Explanations of Latin American development have come from a variety of perspectives, ranging from those that blame the victims to those that blame the system. Setting aside the former, which tend to attribute the lack of development in Latin America to the absence of an entrepreneurial spirit and that deal with symptoms rather than causes, we are left with those explanations that may be labeled standard [end p. 76] social science and those that come from a more radical tradition. The standard social scientific sxplanations will be mentioned briefly because they are more familiar, while those that come from the radical tradition will be more thoroughly compared and contrasted.

Development theories
Standard explanations based on structural-functionalist theory include general modernization and diffusion models (Bauer and Vamey, 1957), along with theories that are more explicitly spatial. These latter are based mainly on the Chicago ecological approach to land use (Schnore, 1965; Amato, 1970) or on models rooted in the distinction between formal and informal sectors (Perlman, 1976. Burgess (1981, 66) summarizes the problems in applying these explanations to Latin America. The underlying assumptions of the urban ecologists include "social heterogeneity, a commercial-industrial city, the predominance of private property, an equally available transport network throughout the city, cheap land on the periphery, and location governed only by the the rules of the market." But the previously built environment and topography, along with violations of many of the other assumptions of the Chicago ecological school, are serious enough to make the model untenable in Latin America. The cost of destroying what was built in the past and replacing it with a new built environment forms a significant constraint on present and future spatial organization of the city. Topography also shapes urban form, especially in many of the larger Latin American cities, given their coastal location and sometimes [end p. 77] rugged sites.

Land speculation, corruption, and squatter invasions negate the assumption that urban form is governed only by rules of the market. Equally available transportation also is an assumption that is violated. The inaccessibility of peripheral, poor populations to public transportation means that many of the poor must travel several hours or walk a long distance to reach their places of work. Absence of heavy industry and the presence of massive numbers of unemployed or underemployed people with no "effective" demand for housing negates more assumptions of the ecological school. Since scholars from this school view urbanization as a phenomenon sui generis valid for all societies, they find it difficult to analyze real cities operating in specific social and economic contexts. Their failure to consider explicitly development and underdevlopment compounds this difficulty.

Three related theories-- marginality (Germani, 1967), spatial dualism of the formal and informal sectors (Bromley, 1979), and the culture of poverty (Lewis, 1966)--have been proposed to explain the existence of large numbers of city dwellers not accounted for by the ecological school. These urban poor, who have little effective demand for standard housing and engage in a series of marginal economic activities when they work, live in settlements on the edges of Latin American cities or on undesirable lands. These lands may be hillsides (Rio de Janeiro), floodplains (Tegicugalpa), or garbage landfills (Lima). The evidence suggests however, that people in these settlements are integrated into the economic life of cities through relations of production and exchange and the urban land market (Lobo, 1982). [end p. 78]

Among the radical approaches, dependency theory has served as the point of reference against which other radical, mainly Marxist, have reacted. Little work has been done in the radical tradition to elaborate the relationship between Latin America's integration into the international economy and the internal spatial structure of its cities (Frank, 1969; Leeds, 1969). Burgess (1981) suggests that the dependency model's failure to explain urban structure results from a confusion of social and spatial processes. The relation between dependency and urban form is further obscured because the nature of dependency has not been clarified; the term has been used to describe a model, a theory, a framework, and a perspective.

At least two strands of dependency analysis should be recognized to understand its role in shaping thought about the relationship between Latin America and the rest of the world. The first strand comes from Raul Prebisch and the Economic Commission for Latin America (ECLA), which identified a decline in terms of trade between countries that produce raw materials and those that produce manufactured goods. Underdevelopment in Latin America is explained by showing the consequences of the transfer of surplus to the industrialized metropolis. The obvious policy prescription was to promote import-substitution that would result in a gradual elimination of this type of dependency. This effort was to be led by local capital in alliance with workers and peasants (Edelstein, 1981, 104). History has shown that this strand has not resulted in fruitful policy. A substantial amount of import-substitution industry has been foreign owned or controlled. Many of the inputs into industry (capital goods and even raw [end p. 79] materials) have to be imported. Other questions about the success of this strategy include the issue of scientific and technological dependence and debt service.

A radical dependency theory, appearing in the mid-1960s, was proposed by writers such as Andre Gunder Frank (1969) and Fernando Henrique Cardoso (1972) to challenge the "bourgeois" dependency theory just described. A major objection to the bourgeois theory was its view that the diffusion of capital and technology from the advanced capitalist countries to the 'less developed' countries leads to the destruction of feudal stagnation and the rise of universal capitalist development, prosperity and democracy" (Edlestein, 1981, 104). Frank held that it is the exchange relationship between capitalist industrialized countries and underdeveloped ones that causes their underdevelopment. In radical dependency theory, participation in the world economic system reinforced the dominance of an agricultural, mining, and commercial elite who, in combination with foreign interests, prevented an industrial capitalism from developing. Radical dependency theory, then, implies that the needs of the center shape the economy and the class structure of Latin America. Within this radical tradition, Frank (1969) promoted a stagnationist thesis. He suggested that Latin America cannot develop but will underdevelop.

The other position within radical dependency theory can be labeled strutural historiography. This position, associated with Cardoso (1972), attempts to account for differences in development in Latin America by describing how these countries have responded to different phases of imperialism. Since Frank's and Cardoso's dependency [end p. 80] theories emphasize the circulation of surplus (exchange), they have not focused attention on internal sources of change.

Another current within radical dependency approaches is represented by Wallerstein's (1975) world systems theory. Wallerstein's approach has led to several attempts to provide a more general theory of Latin American poltical economy.

One of these attempts has been labeled "modes of production analysis," which Henfrey (1981) and others have recently addressed. Chincilla and Dietz (1981, 143) have provided a summary of some of the more recent work. They suggest that four premises are part of the mode of production approach: 1) "Each national economy is a concrete, historically created social formation which is formed by the 'articulation' (interaction, linking, relationship) of two or more modes of production." The interaction of these modes influences the development of that society. 2) Each mode has reached different stages of development or decline and has different relationships with other modes in any given society. For example, differences between competitive capitalism and monopoly capitalism exist and will articulate with other modes of production in different ways. 3) Development is understood as capitalist development, which dominates other modes of production. 4) External influence is shaped by the internal structure (classes, the state) of a national economy. Both internal and external factors exert different forces at different times. For Deitz and Chinchilla and other analysts who consider modes of production as a perspective, a detailed study of a specific country is necessary to illluminate the interaction of these modes.

A second reaction to dependency approaches comes from Warren (1973) and Kay [end p. 81] (1975). Warren, especially, contends that capitalism is progressive and that imperialism implants capitalist relations of production and leads to expanding production. This view holds that the creation of an industrial proletariat that results from expanded production will lead to the overthrow of capitalism. Although from a Marxist tradition, this analysis has much in common with the earlier development theory of bourgeois economists who claimed that Latin America needed more capitalist development.

A third set of approaches reacting to dependency analysis is not easily labeled, but seems based more firmly in the analysis of the accumulation of capital. Various writers, including Barkin (1981), Petras (1981), and Harvey (1982), have discussed the aceumulation of capital in light of the social relations of production at both national and international levels. This approach emphasizes the development of the forces of production within a country rather than the unequal distribution of a surplus among nations analyzed by dependency proponents. The development of the forces of production accompanies an increasing proletariatization of the Latin American population. But this proletariatization is incomplete because many lose their own means of production, especially land, and most are not incorporated as workers into the modern industrial sector. This incomplete process leads to greater unemployment and misery (Barkin, 1981). This group of scholars stresses the importance of a class analysis, with a focus on the drive to accumulate and the relations of production that develop under capitalism.

The four different radical approaches examined here can be collapsed into two major versions of the development or [end p. 82] underdevelopment of Latin America. These two versions do not, however, represent a perfect taxonomy in which two of the four fall in one major group and two in the other. Rather, some proponents of the modes of production approach and some of those rooted in the accumulation of capital approach will be found in both versions. The first of these two larger categories may be labeled "uneven development." For the persons who advocate this view, underdevelopment in Latin America is fundamentally the same phenomenon found in lagging regions of the developed countries. For them, the eventual arrival of an advanced industrial economy is just a matter of time. Within this version, the clearest example is that of Kay and Warren. The second of these large categories may be labeled "underdevelopment." This view suggests that there is a fundamental difference, a qualitative gap, between lagging regions of advanced countries and those countries that are underdeveloped. As evidence, they cite the failure of any country to become a developed country under capitalist relations of production since Japan did so toward the end of the nineteenth century.

This brief summary presents an overview of some of the possible approaches being explored both outside and within geography. A recent exchange in The Professional Geographer (Reitsma, 1982; Smith, 1982; Doherty, 1983) emphasizes the need to go beyond dependency analyses.

Urban Spatial Structure
Relating one or the other of these theoretical approaches to the study of urban spatial structures in Latin America is a task that still needs to be done. Burgess (1981) [end p. 83] has explored urban residential structure in Latin America by using the underdevelopment approach. His writing emphasizes production, exchange, and consumption and uses a class analysis. He thus provides an introduction for understanding the relationships between the broader issue of Latin American development and urban spatial structure.

Castells (1973) also has outlined the relationship between urbanization and Latin America's international position. He identified three main stages. Each of the forms that results from the production system influenced the processes that shape the city in the next stage and also influenced the new form.

Initially, European penetration assumed functions similar to those of pre-Columbian cities that the European settlers and explorers conquered or replaced. These urban centers served as political-religious centers. Economic functions were added to these control functions, which, little by little, assumed greater importance. The first economic functions included serving as collecting points for wealth accumulated in the rural areas of the region. Merchant capital, existing prior to industrial capital, was the agent for the movement of this wealth to Europe.

The spatial structure of colonial Latin American cities reflected political, religious, and economic functions. The colonial city, planned according to a mandate of the Spanish government through the Law of the Indies, had a central plaza around which were located the civil, ecclesiatical, and military authorities. The upper classes resided in homes as near the central plaza as possible (Robinson and Swann, 1976). Artisans and craftsmen lived a little farther away, while the poor people, mainly Indians, [end p. 84] lived on the edge of the city (Stanislawski, 1947). Although spatial differentiation in land use occurred, the pre-industrial Latin American city is notable for its heterogeneity. This land use pattern grew more complex as independence from Spain was obtained and the twentieth century approached. Nevertheless, the cities remained compact units.

The second half of the nineteenth century brought momentous changes to Latin America. Growing demands for raw materials for the industrial and agricultural needs of Europe led to the incorporation of Latin America irto the world market system. Consequently, primacy of cities reached its peak at this time (Roberts, 1978). Part of the wealth created by workers extracting natural resources was siphoned off by urban-based exporters and by government in the form of tax duties. The wealth went to purchase imported manufactured goods and to the rapidly growing public sector. In a few countries, especially Argentina, Brazil, and Chile, the arrival of immigrants stimulated some industrial production.

The transformation of Latin America into a large-scale exporter of raw materials and an importer of manufactured consumer goods had a sginificant impact upon urban function and spatial structure. The functional and spatial heterogeneity of the colonial urban center began to yield to the functional and spatial separation and differentiation of land use characteristic of cities of the industrialized world. The separation of the individual from the means of production is symbolized by the separation of residence from workplace. However, in Latin America, only incipient industry, consisting of small workshops located near the center of the city, appeared (Morris, 1978). Thus, the [end p.85] degree of separation of land use never reached the levels of the advanced industrial countries.

From the middle of the nineteenth century to the early part of the twentieth century, the elites began to break their traditional pattern of central city location. Commercial expansion of el centro (the downtown), a process that resulted from increased wealth and import and export activity, spurred the wealthy to move to attractive sites. In Bogota, for example, coffee exports provided the wealth for the commercial expansion of el centro The wealthy began to move to the north and the poor to the south where less desirable environmental characteristics existed (Amato, 1970).

The spatial trends of the latter part of the nineteenth century continued until the middle of this century, when a third stage may be recognized. At this point, the import-substituting industry and later export-dependent industry accompanied a rapid urbanization process. Large factories demanded more space than the small workshops of the earlier incipient industry of Latin America. As the city expands, land becomes more expensive and larger factories that require more land move to peripheral locations. New industry, much of which is foreign owned, avoids high land costs by taking these peripheral sites for expansion. But the burden of accessibility is placed on the workers who mayor many not live in the immediate vicinity of the new plant. Industrial capital, maximizing its private benefits, inflicts great social costs on the population of the cities. Rapid accumulation of capital, as in Brazil, has led to pauperization of the working class. The workers are pushed to the far margins of the [end p. 86] cities where land is cheap, but where services are not readily available. Land speculation accompanies and even causes this decentralization of the working class because workers are forced beyond the urban fringe. Vacant land between workers and the city increases in price because the infrastructure, a result of worker and community demands, has increased the incomes of the property and construction sectors (Kowarick, 1980). The relative weakness of the working class in Latin America suggests that the decentralization of industry has responded to the needs for rapid accumulation rather than to the sharpening of the class struggle.

Dependency models propose that Latin America became capitalist, but leave little room for understanding the class structure that has developed there. Peasants expelled from the land by penetration of capitalist agriculture or by a growing population, but not absorbed as wage labor in either rural activities or industrial processes are not, at least in the classic Marxist sense, workers. Yet, they make up a large and growing share of the population in Latin American cities and constitute a large reserve army available for work at very low wages. A framework for treating these complexities needs to extend beyond dependence to one or several of the other analyses presented here. A start in this direction is provided by Gilbert (1982) in his collection of readings. However, a more explicit link between the nature of economic relationships and Latin American urban spatial structure is still necessary.

References cited
[end p. 87] Amato, P. "Elitism and Settlement Patterns in American City,"Journal of the American Institute of Planners Vol. 36 pp. 96-105.

Barkin, D. "Internationalization of Capital: An Alternative Approach," Latin American Perspectives Vol. 8 (1981), pp. 156-161.

Bauer, P.T. and B.S. Vamey. The economics of Underdeveloped countries (Chicago: University of Chicago Press, 1957).

Bromley, R. (ed.) The Urban Informal Sector (Oxford: Pergamon Press,

Burgess, R. "Ideology and Urban Residential Theory in Latin America," in H. and R.J. Johnson (eds.) Geography and the Urban Environment, vol. 4 (Chichester, England: Wiley, 1981), pp. 57-114.

Cardoso, F.E. "Dependency and Development in Latin America,"The New Left Review Vol. 74 (1972), pp. 83-95.

Castells, M. (ed.) Imperialismo u urbanización en América Latina (Barcelona: Gustavo Gille, 1973).

Chinchilla N. and J. Dietz Toward a new Understanding of Development and Underdevelopment," Latin American Perspectives Vol. 8 (1981), pp. 138-147.

Doherty, J. "Beyond dependency", Professional Geographer Vol. 35 (1983), pp. 81-83. [end p. 88]

Edelstein, J. "Dependency: A Special Theory Within Marxist Analysis," Latin American Perspectives Vol. 8 (1981), pp. 103-107.

Elbow, G.S. "Determinants of Land Use Change in Guatemalan Secondary Urban Centers," The Professional Geographer Vol. 35 (1983), pp. 57-65.

Frank, A.G. Latin America: Underdevelopment or Revolution (New York: Monthly Review Press, 1969).

Germani, G. "The City as an Integrating Mechanism: The Concept of Social Integration," in G.H. Beyer (ed.) The urban Explosion in Latin America (Ithaca, New York: Cornell University Press, 1967), pp. 175-214.

Gilbert, A. (ed.) Urbanization in Contemporary Latin America(London: Wiley, 1982).

Griffin, E.C. and L. Ford. "A Model of Latin American City Structure,"Geographical Review Vol. 70 (1980), pp. 397-422.

Harvey, D. The Limits to Capital (London:Basil Blackwell, 1982).

Henfrey, C. "Dependency, Modes of Production, and the Class Analysis of Latin America," Latin American Perspectives Vol. 8 (1981), pp. 17-54.

Kay, G. Development and underdevelopment: A Marxists analysis (London: Macmillan, 1975).

Kowarick, L. A espoliaçãoUrbana (São Paulo: Paz e Terra, 1980).

Leeds, A. "The significant variables [end p. 89] in Determining the Characteristics of Squatter Settlements," América Latina,Vol. 12 (1969) pp. 44-86.

Lewis, O. "The Culture of Poverty," Scientific American Vol. 215 (1966), pp. 19-25.

Lobo, S. A A house of my own( University of Arizona Press, 1982).

Morris, A. S. "Urban Growth Patterns in Latin America with Ilustrations from Caracas," Urban Studies Vol. 15 (1978), pp. 299-312.

Perlman, The myth of marginality (Univ. California Press Berkeley: 1976).

Petras, J. "Dependency and world system Theory, " Latin American Perspectives Vol. 8, 1981, pp. 148-155.

Reitsma, H-J. A. "Dependency geography, dependency relations and the Capitalist Scapegoat,"The Professional Geographer 34 (1982), pp.125-130.

Roberts, B. Cities of Peasants (London: Arnold, 1978).

Robinson, D. J. and M. M. Swann. "Geographical Interpretations of the Hispanic-American Colonial City," in R.J. Tata (ed.) Latin America: Search for Geographic Explanations (Chapel Hill, North Carolina: Conference of Latin Americanist Geographers, 1976), pp. 1-15.

Schnore, L. "On the spatial structure of [end p. 90] Cities in the Two Americas," in and L. Schnore (eds.) The study of urbanization (New York: Wiley, 1965), pp. 347-399.

Smith, N. "Theories of Underdevelopment: A Response to Reitsma," The Profesional Geographer Vol. 34 (1982), pp. 332-337.

Stanislawski, D. "Early Spanish Town Planning in the New World ," Geographical Review Vol. 37 (1947), pp.

Wallerstein, I. World Inequality: Origins and Perspectives on the World System (Montreal: Black Rose Books, 1975).

Warren, B. "Imperialism and Capitalist Industrialization", New Left Review (1973), pp. 3-44. [end p. 91]