Dependency Theory: A Critique

Author(s): Harriet Friedmann and Jack Wayne

Source: The Canadian Journal of Sociology / Cahiers canadiens de sociologie, Vol. 2, No. 4
(Autumn, 1977), pp. 399-416
Published by: Canadian Journal of Sociology
Stable URL: https://www.jstor.org/stable/3340297
Accessed: 08-12-2018 19:08 UTC

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 support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Canadian Journal of Sociology is collaborating with JSTOR to digitize, preserve and extend
access to The Canadian Journal of Sociology / Cahiers canadiens de sociologie

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
 Dependency theory: a critique

Harriet Friedmann Jack Wayne

University of Western Ontario University of Toronto

Abstract. Although dependency theory represents an advance over evolutionary theor

development and underdevelopment, it has confronted serious difficulties in attempting
social relationships based on geographical observations and in attempting to construct a
the world capitalist system. It is argued that these difficulties are rooted in four basic as
dependency theory: (1) the basic unit of the world system is seen as a dyadic relation be
nations; (2) models of the world system are cumulations of these units into hierarchicall
sets of roles; (3) geographical and social relationships are collapsed in description, preven
independent analysis of the latter; and (4) basic relations among units of the system are
exchange instead of production. The major contribution of dependency theory - the pow
description of consequences within dependent regions of dynamics within the world sys
be enhanced, and the difficulties resolved, through its careful integration with Marxian
capitalism and imperialism.

Resume. Bien que la theorie de la dependance fait figure de progres quand on la comp
theories evolutionnaires du d6veloppement et du sous-developpement, elle s'est heurtee a
serieuses difficultes en essayant d'analyser les rapports sociaux bas6s sur des observation
g6ographiques et d'edifier un modele du systeme capitaliste mondial. II est prouv6 que ce
difficultes sont reli6es a quatre aspects fondamentaux de la th6orie de la dependance A s
l'unite de base du systeme mondial est percue en temps que relation compl6mentaire ent
nations; (2) les modeles du systeme mondial sont des accumulations de ces unites dans d
ensembles de roles ordonnes de faCon hierarchique; (3) les rapports g6ographiques et soci
reduits a une simple description empechant toute analyse independante de cette derniere
rapports de base entre les unites du systeme sont ceux de 'echange et non de la productio
L'apport principal de la theorie de la dependance - soit une description poussee des co
de la dynamique du systeme mondial au sein des regions dependantes - peut etre mise en
les difficultes aplanies, grice a une integration prudente des theories marxistes portant su
capitalisme et l'imperialisme.

399

Canadian Journal of Sociology/Cahiers canadiens de sociologie 2(4)1977

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
I. Introduction
Until the last decade sociological and other social scientific work on
development and underdevelopment was carried out largely within a model
that posited a sequence of evolutionary stages for societies. These stages were
identified for the economy (Rostow, 1960), for the polity (Almond and
Coleman, 1960), and for the structure of social relationships within the society
(Parsons, 1966). In this body of work the "prime mover" of social change was
sought in the characteristics of the society itself. When "underdeveloped"
societies were examined, the underdevelopment was seen as a consequence of
specified attributes of the population, often at the level of cultural practices and
religious beliefs (McClelland, 1962; Banfield, 1958).
In recent years there has been a serious questioning of the evolutionary
approach. Critiques such as those by Frank (1969b) and Brett (1974) have led
to a search for a new theoretical framework for interpreting and explaining
development and underdevelopment. The most consequential result of this
search has been the emergence of a new model, generally called dependency
theory. Special issues of journals (Latin American Perspectives 1, 1; Bulletin of
the Institute of Development Studies 3, 4; and Social and Economic Studies 22,
1) have been devoted to the examination of dependency theory, and this
perspective has generated a great deal of comment in scholarly and academic
circles.
The attractiveness of the dependency approach, in our view, stems from the
fact that it represents an advance over the earlier model in four major ways:

1. Unlike the evolutionary model, dependency theory does not assume the
"society" or "nation" to be a self-contained unit. Rather it recognizes the fact
that there are political and economic relationships among social formations
and that these relationships have different consequences for each of those
formations.

2. Dependency theory attempts to be holistic. Whereas in the earlier,

evolutionary model there was a division of labor among economists, political
scientists, and sociologists, each looking at different characteristics of the unit,
dependency theorists have been arguing forcefully (Frank, 1969b; Amin, 1974)
for an integration of the social sciences and their subject matter.

3. While evolutionary theory was idealist in its attempts to explain social

change (cf. Parsons, 1966), dependency theory has looked closely at the
material bases of organized social life, in particular at the growth and extension
of the world capitalist system.

4. At its best, dependency theory has been historically specific, replacing the
earlier evolutionary stages (which have clearly never been experienced by most
of the social formations in the world) with concrete analyses of historical
material.

These characteristics of analysis may be found in the work of Andre Gunder

Frank, the most influential of the dependency theorists. Frank has argued that
there has been a "development of underdevelopment" outside of Western
Europe, North America, and Japan as a consequence of emerging relationships
among political-economic formations which were brought into being as
capitalism expanded. Thus, according to Frank,

400

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
.. the expansion of the capitalist system over the past centuries effectively and entirely penetrated
even the apparently most isolated sectors of the underdeveloped world. Therefore the economic,
political, social, and cultural institutions and relations we now observe there are the products of the
historical development of the capitalist system no less than are the seemingly more modern or
capitalist features of the national metropoles of these underdeveloped countries. (1969b:5)

In other words, for the period of time that capitalism has formed the dominant
mode of production and distribution in the world, the most important fact
about any nation has been its relations with other nations. The modern history
of any given country, Frank asserts, makes sense only as part of the history of
the organized world system that capitalism historically has created. As a result,
any observations of contemporary societies must be explained in terms of their
historical relations to other societies. If they appear backward now, their
relationships are the cause, just as these same relationships are the cause of
what has been labelled "development" in other societies.
Frank's unit of analysis is not the individual society, but the relations
between societies. The set of all such relationships forms a world system, and
the holistic study of this system is the place to begin in studying its
consequences for individual countries. If the relationship between two societies
results in inequalities, it makes sense to study the nature of the relationship in
order to understand both wealth and poverty; neither can be understood alone,
in its own terms, apart from the one relationship which simultaneously brought
both about. This is what Frank (1968:32-39) means by structuralism as a
measure of theoretical adequacy.
Further, Frank and other dependency theorists have argued that the main
relatonship between societies has been an exploitative one, simultaneously
creating wealth at one of its poles, the metropolis (Frank), or the centre
(Galtung, 1972; Amin, 1974), and poverty at the other pole, the satellite
(Frank), or the periphery (Galtung, 1972; Amin, 1971; Wallerstein, 1973).
Furthermore, according to Frank and Amin, such developing-
underdeveloping, or exploitative, relations also exist at levels other than the
nation state, such as sub-national or cross-national regions.
The value of the dependency approach, then, lies in its recognition that
development and underdevelopment have taken place in the context of the
growth of capitalism as a world system. The approach usefully analyzes
relationships between nations and sees both development and
underdevelopment as historically observable consequences of those
relationships, and it attempts to be holistic in its perspective. The particular
success of the approach lies in its view of underdevelopment as a product of the
domination of one national economy by another. A primary mechanism in all
cases of underdevelopment, we learn in this literature, has been the emergence
of a dominant social class within the dependent underdeveloping nation, which
participates in the exploitation of the nation but is itself dependent on the
metropolis or centre. It is in the tension between the concepts of inter-class and
international relations that problems first arise.

II. The critique of dependency theory

While dependency theory provides a useful starting point for analyzing the
structural changes, particularly the changing class structure, created within an
underdeveloping nation or region (Wayne, 1975; Amin, 1971), its difficulties
become apparent once one moves beyond these historically observable

401

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
consequences of the dependency relation within individual satellites or
dependencies. Frank (and many of the other theorists) set themselves the task
of analyzing the world system of relations in order to understand the causes and
structures of underdevelopment. As Frank argues, only a complete, holistic
analysis will be an effective guide to the understanding of exploitative,
underdeveloping relations among national economies (Frank, 1969b:32-39).
Having set these standards for itself, however, the dependency literature has
not lived up to them. Beginning with assumptions and data based on dyadic
relationships between nations, dependency theorists have come up against
formidable obstacles in their attempts to produce an adequate model of the
world social and economic structure. In the dependency perspective, the world
system is seen as (1) a cumulation of bilateral relationships between nations, (2)
a hierarchically ordered set of roles, (3) a confused product of relationships
which are in part geographical and in part social, and (4) a product of exchange
relations rather than relations of production. Let us take each of these points in
turn.

(1) Bilateral relations. Frank (1969a:67-73) describes Chile's change from

being a satellite of Spain to being a satellite of England when the English allied
themselves with the Chilean landowning class. This change was predicated on
Spain's own transformation from world metropolis to satellite. How did this
transformation in Spain's structural position come about? Frank leaves this
transformation unanalyzed.
Samir Amin (1971:109) finds West Africa underdevelopment affected by
"the shift of centre of gravity of trade in Africa from the Savannah hinterland
to the coast" which was itself "a direct consequence of the change of centre of
gravity in Europe, from the Mediterranean to the Atlantic." At the same time
there was a shift to mercantile capitalism. Why did the centre of gravity shift
within Europe? How did mercantile capitalism become the most prominent
way of doing business? These questions are unanswered in Amin's account.
The work of the dependency theorists provides compelling evidence of the
descriptive power of the underdevelopment and dependence framework: it
leads to questions about the historical transformation of class relations within
dependent societies as they are incorporated into a changing world capitalist
system (Frank, 1972:30), questions which cannot be asked within the
functionalist, evolutionary framework and whose answers lead to much greater
insight into contemporary forms of underdevelopment. Yet this dependent
incorporation into a world system has been treated in these empirical studies as
a given, itself unanalyzed. Each satellite had a series of bilateral relations of
dependency, and unexamined changes from one dependent relationship to
another are seen as causal in analyzing changes in its internal organization. But
this set of bilateral relations poses two overriding theoretical and
methodological problems. First, how are these bilateral relations structured at
the level of the world capitalist system? And second, given a commitment to
holism, what are the dynamics in the world system which determine "shifts at
the centre" with their dramatic consequences for the peripheries of the world?

(2) Hierarchy. Frank and other dependency theorists have had difficulties in
developing a coherent picture of a world system for which the analytical
starting point is bilateral relationships among nations. If the analyst takes a set
of two nations in which one nation exploits the other, then adds to that a

402

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
second such set, then a third such set, and so on until all nations are included,
the picture is an unorganized aggregate of sets, not a coherent structure. Thus,
attempts are being made in the dependency literature to cumulate these
bilateral relationships using different strategies for linking the sets of nations.
Frank's strategy is to build on the observation that in each bilateral
relationship one or both of the nations can be linked to another nation in a
comparable relationship of exploitation and surplus appropriation. Nation A
may exploit nation B, but B in turn may exploit C while A is being exploited by
Q. Thus, when talking about the relationships among France, Spain, Chile, and
Peru, and about relationships within Chile itself, Frank steers our attention to
"the capitalist structure of chain-linked metropolis-satellite constellations"
(1969a:38-39). This chain, we are told elsewhere, "extends from the
macrometropolitan centre of the world capitalist system 'down' to the most
supposedly isolated agricultural workers . ." (1969a:16). This chain of
relationships is a chain of exploitation and of surplus appropriation; surpluses
which originate in the mines and agricultural areas are transferred along a
chain until they come to the macrometropolis where, presumably, they are
recycled into new investments.
This formulation leads to some insights into the dynamics of capitalism
within the satellite; for example, portions of the surplus are retained by the
dependent capitalist class in the satellite nation. However, the formulation is
less satisfactory if one follows the chain through to its macrometropolitan end.
There are many chains of dependency corresponding to each micro-satellite.
Either all chains are connected at the top to one ultimate macrometropolis,
forming one world hierarchy; or several chains are joined to one of several
macrometropoles, forming several hierarchies; or a series of independent chains
exist. If there is a series of independent chains, either there is not a world
system, in which all parts are connected, or we are unable to determine in any
logically consistent fashion what the relationship is between Q (in the chain Q,
A, B, C) and each link in any other chain. If there are several hierarchies, each
composed of several chains culminating in macrometropoles, we are unable to
derive consistently the relationship between the macrometropoles. If there is a
single world hierarchy, culminating in one ultmate macrometropolis, then
problems of structural dynamics become acute; as we shall see when we
examine the work of Wallerstein, it becomes impossible to explain the changing
of world macrometropoles except within a static overall structure.
Thus, questions concerning anything but direct vertical relations between
two national economies cannot be provided without weakening, or abandoning
altogether, the notion of chains of exploitation. One solution, therefore, is to
abandon relationships as the unit of analysis and to focus on the nation. Then
questions about the degree to which the nation exploits others and is itself
exploited can be asked. Subsequently, one can rank nations according to the
degree that they exploit/are exploited. These rankings do not, of course, point
to any structured set of relations among nations but merely refer to
characteristics of nations taken individually.

1. A second and related solution is to make the "metropolis" the term applied to the
"environment" which causes changes in the particular satellite being studied. The "metropolis"
then becomes an increasingly diffuse and non-analytical term, referring to the universal
"exploiter" which benefits from the "surplus" somehow extracted from the dependency.

403

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
 This ranking of individual nations does not fit with the notion of a system of
relationships and the problem is to reintroduce relationships in such a way that
the observer can regain a sense of systematic connection among nations in the
world. Immanuel Wallerstein (1973, 1974) attempts to solve the problem by
introducing a world "system of stratification" which is composed of core, semi-
peripheral, and peripheral countries (1973:1-2, 1974:403-5). All of these
countries are capitalist, because all participate in the world capitalist market.
Under certain circumstances (such as a fight over markets in times of
contraction), there are "possibilities of semi-peripheral countries moving
towards core status and peripheral ones moving towards semi-peripheral
status" (1973:16). Advancement is made at the expense of (other) peripheral
countries, so that movement up (or down) the status ladder is made within the
existing system. Wallerstein's approach to introducing systematic connections
among nations is, then, to create the ladder, or status hierarchy, in which all
nations have a place (1974:404).
Sociologists are accustomed to finding such hierarchies in descriptions of
the formal organization of corporations and government bureaucracies.
Wallerstein's world of nations is similarly divided into three levels, in which
terms like "staff' and "cadre" reinforce the implicit analogy with formal
organizations (1974:404). "Core," "periphery," and "semi-periphery" are
analogous to "top management," "workers," and "middle management."
Moreover, there is "constant and patterned movement between groups of
economic actors as to who shall occupy various positions in the hierarchy"
(1973:1). This formulation is, in our view, highly problematic. It is difficult to
conceive of nations as conscious actors jockeying for positions; at a minimum,
capitalist nations must have a class structure in which there is anything but a
coherence of interests. If it is really the capitalist class within the nation that is
jockeying for a new position relative to other national capitalist classes, then we
need some sense of the relationships between capitalists in different nations and
the intersection between these and the class relations within each nation.
At bottom, the conception of the contemporary world economy as a single
corporate hierarchy writ large raises more questions than it answers. How does
the corporation change its structure? In the sociological literature corporations
change in response to the "environment." But the environment for the whole
world cannot, by definition, exist; where could it be? Thus we are left with the
belief that the world structure cannot change.
Wallerstein, as we have seen, argues that all countries are capitalist because
they participate in a unified world capitalist market. Wallerstein's argument
rests on an analogy between the national economy and the individual firm: all
competitors in the world market are by definition comparable social actors,
whatever the differences in their success. But this definition begs the question of
the class structure of the "socialist" countries who trade on the markets and,
hence, of the possibilities for other countries to reorganize their social
structures in order to minimize exploitative relations within national
economies. If the class structure of socialist countries is a priori capitalist
because they trade in world markets, does trade in such markets lead to
capitalist class formation or stem from it? To avoid circular reasoning, we must
have an independent definition of class relations, and this must be the basis for
conceiving of national economies as capitalist or socialist. Moreover, the
struggle for independence from exploitative capitalist relations, through

404

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
reorganization of internal class relations, must not be cast off lightly. Such
struggles are a potential source of change in the world capitalist system as a
whole, not simply in the relative status of nations.
Johann Galtung (1972) has attempted to introduce systematic connections
among nations by pointing to divisions within nations, and by pointing to
competing empires composed of relations between the divisions of several such
nations. These divisions within nations are not classes (despite his bow to Lenin
in a footnote, 1972:97n) but social groupings identified by spatial metaphor:
centre and periphery. Given that each nation can also be central or peripheral,
the outcome is a four-fold table with a centre in the centre, a centre in the
periphery, a periphery in the centre, and a periphery in the periphery.
Imperialism is defined as that type of bilateral relationship between a central
nation and a peripheral one that features a particular harmony of interest
(between the two centres) and particular disharmonies of interest (between each
centre and its periphery and between the two peripheries). The bilateral
relations thus cumulate into a series of hierarchies of nations ("feudal
interaction structures"), and there are, apparently, competing empires.
There are perhaps more possibilities for change in Galtung's model, as
harmonies and disharmonies of acting units become redefined and as empires
compete with one another. But Galtung's formulation has the effect of
separating imperialism and underdevelopment from capitalism, thus moving
away from the concept of a capitalist world system altogether and toward a
simple Great Power conception. Galtung's centre and periphery within each
nation, groupings identified in terms of a spatial metaphor, do not lead to a
concrete view of class dynamics. Imperialism becomes a separate concept which
supplants capitalism as a description of the dynamics of the world system,
rather than being its historical outcome, and the materialist basis of the analysis
is lost. With independently defined, ahistorical categories, the whole analysis
becomes an abstract formalist exercise in logical possibilities:
Two capitalistic empires may be in competition, but they may also sub-divide the world between
them into spheres of interest so that periphery relations become more neutral. In this first phase one
empire may fight to protect itself in the competition with another capitalistic empire, but in a
second phase they may join forces and more or less merge to protect not this or that particular
capitalist empire, but the system of capitalism as such. And we could also easily imagine a third
phase where non-capitalist empires join with capitalist empires in the pattern of 'united
imperialism,' for the protection of imperialism as such. (1972:130)

In sum, the attempt to reintroduce systematic connections among nations after

they have been categorized and ranked has not been successful. Frank's chains
cannot be maintained if one contemplates the necessity of understanding
relationships among links in different hierarchies or among different chains
connected at the top into the same hierarchy; Wallerstein's corporate analogy
leaves us without the possibilities of structural change in the system as a whole
as well as within any of its parts; and Galtung makes explicit what Wallerstein
has left implicit, namely, that capitalism and imperialism are separate
phenomena and that a priori tendencies to predation on the part of dominant
groups in centre countries can replace class structure based on material
relations of production as the starting point of the analysis. Thus, universal
exploitation of nations by other nations is taken as a given, and relations
among states becomes the focus of analysis of a purportedly economic relation.
Any reference to "capitalism" is simply a label attached to the existing system
and plays no part in the analysis.2

405

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
 But why do imperialism and underdevelopment take place? Surely if there is
imperialism in a "world capitalist system," it must have something to do with
capitalist relations of production. If all forms of organization of the labor
process, including slavery and feudal obligations, are "capitalist" by virtue of
the existence of a world market, as Frank and Wallerstein argue, then the
examination of these different forms is secondary at best. Unless the dynamics
of capitalist production are examined, it is difficult to understand why - as
opposed to how - dependency relations are generated. To conceive of nations,
or even groups within nations, as social actors, motivated by self-interest to
achieve higher positions in a status hierarchy, is to fall short of constructing an
adequate explanation of the world capitalist system.

(3)Class and region. While it is useful to describe the spatial relationships

among cities, towns, and regions, these descriptions of spatial relationships do
not explain the nature of the interaction between and among social units. We
have seen in Galtung's analysis that a spatial metaphor can be taken quite
seriously, and ultimately replace class dynamics and the dynamics of social
organization in general, as the starting point for understanding
underdevelopment. Although it is clearly the case that there is a spatial
distribution of population, wealth, technology, and many other variables over
the world's landscape, it is our argument that this spatial distribution is a
product of structured relationships rather than a cause. The world's wheat
supply can be grown in a very great many places, and the fact that today a very
great proportion of it is grown in Canada, the United States, Argentina, and
Australia, rather than Russia, Hungary, or India, has much more to do with the
organization of capital and labor internationally than with topographical and
soil features on the one hand or transportation costs on the other (Friedmann,
1976). Again, the fact that one productive activity rather than another came to
be initiated in a particular location may have very little to do either with the
precapitalist history of that location or with its geographical potential (cf.
Wayne, 1975).
Although this argument underlies dependency analyses at first glance, close
examination reveals inconsistencies. Frank's categories are a confused and
confusing combination of the social and the geographical: cities, such as
London, New York, or Sao Paolo, exploit other cities, as well as agricultural
workers who frequently form the end of his exploitative chain. The problems of
the hierarchy of nations and regions are thus compounded by the confusion
about whether it is social groups or geographical areas which occupy positions
in the hierarchy (or links in the chain). Frank's reply to critics accusing him of
replacing class with geographic analysis is well-taken only in part. Given his
advocacy of the study of metropolis-satellite relations from the perspective of
the satellite (1969b: 111), he is able to describe "underdevelopment in a
dependent region such as Latin American ... as the product of a bourgeois

2. Wallerstein defines capitalism as commodity production ("for profit") both in the summary
article criticized here (1974) and in his more extensive historical study. But states do not
exchange goods; they govern regions across whose boundaries goods are exchanged.
Consequently the mechanism connecting political and economic relations should be more
clearly specified, and the conception of economic terms such as "capitalism" developed more
fully. For a further critique of Wallerstein's work see Skocpol (1977).

406

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
policy formulated in response to class interests and class structure, which are in
turn determined by the dependence of the Latin American satellite on the
colonialist, imperialist metropolis" (1972:1). But what emerges is a confusing
combination of class analysis and geographical description. Frank appears to
owe much to the geographers in his description of the chains of metropolis-
satellite relationships. Here is Skinner's description of central place theory, to
which we shall add (in brackets) a translaton into Frank's wording:
Central places - the generic term for cities, towns and other nucleated settlements [metropoles]
with central service functions [surplus appropriating roles] - may be classified in a variety of ways.
The approach taken here follows the lead of Christaller and Losch. In the analytical tradition which
stems from these scholars, a given central place [metropolis] may be typed according to its position
in interlocking spatial [exploitative] systems, within which economic function [degree of
exploitation] is associated with hierarchical level. (1972:563)

Similarily, the Chicago school of urban sociology conceived of hierarchies of

regional domination:
An example of this ... type of region is the trade area, which is delineated by the network of
economic interconnections that holds it together and which can be described in terms of the radii of
influence which extend from the center outward. The importance of such regions in historical and
contemporary societies may be indicated by pointing to the role of rural trade centers,
metropolitian regions, and cultural and political capitals, which extend their tentacles out in all
directions and influence the life of their regions in varying degrees. (Wirth, 1964:210)

Countless studies within social science have noticed these unequal spatial
relationships. To describe them, however, is a different matter from explaining
them, which, we argue, can most compellingly be done through an examination
of structured social relationships, particularly social class. Frank does attempt
to do this, as phrases such as "imperialism," "national bourgeoisie," and
"peasants" indicate. But he does not distinguish between his description of
regional inequalities and his analysis of the class relations within and among
them, which result in the observed distributions of functions. While the
geographical description logically results in a hierarchy of bilateral relations, it
is invalid to transfer this perception of the regional structure of the system to
the class relations which underlie it.
Frank's central analytical question concerns the actual and potential role of
the national bourgeoisie in the satellite. Thus, from the perspective of the
satellite, it is presumed to be enough to know that surplus is appropriated from
outside in order to be able to examine class relations within the satellite.
However, Frank's great contribution lies exactly in his insistence that
"development" and "underdevelopment" are opposite sides of the same coin,
that both are caused by the same dependent, exploitative relation. Thus, he
succeeds in examining class relations within the dependency, which result from
relations of dependency, but he cannot examine the class relations within the
imperialist metropolis. These must logically bear at least the same importance
to the metropolis-satellite relation as the class relations within the satellite. His
first "contradiction" of "expropriation/appropriation of economic surplus" is
based on the existence of external and internal monopoly in a satellite such as
Chile. But the actual analysis is of the organization of the internal monopoly
only, with a demonstration, but not analysis, of the fact that "external
monopoly has always resulted in the expropriation of the significant part of the
economic surplus produced in Chile and its appropriation by another part of
the world capitalist system" (1967:6-7).
Thus, one can describe and demonstrate mechanisms of transfer out of

407

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
Chile, and one can analyze the resulting class relations within Chile. But one
cannot, with a bilateral dependency model and this perspective, explain the
internal class dynamics of the metropolis leading to external exploitation, that
is, imperialism. Nor can one explain how one imperialist nation replaces
another as the appropriator of the surplus of a given satellite. Frank has
responded to these analytical problems, not by sorting out the confusion
between region and class as analytical categories nor by attempting a truly
global analysis of imperialism as he in principle advocates, but by making the
metropolis an ever greater abstraction. In a work directed to answering his
critics, Lumpendevelopment and Lumpenbourgeoisie (1972), processes of
underdevelopment are attributed decreasingly to specific metropolis-satellite
relations and increasingly to an abstract "world metropolis." For example,
whatever the changes in its metropoles, and these have been three, the whole
history of Argentine underdevelopment is summarized as follows:
The bulk of the capital available for investment was channeled, by the institutions of
underdevelopment, into mining, agriculture, transport, and commercial export enterprises linked to
the metropolis; almost all the rest went to luxury imports from the metropolis, with a very small
share left for manufactures and consumption related to the international market. Because of
commerce and foreign capital, the economic and political interests of the mining, agricultural and
commerical bourgeoisie were never directed toward internal economic development. The relations
of production and the class structure of the latifundia, and of mining and its economic and social
"hinterlands," developed in response to the predatory needs of the overseas and the Latin American
metropolis.(1972:23) (emphasis added)

If the italicized phrases are removed, what is left is a simple description of

what happened, not analysis. The imprecise nature of the explanatory concepts
makes them little more than a label for unexamined and unexplained "causes"
of observed consequences. A later description relies on the transportation
network as the model for the structure of exploitation within Latin America
(1972:69-70), indicating once again that the source of the problem is the
confusion of geographical description of consequences of the structure, with
analysis of the structure itself.
(4)Relations of production and exchange. The basis of this conceptual
confusion is the primacy given to relations of exchange rather than relations of
production. It is certainly true that capitalist exchange relations have
penetrated the remotest areas and populations, conceived both geographically
and socially. It is also true that in order to monetize economic relations, thus
providing the basis for exchange throughout the world, capital has
systematically destroyed the basis of independent social organization of every
population previously outside its orbit. Once these bases of self-sufficiency were
removed, the population was forced to play a part within world markets. But
while the movement of capital into hitherto virgin territory can be described in
geographical terms, the creation of a peasantry and/or a proletariat must be
analyzed in terms of social class. In other words, the newly established
relations of production are the basis for understanding the exchange relations
among geographically defined units.
Unless attention is paid to relations of production, a number of questions
will go unanswered. The most important of these questions concern the
dynamics of capitalism, which has historically expanded in terms of exchange
and commodity production. It has drawn more and more people into
productive roles within the system. Why was this the case? What is the

408

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
connection between the growth of monopoly or the relations between merchant
and industrial capital on the one hand and colonialism or dependence on the
other? Why is there underdevelopment within the metropolitan countries, and
indeed, within those very metropoles that are identified as the ultimate termini
of the surpluses extracted from the underdeveloped regions and nations? We
suggest that these questions cannot be answered without focusing on class
relations. As we have seen, analyses of underdevelopment that do not make
class the central analytical category leave the "world capitalist system" as
unspecified as the functionalists' "environment."
The problems generated by attempting to analyze dependence without
making class a part of the analysis are demonstrated in the influential work of
Arghiri Emmanuel. Emmanuel (1972) presents us with a world order composed
of hierarchically arranged rich and poor countries, each "exploiting" the one(s)
below. Basing his argument on the two assumptions of labor immobility and
capital mobility internationally, Emmanuel redefines Marx's labor theory of
value (Bettelheim, 1972:110-115) in neo-classical terms of "factors of
production," transforms Marx's cost of production formalae accordingly and
"proves" that "unequal exchange" between nations is based on differential
wages (that is, a smaller proportion of the product produced in underdeveloped
countries goes to the labor, as opposed to the capital, factor). It is unnecessary
to delve into the technicalities of the formalae, since Emmanuel's analysis falls
down on two grounds. First, his central assumption of labor immobility is
incorrect. Capital organizes labor for production internationally no less than
nationally. Immobility of labor internationally superficially appears to be the
case at some periods, but this must be seen in the light of the fact that capital
has organized mass migrations both to colonies at the end of the nineteenth
century (Bettelheim, 1972:299) and from southern to northern Europe today. In
periods when labor power is organized more or less stably within national
boundaries, this is no less a consequence of capitalist relations of production
than are migrations. Second, monopoly organization of capital internationally
is an empirically demonstrable fact. Emmanuel bases his argument of the
division of the product between capital and labor on the assumption of a free,
though nationally segmented, market. But monopoly organizes production
and, as a consequence, exchange internationally. Emmanuel completely
misunderstands monopoly, as indicated in his one discussion of it:
As for the actions of the monopolies, of which Marxist authors talk so much, this question is as
remote from our subject as any other form of direct plunder of the underdeveloped countries by the
rich and strong ones. (1972:93)

If monopoly initially emerges through capitalist dynamics in the imperium, by

this logic the imperium is no longer capitalist but some system of "direct
plunder" domestically as well. He continues the discussion by adding that
"some Marxist writers find that the monopolies cannot figure as a factor in
non-equivalence since they are rife on both sides of the barrier" (1972:94). But
monopolies are not exclusively national; it is exactly the "multi-national
corporations" which invest capital in underdeveloped countries and provide the
major outlet for their products.
Bettelheim's critique of Emmanuel is penetrating. Consistent with the
Marxist analysis of imperialism, in which world economy is defined as "a
system of production relations and, correspondingly, of exchange relations on
a world scale" (Bukharin, 1973:26), he argues that

409

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
... the poverty of the 'poor countries' and the wealth of the 'rich countries,' that is, their economic
inequality, is 'prior' to exchange between them and to what is called the 'inequality' of this
exchange. It results from world relationships of domination and exploitation and from the effects
of these relations on the development of the productive forces and on the transformation of the
relations of production. (1972:292)

In other words, capital is organized on a world scale, with both international

and national aspects. Through investment in the production of different
commodities at different technological levels, it organizes different class
structures (relations of production) in different areas. The resultant exchange
of these commodities at any given moment may be analyzed, but it will never
lead in its own right to an understanding of the different relations of production
which resulted in the commodities being exchanged.
It is now possible to identify the contribution made by dependency theory
and to see the ways in which one might build upon it to create a more holistic,
structural account of underdevelopment. The dependency literature usefully
focuses our attention on relations among nations and regions. It focuses our
attention on the fact that development and underdevelopment are linked and
that both have taken place in a world integrated through capitalist relations of
production and exchange. And it provides us with a model for understanding
the structural changes created within underdeveloped areas. Dependency
theory as it now stands, however, has not moved beyond analysis of bilateral
relations between nations and, hence, has tended to conceptualize the world of
nations as a static hierarchy which does not change, save for movements of
individual nations. The body of work in this tradition has taken spatial
metaphors and spatially defined units much too seriously. Consequently, the
analysis has rested on exchange relations, with relatively little attention paid to
the relations of production, which, we argue, underlie them.
It is our argument that the major contribution of dependency theory can be
enhanced, and the problems eliminated, through its careful integration with
Marxian theories of capitalism and imperialism. The works of Lenin,
Bukharin, and Luxemburg present a view of capitalist dynamics that gives
more precise meaning to the "world capitalist system" to which Frank and the
other theorists make reference.

III. Imperialism and underdevelopment

It is important at the outset to identify one of the major shortcomings of the
Marxist literature on underdevelopment, namely, the modified "stages" theory
so often employed. For example, Bettelheim frequently refers to the
"hindering" or "blocking" of the development of the productive forces in the
colonies (1972:289 for example). Similarly, Luxemburg postulates a forced
progression by imperialism from "natural economy" to "simple commodity
production" to capitalist relations of production (1968:466-467). These stages
are not consistent with actual historical developments. Frank, by viewing
underdevelopment as a distortion of pre-contact relations of production and
exchange within the dependent area, has provided an invaluable replacement of
the blocking concept in the classical literature.
On its part the classical Marxist literature on imperialism furnishes us with
a conceptual framework and a set of historical studies that allow us to
overcome the difficulties encountered by dependency theory as it now stands.
In this section we shall draw on some of the theoretical propositions and

410

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
empirical generalizations generated by these writers in order to present, in
condensed and simplified form, a perspective on the world capitalist system
that will be productive in terms of research questions on development and
underdevelopment.
We shall begin with the proposition that historically capital has tended to
expand to virtually all parts of the world and all populations. This proposition
is historically evident: the story of trade and production on every continent
since the industrial revolution is one of increasing interdependence with
capitalist organizations based in Europe, and, more recently, North America
and Japan. While there have been national economies at specific times which
have disengaged themselves to varying degrees from world markets, such as the
USSR and post-revolutionary China, this resulted from anti-imperialist
struggles and thus reinforces the main point, namely, that through inducing
commodity production for a world market capital has come to permeate
virtually every section of the globe.
This expansion of capital has taken place (a) by the export of capital, (b) by
the enlargement of markets, and (c) by the creation of peasantries and
proletariats, as well as the creation of dependent bourgeoisies (that is, by
bringing populations with pre-capitalist relations of production into exchange
relationships with capitalist organizations). These expansions have taken place
both within predominantly capitalist national economies and into separate pre-
capitalist areas. At the highest level of generality, there is a structural similarity
between English peasants forced off the land, thereby providing a pool of labor
for the factories, and African peasants similarly forced temporarily or
permanently to labor in mines or plantations. While the movement of African
peasants comes at a later time in history and while many Africans have come to
play a different version of the proletarian role, structurally they stand in the
same relationship to capital as English workers, then or now.
Capitalism has expanded for a number of reasons. Luxemburg (1968)
argues that capitalism cannot sustain itself without constant expansion. Any
particular capitalist organization and, as we shall argue, therefore all capitalist
organizations, must constantly seek to expand profits and hence expand (a) the
amount of capital it has invested, (b) the size of its markets, and (c) the number
and productivity of its employees. It must expand because, at the simplest level,
no capitalist organization exists in a vacuum; each has competitors, and
increasing size and scope is the only way of ensuring survival, as much as that is
possible.
The trend, as both Bukharin (1973) and Lenin (1963) have argued, is for
increasing concentration of capital; more and more capital is owned and
controlled by fewer and fewer capitalist organizations. This trend, facilitated by
new organizational forms (such as cartels and trusts), in which bank and
industrial capital are merged, has brought the world to an age of monopoly
capitalism. Organizations previously independent are either driven out of
markets and hence bought up or brought to bankruptcy or integrated vertically
with other organizations. Monopoly capitalism does not, however, mean an
end to competition. At its most extreme, state capitalism, the capitalist
organization must still compete with others organized within other countries
(Bukharin, 1973:122 ff). Even before this degree of capital concentration is
reached, however, the state is clearly involved, militarily and in other ways, in
protecting and extending the domestic and foreign interests of national

411

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
capitalist organizations.
Imperialism and underdevelopment can only be understood, it follows, by
paying attention to competition among capitalist organizations and the role of
the state in promoting the welfare of specific groups of capitalists under specific
historical conditions. As Brett (1974) has shown in his study of East Africa in
the period following the First World War, the British government sought, by
means of its direct control over its East African colonies, to promote the
welfare of British capitalists.
British attitudes to colonial development were decisively conditioned by her needs as a major
manufacturing and capital exporting country. The resulting demand for external markets and
cheap sources of raw materials had always influenced policy; these demands were greatly intensified
by the effects of the War, which suddenly exposed her competitive weakness and heightened the
importance of markets which could be directly controlled through the colonial system. (1973:71)

While colonialism in general was regarded as good for the welfare of capitalist
enterprise, Brett goes on to show that specific government action was
undertaken to help specific groups of manufacturers who were in particular
difficulty and who were structurally related to government decision-making
bodies. Thus the textile industrialists, faced with rises in raw cotton prices,
received a great deal of governmental help. "Their needs were to be of prime
importance in determining both the nature of the aid programme ultimately
evolved in East Africa and the areas in which it was to be spent" (1973:121).
The British government increased the profit position of the textile industrialists
by taking a number of steps: they underwrote research into tropical cotton
growing, introduced preferential tariffs, furnished an infrastructure at great
capital cost (largely railways), and paid a great deal of attention to working
with peasants in order to develop cotton cultivation in the East African
colonies.
Brett's example reinforces the point taken from the classical literature: in
order to understand underdevelopment we must understand the search for
competitive advantage undertaken by specific blocs of national capital at
specific historical periods. The fact that vast areas of Tanzania and Uganda are
today under cotton cultivation, rather than devoted to subsistence agriculture
or an alternative cash crop, has to do directly with the crisis in the British textile
industry following the First World War and the steps undertaken by the British
government to solve that crisis. Thus, as Wallerstein (1974) argues, the specifics
of drawing a previously unintegrated population into exchange relations with
specific capitalist organizations, and hence into relationships which result in
underdevelopment, can be understood by looking at capitalist rivalries and the
role of the state in undertaking actions on behalf of capitalists. But the fact that
African cotton is produced through peasant agriculture rather than a
plantation or tenancy system as in the US South must be explained through the
specific intersection among capitalist organizations, states, and existing social
relations. Imperialist expansion to secure the supply of cotton for English
factories took different forms in different colonies or dependencies at different
times, a problem that can be addressed only through simultaneous analysis of
each case of underdevelopment within the holistic perspective of imperialism.
This requires an examination of changing social relations, especially productive
relations, in their own right.
The result also has a great deal to do with class contradictions within the
metropolitan economy. Imperial initiatives in East Africa were taken with one

412

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
eye on growing unemployment and underemployment in Britain (Brett,
1973:117). Imperialist initiatives are, however, capitalist initiatives, although
"bamboozled" workers may identify with the expansionist interests of national
capital (Bukharin, 1973:162-163). Imperialism feeds the concentration of
capital and provides the additional profits that make capitalist organizations
viable, adding to the power of capitalist organizations at the same time as their
interests and the interests of the state become intertwined.
In the process of extending capitalist interests into new lands and territories,
the metropolitan class structure reproduces itself, with the difference (as Frank
has made very clear) that the colonial bourgeoisie is a dependent bourgeoisie,
whose position depends on the control over the economy maintained by the
external capitalists. To be sure, labor need not be organized along the lines
normally found in the metropolitan countries (Luxemburg, 1968:363-364). It is
always necessary, however, to destroy the pre-existing economy (what
Luxemburg called the "natural economy") in order to integrate the population
into the world capitalist economy. But neither the organization of peasants and
workers as colonial producers nor the creation of consumer markets can be
understood without making reference to changing relations of production
within the metropolitan economy, with particular reference to intra-national
and (with growing relevance) international capitalist rivalries accompanied by
state intervention.
To see underdevelopment as a product of relationships governed by capital
is a great advance over widely held sociological theories of social change. But
even the abbreviated argument of the classical theorists of imperialism which
we have presented adds greatly to the understanding of underdevelopment
because it directs us to historically specific studies of such relationships. These
studies suggest, among others, the following lines of research:

1. Before the mid-nineteenth century, capital was organized nationally and

was changing from chartered monopolies to formally independent
corporations. Toward the end of the nineteenth century, colonialism was based
on a partial reversal of this tendency, with corresponding changes in the role of
the state. Thus, the assumption of colonial rule by the British state in India in
the mid-nineteenth century, replacing the rule of the East India Company,
marked a shift in relations among capitalist groups, among nations, and
between capital and the state. Canadian independence and incorporation of
territory previously governed by the Hudson's Bay Company represented a
different outcome in international relations resulting from the same basic shift
in the capital-state relation. Several decades later, the partition of Africa
represented a renewed establishment of chartered companies and colonial
relations among regions, but within a changed context of relations among
capitalist groups and among states.

2. Under certain conditions capitalist organizations have specific

requirements which can be satisfied only by extending the domination of
European/American/Japanese state and capitalist power into new territories
and among new populations. In terms of the requirements for raw materials,
these arise at determinate times and are limited in range for each group. Insofar
as the extraction and / or growing of these raw materials is concerned, we expect
to find in each case that (a) the time at which the dependency is drawn into the
capitalist order, (b) the organization of production undertaken, and (c) the

413

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
existence and/or characteristics of political domination are all likely to be the
outcome of the intersection between the specific needs of the capital groups for
whom the raw materials are being developed and the specific conditions
governing social relations in actual or potential producing areas. For example,
the organization of production of agricultural materials on a settler, plantation,
or peasant basis cannot be understood without simultaneous reference to the
capital involved in organizing production of the wheat, the sugar, the sisal,
et cetera and to the prior social conditions in the North American prairies, the
Caribbean, and East Africa.
It may be useful here to speak of underdevelopment strategies that are
consciously undertaken by capitalist and state organizations. Given that
underdevelopment leads to increased profits by the capitalist organization, we
can chart (insofar as data are available) the development of strategies for
destroying existing economies and tying the population to capitalist relations of
production and exchange. These strategies, it is suggested, are formulated
much as other strategies for increasing profits are formulated. The history of
underdevelopment can be seen as the history of executing these strategies,
within the limits set by state and rival capitalist organizations, as well as, from
time to time, within the limits set by populations roused to the struggle for self-
determination.
Once the problem is studied in this light, dependency theory can become a
much more useful tool of analysis. Rather than studying bilateral relationships
between nations, we must focus our attention on activities undertaken by
capitalist organizations and on the populations that are brought into a
relationship with capital as workers or peasants, wherever they are found.
Workers, materials, and markets are necessities for capitalist organizations,
and there is competition for these necessities among capitalist groups. The
state, as an organization, helps to secure these elements through political
control and military intervention on behalf of the capitalist organizations
intertwined with it, but this relationship is a complex one, not necessarily
contained within the nation.
We resist, therefore, the notion that cities exploit hinterlands or that nations
exploit nations. One spatially defined unit does not exploit another spatially
defined unit. Capital does, however, exploit workers and defines the conditions
of production and exchange for peasants. While both are located in space, the
map of where they are located does not obviate the need for analysis of the
productive relations - as well as exchange - that cuts across national
boundaries. There is no reason to assume that social relations follow the same
pattern as more clearly visible international relations. On the contrary, the
complex interrelations among capitalists in different branches of production,
different forms of productive organization, and different states must be matters
for empirical research. At the same time, a model of their evolving interaction
at the global level must take into account as a central factor the organization of
production of all commodities in all regions. Thus, the route to the common
goal of historical specificity combined with analytical universality (Wallerstein,
1974:391) must include a focus on the relations of production. If this view has
merit, it would not be out of place to suggest that even within the constraints of
a world system, people can affect their conditions of existence in all nations by
altering their social and political relations.

414

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
References
Almond, Gabriel A., and James S. Coleman, eds.
1960 The Politics of the Developing Areas. Princeton: Princeton University Press.
Amin, Samir
1971 "Underdevelopment and dependence in Black Africa: historical origin." Journal of
Peace Research 2:81-117.
Banfield, Edward C.
1958 The Moral Basis of a Backward Society. New York: Collier-Macmillan.
Berry, Brian J.L.
1972 "Hierarchical diffusion: the basis of developmental filtering and spread in a system of
growth centers." In P.W. English and R.C. Mayfield, Man, Space and Environment.
New York: Oxford University Press.
Bettelheim, Charles
1972 Appendices I, III and IV in Arghiri Emmanuel, Unequal Exchange: A Study of the
Imperialism of Trade. New York: Monthly Review Press.
Brett, E.A.
1974 Colonialism and Underdevelopment in East Africa. London: Heinemann.
Bukharin, Nikolai
1973 Imperialism and World Economy. New York: Monthly Review Press.
Clairmont, Donald, and Dennis Magill
1974 Africville: The Life and Death of a Canadian Black Community. Toronto: McClelland
and Stewart.
Cohen, Abner
1969 Custom and Politics in Urban Africa. Berkeley and Los Angeles: University of
California Press.
Emmanuel, Arghiri
1972 Unequal Exchange: A Study of the Imperialism of Trade. New York: Monthly Review
Press.
Frank, Andre Gunder
1969a Capitalism and Underdevelopment in Latin America. New York: Monthly Review
Press.
1969b Latin America: Underdevelopment or Revolution. New York: Monthly Review Press.
1972 Lumpendevelopment and Lumpenbourgeoisie. New York: Monthly Review Press.
Friedmann, Harriet
1976 "The Transformation of Wheat Production in the Era of the World Market, 1873-1935:
A Global Analysis of Production and Exchange." Unpublished Ph.D. Dissertation,
Harvard University.
Galtung, Johan
1972 "Structural theory of imperialism." The African Review, 1(4):93-138.
Gerth, Hans, and C. Wright Mills
1958 From Max Weber: Essays in Sociology. New York: Oxford University Press.
Hempel, Carl S.
1965 Aspects of Scientific Explanation. New York: The Free Press.
Hill, Polly
1963 Migrant Cocoa-Farmers of Southern Ghana. Cambridge: Cambridge University Press.
Kirwein, Michael
1974 "The Prohibition of the African Leviatic Custom: An Interdisciplinary Study of the
Conflicts Between Christian Teachings on Marriage and the Leviatic Care of Widows in
Four African Societies." Unpublished Ph.D. Dissertation, University of Toronto.
Leeds, Anthony, and Elizabeth Leeds
1974 "Brazil and the myth of urban rurality." In Arthur J. Field, ed., City and Country in the
Third World. Cambridge, Mass.: Schenkman Publishing Company.
Lenin, V.I.
1970 "Imperialism, the highest stage of capitalism." Selected Works, Volume 1. Moscow:
Progress Publishers.
Lerner, Daniel
1968 "Toward a communication theory of modernization." In S.N. Eisenstadt, ed.,
Comparative Perspectives on Social Change. Boston: Little, Brown and Company.

415

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms
Luxemburg, Rosa
1968 The Accumulation of Capital. New York: Monthly Review Press.
McClelland, David
1964 "Business drive and national achievement." In Amitai and Eva Etzioni, eds., Social
Change. New York: Basic Books.
Parsons, Talcott
1966 Societies: Evolutionary and Comparative Perspectives. Englewood Cliffs: Prentice-Hall.
1971 The System of Modern Societies. Englewood Cliffs: Prentice-Hall.
Parsons, Talcott, and Edward Shils, eds.
1951 Toward a General Theory of Action. New York: Harper and Row.
Rostow, W.W.
1960 The Stages of Economic Growth. Cambridge: Cambridge University Press.
Skinner, G. William
1972 "Marketing and social structure in rural China." In P.W. English and R.C. Mayfield,
eds., Man, Space, and Environment. New York: Oxford University Press.
Skocpol, Theda
1977 "Wallerstein's world capitalist system: a theoretical and historical critique." American
Journal of Sociology 82(5): 1075-1090.
Smelser, Neil
1963 The Sociology of Economic Life. Englewood Cliffs: Prentice-Hall.
Wallerstein, Immanuel
1973 "Dependence in an interdependent world: the limited possibilities of transformation
within the world capitalist economy." Paper presented at the Conference on Dependence
and Development in Africa, Ottawa, 16 to 18 February.
1974 "The rise and future demise of the world capitalist system: concepts for comparative
analysis." Comparative Studies in Society and History, 16(4):387-415.
Wayne, Jack
1975 "Colonialism and underdevelopment in Kigoma Region, Tanzania: a social structural
view." Canadian Review of Sociology and Anthropology, 12(3):316-332.
Wirth, Louis
1964 On Cities and Social Life. Chicago: University of Chicago Press.

416

This content downloaded from 182.18.238.249 on Sat, 08 Dec 2018 19:08:13 UTC
All use subject to https://about.jstor.org/terms