investments have created jobs in cities and as workers, displaced from the land because of the switch from subsistence to commercial agriculture, have migrated to urban areas. Such changes are seen as consequences of the progressive incorporation of their economies within the global corporate capitalist economy. Attempts to explain global patterns raise many contentious issues that merit wider consideration. Theorists will debate the concept of 'underdevelopment', the reasons for the rise and reproduction of capitalism, and the extent to which individual countries or parts of countries in Africa and Asia are incorporated within the global economy and world system of nations. They may suggest that the mode of regulation is a more important factor in urbaniza-

tion than the stage of capital accumulation. Associations between global change and urbanization have been explored for evidence of the validity of the interdependency theory, but further research is required before the links can be regarded as concrete and causal rather than ephemeral and coincidental. Empiricists will query the reliability of urban data, the definition of urban places and the extent to which urbanization is a localized or widespread phenomenon within countries. Urbanization represents the largest shift in the distribution of population in history. Such are its complexities that many will question the purpose and value of trying to make meaningful statements about the location of over 2.7 billion people. The need for high-order generalization and explanation, however, is likely to increase as the pace of urban change quickens, especially in Africa and Asia, with farreaching implications for environmental sustainability and social welfare. It took over eight millennia for half the world's population to become urban. Present predictions suggest that it will take less than 80 years for this process to encompass most of the remainder. Endnote 1This paper is based on estimates of urban populations abstracted from the United Nations 1991 report on World urbanization prospects.The highly variable quality and reliability of world urban data are emphasized in the United Nations' Demographic for deveyearbook 1994 and in the World Bank's World for 1993. For a general discussion see lopment report Goldstein (1994).

resources, of policyandeconomic Auty, R. 1995 Patterns development: London:Arnold. growth. New consequences ofurbanization. York: Berry,BJ.L. 1973 Thehuman St Martin'sPress. introduction. York: New school: Boyer,R. 1990 Theregulation a critical Columbia UniversityPress. world Brunn, S.D. and Williams,J.F. 1993 Citiesof the world: urban New development. York:Harper Collins. regional Carter, H. and Lewis, C.R. 1991 An urban of geography England in London: EdwardArnold. and Wales thenineteenth century. a London: Castells,M. 1977 Theurban approach. question:Marxist EdwardArnold. society cities,theinformational andtheglobalecon-, 1992 European Amsterdam:Centre for MetropolitanResearch. omy. structures the worldChase-Dunn, C. 1989 Global formation: of London:Blackwell. economy. Chen, N.P. and Heligman, L. 1994 Growth of the world's megalopolises.In Fuchs, RJ., Brennan, E., Chamie,J., Lo, F. and Uitto, J.I. (eds) Mega-city growthand thefiture. New York:United Nations UniversityPress: 17-31. Childe, V.G. 1950 The urban revolution Tn Plann.Rev. 21: 3-17. world: Clark,D. 1996 Urban globalcity.London: Routledge. Cohen, R.B. 1981 The new international division of labour; multinational corporations and the urban hierarchy. In and Dear, M. and Scott, AJ. (eds) Urbanization urban planning in a capitalist London:Methuen: 287-315. society. Corbridge, S. 1993 Colonialism, post-colonialism and the political geographyof the Third World. In Taylor, PJ. (ed.) Political London: Bellhaven: century. of geography the twentieth 58-89. Cumming, S.D. 1990 Post-colonialurban residentialchange in Zimbabwe: a case study. In Potter, R.B. and Salau, A.T. and in the London:Mansell: (eds)Cities development ThirdWorld. 32-50. and internaDaniels, P.W. 1991 Services metropolitan development: London:Routledge. tionalperspectives. industries theworld in Oxford:Blackwell. -, 1993 Service economy. Davis, K. 1965 The urbanization of the human population Scient. 213: 40-53. Am. urbanization. Angeles:Universityof California. Los -, 1969 World of Dicken, P. 1992 Globalshift: the internationalisationeconomic London:Paul Chapman. activity. era. 1: Dogan, M. and Kasarda,J.D. 1989 Themetropolis Volume a world cities. London: Sage. ofgreat era. London: Sage. 2: , 1990 Themetropolis Volume mega-cities. in Drakakis-Smith,D. 1992 Urbanand regional change Southern Afiica.London:Routledge.



and Dunning,J.H. 1992 Multinational enterprises theglobaleconomy. J.I. (eds) Mega-city growthand thefuture.New York: United Nations UniversityPress: 103-30. Reading, MA: Addison-Wesley. Feagin,J.R. and Smith, M.P. 1987 Cities and the new interna- Painter, J. 1995 The regulatory state: the corporate welfare state and beyond. InJohnston, RJ., Taylor, PJ. and Watts, tional division of labour: an overview. In Smith, M.P. and city. of MJ. (eds) Geographies global change.Oxford: Blackwell: Feagin,J.R. (eds) Thecapitalist Oxford:Blackwell:1-37. 127-44. Forbes, D. and Thrift, NJ. 1987 Internationalimpacts on the economies. London: urbanisationprocess in the Asian region: a review. In Fuchs, Pred, A.R. 1977 City systemsin advanced and Hutchinson. RJ., Jones, G.W. and Pernia, E.M. (eds) Urbanization in urban Asia.Boulder, CO: Westview:23-42. Roberts, S.M. 1995 Global regulationand trans-stateorganisapolicies Pacfic and in Latin tion. In Johnston, RJ., Taylor, PJ. and Watts, MJ. (eds) Frank,A.G. 1967 Capitalism underdevelopment America Oxford:Blackwell:111-27. studiesof Chile and Brazil. New York: Monthly historical Geographies ofglobal change. Review Press. London: Pine Forge Sassen, S. 1994 Citiesin a worldeconomy. or revolution. York: New Press. -, 1969 LatinAmerica: underdevelopment Sit, V.F.S. 1993 Transnational capital flows, foreign investMonthly Review Press. ments and urban growth in developing countries. In Frobel, F., Heinrichs,J. and Kreye, 0. 1980 The new internadivision labour. tional Kasarda, J.D. and Parnell, A.M. (eds) Third Worldcities. of Cambridge:C.U.P. Gilbert, A. and Gugler, J. 1992 Cities, povert and development. London: Sage: 180-98. Oxford:O.U.P. city. Sjoberg,G. 1960 The pre-industrial Dubque, IA: Free Press. Goldfrank, 1979 The world-system capitalism and present. Stoneman, C. 1979 Foreign capital and the reconstructionof of past ZimbabweRev. Afr.Polit.Econ.11: 62-83. BeverleyHills, CA: Sage. Goldstein, S. 1994 Demographic issues and data needs for Susman, P. 1989 Exporting the crisis:US agricultureand the Third World Econ.Geogr. 293-313. 65: megacity research.In Fuchs, RJ., Brennan, E., Chamie,J., and New nation-state and Lo, F. and Uitto,J.I. (eds)Mega-city growth thefuture. Taylor, PJ. 1993 Political geography: world-economy, localiy.London: Longman. York:United Nations UniversityPress:32-61. of Gugler, J. 1988 The urbanisation the Third World.Oxford: Thrift, NJ. 1987 The fixers: the urban geography of international commercialcapital. In Henderson,J. and Castells,M. O.U.P. and London: Sage: justiceand thecity.London: Edward restructuring territorial (eds)Global development. Harvey, D.W. 1973 Social Arnold. 203-33. and worlds.London: -, 1989 The geographyof internationaleconomic disorder.In theory the three Hette, B. 1990 Development in Oxford: Johnston, RJ. and Taylor, PJ. (eds)A world crisis. Longman. Blackwell:16-79. Jessop, B. 1990 Regulation theories in retrospectand prospect Econ.Soc.19: 153-216. in Timberlake, M. (ed.) 1984 Urbanization the world-economy. London:Academic Press. -, 1992 Fordism and post-Fordism:a critical reformulation. In Storper,M. and Scott, AJ. (eds)Pathways industrialisation-, 1987 World-systemtheory and the study of comparative to London:Routledge:46-69. urbanization.In Smith, M.P. and Feagin,J.R. (eds) Thecapiandregional development. talistciy. Oxford:Blackwell:37-64. for Johnston, RJ. 1980 Cityandsociey:an outline urban geography. United Nations, 1991 Worldurbanization Harmondsworth: New York: prospects. Penguin. United Nations. Kasarda, J.D. and Parnell, A.M. 1993 Third Worldcities. London: Sage. New -, 1994 Demographicyearbook. York:United Nations. colonialism the world-economy.UNCTC, 1993 World and investment 1993: transnational report, corporaKing, A.D. 1990 Urbanism, and urban London: tionsand integrated Cultural spatialfoundationstheworld international New York: United system. of production. Nations. Routledge. citiesin a world Wallerstein, 1979 Thecapitalist I. New Knox, P.L. and Taylor, PJ. (eds) 1994 World world-economy. York:C.U.P. Warf, B. 1989 Telecommunications and the globalisation of system. Cambridge:C.U.P. financialservicesProf.Geogr. 257-71. 41: Lampard, E.E. 1965 Historical aspects of urbanisation. In New century. Hauser, P.M. and Schnore, L.F. (eds) Thestudy urbanisation. Weber, A.F. 1899 Thegrowth citiesin thenineteenth of of York: Macmillan. 1962 reprint. Ithaca: Cornell University Chichester: John Wiley and Sons: 519-54. a order: Press. ecology Lipietz, A. 1992 Towards neweconomic post-Fordism, anddemocracy. Wheatley, P. 1971 Thepivot of thefour quarters. Cambridge:Polity Press. Chicago: The Lo, F. 1994 The impacts of current global adjustment and Universityof Chicago Press. New York: World shiftingtechno-economicparadigmin the world city system. World Bank, 1993 World development report. In Fuchs, RJ., Brennan, E., Chamie, J., Lo, F. and Uitto, Bank.