Global Formation:
Structures of the World-Economy
Dedicated to my daughters
Cori, Mae and Frances
First edition published by Basil Blackwell, 1989.
Second Edition published by Rowman and
Littlefield, 1998.
ABSTRACT
This book develops a structural approach to the
study of the modern world-system. The stress is on systemic patterns and deep
structural logic rather than on conjunctural situations or particular areas or
periods.
A new introduction to the second edition discusses
recent world events (such as the demise of the
My approach differs from the work of most other
scholars utilizing the world-systems perspective in several ways. I believe
that concept formation and theory construction are essential pieces of the
effort to build a cumulative social science. This commitment to theory has led
me to utilize the interpretive works of other world-system scholars to formulate
theoretical models with empirical implications, and to explicitly confront the
conceptual problems raised by critics of the world-system perspective. Also, I
have a fondness for quantitative empirical analysis as a method for testing
theoretical propositions, although this does not prevent me from asking
questions which are difficult to quantify. This predilection has urged me to
review the growing corpus of comparative research which is relevant to our
understanding of world-system processes, and to approach the problems of theory
construction and concept formation with an eye to matters of operationalization
and proposition testing.
Part I outlines the main structures of the
world-economy. Major theoretical debates about world-systems and modes of
production are addressed, and a new formulation is proposed. Then I describe a
set of processes which are hypothesized to operate at the level of the
world-system as a whole. Various approaches to the periodization of the
development of the modern world-system are compared, and recent developments
since World War II are examined from the perspective long run systemic cycles
and trends. Finally I discuss world culture and the role of normative
integration and cultural domination in the reproduction of the global system.
Part II examines the relationships between states
and the capitalist world-economy. Issues raised in the current literature on
the political economy of capitalist states are considered, and explanations of
differences in state strength and regime form between core and peripheral
states are formulated. The multicentric interstate system of unequally powerful
and competing states is conceived as the main political structure behind the
reproduction of capitalist relations of production. Geopolitics, warfare, and
commodity production are interdependent forms of competition which reproduce
one another. Chapter 9 formulates an explanation for the hegemonic cycle--the
rise and decline of hegemonic core powers--and reviews related research.
Part III focuses on the nature of the
core/periphery hierarchy in the modern world-system. The analytic definition of
core production is specified as relatively capital intensive commodity
production, and contending conceptualizations are discussed. The definitions of
semiperipheral areas are considered, and the notion of discrete boundaries
between core, peripheral and semiperipheral zones is critiqued. Various
mechanisms which are alleged to reproduce the core/periphery hierarchy are
described and the crossnational research which examines the effects of these
mechanisms on national development is reviewed. The question of the necessity
of the core/periphery hierarchy for capitalism is confronted. Studies of recent
changes in the structural characteristics of core and peripheral countries are
summarized, and the questions of absolute immiseration of the periphery and the
growing gap between core and periphery are examined empirically. Chapter 13
proposes a causal model of the relationship between certain world-system cycles
and reorganizations of the core/periphery hierarchy, and reviews research which
is relevant to these models.
Part IV considers problems of metatheoretical
stance and methodological approach in the study of world-systems. I locate my
own historical/structural theorizing further toward the nomothetic end of the
idiographic/nomothetic continuum relative to the position of the
My position on method contends that the faults of
mainstream American social science are mainly problems of theoretical substance
rather than research methods. While I abhor raw empiricism as much as the next
fellow, I see the comparative and quantitative methods developed by modern
social science as important tools for studying world-systems and for evaluating
world-system theories.
The last chapter discusses the implications of the
theoretical conclusions reached and the gaps in existing empirical research.
Though the theorization is incomplete, and much more empirical work needs to be
done, I permit myself to comment on possible implications of the world-system
perspective for policy and political action. This is justified by the grave
implications of some of the findings, especially the apparent tendency of the
capitalist world-economy to regularly produce wars among core states, a
systemic feature which, in combination with the proliferation of nuclear
arsenals, portends world-wide tragedy.
Table of Contents
Page
Preface
Introduction to the Updated Edition
Introduction
Part I The Whole System
1 The Deep Structure: Real Capitalism
2 Constants, Cycles and Trends
3 Stages of Capitalism or WorldSystem Cycles?
4 The World-System Since 1945: What Has Changed?
5 World Culture, Normative Integration and
Community
Part II States and the Interstate
System
6 States and Capitalism
7 Geopolitics and Capitalism: One Logic or Two?
8 Warfare and WorldSystems
9 Rise and Decline of Hegemonic Core States
Part III Zones of the WorldSystem
10 Core and Periphery
11 Reproduction of the Core/Periphery Hierarchy
12 Recent Trends
13 Cycles and the Core/Periphery Relationship
Part IV Metatheory and Research
Methods
14 Theory Construction
15 Research Methods
Part V Implications
Glossary
Notes
References
Preface
The structuralist theoretical approach to world-systems analysis
developed in this book owes much to my professors at Stanford University,
especially John W. Meyer and Michael T. Hannan, but also Morris Zelditch, Jr.
and Joseph P. Berger. The ideas emerged
in proximate interaction with Walter Goldfrank, Volker Bornschier, Albert
Bergesen, Albert Szymanski, Joan Sokolovsky, Stephen Bunker, Alejandro Portes,
Craig Murphy, Katherine Verdery, Peter Evans, Michael Timberlake, Richard
Rubinson, David Harvey, and Neil Smith.
They have been presented at meetings of the American Sociological
Association, the Society for the Study of Social Problems and the International
Studies Association. Earlier versions of
several chapters have appeared in the International Studies Quarterly,
Comparative Political Studies, Politics and Society, the Humboldt
Journal of Social Relations, Anthro-Tech and various of the
Political Economy of the World-System Annuals.
An earlier version of chapter 8 was co-authored with Joan Sokolovsky,
part of chapter 11 with Richard Rubinson and part of chapter 15 with Aaron
Pallas and Jeffrey Kentor. I have also
received helpful commentary and critique on various chapters from Giovanni
Arrighi, Robert Wuthnow, Kathleen Schwartzman, William R. Thompson, Christian
Suter, Ulrich Pfister, Pat McGowan, David A. Smith, Patrick Nolan, Terry
Boswell, Patrick Bond, Michael Johns, Phil Vilardo, Linda Pinkow, Roland
Robertson, Richard G. Fox, Phil McMichael, Peter Grimes, and Ken O'Reilly.
I would like to thank
Immanuel Wallerstein, Terence Hopkins, Giovanni Arrighi and the other scholars
and staff of the
collective project to develop a cumulative social science is
often undermined by the pressure to individualize one's theoretical
stance. I have followed Kent Flannery's
(1982) advice against relieving yourself while standing on the shoulders of
giants.
Also I am indebted to
Shirley Sult for her help in typing and retyping the manuscript. My wife, Carolyn Hock, has provided support
and inspiration while prodding me to exercise the body as well as the mind.
Christopher Chase-Dunn
Introduction
This book proposes and begins to implement a structural approach
to the study of the modern world-system.
The stress is on the systemic patterns and the deep structural logic
rather than on the conjunctural situations of particular areas or periods. My approach differs from the work of most
other scholars utilizing the world-systems perspective in several ways. In contrast to the stated position of the
Before outlining the intent and contents of
this book I provide a short summary of the intellectual history of the
world-system perspective for the general
reader. Those unfamiliar with
world-system terms are referred to the glossary. The world-system perspective seeks to analyze
longrun/largescale social change by combining the study of societal-level
processes with the study of intersocietal and transsocietal relations. It challenges the assumption that national
societies (or tribes or city-states) constitute independent units whose
development can be understood without taking into account the systematic
ways in which societies are linked to one another in the
context of a larger network of material exchanges. These inter-societal and trans-societal networks of material
exchanges are termed world-systems.
World-systems
themselves have boundaries. Our modern
world-system is global in the sense that
all, or almost all, of the human beings on earth are strongly linked to it by
virtue of their participation in economic and
political networks which connect every continent. But world-systems of the past were not
global. There have been intersocietal
material interaction networks in small regions,
unconnected with other regions by any sustained material exchange. The processes by which primitive and ancient
world-
systems developed and merged to become the contemporary global
world-system are beyond the scope of this book.2 Suffice it to say that the myriad of smaller
systems became joined into a single
large network by the end of the nineteenth century.
The shift of empirical
focus from single societies to the world-system as a whole constitutes the most
influential aspect of the world-system perspective as it has emerged from the
scholars of the
An important structure
of the modern world-system is the core/periphery hierarchy
a historically constructed stratification system composed of dominant
core states, dependent peripheral areas and intermediate semi
peripheral states. In
Wallerstein's formulation the basic logic of the modern world-system is
capitalism, a mode of production which is defined as a central feature of the
system as a whole. The modern world-system
is a capitalist "world-economy" which emerged in Europe and
Wallerstein defines capitalism as commodity
production in which there are different sorts of class relations wage labor in the core and more politically coerced forms of labor control in
the periphery. Commodity production is
defined in the Marxist sense the
production of commodities for profitable
sale in a price-setting market.
Wallerstein understands all areas within the contemporary global political
economy to be within the bounds of the
capitalist world-economy, including the socialist states.
This perspective on
social change derives from, and is a modification of, Marx's effort to understand modern
society. It focuses on capitalism as a
mode of production, although it denies that capitalism can be understood by looking only at the "advanced"
countries. At the beginning of the
twentieth century many Marxists turned their attention to imperialism. V. I. Lenin [1916] saw imperialism as a stage
of capitalism. Rosa Luxemburg [1913] saw
imperialism as a necessary consequence of the operation of Marx's model of
capitalist development. Nikolai
Bukharin's [1915] Imperialism and World Economy is the most
important early precursor of the world-system
perspective.
Mainstream American
sociologists after World War II used Talcott Parson's digestion of Durkheim,
Weber, and Pareto to construct modernization theory, which was applied to the
"less developed" societies. This
approach saw traditional institutions as
blocking the development of modern economic and political institutions in the
"backward" countries. Karl
Polanyi's work (1944, 1977) during the
postWorld War II years carried on the project of the turn-of-the-century
Marxists in analyzing capitalism on a global scale. Polanyi, a Hungarian non-Marxist socialist,
emphasized the historicity of modes of integration, especially the market
system. This perspective, combined with the Marxist legacy, was to
shape the theoretical approach of the
In
national economic and political institutions. Dependency theory evolved in many different
directions, but some of the scholars who applied and synthesized this approach
moved toward the formulation of a theory of global capitalism, especially Andre Gunder Frank and
Samir Amin. They, and Immanuel
Wallerstein, developed the world-system perspective which this book seeks to theorize, although the
Wallersteinian version will be the one most intensively considered here.
The main intent of
this book is to formulate a structural theory of the capitalist world-economy, and to address the
major theoretical debates which have arisen in the social science literature
about world-systems. As
Sidney Mintz (1985) has written, an explanation should not be
"just one more chorus of the bone song." We must be explicit about which are the
key relationships which compose the
structures of the world-economy, the mechanisms which reproduce these structures,
and the contradictions which create
pressures for structural change. The theory
I develop extends Marx's accumulation model to the world-system as a
whole. It focuses on the world class structure and the core/periphery
hierarchy as the main contradictions around which conflict and competition in
the global political economy turn, and
it reconceptualizes capital as a relational institution of domination and
exploitation which includes states and the interstate system as part
of capitalist relations of production.
The main structures of
the world-economy are:
1the world class system;
2the core/periphery
hierarchy;
3the interstate
system, and
4the world market.
These are defined and interrelated in the chapters that follow.
A second intent of
this book is to summarize the recent empirical research which is germane to our
knowledge of the modern world-system.
This burgeoning corpus is already
large, and any effort to be complete becomes quickly obsolete. Since I am more sympathetic with formal
comparative and quantitative studies than many of the other world-system
scholars, I have tried to pay
special attention to these.
And I have also included many
studies which do not explicitly use world-system concepts, but which
nevertheless examine phenomena which I deem to be relevant to world-system
theoretical questions.
Part I argues that we
can best understand the global system in which we live by defining capitalism
in a way which builds on Marx but which modifies his conceptualization to some
extent. Once we have redefined
capitalism, the continuities of recent centuries become more visible. The major theoretical debates about
world-systems and modes of production are addressed, and a new formulation is
proposed. Then I describe a set of
processes which are hypothesized to operate at the level of the world-system as
a whole. This is presented as a schema
of world-system constants, cycles, and trends, and this schema is then compared
with the idea of stages of capitalism.
Finally I discuss world culture and the role of normative integration
and cultural domination in the reproduction of the global system.
Part II examines the
relationships between states and the capitalist world-economy. The nature of the capitalist state in the
context of the world-system, and the relative internal and external strength of
core and peripheral states are considered.
The multicentric interstate system of unequally powerful and competing
states is analyzed as the main political structure behind the reproduction of
capitalist relations of production.
Geopolitics and commodity production are interdependent forms of
competition which reproduce one another.
Chapter 9 formulates an explanation for the hegemonic sequence the rise and decline of hegemonic core powers and reviews related research.
Part III focuses on the core/periphery hierarchy. The analytic definition of core production is
specified as relatively capital intensive commodity production. Definitions of the semiperiphery and the idea
of boundaries between core, peripheral and semiperipheral zones are
examined. The
mechanisms which reproduce the core/periphery hierarchy are
described and the cross-national research which examines the effects of these
mechanisms on national development is reviewed.
I argue that the core/periphery hierarchy is necessary to the survival
of capitalism. The problems of absolute
immiseration of the periphery and the growing gap between core and periphery
are considered. Chapter 13 proposes a
hypothetical causal model of the relationships among several worldsystem
fluctuations and reviews research which is relevant to this model.
Part IV considers
metatheoretical stance and methodological problems. I locate my own historical/structural
theorizing further toward the nomothetic end of the idiographic/nomothetic
continuum relative to the position of the
On method I hold that
the faults of mainstream social science are mostly problems of theoretical
substance rather than research methods.
While I abhor raw empiricism as much as the next fellow, I see the
comparative and quantitative methods developed by modern social science as
important tools for studying world-systems and for evaluating world-system
theories.
The last chapter
examines the implications of the theoretical conclusions in this book and the
gaps in existing empirical research.
Though this book is only the first step toward specifying and testing
theories of the world-system,
I cannot refrain from making a few remarks about the possibilities
for transforming the capitalist world-economy into a more humane and peaceful
world society, though these statements are not
entirely deduceable from the earlier chapters. This is justified by the grave implications
of some of the findings, especially the apparent tendency of the capitalist world-economy to regularly
produce wars among core states, a systemic feature which, in combination with
the nuclear arsenal, portends world-wide
tragedy.
Here is a more
detailed summary of the main conclusions reached in each chapter:
Part I addresses the substantive theoretical
problems raised by a structural approach to the world-system perspective. It is an effort to go beyond the use of the
world-system as a new empirical frame of reference to an explicit theory of
world-system structures and processes.
Chapter l presents a reformulation of Marx's accumulation model of
capitalist development which utilizes
concepts produced by the
hierarchy.
The world-system is objectively stratified
into social classes (e.g. world bourgeoisie, world proletariat, intermediate
strata) but the political activities of
these classes tend to be oriented toward particular "national" state
apparatuses. This is the reason why many
Marxists argue that the relevant
perspective on class struggle is a national (or "internal") one, but
it is also an important structural basis for the reproduction of divisions
within the world working class. The
core/periphery hierarchy and nationalism crosscut and disorganize workingclass
solidarity, and perpetuate the reproduction of capitalist accumulation.
Chapter 1 stakes out a
position on the mode of production issue which differs from both: (a) orthodox
Marxists (i.e. capitalism = the wage system); and (b) Immanuel Wallerstein's
assumption that a mode of production is a feature of a whole world-system (the
"totality assumption"), and thus there can only be one mode of
production in a world-system.
While I claim that it
is advantageous to include both core and peripheral forms of capitalist
exploitation within the capitalist mode of
production, the identification of the mode of production with the whole
world-system creates grievous problems when we try to analyze the transformation of modes of production. Therefore I introduce the distinction between
the spatial boundaries of a world-system and the logical boundaries of modes of
production. This enables us to study how
a mode of production becomes dominant within a world-system and how modes
of production may be articulated with
one another within a single world-system.
I employ a typology of modes of production suggested by Amin (1980a) and
Wolf (1982), except that capitalism is redefined to include both core and
peripheral forms of labor exploitation for profit.
Chapter 2 describes a
schema of the dynamics of systemic constants, cycles, and trends which are
characteristic of the whole world-system.
The systemic constants are those basic features of the capitalist mode
of production described in chapter 1.
The system cycles include the long business cycle (i.e. the Kondratieff
wave), the hegemonic sequence of the rise and decline of core states, the cycle
of core war severity, and the cycle of core/periphery trade and control. The systemic trends include the expansion to
new populations and territories, the expansion and deepening of commodity relations,
state formation, growth of firms, the transnationalization of capital, the
increasing capitalintensity of production, and
proletarianization.
Chapter 3 compares the
above schema of cycles and trends to the periodizations of capitalist
development advanced by other Marxists.
It is concluded that the various
hypothesized stages of capitalism are more elegantly explained as periodic
combinations of world-system cycles and
trends.
In chapter 4 the
schema outlined in chapter 2 is used to consider the significance of developments which have
occurred in the world-system since World War II, such as the decolonization of
almost the whole periphery, the growth
of transnational corporations, industrialization in the periphery and the
semiperiphery, the growth of international organizations, and the
increasing number of socialist states. It is concluded that no really qualitative
changes have occurred. Rather the
seemingly new institutional and organizational forms which are observed are
functional equivalents of older forms which correspond to long-standing (but
not ahistorical) patterns.
Chapter 5 discusses
the cultural aspects of integration and domination in the contemporary
world-system. It compares the
contemporary capitalist world-economy with earlier world-systems in terms of
the centrality of cultural consensus and
ideological hegemony in maintaining cohesion and order. The particular mix of normative, economic,
and coercive integration which characterizes the contemporary system relies
most heavily on market interdependence and political-military domination,
although cultural hegemony plays a supporting role. The cultural hegemony of the core powers is
systematically undermined by the reproduction of nations and states, and by the uneven development of capitalist
commodity production which sometimes raises nonWestern nations (e.g.
legitimated to some extent by the multicentric interstate system
and its associated support for national identities. The world market and the interstate system
themselves require a substratum of cultural consensus without which
cross-cultural trade is problematic. But
this culturally-based "trade
ecumene" is a minimal level of agreement about media of exchange,
contract, and diplomatic protocol. It
does not constitute a normative order in which Durkheimian processes of moral
boundary maintenance operate.
Part II focuses more
directly on states and the interstate system.
Contrary to the popular notion that capitalism is an economic subsystem
of private firms producing for markets,
and contrary to the characterization of the world-system perspective as
"economistic," I emphasize the importance of states in the capitalist world-system. Chapter 6 discusses the contention
that core states are strong, while peripheral states are
weak. A distinction is made between
internal and external state strength.
With regard to external strength
all parties agree that core states are generally stronger. With regard to internal state strength there
is considerable controversy, but
empirical evidence is presented which supports the idea that peripheral states
are weaker internally than are core states.
I also discuss research on the relationship between regime form and the
position of states in the core
periphery hierarchy.
The structural processes which
sustain relatively democratic states in the core and relatively authoritarian
states in the periphery are examined.
The oscillation between authoritarian and populist regimes in many
semiperipheral states is also considered.
Chapter 7 focuses on
the multicentric interstate system of unequally
powerful states. This is
conceptualized as the main political structure of the capitalist world-economy. It is argued that political-military
competition and competition for shares of world markets are
complementary and interdependent aspects of the struggle to appropriate world
surplus value. These types of competition should be analyzed
as composing a single
systemic logic rather than as two different modes of
interaction. The reproduction of a competitive world market of
capitalist commodity producers is argued to be dependent on the structure of
the interstate system, and conversely,
the reproduction of the interstate system is dependent on the dynamics of
competition among capitals. Chapter 7
also examines certain features of the
interstate system which are rarely discussed in international relations
theory. In addition to the rarity of
attempts to impose imperium over the core and the consistent failure of those
attempts which have been made, we must wonder why hegemonic core powers never
even propose such a
policy. It is argued that
a world-system in which the capitalist mode ofproduction is dominant is a
fertile context for the success of core states which pursue a strategy
combining military control of trade routes with accumulation through
profit-taking, rather than geopolitical
territorial conquest and tribute-gathering.
Thus it is not the hegemons, but rather challenging core powers who set
off on the road of world imperium. But
these challenges do not succeed because they are incapable of obtaining
sufficient support in a system in which expanding and deepening commodity production
and the continuation of the multicentric interstate system hold greater promise
for national ruling classes than the presentiment of shares within a world
imperium.
Chapter 8 further
considers these same problems, arguing that it is not a matter of
"economic versus political-military" power, but rather the
particular institutional features of the capitalist mode of
production which explain the processes of the interstate system. The centrality of markets and capitalist
accumulation undercuts the tendency to empire-formation. This chapter also compares the modern
interstate system with earlier world-empires and interstate systems, and
discusses the functions of world wars in
restructuring the distribution of international power in
response to the uneven development of national economies. It is argued that world wars are a necessary,
cyclical and structural component of the expansion and deepening of capitalism.
Chapter 9 focuses on
the causes of the rise and decline of hegemonic core powers. Contrary to many characterizations, the
modern world-system is not a monolithic
structure of domination with a single center.
Not only is there continual resistance from peripheral areas, but the
core itself is multicentric and is characterized by uneven development
resulting in upward and downward
mobility. Empirical
studies of the hegemonic sequence are reviewed, including Joshua Goldstein's
demonstration of a strong link between Kondratieff economic cycles and
fluctuations in war severity.
Goldstein's research provides the most convincing empirical
demonstration of the existence of world-system processes to date, as well as
chilling support for the notion that world war is a "normal" outcome
of the operation of these processes.
Efforts to measure the rise and decline of hegemons are reviewed, and
the implications of hegemonic sequence studies for the current era of declining
Part III examines the
core/periphery hierarchy as a structural feature of the world-system. This hierarchy is understood as an
institution of socially structured inequality which is reproduced rather than
eliminated by "development."
The whole worldsystem develops but the core/periphery hierarchy, despite
a certain amount of upward and downward mobility, remains a structural feature
of the larger system. Chapter 10 reviews
different terminologies and contending analytic definitions of the
core/periphery hierarchy. An argument is
made in favor of the notion of core production as relatively capitalintensive
production. The spatial nature of the
core/periphery dimension is described as a nested hierarchy of multilevel and
overlapping regional and organizational
boundaries. The notion that core,
periphery and semiperiphery are distinct zones with measureably distinct
boundaries is described as a useful simplifying metaphor for analytic
purposes. But I suggest that the
empirical core/periphery hierarchy more probably corresponds to a
multidimensional set of continuous distributions.
The concept of the semiperiphery
is understood as a designation for two types of intermediate positions. The various efforts which have been made to
measure the position of countries in the core/periphery hierarchy are described
and critiqued. I review research which
reveals important structural
differences among peripheral areas due to differences in the
nature of the indigenous societies before incorporation into the modern
world-system. And I describe important
long-term reorganizations of the institutional and structural nature of the
core/periphery hierarchy which have occurred over the past centuries. I then dispute recent arguments by economic
historians which deny that exploitation of the periphery was an important
factor in the emergence of capitalist industrialization in the core.
Chapter 11 examines
various mechanisms which are argued to reproduce the core/periphery hierarchy such as wage
differentials, trade composition, different forms of core versus peripheral
class formation, the disarticulation of
economic structures in the periphery, transnational corporate exploitation, and
state-centric explanations which focus on differential processes of state
formation and political class struggle.
My own argument is a political theory of differential outcomes of class
struggle which explains why the core/
periphery hierarchy is necessary for the reproduction of
capitalism and how capitalism reproduces the core/periphery hierarchy. Then I review the crossnational research
which has been done on the various proposed mechanisms which reproduce
international inequalities.
Chapter 12 examines
recent trends in characteristics of core and
peripheral countries and the empirical research which documents these
trends. It also discusses the idea of
absolute immiseration as it is applied to the
capitalist world-system.
Research reveals that the magnitude of relative
core/periphery inequalities has not diminished in recent
decades, but there is little evidence to support the claim of periphery-wide
absolute immiseration. The consequences
of world-system inequalities thus stem primarily from relative differences in a
world-economy in which most of the periphery is in fact "developing"
relative to itself over time, but the differences between core and periphery
are not diminishing.
Chapter 13 describes a hypothetical model of
cyclical changes in the core/periphery hierarchy and the temporal and causal
relations between this and the hegemonic
sequence of the rise and decline of core powers and changes in the level of
competition/conflict among core powers.
The chapter describes research which is germane to this model, including
evidence of the existence of cyclical international debt crises.
Part IV confronts metatheoretical and
methodological problems which surround the effort to construct and test
world-system theories. Chapter 14 describes a continuum of metatheoretical
orientations in the social sciences which extends from completely historicist
description at one end to completely ahistorical generalization at the
other. Various intermediate points on
this continuum are discussed, and a case is made in favor of a more structural
(less historicist) approach to world-system theory. The attack on structural Marxism is reviewed
and a defense of theory is proposed. My
own critique of Althusser et al. is both substantive (the failure to use a
worldsystem frame of reference) and methodological (the substitution of
rationalist critique and political debate for systematic empirical research). Nevertheless I find the
distinctions between mode of production and social
formation, and between structural causation and conjunctural events, to be
useful for world-system theory.
The ontological status
of world-systems and other objects of social scientific study are compared and
the frequently made erroneous assumption that there is a necessary relationship
between spatial scale and level of
abstraction is contested.
The potential relevance of
structural world-system theory for political practice is also considered.
Chapter 15 examines
the question of appropriate research methods for studying world-systems. Those who have argued that comparative and
quantitative methods are inappropriate are confronted with arguments about the
assumptions behind formal modeling and comparative data analysis. I contend that we can make good use of the
generalized logic of time series analysis to study processes which operate at
the level of the whole world-system. Several
problems arising from the application of time series analysis are discussed. It is also argued that the quantitative
comparison of smaller units of analysis (e.g. nationstates) is indeed relevant
for our understanding of some world-
system processes. But
findings about relationships among characteristics of e.g. nation-states cannot
be used as evidence for relationships among analogous features of the
world-system as a whole the socalled
aggregation problem. On the other hand,
knowledge of the causality of processes occurring within and between
nation-states is indeed relevant for our understanding of some world-system
processes because not all these operate at the level of the whole.
The existence of
strong worldsystem processes implies a problem for crossnational comparisons:
the nonindependence of cases Galton's
problem. I propose a solution to
Galton's problem which builds worldsystem processes
and interstate relations into the causal model for crossnational
analysis. Several different research
designs for world-system analysis are described and compared, and a new design
for testing multilevel models of causality is proposed. The chapter ends with a consideration of the
possibility of systematic comparative research on a large number of
world-systems, a research design which has conceptual and feasibility problems
but which may nevertheless be quite useful for testing propositions about the
relationship between world-systems and modes of production.
Chapter 16 examines
the implications of this book for future studies of world-systems and discusses the matter of contemporary
political practice. The motivation
behind this work is the feeling that we don't really understand the nature of
the social system in which we are living, a world of our own creation and yet
beyond our control. It was a similar
uneasiness that motivated Marx and the
other social scientists of the nineteenth century. The efforts to synthesize the theories of the
classical sociologists have not produced
a theory which adequately explains the social phenomena we see in the twentieth
century, especially the contradictory
nature of our world-system.
Talcott Parsons performed such a synthesis and the outcome was
modernization theory, a poor vehicle now largely discredited.
And yet simply returning to Marx, as many critics of modernization theory
have done, is not really adequate either.
The twentieth century needs its own Marx before we can have a new
Gramsci or a Lenin.
Many Marxists (e.g. E.
P. Thompson, 1978), have abandoned theory for
historical studies, arguing that theory has been used badly by the
Stalinists. Theory is like
technology. Good theory may be used for
bad ends. And yet
theory is both potentially useful for humane ends, and, I would
contend, an end in itself. I find no
massive contradiction between
scholarship and political practice, although they are certainly not the
same activity. The immediate intent of
this book is theoretical. The
ultimate goal, however, is to help us
understand the world in order to change it.
PART I
The Whole System
This section addresses some basic explanatory problems which
have emerged from the effort to theorize about the capitalist
world-economy. Criticisms of the
world-system perspective by other Marxists are addressed, an explication of the
Wallersteinian version is performed, and certain of its assumptions are critiqued and
reformulated. I will outline a
perspective on modes of production in general, and an argument for modifying Marx's definition of capitalism. The question of world-system boundaries will
be considered, as will the relationship between classes and the core/periphery hierarchy insofar as they are
related to our understanding of capitalism as a mode of production. The Wallersteinian "totality
assumption," which implies that each world-system has only one mode of
production, is criticized and replaced with an analysis of logical boundaries between modes of production within
historical world-systems.
The intent of this
section is to move our analysis of world-systems from an orienting perspective which is used to
interpret history toward a new theory of the underlying tendencies of
development. I do this by explicating
a strongly theoretical
interpretation of the literature produced by the scholars of the
There is a continuum between two polar types
of explanation in the social sciences: at one extreme is ideographic
historicism (the claim that each event, village, country, period, etc. is
unique); while at the other extreme is completely nomothetic ahistoricism (the
attempt to specify one theory which
explains all social systems, small and large, simple and complex).
Immanuel Wallerstein's work is consciously located on this
continuum at a midpoint which is more historical than that of the structural
Marxists, and yet which is more generalizing than that of most historians. Wallerstein interprets the history of
"historical systems," inventing
theoretical concepts that seem to be helpful along the way. By a "strongly theoretical" reading
of Wallerstein I mean to formulate a causal and
structuralist theory out of the insights and loosely defined concepts
which he has produced.
Wallerstein's usage of the term
"historical systems" demonstrates his dialectical view that whole
socio-economic systems are both systemic and
indeterminant to a certain extent.
The real question is: exactly how systemic is the world-system? Fernand Braudel (1984:70) suggests that
Wallerstein's approach is "a little too systematic." My position is that it may be possible to
determine just how systemic the world-system is by clearly formulating models and testing them
against empirical data.
There are those who
argue that a science of society is impossible; that we cannot explain and predict social
change. Human beings, the objects of
analysis, are ostensibly intelligent and have free will and so, it is argued,
attempts to predict their behavior are doomed.
In addition, sociological theorizing is reflexive we are trying to understand ourselves and
therefore objectivity is impossible.
Also once a prediction is known the object of the prediction may
intentionally (and obstinately!) choose to disconfirm it.
The attainments of
social science do not yet completely discredit these claims, although I would
contend that we are not nearly as ignorant as some critics of social science
seem to think. Marx and Weber provide us
giant
shoulders on which to stand, and we ought not underestimate
their value. But no one would dispute
that social science is yet young. Our
paradigms seem more reactive than cumulative, and we haven't yet attained
sufficient consensus to constitute "normal science" in the Kuhnian
sense. This might be because human
behavior is impossible to explain and predict, or simply because we haven't yet penetrated its deep
structure. My effort to theorize about
world-systems posits the usefulness of models of deep social structure and begins to specify one.1
The search for a
social "genetic code" can procede in a number of directions, as Marx
pointed out in his short discussion of the "method of political economy" (Marx,
1973:100-8). The Althusserian
structuralists tend to become encompassed by their own sanitized logical world
of textual interpretation and rational
argument. They never, or rarely, allow
the messy empirical world to penetrate.
Marx pointed out that political economy must alternately move from the
concrete to the abstract, and then back again.
Deduction and induction are both necessary operations.
Why would it be
helpful if we had a formal theory of world-system development?
In addition to all the usual scientific desiderata simplicity, explanation, prediction such a theory might help us to avoid self-
destruction, and to construct a more humane, less exploitative
world society. A theory of social
change, instead of implying a philosophy of fatalistic determinism, can itself
be an aid in our effort to control our own collective future.
This last goal of theorizing should be an intended consequence which
is kept in mind in the course of the theory-building project, although we
must avoid letting political criteria
determine our theoretical decisions. The
right combination of progressive intent and scientific open-mindedness is
difficult to specify, but it is clear that these two should interact
without either dominating the other.
Modes of Production in General
In this introductory section I will discuss some general
theoretical questions which provide a framework for the effort to come to grips
with what is unique to the logic of the modern world-system. What is the general relationship between
historical world-systems and modes of production? And what is the best way to conceptualize the
boundaries of world-systems? These
questions are related, and some of the simplifying assumptions made by
Wallerstein have caused unnecessary confusion and disputation.
By mode of production
we mean the basic underlying logic which any
social system exhibits. For Marx
this was necessarily related to the problem of material production. He focused on the social relations and
institutions which organize production and distribution of the material goods
required for the reproduction of a society and its members. In class societies a very fundamental set of
institutions which are the center of modes of production are those social
relations of production which allow a
class of exploiters to appropriate surplus product from a class of
direct producers. These
institutions forms of
labor control vary across different
modes of production.
A related but somewhat
different way of conceptualizing social logics was developed by Karl
Polanyi. Polanyi (1977) designated three
modes of societal integration which are said to characterize exchange in very
different types of
societies: normative, political and market. Reciprocity, the normative
distribution of resources according to culturally agreed upon rules of justice,
exists in all societies, but is the most important mode of integration in
"communal" or kin-based classless, stateless societies. Redistribution, or the politically
determined distribution of goods becomes more important once specialized
coercive organizations (states) have emerged to monopolize legitimate
violence. States and empires use
political organization and formally organized coercion to extract surplus
product from direct producers.
State-formation proceeds in tandem with class-formation, such that
privileged positions with respect to basic resources become associated with
life-long (not age-based) statuses. The
emergence of hereditary aristocratic lineages usually precedes primary
state-formation. Reciprocity does not
cease to exist within class/state societies, but it becomes articulated with,
and dominated by, the logic of "redistributive" or better, tributary modes of production
and distribution. There are many types
and varieties of the tributary modes of production, some very centralized, some very decentralized (i.e.
feudalism). And the institutional forms
of class relations vary greatly within the tributary modes. But the logic of politically coercive
organizations dominates both class struggle and intersocietal competition.
The third type of
integration proposed by Polanyi is the price-setting market. A market is not identical with all
exchange. Exchange may be regulated
according to custom (reciprocity), according to law (redistribution), or by
competitive buying and selling. The
gathering of people in a village square to exchange goods does not necessarily
constitute a market in Polanyi's sense.
The rates of exchange (prices) must be determined (or greatly
affected)by the buying and selling decisions of a large number of separate
actors trying to maximize their individual returns. A price-setting market is, of course, an
ideal type, the same one which is at the center of classical and neo-classical
economic theory. But Polanyi was at
pains to point out that this form of interaction is not a natural and universal
way in which all people exchange with one another. Rather it is an historical creation which
emerges under certain conditions and which becomes the dominant form of
exchange in a certain type of society.
The process of commodification, the transformation of land, labor and
wealth into commodities which are exchanged in price-setting markets, becomes
the object of analysis rather than a background assumption about human nature
and the propensity to truck, barter, and maximize profit.
A major theoretical
problem has been how to combine Marx's modes of production with Polanyi's modes
of integration. Samir Amin (1980a) and
Eric Wolf (1982) have found similar
solutions regarding the conceptualization of precapitalist modes of
production.2 They agree with Polanyi
that stateless, classless societies have
a mode of production in which normative obligations, usually defined in terms
of kinship ties, regulate production and
exchange. Amin terms these
"communal," while Wolf calls them "kin-based" modes of
production. Ethnographic studies (e.g.
Sahlins, 1972) support the notion that reciprocity is the dominant form of
exchange and that true market relations are rarely found in such societies.
With regard to
Polanyi's redistributive mode of integration, other theorists have pointed out that the term
"redistributive" is somewhat unfortunate. It was derived from the study of chiefdoms,
societies which are somewhat more stratified than most stateless societies, but
not so stratified
as societies in which true states and social classes have
emerged. Neil Smelser (1959) suggests
that the kind of appropriation of resources which occurs with the emergence of
states and empires should rather be
termed "mobilization."
Amin and Wolf propose that social systems in
which political coercion is the dominant relation of production and distribution
should generally be termed
"tributary modes of production."
Feudalism is one such mode of production, and so are the so-called
"asiatic" and "slave" modes designated by other
Marxists. The tributary modes differ
from one another in terms of class relations (i.e. serfdom, clientalism,
slavery, tenantry, corve, helotry, etc.)3 and also in terms of the degree of
centralization of state power feudalism versus absolutism. But they all share the feature that political
power and coercion are the main determinants of outcomes in competition among
classes, states, and empires.
Robert Brenner (1977)
rightly located Wallerstein's theoretical perspective on the terrain of a
dispute about the transition to capitalism in
A second issue which
threaded through the Brenner critique was the matter of "internal"
versus "external" determinants.
The many Marxists who have accepted Brenner's critique of the
world-system perspective associate the alleged emphasis on exchange with the
discussion of international trade. This
is thought to depreciate the importance of internal class relations and class
struggle. The problem of "internal
to what" is usually unexamined.
Some emphasize the firm or the plantation or the village. Others are thinking about the nation-state or
the region.
The Wallersteinian
perspective was misperceived by Brenner and the others who have adopted the
"exchange versus class struggle" criticism. Wallerstein was not arguing that exchange or
trade is a more important aspect of a mode of production than class
relations. Rather he was pointing to the
multilevel character of systems of exploitation and emphasizing the mystifying
consequences of characterizing all international trade as equal
exchange. Class relations and relations
of coercion and exploitation exist at several levels; within firms [as the
important work of Braverman (1974), Burawoy (1979) and Edwards (1979) has
recently illuminated], but also at the level of localities, regions,
nation-states, and the core/periphery hierarchy. These relations can all be understood broadly
as class relations, or production relations in the Marxist sense. And the internal/external distinction is as
mystifying as the idea that international trade (or wage labor) is always equal
exchange. Wallerstein contends that it
is the boundaries between organizations, and nation-states, as well as the
core/periphery hierarchy which divide those oppositional forces which would
challenge the logic of capitalism.
Following Polanyi,
Wallerstein points out that exchange is not simply exchange.
Each material exchange has an institutional underpinning which we must
investigate in order to understand the logic of the system we are studying.
Some exchanges are customary, some are politically determined and some
are in the context of a price-setting market.
And, as Marx clearly believed,
much of the logic of capitalism is contained in the institution of commodity
production for the market.
Commodification is the process by which
formerly non-marketmediated activities come to take the commodity form. It is important to understand that
commodification is not an either/or
matter. There may be more or less of it
in different realms or activities within a social system. We know from studying the historical
development of commodity production that archaic wealth (prestige goods) was
first commodified in the sense of becoming a generalized medium of
exchange money. Land slowly became something which was
alienable and saleable. And the
commodification of food, raw materials, services, and labor occurred in spurts, in interstitial
areas, and to varying degrees within the tributary modes of production. Precapitalist empires, themselves still dominated by the logic of the tributary
mode of production, became more and more commercialized. Wealth, land and labor increasingly became
commodified, but the logic of market integration still played only a
subordinate role.
Labor is never a
perfect commodity. Neither is wealth or
land. We must speak of degrees of labor
commodification. Customary work relations
are the least commodified because obligations are usually personal and
dependent on normative consensus and internalized values. Politically-structured labor
relations based on coercion may become partially
commodified. Thus slave labor is more
commodified than serfdom, and is more manipulablely and congruent with
the logic of profitable commodity production. Consideration of costs and profits, the price
of slaves, and the ability to buy more slaves or sell them off, lent plantation
and latifundium enterprises a rather capitalistic nature even within
precapitalist world-systems.
Wage-labor is the most
perfectly commodified form of labor control
because the price tends to be regulated by the cost of the reproduction
of labor-power; the capitalist can buy only the labor time he needs, and
so wage-labor is even more flexible for
the cost-conscious capitalist commodity producer than is slave labor. Labor-time, rather than the laborer, has become a commodity. And there is the additional benefit that
exploitation is more opaque, because the exchange of wages for labor time is
defined as an equal exchange. Slavery is less easily legitimated.
Both Marx and
Wallerstein see commodity production as necessary to capitalism, but Marx
argued that "fully formed" capitalism can only be based on wage labor, while Wallerstein argues that
peripheral capitalism can be based on less commodified forms of labor
control. Two points should be made about
this definitional disagreement. The
first is that it can be agreed that wage labor is important to capitalism,
because it is the most commodified form of labor control. But it may be unwise to define other types of
labor control as necessarily non-capitalist for two reasons. The first is that
there may be different levels of capitalism. Marx acknowledges this when he discusses
merchant capitalism and simple commodity
production. And secondly these allegedly
less capitalist forms of capitalism may be articulated or integrated in
systematic ways with "full"
capitalism.
Even more importantly,
as I shall argue in chapter 1, the survival of capitalism may depend on the
articulation of different forms of labor control. In chapter 2 I build upon earlier work to
hypothesize a schema of empirically observable cycles and trends which are thought
to be characteristics of the
contemporary world-system as a whole.
Chapter 3 compares the stages of capitalism suggested by some Marxists
with the idea of cycles of world-system
development. In chapter 4 the schema
developed in chapter 2 is employed in a consideration of structural changes in
the world-system which have occurred
since World War II. Chapter 5 examines
the nature of global culture and the role it plays in the reproduction and transformation of the contemporary
world-system. Let us now examine further
the question of how capitalism is most usefully defined.
Chapter 1
The Deep Structure: Real Capitalism
The goal of this chapter is to reformulate Marx's theory of
capitalist accumulation using insights provided by the analysis of the modern
world-system. We begin by exploring
three approaches to theorizing about the deep structural dynamics of the
world-system: (a) a purely formal approach to the specification of logical
boundaries between modes of production;
(b) the process of commodification and its limits; and (c) the problem
of world classes. The epistemological
assumptions behind this effort to theorize the world-system are described and
defended in chapter 14. After
reformulating Marx's theory we will critically evaluate one of Wallerstein's
simplifying assumptions, the idea that the mode of production is necessarily a
feature of a whole world-system, and the corollary that each world-system has
only one mode of production.
My ultimate goal is to
reformulate Marx's theory to take into account those systematic aspects of
capitalist development which Marx neglected.
This requires:
1 a clear specification of Marx's model;
2a critique of its
inadequacies in the light of our knowledge of world-system processes;
3a reformulation of
the concepts and basic axioms, and
4testing the new
formulation against contemporary and historical social reality.
This book begins these tasks, but it certainly does not finish
them.
Marx's Model of Capitalist Accumulation
Marx began his explication of the laws of capitalist development
with a dialectical analysis of a fundamental institution of capitalism the
commodity. From this
institutional form, which allegedly contains the secrets of the deep structure
of capitalism, he derives the law of value, the roles of capital and labor, and
the accumulation of capital through the production and appropriation of surplus
value. This theoretical formulation has
several advantages. It is elegant. It focuses on what are indubitably essential
features of the capitalist mode of production
market exchange, commodified labor, the concentration of means of
production in private hands, and the accumulation of capital by means of the
production of commodities and the exploitation of commodified labor.
There are two main
issues which divide Marxists over the definition of the basic characteristics
of the capitalist mode of production: (a) the nature of class relations in
capitalism, and (b) the importance of the state and the interstate system for
capitalism. Marx's most abstract model
of the capitalist accumulation process, as presented in volume 1 of Capital
(1967a), assumed a closed system in which there is a single laissez-faire
state standing behind property relations, but not directly engaging in the
accumulation process. The model also
assumed only two classes: capitalists
owning and controlling the major means of production, and proletarians selling
their labor-power for wages in a competitive labor market.
Marx defined
capitalism as a system in which ownership and control of the major means of
production are in the hands of private (not state) entrepreneurs who produce
commodities for a competitive market.
For Marx the commodification of labor is primarily through a competitive
labor market the wage system in which proletarians who own no means of
production are "free" to sell their labor-power to capitalists. He developed his model on the basis of his
observations of nineteenth-century British capitalism. He assumed that
Thus, according to
Marx, the basic characteristics of fully developed capitalism are:
1 Generalized
commodity production: The production of commodities for profitable sale on
a price-setting (competitive) market.
2 Private ownership
of the major means of production: Private capitalists accumulate capital by
making investment decisions within a logic of profit maximization. This implies that the capitalist state does
not directly interfere in investment decisions or in the market, but rather
provides legitimation and order, using its power primarily to guarantee
external defense and internal peace consistent with the institutions of private
property.
3 The wage system:
Labor-power is a commodity sold by proletarians (who do not own means of
production) to capitalist owners of the means of production in a competitive
labor market.
The problem with
Marx's formulation is not so much what it includes as what it leaves out. Marx seeks to overcome the vacuities of
classical political economy by conceptualizing capitalism as an historical
system which
came into existence through the use of force (see Part 8, volume
1 of Capital)and which will pass out of existence through the
development of its own internal contradictions.
But, in seeking analytical elegance in his specification, Marx abstracts
from a number of processes which should be included within the specification of
the capitalist mode of production. For
example in volume 1 of Capital Marx assumes:
1 the existence of an
English-type caretaker state which does not directly interfere in the process
of development;
2 no international
trade a closed system;
3 a completely
competitive relationship among capitals; and
4 the complete
commodification of labor-power such that there exists only a class of workers
with no institutional power to obtain more than the subsistence wage, and a
class of capitalists who own and control all means of production.
Certainly simplifying
assumptions are required in any theory which attempts to specify the essential
tendencies of a mode of production. We
should not include sun spots or climate change in our theory. Some processes must be designated as
exogenous, while others, hopefully, capture the kernel of the social system we
are studying. The problem I am raising
is that Marx may have distorted the kernel somewhat by his choice of
simplifying assumptions.
My reading of the
world-system literature leads me to question the wisdom of several of Marx's
theoretical decisions. It is not a
matter of the simplifying assumptions being incorrect in one or another
concrete empirical situation. This is
true of every theoretical abstraction.
It is rather that
the essential processes of capitalist development may be
distorted by the
particular assumptions made by Marx. Marx attributed a great deal to
historical specificity, as have many Marxists since. But his theory was an abstraction from the
complications of history. The approach I
will outline here will do the same thing, except that it will draw the
boundaries between the essential endogenous process of capitalism and the
historical excrescences in a different way.
The hard task is to
reformulate a new theory of the essential kernel of capitalism. Bertell Ollman (1976) has convincingly argued
that Marx held to a philosophy of "internal relations" in which an
essential part (termed a "monad" by Leibnitz) contains relations that
express the basic nature of the whole system under analysis. This consists in identifying the crucial or
kernel set of relations. Some Marxists
(Dobb, 1947; Brenner, 1977) have argued that the key social relation for
capitalist society is the relationship between capital and labor as it occurs
within the firm, or as Marxists say, at the point of production. This is undoubtedly an important relationship
and a spate of excellent studies have focussed on the labor process as it has
developed in contemporary core capitalism (Braverman, 1974; Edwards, 1979;
Burawoy, 1979). The world-system
perspective encourages us, however, to notice how control institutions
(relations of production) are structured beyond the point of production, in
states, and, indeed, are institutionalized in the core/periphery hierarchy.
Additions to Marx's Model
The conceptualization of the world-system as a multilayered
system of competing groups has been very helpful in accounting for the
historical development of
capitalism. Here I would
suggest that the processes of state formation, nation-building, class
formation, and the reproduction of the core/periphery hierarchy can be
theorized as fundamental to the capitalist mode of production itself. Obviously these processes are beyond the
scope of a narrowly economic view of capitalism, but it is precisely the
transcendence of such an
economistic theory which is necessary if we are to theorize the
development of capitalism as a whole system.
The disadvantage of
this inclusion of processes formerly thought to be historical into the basic
model of capitalist development is that it
complicates the model greatly.
Instead of a kernel social relation located at the point of production
we have a much more complicated set of organizational, political, market,
interstate and world class relations.
What is needed is a new synthesis of these processes which has the
virtues of Marx's original theory: simplicity and the identification of a
relational kernel.
One of the key
insights into capitalist development stimulated by Wallerstein's (1979a)
theoretical writings is that commodity production regularly takes place in an
arena that is importantly structured by noneconomic relationships. There has never been an empirically existing
perfect market within the capitalist system.
Instead capitalism is structured as a set of power relations, sometimes
taking the form of price-setting markets, but just as often constituted as
institutionalized power or authority relations
among classes and states.
Thus capitalism is a competitive system in which no single organization
exercises monopoly control over production and consumption, but within certain
organizational realms monopoly power is temporarily
exercised. This
organizational power is most often institutionalized within state structures or
guaranteed by property laws which are backed up by states. Thus mercantilism is not a stage of
capitalist development, but, with some variations in form and extent, is a
constant feature of capitalism. In
thelong run, however, these extra-economic sources of control are themselves
subjected to competition in the arena of the world-system.
Thus the
organizational structures of states and firms, and the structured relations
among classes are subjected to a competing-down process which occurs in the
interstate system and the world market (see chapters 7 and 8). This accounts for certain regularities which
can be observed in the world-economy.
Not only are peripheries underdeveloped, but the uneven development of
core countries results in the rise and fall of hegemonic core powers. This is because the correct combination for
success in the capitalist system depends, not only on efficient production for
the market, but also on the right mix of state investment in infrastructure,
regulation of classes, and the exercise of military power and diplomacy in the
interstate system.
Marx's simplifying
assumption that workers in a purely capitalist system receive only the wages
necessary for the reproduction of the labor force does not accurately reflect
the process of class struggle which occurs within the capitalist system. Many core workers undoubtedly receive wages
beyond reproductive necessity, and many peripheral workers receive wages which
are below what they need to reproduce themselves and thus must rely on other
resources. Marx assumed
that trade union movements would more or less automatically develop into
socialist challenges to the logic of capitalism.
By now it is obvious that trade unions by themselves do not
challenge the basic logic of capitalism, although they do raise the wage bill
paid by capital. This suggests that the
process by which workers resist their perfect commodification should be seen as
a normal part of capitalist development itself.
Their differential success in this is known to be mediated primarily by
the extent to which they are able to gain access to state power, and toutilize
this access to guarantee their right to bargain collectively with capital.
While the forms of the
welfare state and the political legitimation of labor unions exist not only in
the core but in the periphery, and while the real level of protection which
workers are able to receive from their own unions and their states varies over
time in all states, it is still the case that there is a significant
differential between the core and periphery in terms of the political
protection versus the use of political coercion in class relations. These systematic variations must be taken
into account in any theory of accumulation, uneven development, and crisis. Erik Wright's (1978:147-54) explication of
the different theories of crisis refers to one cause of the decline of the rate
of profit as the "profit-squeeze" model, in which workers are
effective in maintaining a wage level which discourages new capital
investment. This is an example of the
inclusion of the consequences of class struggle into the model of accumulation
itself.
Another consequence of
the systematic inclusion of extra-economic determinants of class position is
that it enables us to better understand exploitation in the periphery. A great debate has emerged among Marxists
about the definition of the capitalist mode of production. Many reassert Marx's claim that the fully
developed capitalist mode of production can only exist in the context of the
wage system. Thus the slavery and
serfdom which were
created in peripheral areas during the expansion of the European
world-
system are classified as precapitalist modes of production which
were
articulated with capitalism.
This debate about the
articulation of modes of production has been reviewed and clarified by Aidan
Foster-Carter (1978). He suggests
thattheorization of the capitalist mode of production at the world-system level
may bring about a synthesis of the currently somewhat Balkanized discussions of
particular theoretical questions by Marxists concerned with problems of
development. Foster-Carter recommends
that the theorization of capitalism should go beyond an economistic approach to
include political dimensions. I think he
is mistaken, however, as were Brenner (1977) and Frank (1979a), to suggest that
world-system articulation ought to be conceptualized as
"exchange."
Certainly the form of exchange (commodity production versus gift-
giving or tribute payments) is important, but international
trade and the core/periphery hierarchy should be analyzed in terms of
production relations as well. As Albert
Bergesen (1983) has strongly argued, it is institu
tionalized power (in the form of private property, colonialism
and neo-colonial geomilitary influence) which stands behind the fictively equal
trade between core and periphery.
When we include
noneconomic dimensions of class into our definition of capitalist production
relations we must abandon the assumption of a perfect price-setting labor
market. Intraclass stratification into
"segmented labor
markets" is often structured by extra-economic institutions
e.g. nationalism, racism, sexism, ethnic solidarities, or trade unions, and
immigration laws.
In the periphery
extra-economic coercion plays a much greater part in production relations. Even so, the cost of slave or serf labor and
their relative efficiencies or inefficiencies entered into the calculation of
profitability and investment decisions.
The elimination of these extreme forms of labor coercion in the
periphery has by no means equalized the levels of coercion exercised over core
and peripheral workers. The wage
differential beyond the difference in productivity analyzed by Arghiri Emmanuel
(1972) isbased on the exercise of coercion in the world-system. Peripheral states generally exercise more
repressive controls over worker organizations.
If legal protections for workers exist, they are usually not
enforced. And the core/periphery
division of labor itself (differences in productivity as well as wages)
contributes to the great income inequalities which exist between core and
peripheral workers. These differences,
which may be partly conceptualized as stratification within the world
proletariat, are produced both by the direct exercise of core power through
state action and transnational corporate politics, and as the indirect
consequence of the core/periphery division of labor.
The Wallersteinian
world-system perspective contends that the core/
periphery hierarchy and the exploitation of the periphery by the
core are necessary to the reproduction of capitalism as a system. Thus, rather than a temporary stage on the
road to fully developed core capitalism, the "primary accumulation"
(Frank, 1979a) by which the core exploits the periphery is one of the main
mechanisms which allow the continuation of expanded reproduction in
the core. The relative
harmony of capitalists and workers in the core (which can be observed in the
interclass alliances typified by social democratic regimes or the
"business unionism" form of class struggle in the
It is undeniable that
the greater proportion of surplus value produced in the world-economy is, and
has long been, produced in the core, but the exploitation of the periphery both
creates extra amounts of surplus value which can be redistributed in many
indirect forms to core workers, and also rein
forces the ideologies of nationalism and "national
development" which facilitate class alliances in the core. Albert Szymanski (1981:chapter 5) shows
evidence against several propositions which support the "necessity of
imperialism" argument. He does not,
however, contradict the claim that the core/periphery hierarchy allows
capitalist accumulation to proceed as a result of its effects on class peace in
the core. Szymanski's critical
evaluation of the "aristocracy of labor" thesis (1981:chapter 14)
does not disprove that core workers benefit from the core/periphery hierarchy,
and in fact Szymanski admits that the concentration of cleaner, betterpaid jobs
in the core is an important contributor to the depolarization of class conflict
within core countries. This suggests
that decreases in the exploitation of the periphery by the core may have
potentially revolutionary consequences for core countries, and thus for the
capitalist world-system as a whole.
Logical Boundaries Between Modes of Production
The discussion of structures often proceeds at a metatheoretical
level in which great paradigms clash but little is accomplished for cumulative
social science. A theoretical holism
which simply asserts that the essential features of a
social system (the monad or kernel) exist at the level of the
whole system, is such a metatheoretical claim.
As an heuristic device it has the same
scientific status as the claim that the point of production is
the spatial location of the systemic monad.
Both of these metatheoretical points of departure try to conceptualize
the deep structure spatially, and they do so in order to help us sort out the
central qualities from which the more complicated world of appearances and
concrete realities can be explained.
But the attempt to
identify the qualities of modes of production with spatial entities may produce
more confusion than clarification. If we
assume that each world-system has one and only one mode of production, as
Wallersteindoes, how do world-systems change?
It is more useful to conceptualize modes of production in terms of logical
boundaries rather than spatial boundaries.
This allows for the articulation between different modes, and for the
competition between modes within a single socio-economic system. It may be the case that most spatially
designated socio-economic systems have one mode of production which is
dominant, but if we eliminate the possibility of the coexistence of modes, we
cannot discuss situations in which modes of production may be vying with each
other for domination, and thus our ability to analyze transformation is
accordingly limited.
A clear specification
of the underlying structural tendencies of
capitalist development is necessary to distinguish institutional
forms and social movements which reproduce capitalism (or allow it to intensify
or to expand to a larger scale) from those forms and movements which may act to
transform the capitalist system into a qualitatively different system. Here Althusser (Althusser and Balibar, 1970)
has given us a useful distinction
that between the mode of production (the basic essence of
capitalism as a system) and the social formation (the concretely existing set
of social institutions which contain historical survivals of earlier modes of
production and nascent elements of modes of production of the future). The idea of "social formation" will
here apply to the whole world-system rather than to separate national
societies. Thus the book is titled Global
Formation.
The above posits the
existence of a systemic essence the
basic
tendential laws of capitalist development. My discussion focuses on the logical "boundaries"
between different modes. It is argued
that these boundaries must be specified clearly in order to understand how
fundamental social change the
qualitative transformation of social systems
occurs. We must consider the
possibility that two or more modes of production may coexist within a single
world-system. Foster-Carter's (1978)
review of the mode of production controversy discusses several important
conceptual distinctions. He points out
that both articulation (in the sense of the complementary interpenetration of
two modes), and contradiction (in the sense of conflict and competition among
modes) are necessary in order to understand the interactions among modes of
production. As Foster-Carter puts it,
"Each concept needs the other: articulation without contradiction would
indeed be static and anti-Marxist; but contradiction without
articulation...fallaciously implies that the waxing and waning of modes of
production are quite separate activities, each internally determined, whereas
in fact they are linked as are wrestlers in a clinch" (1978:73).
An Analytic Narrative
Here follows a rather abstract and brief sketch of the history
of the
capitalist world-system which will be helpful in sorting out the
relationship between modes of production and world-systems. Wallerstein contends that the transition to
capitalism occurred for the first time in the European world-
economy during the "long sixteenth century"
(14501640). This system was
imperialistic from the beginning in that it was composed of a hierarchical
division of labor between core and peripheral areas. It emerged from European feudalism, a
somewhat unique social system which was itself a devolved combination of the
Roman Empire and the Germanic tribal societies (
Fernand Braudel's
(1984) narrative of the European world-economy differs from Wallerstein's in
theoretically relevant ways. Though he
avoids committing himself to formal definitions, Braudel makes a distinction
between capitalism and the market economy.
For him capitalism is haute finance and mercantile monopoly
spreading over the top of market exchange like a parasite, able to influence
the development of states and economies by subtle manipulations. Braudel differs from Wallerstein by
conceiving of the European world-economy as a set of layered modes of
production with capitalism at the top and precapitalist modes such as slavery
and serfdom in the periphery. This
layered approach leads him to include the city-state Mediterranean
world-economy of the twelfth century in his narrative of the development of the
capitalist world-system. He also
contends that a world-economy necessarily has a single city at its center, and
he describes the see-sawing hegemonies of
What we have here are
two different notions of the transition to
capitalism. For Braudel
just the top layer was capitalist, and it was merchant or commercial
capitalism, not industrial capitalism.
For Wallerstein each world-system has only one mode of production, so
the entire European world-
economy (both core and periphery) experienced the transition
from feudalism to capitalism during the "long sixteenth century"
(Wallerstein, 1974:chapter 8).
Braudel's notion of a
layered world-system with capitalism at the top and precapitalist modes in the
periphery is similar to the formulations of many other theorists (e.g. LaClau,
1977). But unlike LaClau, Braudel
implies notonly that capitalism creates and sustains peripheral exploitation,
but that capitalism is dependent on the existence of a periphery. In this he follows the lead of Rosa Luxemburg
(1968). Here is where Wallerstein's
formulation is superior. The idea of a
mode of production as a self-sustaining logic of reproduction would seem to
require that sectors or layers which are necessary to the reproduction of the
logic be defined as part of that logic.
According to
Wallerstein the period of the fifteenth and sixteenth centuries was a period of
transition in which the logic of capitalism came to dominate European
feudalism. The key struggle was the
defeat of the Habsburg attempt to convert the nascent capitalist world-economy
into a tributary world-empire.
The peripheralization
of Eastern Europe was possible in part because the timing of the process of
infeudation there was delayed due to the weaker influence of the
From its base in
Europe and
economy. In the core this
meant the more thorough subjugation of the labor process to the logic of
efficient and profitable commodity production (Marx 1976:1025-38) and in the
periphery it meant the partial decline of the directly coercive aspects of
labor control and the rise of more opaque market-mediated wage labor relations.
Capitalist slavery and
serfdom were eventually replaced by less directly repressive forms of labor
control. The process of proletarianization
deepened in both the core and the periphery as subsistence production and home
production (worker's gardens, nonmonetized family labor, etc.)
became
commodified, making workers more and more exclusively dependent
on the sale of their labor power. In the
periphery at times "precapitalist" forms of labor control were either
created anew or revitalized to produce commodities for export to the core. In addition, many "part-time
proletarians" received low wages made possible by their partial dependence
on village communities which served as labor reserves for the sector of the
economy producing peripheral commodities (
Thus the expansion of
capitalism during some phases reproduced "precapitalist" forms of
production, while during other phases it broke these forms down and replaced
them with production relations more similar to those of the core. Proletarianization and the commodification of
life has therefore increased in both the core and the periphery, with a certain
lag in the periphery, and the retention in the periphery of more coercive types
of labor control. These are more
coercive relative to the core but less coercive relative to the earlier
periphery.
At the level
of the core/periphery hierarchy capitalist production relations have similarly
become less directly coercive and more formally organized as market relations
between equals, even though the unequal power of core and peripheral states in
the interstate system, and the "market
advantages" of core capitalism (employing capital-intensive
technology and skilled highly paid labor) remain great. Nevertheless, neo-colonialism is
not colonialism. The
sovereignty of contemporary peripheral nation-states may often be in question,
but it is undoubtedly greater than when colonial empires carved up Asia,
Africa, and the
of economic domination by the core, but they certainly involve
more core-like capitalist relations of production than did the purely
extractive peripheral industries of earlier centuries.
In this brief
description of the spread and deepening of capitalist relations of production
we can see the importance of specifying both the spatial and logical boundaries
of capitalism. The intensification of
commodified relations within both the core and the periphery of
the capitalist system is an important process which has consequences for the
ability of capitalism to reproduce itself.
An exclusively spatial metaphor for conceptualizing production relations
blurs the distinction between the logics of different modes of production. I am arguing that, though articulation of
wage-labor production with non-wage forms of commodified labor is an endogenous
feature of capitalism, we must analyze the logical boundaries of modes of
production without relying too heavily on the spatial dimension.
Socialism is a mode of
production which subjects investment decisions and distribution to a logic of
collective use value, and as such, socialistmovements reintroduce
non-commodified relations into the interstices of capitalist relations. But the complete institutionalization of a
socialist mode of production awaits the day when this qualitatively different
logic becomes dominant in the world-system.
Thus the question of logical boundaries is important for allowing us to
see how a particular organizational form may reproduce capitalist relations or
contribute to the transformation of
capitalism.
A particular
institution may, of course, do both at the same time. Thus labor unions have forced capital to
expand mechanization at the same time as they have placed constraints on
capitalist control of a certain share of
surplus value. Similarly,
"socialist" states may reproduce the logic of commodity production on
a national scale (state capitalism) while at the same time creating a more
socialist logic of distribution, at least within their own political boundaries
(Chase-Dunn, 1982b). Thus logical
boundaries must be specified which enable us to understand the qualitative
distinction between capitalism and socialism even when these are not
differentiated spatially.
We can borrow a
metaphor from biological systems in order to begin this task. Alker (1982) and Lenski and Lenski (1982)
have argued that social systems are reproduced by "genetic"
codes. These theorists share a
cybernetic conception of systemic structure in which social information becomes
coded in cultural and symbolic systems.
Materialists may utilize the imagery of the genetic theory of deep
structure and reproduction without adopting an idealist philosophy. For Marx the "genetic structure" of
social systems was inherent in the institutions by which material life was
reproduced. He posited a tension between
the technical and social forces of production (technology, labor process) and
the relations of production (those political institutions whichallow
exploitation to take place). Cultural
institutions were thought to be reflections of the more basic struggle over
material production. Thus the essential
nature of a particular mode of production could be understood from an analysis
of the typical institutions of production and class exploitation. A deep structure theory which follows the
materialist approach will start with an analysis of class relations. These are conceptualized in various ways by
different Marxists and these differences have profound effects on the
understanding of modes of production and their transformation.
A Purely Formal Approach
What is a model of deep structure? We have various formal alternatives which may
be useful. A theory may be specified
axiomatically as a series of propositions and derived hypotheses. Nowak (1971) has thus formalized parts of
Marx's accumulation model of capitalist development. Such an axiomatic theory could employ the
kind of dialectical causal logic suggested by Wright (1978:
15-26). Or a theory may
posit the existence of a set of basic processes in terms of causal relations
among a number of variables. The
equations which describe this kind of model may be written, and alternative
specifications are easily posited. Let
us take this second form and imagine that we have decided upon a set of key
variables for representing the underlying model of capitalist development. We then have a number of equations which
posit causal relations among basic variables under specified scope
conditions. If we want to use
dialectical logic we may model complicated non-Aristotelian relations among
variables (see Alker, 1982).2
Let us think about
what it would mean to formally specify the logical boundaries of
such a formalization of the basic processes of the capitalist system. We can ask what would have to change to
constitute a qualitativelydifferent system.
If we could somehow actually know the exact specification which
corresponds to the actual capitalist mode of production at a single point in
time we would likely find out that the historicists are partly right. Even ignoring those aspects of social reality
which are largely conjunctural, the most appropriate structural model is likely
to be always changing to a certain extent. Let us assume that this
"background" change is fairly constant over time and that it is
relatively small. If, indeed, the basic
structure changes rapidly, then generalizations about modes of production are
inappropriate, and the best we can do is to write history.
The problem of
"stages of capitalism" might be understood in such a formalistic
approach by focusing on larger changes in the parameters
specifying relations among variables. Qualitative transformation in such a
formalistic approach could be designated as more fundamental change in the
model such that wholly different processes with different variables are created
and become dominant in the overall determination of development. This would correspond to transformation of
structural "constants" which are then seen not as constant relative
to all socio-economic systems, but only with regard to a particular mode of
production.
Many problems are
raised by such a purely formal approach to system transformation. How can we distinguish between changes in the
model which are unimportant for the basic structure, or which reproduce the
same system logic using new organizational forms (perhaps on a greater spatial
scale), from changes which truly transform the system logic? The distinctions between background, stage,
and transformative change suggested above are based on formal distinctions within
mathematically or logically specified models.
Certainly such an approach would need to combine these formal
distinctions withsubstantive considerations about the nature of modes of
production. If we want to be able not
only to say that fundamental change has occurred, but also to say in what
direction change is occurring, we need to have substantive models of the
various possible modes of production: kinbased, tributary, capitalist,
and socialist. Here we
will work mainly on the specification of the
capitalist mode, although it should be remembered that the
problem of logical boundaries implies the necessity of specifying the other
modes as well.
We require a
substantive core of assertions on which to base our theory of world-system
development. It might be possible to
work inductively from empirically observable cycles and trends (see chapter
2). It should be recognized that quite
different deep structure theories might account for any particular set of
empirical patterns, though an inductive approach would probably rule out some
theories. At this point I will proceed
in the other direction, from prior theorizing, although it should be emphasized
that this deductive kind of theoretical development should not remain a
selfcontained universe. Different
formulations must eventually contend with one another to account for empirical
evidence. We are a long way from
sufficient clarity, however, and the first task is to think through some
general theoretical problems.
A question which poses
nicely the problem I wish to confront is "what is capital?" Marx understood capital to be a social
relationship, an institutionalization of control through property ownership of
the means of production. For Marx
capital contained within itself the relationship with labor, a relationship in
which property-less workers must sell their labor time to capital-owners. How might we modify this definition in the
light of the insights provided by viewing capitalism as a world-system? I will approachthis problem from two
directions. The first uses insights about
the state and the interstate system to reformulate the definition of
capital. This involves the analysis of
the process of commodification and its limits.
The second
approaches the problem of domination more directly by analyzing
the relationship between classes and the core/periphery hierarchy. Let us consider these in turn.
Commodification and its Limits
Marx began with the commodity.
The commodity is a standardized product produced for exchange in a
price-setting market, in which the conditions of production are subject to
reorganization as a result of changes in input
costs. A work of art is
not a commodity because it is not (by definition) standardized or
reproducible. Your mother is not a
commodity because she is not replaceable in the market. There are no perfect commodities, but many
social objects approximate commodities.
It is uncontestably the case that the deepening of capitalist relations
involves the commodification of ever more areas of human life. Production for use in the home becomes
replaced by the purchase of commodities.
Economists commonly project the character of commodities onto
everything. Children become consumer
durables. Some socio
logists carry on in the same manner, although they sometimes
differentiate purely economic from "social" exchange (Blau,
1964). The "market mentality"
tends to analyze all interaction as commodified exchange.
All of human life is
not organized in the commodity form.
Some things represent collective "goods" or transcendent
values which have no calculable market price.
Many relationships, even in "advanced" countries, involve real
transcendent solidarity which is not reducible to a business deal. Friendship, love, patriotism are often
described in the language of "rational choice," butpart of the nature
of these relationships involves a merging of the individual interests (the
constitution of a corporate entity) such that the terms of exchange among parts
are undefined, and undefinable.
Marx observed that the
market, especially the market for labor, while it formally represents an
exchange among equals, is actually an institutional mystification of the main
type of exploitation which typifies capitalist society: the appropriation of
surplus value. One of his great
contributions was to show how the formal equality of the market could produce
socially structured inequality. I have
mentioned above that Marx's model of capitalist
accumulation assumes a perfectly operating system of markets for
labor and other commodified objects.
This assumption has led to the narrow view, held by both Marxists and
non-Marxists, that capitalism is limited to a market system in which
"private" owners of the means of production employ commodified labor
to accumulate capital.
The public/private
distinction is fundamental to Marx's definition of "independent
producers," and thus also to his distinction between production for use
and production for exchange. It is
assumed that, in a market economy, independent producers seeking to maximize
their own returns exchange
commodities with one another.
The production of use values, on the other hand, assumes some unity of
interests, either because an individual or family is producing for itself,
rather than for the market, or because the unit producing use values does it
according to a calculus which includes other principles in addition to strict
economic efficiency. Thus use-values may
be produced by a traditional division of labor in an Indian village in which
"prices" are customary (Mandel, 1970), by a state within a tributary
mode of production according to a logic of political domination, or by a
socialist society inwhich investment decisions and exchange terms are
democratically determined using a calculus of social need.
A historicalstructural
perspective on capitalism suggests that we should seek its kernel in the main
specific types of competition and integration which operate in the whole
terrain of interaction. Since the market
principle is historically only part of the determination of success and failure
in the worldsystem, we should incorporate the dynamics of those non-market
processes which are important to successful accumulation into our
conceptualization of the kernel relations.
It has been argued by some critics of Wallerstein that states and the
interstate system (which form the political basis of world capitalism) operate
according to somewhat different principles than do firms and classes struggling
to survive and succeed in the arena of the world market (Skocpol, 1977;
Zolberg, 1981). In chapter 7 I argue
that the logic of competition and conflict in the interstate system
(geopolitics) and the logic of competition in the world market should be
understood as inter-dependent determinants of the dynamics of capitalist
development. The present task is to
ground this idea in an analysis of the process of commodification and its
limits.3
The expansion of
capitalism has largely consisted of a deepening and widening of commodified
relations. More and more aspects of life
become commercialized. More and more
interactions are mediated by markets.
And the spatial scale of market integration grows. But there are clearly limits to the degree of
commodification which is possible for capitalism to survive as a social
system. Capitalists themselves, workers'
organizations and especially states sometimes resist market forces and act to
erect institutional barriers to the commodification of certain relations. Much of politics in capitalistsociety is a
seesawing struggle among different groups over the process of commodification
and its limits. This struggle is a
normal part of capitalism itself, although the resistance to commodification
holds the possibility of challenging the basic logic of capitalism.
Some of the limits on
commodification are erected by political groups operating explicitly in their
own interests. Thus cartels, guilds, and
trade unions are organizations which resist market forces in the name of
"special interests." But the
most extensive efforts to regulate market forces are carried out by states in
the name of "society." Public
health regulations, minimum wage legislation, national parks and many other
institutional features of capitalist societies are legitimated in terms of
state provision of "public goods" which market forces either would
not provide or which must be protected from commercialization in the name of
those "universal interests" represented by the State.
Karl Polanyi's (1944)
sweeping portrayal of the "great transformation" outlined the growth
of markets and the commodification of wealth, labor, and land in
Polanyi's depiction of
the transition to socialism is more historical and openended than the
functionalist notion of an equilibrium between differen
tiation and integration or an automatic dialectic which
mechanically produces social transformation.
Obviously the political tussle over the extension and limitation of
commodification involves class struggles, ideological hegemony, complicated
matters of consciousness formation, etc. and is anything but automatic. But Polanyi's analysis falters because he
fails to consider the importance of the fact that these struggles take place
within a larger terrain formed by the world economy and the interstate
system. Despite his focus on many
international aspects of capitalist development, Polanyi does not see that the
structure of the interstate system itself poses a major difficulty for his
theory of the transition to socialism.
The limits on commodification, and also the limits on collective
rationality are set in a competitive struggle in which firms, classes, and most
importantly states are the players against one another.
The myth of the Nation
as a transcendent solidarity is an important determinant of success or failure
in this struggle. States must be able to
legitimate their actions and to mobilize participation (taxpaying and warfare)
in order to survive or prevail in the worldeconomy. Those that do this less efficiently or less
effectively are likely to lose out in the competitive struggle. Thus both political mobilization and
comparative advantage in the world market are important to the logic of
capitalism.
The ideological
mystification which holds that the State represents the "general
will" or the "universal" interest is not false only because
states are controlled more by capitalists than by workers. It is also false because states represent
only the interests of their own citizens.
Thus the State is only one more political organization in a larger arena
of competing political organizations.
Even if a state did represent all of its citizens these would
not be universal interests from the point of view of the
socioeconomic systemas a whole. The
"society" represented by each state is a national society, a subgroup
in a larger whole. And there is no world
state to aggregate the interests of all participants in the worldsystem.
The political
structure of capitalism is not the
The expansion and
deepening of commodification has always created reactions and stimulated the
formation of political structures to protect people from market forces. Capitalists themselves organize political
institutions which limit market forces.
Contrary to much ideology most real living capitalists usually prefer
monopolistic certainty to the vagaries and risks of market competition. Thus capitalist states have always tried to
protect the capitalists who control them.
States act to expand markets or to destroy barriers to market
competition when their own capitalists will benefit because they enjoy a
competitive advantage. And workers and
peasants have tried to protect themselves from market forces through guild
organizations, community structures, workers' coops, labor unions and political
parties. Marx's (1967a) discussion of
the struggle over the length of the working day shows how middlestratum
professionals sometimes become involved in welfareoriented regulation of market
forces. The abolition of child labor and
slavery are additional examples of limits to commodification created by
movements with complex class identities.
The constraints on
market forces imposed in one region, state or industry are often one of the
most important driving forces of the expansion of commodification to new
areas. Successful labor organizing
causes capital to look elsewhere for labor.
Monopolies organized locally or nationally encourage consumers to try to
gain access to outside markets where goods may be cheaper, and cheaper
production in these outside markets is also encouraged. Thus commercialization and regulation
interact in a spiral which drives a number of the longrun trends visible in the
worldsystem. Commodification causes
political organizations to emerge, which then result in incentives for the
further extension of commodification.
This may undermine earlier, smallerscale political organizations by
pricing them out of competition or by causing them to internally impose new
conditions of "market efficiency."
So far the scale of market forces has always been able to escape
political regulation, now even encompassing the "socialist" states
and encouraging their movement toward deregulation and levels of allocation
more typical in the larger worldeconomy.
The size of firms, the
spatial scale of markets and the depth of market integration increase partly in
reaction to political efforts to regulate market forces. The "internationalization of
capital" is partly explained by escape from political regulation and labor
unions and a good portion of the trend
toward evergreater stateformation the arrogation of powers over more and more
areas of life can be attributed to state
action vis a vis market forces.
Increasingly states not only react to market forces with regulation but
theyintervene to create market forces.
As firm size and the spatial scale of markets has grown, state
responsibility for economic development has increased. The state not only serves private
entrepreneurs and tries to attract capital, sometimes it itself is the
entrepreneur. This occurred first in
several second tier core and semiperipheral countries trying to catch up with
British industrialization, but now peripheral capitalists regularly use the
only organization they have access to which is large enough to compete in the
global economy the State. State capitalism is thus a trend, not of
socialism, but of the increasing scale of production and market integration in
the capitalist worldeconomy.
What are the
implications of the above for our effort to redefine capital in the light of
the historical patterns of capitalist development in the worldsystem? For one thing we need to reconsider the question
of forms of property and capitalism. The
emphasis on commodification and the importance of markets in the context of
multiple competing and unequally powerful political organizations leads us to
question a narrow interpretation of "private" property as the only
appropriate form of ownership for capitalism.
General Motors would still be a capitalist firm competing in the world
automobile market even if it were to become owned and controlled by its
employees. Even stateowned firms are
subject to constraints based on the ability of the state which owns them to
compete in the larger global political economy.
Capital is that sort of power which is able to combine the operation of
profitable commodity production with the relatively efficient provision of
those "public goods" which are necessary or advantageous to
profitability. Thus capitalists, in this
sense, are neither private owners of productive wealth nor state managers, but
rather the combined operations which produce success or failure in the capitalist
worldsystem. Iam not suggesting that we
should abandon the distinction between privately held capital and the state,
nor am I suggesting that these do not often have contradictory interests. What I mean is that "success" in
the capitalist system requires both of these and their cooperation whether or
not the jobs be differentiated to separate organizations or more integrated, as
when states directly control commodity producing firms. Thus states are part of the relations of
production in capitalism, and capitalism cannot be understood as separate from
the logic of nationbuilding, stateformation and geopolitics as they typically
operate in the context of an expanding and deepening process of
commodification.
In order to understand
how geopolitics and the competition among states in the interstate system is
integrated with capitalist commodity production we
must analyze the form and content of interstate competition
within the
capitalist world-economy.
This subject is taken up in the second part of this book.
World Classes and the Core/periphery Relationship
Let us now turn to the way in which exploitation and domination
are
structured the
interaction of interclass and intraclass conflict in the
world-system. Albert
Bergesen (1983) has argued that the core/periphery relationship can be
conceptualized as a kind of class relationship.
He shows
convincingly that the relationship between the core and the
periphery is not simply an exchange between equal partners. Structured power has created and sustained
the hierarchical division of labor between the core and the
periphery, and continues to do so. Bergesen's argument is an important response
to Brenner's (1977) characterization of the world-system perspective as
"circulationist." Nevertheless, rather than collapsing the
categories of class and core/
periphery, it may be more helpful to use them both and to study
their inter-
action. First, what is
the difference between the two?
According to Marx, the capitalist/proletarian relationship is based on
ownership and/or control of productive property versus a class of workers who
do not own means of production and must sell their labor-power. This institutional situation is understood to
have come into existence through the use of extra-economic coercion, and to be
largely sustained by the normal processes of capitalist economic
competition. The core/periphery
relationship is analytically understood as a territorial division of labor in
which core areas specialize in capital intensive production using skilled
highly paid labor, and peripheral areas specialize in labor intensive
production using low wage (or coerced) and relatively unskilled labor.4
The world economy is
composed of "commodity chains," forward and backward linkages of
processes of production (Hopkins and Wallerstein, 1986). These commodity chains link raw materials,
labor, the sustenance of labor, intermediate processing, final processing,
transport, and final consumption. The
great bulk of consumption in the capitalist world-economy is of products whose
commodity chains cross national boundaries, and a large portion of these link
the core and the periphery.
Wallerstein contends that core activities take place at those
"nodes" on commodity chains where capital-intensive technology and
skilled highly paid labor are used, and where relatively greater surplus value
is appropriated (see chapter 10).
The core/periphery
relationship was brought into existence by extra-
economic plunder, conquest, and colonialism, and is sustained by
the normal
operation of politicalmilitary and economic competition
in the capitalist world-economy.
The world class
structure is primarily composed of capitalists (owners and controllers of means
of production) and property-less workers.
This class system also includes small commodity producers who control
their own means of production but do not employ the labor of others,5 and a
growing middle class of skilled and/or professionally certified workers.
The territorial
core/periphery hierarchy cross-cuts this world class structure and interacts
with it in important ways. Thus the
categories of core capitalist/peripheral capitalist, and core worker/peripheral
worker are useful for an analysis of the dynamics of world capitalism. Core labor is usually conceived as organized
by the wage-system in which competition in a labor market determines
wages. But at least some core workers
have long been in the category of protected labor in the sense that
trade unions, welfare legislation and/or other politically articulated
institutions (such as immigration controls) give them some protection from
competition in the world labor market and some advantages in the struggle with
capital. Guilds protected workers and
producers from competition within medieval cities,
encouraging the development of capitalist production outside
city walls. Core states, and other
states as well, develop politically-mediated mechanisms for
protecting workers, but the uneven process of capitalist
accumulation manages to find ways around many of these protections. The point is that the world class system may
be best understood as a continuum from protected labor through wage labor to
coerced labor which roughly corresponds to the core/periphery
hierarchy.
It is
important to realize that labor may be commodified without receiving a formal
wage. The serfs and slaves who produced
surplus value for peripheral capitalists in earlier centuries were treated as
commodities in the sense that their labor process was directed by a logic of
capital accumulation, even though they did not have the juridical freedom to
sell their labor-time to capital. There
are many degrees and forms of the commodification of labor.
Sidney Mintz (1977)
addresses the question, "was the plantation slave a proletarian?" and
his answer, after examining the interconnections between industrial capitalism
in
Marx clearly
distinguishes proletarian from slave labor in his model of fully-developed
capitalist accumulation. In volume 1 of Capital
Marx (1967a: 168) outlines his definition of commodified labor as the buying
and selling of labor-power. The
proletarian is defined as one who is free to sell his labor time, but is not
himself a commodity in the sense that a slave is. He is also "free" of the ownership
of the means of production and so is institutionally compelled to sell his
labor-power because he cannot produce for subsistence. We would expect to find
an argument by Marx as to why this narrow definition of commodified labor is
important in his theory of capitalism, but instead we find a kind of
circularity in his definition of a commodity.
He declares that "the exchange of commodities of itself implies no
other relations of dependence than those which result from its own nature"
(Marx, 1967a:168).
It is implied by Marx
that slave labor is not itself a commodity because slaves are not specifically
produced for sale. Actually
slave-breeding in mid-nineteenth century
system. He says,
"Otherwise with capital, the historical conditions of its existence are by
no means given with the mere circulation of money and commodities. It can spring into life only when the owner
of the means of production and subsistence meets in the market with the free
labourer selling his labour-power. And this
one historical condition comprises a world's
history" (Marx, 1967a:170). Actually it comprises the
history of the core zone, which is itself not comprehensible without
consideration of the
accumulation carried out in the periphery through forms of labor
which were partly commodified while also containing a large dose of politically-mediated
coercion.
Coercion beyond the
operation of free labor markets remains an important condition for reproducing
wage differentials in the contemporary world.
After all, if there were no extra-economic barriers to labor migration,
day laborersin the
The combination of
capital/labor relations and core/periphery relations produces many of the
consequences which are fundamental to the capitalist development process. The dynamics of the interstate system and the
process of uneven development are the result of interclass and intraclass
conflict and competition as well as international competition. Class alliances or "relative
harmonies" between capital and sectors of labor within core countries
cause, and are caused by, the internationally strong states and internally
relatively well-integrated nations of the core.
The more coercive and exploitative interclass relations of the periphery
are partly the result of alliances between peripheral and core capitalists, as are
the relatively weaker states and less integrated nations of the periphery. State socialism, the most important of the
anti-capitalist movements which have emerged within the capitalist
world-system, was possible because of the "combined and uneven
development" which occurred in semiperipheral areas
(Trotsky, 1932), where cross-cutting contradictions (conflicts among
capitalists and between
capitalists and workers) were exacerbated by an intermediate
position in the core/periphery hierarchy.
The interstate system,
centered on powerful core states which contend with one another for hegemony,
is an important structural basis of the continued competition within the world
capitalist class. No state represents
the "general" interests of the capitalist class as a whole. Rather subgroups of the world capitalist
class control particular states. This multicentricstructure of the world
capitalist class allows the process of competitive capitalist uneven
development to continue. The rise and
fall of hegemonic core states and changes in the structure of international
alliances allow for flexibility in the world polity and accommodate changes in
the distribution of comparative advantages in commodity production. A single world capitalist state would be much
more likely to politically sustain the interests of those capitalist subgroups
which controlled it, and thus to impede the shift in productive advantage from
less to more "efficient" producers.
Similarly such a state would become the single object of orientation of
anti-capitalist movements. Thus, unlike
the present interstate system, in which successful social constraints on
capital often lead to the state's loss of centrality in the world market and
capital flight, in a world capitalist state the
aggregation of workers' interests would likely be more effective
in actually transforming the logic of the political economy toward a more
democratic and collectively rational system.
The world-system
perspective contends that the core/periphery hierarchy, rather than being a
passing phase in the transition of "backward" areas toward core-type
capitalism, is a permanent, necessary and reproduced feature of the capitalist
mode of production. This contention is
supported by historical studies and crossnational comparative research (see
Bornschier and Chase-Dunn, 1985) which have substantiated the continuation of
mechanisms that reproduce core/periphery inequalities. The periphery of the world-economy has
certainly "developed" and changed greatly since the incorporation of
Latin America, Asia, and
hierarchical relationship between the core and the periphery
remains.
What is the
function of this territorial hierarchy for capitalism as a system? It has probably never been the case, even in
the heyday of pure plunder, that more surplus value was extracted from the
periphery than was produced in the core.
Most of the surplus value accumulated in the core is produced by core
workers using relatively more productive technology. Never-
theless, the surplus value extracted from the periphery has
played a crucial role in allowing the relatively peaceful process of expanded
reproduction in the core to proceed.
This has occurred in three ways: a) by reducing the level of conflict
and competition among core capitalists within core states; b) by allowing
adjustments to power relations among core states to be settled without
destroying the interstate system; and c) by promoting a relative harmony
between capital and important sectors of labor in the core.
This last point is
politically sensitive because it is contrary to much Marxist class
analysis. It also raises the admittedly
difficult question of longrun versus shortterm class interests. There has been much discussion
about whether or not core workers exploit peripheral workers, or
benefit from their exploitation (Emmanuel, 1972:271-342). It is clear that many core workers do benefit
from the exploitation of the periphery in a number of different ways. They are able to buy peripheral products
cheaply. The territorial division of
labor between the core and the periphery enables a larger proportion of core
workers to have cleaner and more skilled jobs.
The profits from imperialism enable some core capitalists to respond
more flexibly to worker demands for higher wages. The greater affluence of the core allows core
states to devote more resources to welfare, and to maintain a relatively
greater degree of pluralism and democracy.
Thus the
Marxists who have maintained that lack of socialist militancy among core
workers is entirely due to "false consciousness" based on
nationalism and anti-socialist propaganda are wrong, although
the status-based mechanisms of ideological hegemony are also important in
maintaining class harmony in the core.
The objective structuring of interests based on the cross-cutting nature
of class and core/periphery exploitation has stabilized the capitalist
world-economy. This structural basis of
capitalism implies that constraints on the ability of core capitalists to keep
on exploiting the periphery may have revolutionary consequences for class
relations within core countries. If
core/periphery wage differentials were to decrease, core workers would be less
subject to the "job blackmail" backed up by the threat (and reality)
of capital flight.
What is Real Capitalism?
Let us combine the above discussions of commodification and
class-core/
periphery together to postulate a new specification of the
capitalist mode of production. Real capitalism
can be defined as:
1 Generalized
commodity production in which land, labor, and wealth are substantially
commodified.
2 Private ownership
and/or control of the means of production, which may be exercised by
individuals or organizations, including single states, which are themselves
players in the larger competitive arena of commodity production and
geopolitics. This allows for "state
capitalism."
3 Accumulation of
capital based on a mix of both competitive production of commodities and politicalmilitary
power, in which commodity
production has
the greater weight in the determination of outcomes in the system as a whole.
4 Exploitation of
commodified labor which is, however, not always paid a wage.
5 The combination of
class exploitation with core/periphery exploitation such that the former is
more important quantitatively in the accumulation of capital, but the latter is
nevertheless essential because of its political effects on the mobility of
capital and in reducing class conflict and weakening anti-capitalist movements
in the core.
This adds considerably
to Marx's more elegant formulation, but the elements added allow us to account
for many of the structural features of the contemporary world-system by
referring to its mode of production.
Samir Amin
(1974) has suggested a slightly different formulation which
defines core capitalism the way Marx did, except that it is articulated with
and reliant on peripheral capitalism.
Core capitalism is defined by Amin as "autocentric"
self-expansion of capital, while peripheral capitalism is externally determined
(by the core) and uses an extra dose of coercion in its class relations. The definition suggested above is intended to
analytically combine the interstate system and the core/periphery hierarchy
into our definition of capitalism in order to explain certain features of the
modern world-system which are quite problematic when Marx's definition is used
in its unmodified form.
Now we have an
explanation, based on the capitalist mode of production itself, for why
socialism has not emerged in the "most developed" (core) areas. The core/periphery hierarchy operates to
reduce class antagonisms or to suppress class conflict in both the core and, to
some extent, in the periphery. In
peripheral areas a comprador bourgeoisie often rules with coresupport, but when
anti-imperialist movements emerge they are most often broad
anti-core class alliances.
But in the semiperiphery class antagonisms are not cross-cut by the
core/periphery hierarchy, and it is there that the strongest attempts to create
socialism have been made.
The above definition
also allows us to theoretically account for the general features which many
different peripheral areas have in common.
We do not mean to contend that all peripheral areas are the same. The original social structural
characteristics and cultures of the societies which were incorporated into the
expanding European world-economy undoubtedly affected the particular
institutional forms which emerged in different areas (see chapter 10). Yet the process of peripheralization and
exploitation, with its
phases of alternately sustaining and destroying
"precapitalist" institutions and forms of labor control (depending on
changing opportunities in the world market) has a certain unity which the
world-system definition of capitalism captures.
The processes of
uneven development, the rise and fall of hegemonic core powers, and the upward
and downward mobility of areas in the core/periphery hierarchy, are more
clearly understood when we incorporate the interstate system and the
core/periphery hierarchy into our notion of capitalism. All these advantages are strong reasons to
adopt the above definition of the capitalist mode of production which is
suggested by Wallerstein. But some of
Wallerstein's simplifying assumptions, like those of Marx, may have created
more problems than they solved.
World-System Boundaries and Modes of Production
As has been mentioned above, Wallerstein's totality assumption
holds that the mode of production is a feature of a whole world-system, and by
extension theneach world-system has only one mode of production. This simplifying assumption has allowed us to
examine the systematic ways in which capitalism and imperialism are linked, and
to reconceptualize the capitalist mode of
production as including both core and peripheral types of
capitalist
exploitation. Now that
this is accomplished we can reexamine the totality assumption in order to
resolve certain problems which it causes.
Many critics of the
world-system perspective have pointed out that the equation of a mode of
production with a whole world-system makes problematic the understanding of the
transformation of modes of production.
If each
world-system can have only one mode of production, how do modes
change? The simplifying assumption which
equates spatial boundaries with logical boundaries runs into difficulties when
we try to understand how modes of production have been transformed in the past
and how they might be transformed in the future.
The simplest
Wallersteinian definition of a world-system focuses on a network of material
exchange, the exchange of fundamental or necessary goods food and raw materials. In order to use this definition we need not
consider the question of the mode of production or the institutional forms of
the exchanges. We only need to know the
extent of direct and indirect material flow densities. This is a convenient and empirically useful
definition which allows us to investigate the relationships between worldsystems
and modes of production rather than assuming that each worldsystem has only one
mode of production. Its application to
the idea of a European world-economy raises difficulties.
The idea of
Wallerstein
substitutes the mode of production criteria for the network of bulk goods criteria
when he analyzes the interaction between Europe and
(Wallerstein, 1979c).
These are alleged to be separate world-systems because Europe is
capitalist while
To be fair
Wallerstein's (1974:59-63) analysis of why capitalism did not emerge in
discussions of articulation which have led to the proliferation
of modes of production (Hindess and Hirst, 1975; Taylor, 1979; Wolpe, 1980).
The main
thrust of Wallerstein's analysis has been to demonstrate the
connectedness ofcore and peripheral kinds of exploitation. Once this is accomplished by
reconceptualizing the capitalist mode of production as above we no longer need
the totality assumption.
This assumption also
creates difficulties when we attempt to understand the transition to socialism
in the twentieth century. Although I
have argued that existing socialist states have not yet developed an autonomous
socialist
mode of production (Chase-Dunn, 1982b), I would not want this to
be true by definition. As difficult as
it may be in practice for a city or a nation-state to develop an autonomous and
self-sustaining socialist mode of production in the context of the contemporary
capitalist world-system, I would not argue that it is impossible in principle.
In conclusion let me
summarize. I have redefined the
capitalist mode of production to allow the explanation of systemic features
which are not easily explained using Marx's definition. Incorporating the interstate system and the
core/periphery hierarchy as central features of capitalism allows us to
understand the general patterns of development in the global political economy
and the general features of class struggle and international uneven
development. It also allows us to
understand the general features of development in peripheral areas. By redefining capitalism as I have, it
becomes possible to drop Wallerstein's totality assumption while retaining the
above explanatory advantages. It also
allows us to separately analyze the logical boundaries of modes of production
and the spatial boundaries of world-systems.
Does this suggest that
the articulation of modes of production is "moved out" to the
boundary between peripheral capitalism and other modes of production? This spatial analogy is improper if it
implies that other modes of
production can exist only on the perimeter of the
world-system. It is possiblethat
elements of socialism, or the tributary mode of production, exist in the
core or the semiperiphery as well as in the periphery and
beyond. Ernest Mandel (1977) has
contended that street lights, which are free, non-commodified
"public" goods produced for the use of everyone, are socialist
institutions even within the heart of capitalism.
As has been mentioned,
however, the general analytic framework which allows us to have more than one
mode of production within a concrete social system does not by itself tell us
how these may be in interaction, or in contradiction, with each other. For this we must have a substantive theory of
the nature of each, and knowledge of their compatibilities and
contradictions. I have surmised
elsewhere that socialism, a holistic mode of production in its very nature, may
have a more difficult time emerging and growing in the interstices of capitalism
than capitalism did within the terrain of the tributary mode of
production. This is because capitalism
thrives on competition and conflict, and only grudgingly constructs collective
rationality, whereas socialism requires cooperation, and has difficulty
sustaining it in the context of a still strong and fiercely competitive
capitalist arena.
Let us now turn to a
consideration of certain observable cycles and secular trends which are
characteristics of the whole capitalist world-system.
Chapter 2: Constants, Cycles and Trends
This chapter describes a number of features of the capitalist
world-system as a whole which are, in principle, more directly observable than
its deep structural essence. In fact,
however, much of the operationalization and evidence-gathering which needs to
be done in order to verify these hypothesized surface level patterns has not
been done. I will discuss those studies
which have been carried out, and their implications for the schema I am
proposing in later sections, especially chapter 13.
I will argue in chapter
4 that a proper understanding of the structural features, systemic logic and
normal cycles of the capitalist world-economy reveals that the global system
has not undergone any important transformative change in the period since the
Second World War. The rapid changes of
scale and apparently new institutional forms are interpreted as continuations
of processes long in operation. This
argument requires that we have a fairly clear idea of the deep structural logic
of capitalism, and an accurate specification of the processes which maintain
the systemic features of the modern world-system and drive forward its cycles
and trends.
What follows is a list
of structural constants, cycles, and trends which are asserted to be features
of the whole world-system. The constants
are those deep structural features of the capitalist mode of production adduced
from our discussion in chapter 1. The
cycles and trends are suggested by our empirical knowledge of the modern
world-system, although only some of these have been quantitatively studied over
long periods of time. The present list
is a revised version of that presented in Chase-Dunn and Rubinson (1977),
modified by consideration of the similar discussions contained in Hopkins and
Wallerstein (1982: chapters 2 and 5) and the empirical studies cited below.
The term
"cycle" as utilized in worldsystem studies does not imply a perfect
sine wave with unvarying amplitude, symmetry and period. Joshua Goldstein (1988:chapter 8) contends
that social cycles typically involve processes which are sufficiently
indeterminant to preclude exact mathematical specification, and thus we analyze
sequential changes with only approximately specified periods. Robert Philip Weber (1987) maintains that
sequences with "quite irregular" periodicity should rather be termed
"fluctuations" and the word "cycle" should be saved for
more rigorous occasions. I will not
worry about these matters for now since most of the "cycles"
discussed in this chapter are hypothetical.1
Systemic Constants, Cycles, and Trends
In order to study change we must have a clear idea of the
structural constants, cycles and trends operating in the capitalist
world-economy. For the moment particular
historical contingencies will be ignored.
Structural Constants
As explained in chapter 1, the capitalist world-economy is a
worldsystem in which the capitalist mode of production has become
dominant. Thus this historical system
has a deep structural logic of capitalist capital
accumulation which drives it to expand and which contains
systemic contradictions which will provoke its transformation to a different
logic once expansion and deepening of capitalist social relations have
approached their limits.
The constant
structural features of the capitalist world-system are:
1 The interstate system a system of unequally powerful nationstates
which compete for resources by supporting profitable commodity production and
by engaging in geopolitical and military competition.
2 A core/periphery hierarchy in which the
countries which occupy a core position specialize in core production relatively capitalintensive production
utilizing skilled, high-wage labor.
Peripheral areas contain mostly peripheral production laborintensive, low-wage, unskilled labor
which has historically been subjected to extra-economic coercion.
3 Production relations in the capitalist
world-economy are more complex than Marx (1967a) assumed in his basic model of
capitalist accumulation. The direct
producers differ in their access to political organizations, most importantly,
states. Thus there is a reproduced
differentiation between core labor and peripheral labor. Labor is commodified, but it is not a perfect
commodity. Direct producers (workers)
vary in terms of the degree to which their interests are protected or coerced
by political organizations. Core workers
often enjoy the protection of state-legalized labor unions and welfare laws,
although many remain in the condition of "free" laborers more subject
to the vicissitudes of the labor market.
At the other extreme are those peripheral workers who are directly
subjected to extra-economic coercion
historically: serfs, slaves, and contract laborers as well as workers in
countries where independent trade unions and labor parties are suppressed by the
state. Thus the continuum from protected
to coerced labor is a constant differentiation within the world work force
although, as we shall see below, the trend toward proletarianization moves a
greater proportion of the work force toward full-time dependence on capitalist
commodity production.
4
Commodity production for
the world market (which includes both national and international markets) is
the central form of competition and source of surplus value in the capitalist
mode of production. This form of
competition is fundamentally interwoven with the competitive political
processes of state-formation, nation-building and geopolitics in the context of
the interstate system. The world market
is not a perfect price-setting market, although it has long been, and remains,
a very competitive arena. Monopolies are
politically guaranteed within subunits (single states or the colonial
empires of individual core states), and super-profits deriving from these
monopolies are subjected to a longrun competition as the political conditions
for maintenance of monopolies are themselves subjected to the forces of
economic and geopolitical competition.
Systemic Cycles
1 The long business
cycle (Kwave). This is a world-wide
economic cycle (see VanDuijn, 1983) in which the relative rate of capital
accumulation and overall economic activity increases and then decreases toward
stagnation in a 40 to 60 year period.
This cycle was first discovered through the analysis of price series by
N. D. Kondratieff (1979). The causes of
the Kwave are explained within a Marxist framework by Mandel (1980). Goldstein (1988) shows how the long wave is
composed of two cycles, a price cycle of inflation and deflation, and a
production cycle of growth and stagnation.
According to Goldstein (1988:chapter 10) the price cycle lags behind the
production cycle by about ten to fifteen years.
Suter (1987) has recently presented an analysis of world-system debt
cycles which integrates the Kwave with the shorter (15-25 year) Kuznets cycle
as they have interacted in the nineteenth and twentieth centuries.2 The hegemonic sequence. This refers to a fluctuation of hegemony
versus multi-centricity in the distribution of military power and economic
competitive advantage in production among core states. Hegemonic periods are those in which power
and competitive advantage are relatively concentrated in a single hegemonic
core state. Multicentric periods are
those in which there is a more equal distribution of power and competitive
advantage among core states. In only a
very rough sense is this a cycle because its periodicity is very uneven. It should rather be termed a sequence or
fluctuation. There have been three
hegemonic core states since the sixteenth century: the United Provinces of the
3 The cycle of core war
severity. Recent research by Goldstein
(1985, 1986) confirms that the severity of world wars wars among core states contending for
dominance in the world-system is
periodic in a 40 to 60 year cycle which is strongly related in time to the
Kondratieff wave. There have been
several peaks of severity (battle deaths per year) of core war since 1500, but
the three most severe periods of world war
the Thirty Years War, the Napoleonic Wars, and World Wars I and II were followed by the emergence of a new
hegemonic core power. These processes
are discussed in more detail in Part II of this book.
Samir Amin (1980a) has
suggested that world wars represent normal forms of competition within
capitalism as a system. The political
structure within which peaceful capitalist accumulation proceeds becomes unable
to provide stable support after a period of uneven economic growth, and world
wars establish a new power framework for continued capitalist
accumulation. Albert Bergesen (1985) discusses
this idea in an analysis of the relations among several world-system cycles.
4
The structure of
core/periphery trade and control. A
periodic change in the pattern of control and exchange between the core and the
periphery has characterized the world-system since 1450. Periods of relatively free market
multilateral exchange have been followed by periods in which trade was more
politically controlled and tended to be contained within colonial empires
(Krasner, 1976). Bergesen and Schoenberg
(1980) have also demonstrated that waves of expansion of colonial empires are
correlated in time with the existence of warfare among core states. Models of the relationships between these
cycles are formulated in chapter 13.
Systemic Trends
A number of systemic trends increase in waves which roughly correspond
in time with some of the cycles described above.
1 Expansion to new populations and
territories. The capitalist world-
system has expanded to incorporate and (usually) peripheralize
formerly external arenas in a series of waves since the sixteenth century. These waves have been documented in terms of
the expansion of formal colonial administration by Bergesen and Schoenberg
(1980). The limits of this type of
expansion were reached at the end of the nineteenth century when nearly the
whole globe became integrated into a single hierarchical division of labor.
2 The expansion and deepening of commodity
relations. Land, labor, and wealth have
been increasingly commodified in both the core and the periphery. More spheres
of life have become mediated by markets in the core than in the
periphery, but all areas have experienced a secular increase in
every epoch ofthe modern world-system.
This trend, like others, is somewhat cyclical in that during periods of
economic stagnation commodification slows down, or even reverses as some people
fall back on subsistence production and/or reinvent mutual aid forms of
support.
3 State formation. The power of states over their populations
has increased in every period in both the core and the periphery (Boli,
1980). States have increasingly
expropriated the authority and resources of other societal actors and
organizations, although this trend has been uneven, and there have been periods
when state control temporarily decreased or became more decentralized in
particular areas.
4 Increased size of economic enterprises. The average size in terms of assets and
employees controlled by economic enterprises has increased in every epoch. Agricultural enterprises and industrial
enterprises have gone through periods when this trend has slowed or even
temporarily reversed. The causes of this
concentration of capital are different in upward and downward phases of the
Kwave (Bergesen, 1981).
5 The transnationalization of capital. Much contemporary literature discusses the
"internationalization" of capital (Lapple, 1985; Hymer, 1979). This usage is incorrect, because the agents
of exchange are never nations, and neither are they usually states. Rather capital flows across the borders of
states, and is thus transnational (or trans-state). Capital has been trans
national at least since the long sixteenth century, when
substantial direct investments by productive and merchant capital were made
across core states (Barbour, 1963) and by core capitalists in peripheral areas
(Frank, 1979b). Since then capital has
become increasingly transnational in the sense that the proportion of total
world investment which crosses state boundaries has risen. The most recent
expansion of transnational corporations is thus a continuation of a trend long
in operation (Bornschier and Chase-Dunn, 1985:chapter 3).
6 Increasing capital intensity of production
and mechanization. Several
"industrial revolutions" in both agriculture and manufacturing since
the long sixteenth century have increased the productivity of labor in both the
core and the periphery, although the core retains its relatively higher level
of productivity. Capitalist relations of
production have deepened across the system, subjecting the labor process to
greater amounts of direct control, although decentralization in the form of
small commodity production, subcontracting (putting-out), or bureaucratization
of labor control (Edwards, 1979) reproduces a "competitive sector" of
independent producers and forms of autonomy.
The longrun trend among transnational firms has been toward more direct
control over production and the expansion of the scale of coordination of
production, but some industries and sectors remain decentralized, and in
certain periods centralization slows or even reverses in some regions.
7 Proletarianization. The process of class formation has increased
the dependence of the world work force on participation in labor markets. Subsistence redoubts, urban informal sectors,
and domestic economies have
functioned to sustain a substantial semi-proletarianized sector
of the world work force,3 but the
longrun trend has been to move a greater and greater proportion of direct
producers into full-time dependence on commodity production. This has been true of both the core and the
periphery, with a greater lag in the periphery.
Similarly, the extent of extra-economic coercion used to compel labor in
the periphery has decreased somewhat over time, as the alternatives to labor market
participation have been reduced. Capitalist
slavery, serfdom, and contract labor have been largely
eliminated, although therelative degree to which political coercion is applied
in class relations
remains greater in the periphery than in the core. This includes the
suppression of trade unions and peasant and/or labor parties by
authoritarian peripheral states, often supported by core states.
8 The growing gap. There has long been a trend toward a growing
gap in average incomes between core and peripheral areas. Within core areas the middle strata of cadres
necessary to capitalist core production has expanded its relative size as a
proportion of the labor force, and the wages of important sectors of the
working class have increased relative to incomes and levels of living in peripheral
areas. Mandel (1975) and Amin (1975)
contend that both core and peripheral workers received only subsistence wages
until the 1880's when, they allege, the gap between these incomes first
emerged. But if we are sensitive to
smaller differences and include middle strata in our estimates of average
income, we are likely to find that the trend toward a growing gap has been in
operation since the sixteenth century.
The above schema of
structural cycles and trends has been combined into a "descriptive model"
which depicts relations in time among the different features (see ChaseDunn,
1978:170). A causal model of the
relationship between the hegemonic sequence and the core/periphery hierarchy is
specified in chapter 13 (see figure 13.2).
In chapter 4 we use the above schema to examine claims that fundamental
changes have occurred in the world-system since World War II. Now let us compare the idea of stages of
capitalism with the above schema of world-system cycles.
Chapter 3: Stages of Capitalism or World-System Cycles?
The periodization
of capitalist development has proven to be a controversial topic in both
classical and contemporary theories. The
conceptualization of stages of development is important because whatever
position one adopts has implications for our understanding of both the past and
the possible futures of systemic transformation. Most stage theories, including those of some
Marxists and the modernization school, focus on national societies as the unit
of analysis. National societies are
thought to evolve through historical stages, being relatively
"advanced" or "underdeveloped." Other Marxists and dependency theorists have
focused our attention on the fact that national societies interact with one
another in systematic ways and form together a larger system which itself
evolves. This reconceptualization solves
many of the problems created by the assumption that national societies are
independent, but it raises a number of new issues. How are we to conceptualize this larger
system and what is the best way to periodize its development? When has it simply adjusted to its own
contradictions and when has it become a fundamentally different type of
system?
I would like to
dismiss outright the notions which entertain a unilinear development of
national societies and discuss different approaches to stages of world-system
development. Here I will contrast
two contemporary views. The first focuses on a series of stages in
which the relations between the core and the periphery are thought to be
qualitatively different, based on the
predominance of different kinds of capital in the core. The second
acknowledges
these differences but focuses instead on characteristics of the system as a
whole which vary over time. It argues
that some system
characteristics
are cyclical in nature, while others are secular trends, and thereby our
attention is focused on the similarities of the different periods rather than
their differences. Neither of these
approaches is "historicist" in the sense of emphasizing what is
entirely unique about a particular period.
Rather they both focus on the way in which systematic laws of motion of
capital accumulation produce different institutional forms in different
periods.
Stages of
Capitalism and Dependency
A number of
recent Marxist works focusing on the world-system argue a periodization of
capitalist development which distinguishes between three (or four) stages in
which the characteristics of core capital and its relationship to peripheral
areas vary. I will review and compare
these periodizations, and then outline a
different approach.
Albert Szymanski
(1981:95) summarizes his version of periodization as follows:
Imperialism has gone through four qualitatively distinct stages: first, noncapitalist mercantile
imperialism from around 1500 to around 1800; second, competitive capitalist
imperialism from around 1840 to around 1880; third, early monopoly capitalist
imperialism from around 1890 to around 1960; and fourth, late monopoly
capitalist imperialism since the 1960s.
In another
version Andre Gunder Frank (1979a:9) approvingly quotes Samir Amin as follows:
I distinguish 3 periods: (1) mercantilist, (2) developed (achevé) capitalist (post industrial
revolution, pre-monopolist) and (3) imperialism. To each of these periods there correspond
specific functions of the periphery at the service of the essential needs of
accumulation at the center. In stage (1)
the essential function of the
periphery (principally American, supplementarily African which
supplied the former with slaves) is to permit the accumulation of money wealth
by the Atlantic merchant bourgeoisie, wealth which transforms itself in real (achevé)
capitalism after the industrial revolution. Hence the system of plantations
(after the pillage of mines) around which all of
Amin's position
on periodization is further explained in his Imperialism and Unequal
Development (1977:229-35).Let us focus on the characteristics attributed to
each stage and the differences which are
alleged to exist between stages. The
first transition is that between mercantile (or merchant) capitalism and
industrial (competitive, developed)
capitalism. The distinctions made here
involve assertions about class relations, exchange relations, and the
relationship between the state and the
economy. Much of this follows from
Marx's discussion of merchant capitalism in volume 3 of Capital
(1967b:323-37) and his distinction between merchant capitalism and the fully
developed capitalist mode of production ─ industrial capitalism. Marx's abstract model of capitalist dynamics
as presented in volume 1 of Capital assumes a system in which there is
competition between capitals, sale of commodities at their labor value,
non-interference by the state in the economy, and a closed system in which
there is no international trade. Of
course Marx did not claim that this model directly
represented any
concrete society. It is rather an
abstraction employing simplifying assumptions which is intended to explicate
the underlying developmental tendencies of the capitalist mode of
production. British national society in
the nineteenth century was Marx's inspiration and is probably the case which
closest approximates the abstract model.
Marx believed that he had teased out the underlying nature of capitalism
by examining what he considered to be its purest and most highly developed
concrete case, and that other countries would follow Britain's path of
development.1 It was argued
in chapter 1 that some of his simplifying assumptions were mistaken, not
because they were empirically inaccurate, but because they missed much of that
which is fundamental to capitalism as a system.
Marx's model of
capitalist accumulation was not intended to apply to the stage of mercantile
capitalism. In this stage commercial
capital (involving the buying and selling of commodities produced by
independent producers) is the most prevalent and important form of
enterprise. Merchant capital does not
directly coordinate production and sales.
Production is left to independent producers. Thus merchant capital does not directly
subject production to the logic of capitalism.
Merchants make profit by buying cheap and selling dear, often by exploiting
price differentials between areas which are not integrated into a single market
economy. Production capital, on the
other hand, directly combines raw materials and labor purchased from
proletarians to produce commodities for sale.
This integrates the value relations between different types of production such that "abstract
labor" comes to exist. This means
that market integration is complete and the division of labor is subjected to
a single price system. Qualitatively different types of work come to
be related to one another in a single quantitative dimension, that of
labor value. The commodification of labor is necessary for
this to occur, and most Marxists believe that this is only accomplished under
the wage system.
Thus merchant capital (before the existence of integrated
markets) exchanges "unequals" in the sense that goods are produced in
economies in which a single standard of value has not been formed. Merchants make profit by moving commodities
from areas in which they have low prices to areas in which they have high prices. It is only within fully developed industrial
capitalism that commercial capital (operating within a single price
system) makes profit by receiving part of the surplus value produced in the
production process, not by exchanging unequals.
Merchant capital operates solely in the "sphere of
circulation," although Marx asserted that it acted as a solvent on
pre-capitalist relations of production under some conditions.
It is similarly
asserted that proletarianization is only rudimentary during the period of
mercantile capitalism, that is, the wage-earning class is small and there are
many extra-market constraints which prevent labor (and other "factors of
production") from taking the commodity form. Marx's
(1967a:717-33) discussion of primitive accumulation describes how
English and Scottish peasants were forcibly separated from the means of
production and turned into sellers of labor-power from the last part of the
fifteenth century on.
The stage of
mercantile capitalism is also characterized as a period in which political relations were much more
determinant of the terms of exchange and the conditions of production than in
later industrial capitalism. The guilds,
municipal monopolies, noble privileges, and state policies of
mercantilist
"armed" trade (Parry, 1966) are understood as proof that market
mechanisms were as yet only weakly operating.
States operated as "violence-
controlling"
organizations providing protection for economic operations enabling their
benefactors to realize what
Primitive
accumulation in the periphery (plunder, forced labor, monopoly trade) is
understood by many Marxists, including Marx himself, as one of the main
processes which created the capitalist mode of production, but which is not in
itself part of that mode of production.
In the emergence
of industrial capitalism a number of important changes are alleged to have
occurred. For one thing the
"bourgeois revolutions" in
within the
national markets of the core countries.
And in the middle of the
nineteenth century there was a temporary reduction of political
constraints on international trade with the lowering of European and American
tariff barriers. According to many histories of capitalist development this is
the period when industrial (or productive) capital came to dominate the economy
for the first time. The wage system and
the process of expanded accumulation
became self-reproducing. Industry became
organized as a competitive interaction of small firms, and the rate of
technological change increased dramatically.
The factory system in which labor was brought together with machinery
was created in the English Midlands.
The
core/periphery relationship is alleged to have been predominantly a matter of
commodity trade during this period, and Ernest Mandel (quoted in Frank,
1979a:8) asserts that this constituted the "equal exchange of
equals." A single measure of world
value was constituted by the integration of the world division of labor into a
single interactive market. It is alleged
that trade between the core and the periphery contained equal amounts of labor
value in the period of competitive capitalism.
And Mandel (1975:49-61) argues that competitive industrial capitalism
encouraged indigenous capital formation in the periphery.
There is some
disagreement about the effects of core/periphery trade during the period of competitive capitalism
on the periphery. Some Marxist scholars
argue that merchant capital tended to dissolve precapitalist relations of production by drawing people
into commodity production for the world market.
On the other hand, Geoffrey Kay (1975) argues that merchant
capital tended to
underdevelop the periphery by perpetuating precapitalist forms of production,
shoring up traditional rulers, and extracting surplus product. He contrasts this to what he sees as the
developmental effects of industrial
capital, which directly subjects labor in the periphery to capitalist
production, creating the potential for autonomous capitalist growth in the
periphery. A similar argument is made by
Bill Warren (1980).
Most stage
theories envision an important new period of capitalism emerging in the late nineteenth century with
the formation of the "monopolies."
Since Lenin (1965) this allegedly new development in the core has been
causally linked with "imperialism," i.e. the exploitation of
peripheral areas by the monopoly capitals of the core. Amin (1977) reserves the term imperialism for
this kind of exploitation, and calls earlier capitalist exploitation of the
periphery "expansionism." The
emergence of "monopoly capitalism" is thought to involve a change in
the operation of the process of accumulation.
Monopoly pricing within national markets and the extraction of surplus
profits by large firms has been understood as a new departure in which the
distribution of surplus value is distorted.
The so-called "monopoly
sector," the sector in which the largest firms dominate the most
profitable types of production, is thought to obtain a larger share of the total surplus value by draining
surplus value from the "competitive sector."
In addition, the
growth of state expenditures and increases in the social wage are understood as
subsidizing the costs of reproducing labor
power, and thus lowering the wage bill of "monopoly capital." State spending, especially on military
forces, is thought to provide a necessary outlet for "economic
surplus" (Baran and Sweezy, 1966) and to help resolve the irrationalities
of capitalist production by providing employment and investment opportunities
not generated in the private sector. The
domination of "finance capital" (a coalition of banking and
industrial capital) (Hilferding, 1981) in combination with the much more
direct involvement of states in
sponsoring economic development (Poulantzas, 1973) is alleged to have had
important effects on the tendencies of the accumulation process. Arguing the direct opposite position from
that of Kay (1975) discussed above, Mandel contends that the emergence of
monopoly capitalist imperialism and the export of investment capital to the
periphery put a brake on the indigenous capitalist development of the periphery
which had begun during the period of competitive capitalism. The declining rate of profit in the core led
to the massive export of investment capital to the periphery, but the monopoly
nature of this capital and the use of state power in formal colonialism led to
the exploitation and underdevelopment of the periphery according to Mandel.
Mandel, Amin, and
Arghiri Emmanuel (1972) contend that the core/periphery relationship altered during
the late nineteenth century due to the
emergence in the 1880s of a wage differential between core workers and
peripheral workers. Previous to that,
workers in both the core and the
periphery had received subsistance wages, but in the late nineteenth
century, due to the diminishing reserve army of labor in
equalization are
greater than the frictions encountered by exports of investment capital.
Many Marxists
argue that the middle of the twentieth century brought the emergence of yet
another stage of capitalist development.
This has been termed "late capitalism" by Mandel and is
thought to be characterized by the
increased
importance of "technological rents" and the further institution-
alization of
science in the profit-making process (see also Habermas, 1970). This period is also characterized by the
great expansion of the transnational corporations that coordinate production
and profit-making on a global scale.
The alleged
consequences of this new stage for core/periphery relations are described in
chapter 4. Whereas Kay (1975) argued
that production capital of any kind involving direct investment and control of
peripheral production caused autonomous capitalist accumulation in the periphery,
The thesis that
the latest stage of capitalism is less detrimental to the economic growth of
the periphery is shared to a certain extent by some dependency theorists (e.g.
Dos Santos, 1963; Cardoso and Faletto, 1979).
Peter Evans (1979) discusses the emergence of "dependent
development" in the Latin American
periphery based on the production of manufactured goods for the home markets of
Latin American countries by transnational corporations. While these authors agree that the "new
dependency" (or "associated-dependent development") perpetuates
the core/periphery structure of power in many ways, they also believe that
overall economic growth is increased by this new form of core penetration. This contention is disputed by Frank, Amin,
and Mandel, who believe that real economic development in the periphery is only
possible in the context of a socialist revolution.
As discussed in
chapter 4, Folker Fröbel, Jürgen Heinrichs and Otto Kreye (1980) have argued
that a "new international division of labor" has been created by
transnational corporations using free production zones in the periphery to
employ cheap labor to produce manufactured goods for the world market. John Borrego (1982) contends that the core/periphery
hierarchy itself is disappearing with the emergence of a new stage of
"meta-national"
capitalism. This kind of emerging stage of global
capitalism is also described by Robert Ross and Kent Trachte (forthcoming).
Cycles of World-system Development
Another view of
the stages of development of the capitalist world-system is that proposed in
chapter 2. This view periodizes
development in terms of characteristics of the system as a whole: constant
features, cyclical processes, and secular trends. It suggests a modified model of the capitalist
mode of production which is true to the spirit, but not the letter, of Marx's
accumulation model. The capitalist mode
of production is understood as commodity production for profit on the world
market in which labor is a commodity but is not always paid a wage. Capitalist production relations are
understood as the articulation between wage labor in the core and coerced labor
in the periphery. The interaction
between core and periphery is understood as fundamental to capitalism as a
system. Thus imperialism (generally
understood, including colonialism,
direct investment, core/periphery trade, and neocolonialism) are conceived to
be integral to the functioning of the capitalist mode of production. Capitalism thus includes both expanded
reproduction in the core and primary accumulation in the periphery. Primary accumulation is not merely a process
of creating capitalist production relations, but is rather a permanent and
necessary feature of capitalism as a mode of production. The state and the interstate system are not
separate from capitalism, but are rather the main institutional supports of
capitalist production relations. The
system of unequally powerful and competing nation-states is part of the competitive
struggle of capitalism, and thus wars and geopolitics are a systematic part of
capitalist dynamics, not exogenous forces.
According to
Wallerstein (1974), the first epoch of capitalist development, from 1450 to
1640, was the period of the transition from feudalism and the somewhat precarious
emergence of the institutions and class relations of the capitalist mode of
production in
The second epoch,
from 1640 to 1815, is understood as one in which the European capitalist
world-economy stagnated somewhat and then was stabilized (Wallerstein, 1980a). This was the period of Dutch hegemony and the
fierce competition between the Dutch, English, and French for the colonial
profits of the older Portuguese and Spanish empires. The competition between the English and the
French for hegemony was finally resolved after the failure of
Napoleon's bid
for world empire (or earlier, see Braudel ─ 1977:102).
The third epoch,
from 1815 to 1917, is understood as the period of the final expansion of the capitalist system to
the whole globe and its consolidation.
This period saw another economic reorganization known as the
industrial revolution, which was actually just another expansion of capital
intensity in industrial and agriculture production. The "factory system" further
extended the trend toward rapid urbanization of the work force in the
core. The Pax Britannica was
followed by the "new imperialism" in which the newly rising core and
semiperipheral powers ─ the
The fourth epoch,
from 1917 to the present, is understood as the final consolidation of the
system and the beginning of the period of its crisis and transformation to a
fundamentally different kind of system.
This is the period of the Pax Americana, the decolonization of
almost all of the periphery, the increased integration of global production by
capitalist firms (public and private), and the emergence of anticapitalist
forces which increasingly limit the manuevering room for capitalist
accumulation. The period since 1945 is
scrutinized more closely in the next chapter.
The four epochs
outlined above can be understood analytically as involving three kinds of structural elements:
(1) those underlying institu-
tional features
and developmental laws that are basic to capitalism as a system; (2) cyclical
processes which repeat themselves in each of the epochs; and (3) secular trends
which increase at a varying rate across all of the epochs. These elements and some of their interrelations
have been described in chapter 2.
The cycles and
trends listed in chapter 2 must be applied to each epoch with some
qualifications. The first epoch
(1450-1640), due to the as yet poorly formed and unstable institutional basis
of capitalist development, did not exhibit all the features characteristic of
later epochs. Thus
During the first
epoch the center of economic hegemony moved from
The third epoch
(1815-1917) may be understood as the "classic" period in terms of the
trends and cycles described above. The
British came into both politico-military and economic hegemony after the defeat
of the French in a series of wars from 1756 to 1815. Napoleon's failure to impose imperium on the
world-economy allowed the British hegemony in production to expand into the
markets of the continent and the newly independent Latin American states.
The fourth epoch
(1917 to the present) is confusing insofar as it is a period in which the
contradictions of capitalist development meet certain "ceiling effects"
or obstacles which constrain readjustment of the capitalist accumulation
process (Chase-Dunn and Rubinson, 1979).
For example, a number of the
secular trends described have run up against limits. There are no more human societies for the
capitalist mode of production to invade and conquer. There is no more territory to be grabbed by
colonizing core states. As Lenin pointed
out, further expansion by competing core states must involve the redivision of
territory already conquered. This
constraint on expansion has been further exacerbated by the achievement of formal sovereignty in almost
the whole periphery. Thus competing core
states must often vie with one another for access to the periphery by offering
better terms. This is a far cry from the
1885 Berlin Congress on
These ceiling
effects which limit the further expansion of the capitalist system to new
populations and territories do not, of course, preclude its deepening. They do, however, tend to limit the degrees
of freedom available to capital in its attempt to adjust to the contradictions
created by its own development. As well,
the "socialist" revolutions which have taken state power in the
semiperiphery and the periphery create obstacles to the maneuverability of
capital. If the present period is one of
transformation we should expect a combination of the old cycles characteristic
of previous epochs with newly emerging institutions and forms of struggle. One of the most important tasks is to sort
out the processes of the old system from
the harbingers of a new one.
For example, the
schema in chapter 2 would predict that the decline of
With the above
qualifications to the cyclical schema in mind let us now compare the
world-system approach to periodizing capitalist development with the
"stages of capitalism" approach outlined in the first part of this
chapter. Differences Between the Two Types of Periodization
Broadly speaking,
the world cycles approach emphasizes the continuity of the basic capitalist
processes in each of the periods, while the "stages of capitalism"
approach emphasizes differences of form and consequence. Wallerstein's approach implies that
production capital (both agricultural and industrial) has always been the
driving force of development and thus
"industrial
revolutions" have occurred in every epoch.
In the sixteenth century it was not merchant capital but production
capital in agriculture and town manufactures which, with revolutionized class
relations, created the basis for increased productivity, and a strengthened set
of states in the core. In the
seventeenth century it was Amsterdam's productive advantage in the herring
industry, dairy farming, and shipbuilding which was the basis of its hegemony,
not only mercantile "armed trade" (Wallerstein, 1980:chapter 2). Similarly, investment in the periphery by the
great chartered companies, the establishment of plantations worked by slaves,
and the extensive export of capital from
Sidney Mintz's
(1985) analysis of the place of sugar production and consumption in the development of the modern
world-system describes sugar cane plantations in the
Some authors have
contended that the slave-based plantation system was a form of merchant
capitalism, a contention which would seemingly be contradicted by the fact
that plantations involved the direct investment of core capital to produce
commodities. For Marx merchant
capitalism was defined as the purchasing of goods which are resold for a
profit. The idea of merchant capital has
been extended beyond this definition, especially by Fox-Genovese and Genovese
(1983). They link their idea of merchant
capital with slave production and stress the anti-capitalist world view of the
slave─holders as evidence that they were not capitalists. Eugene Genovese's earlier work (1971, 1974)
demonstrates fascinating linkages between consciousness, ideology, resistance,
and economic structures, but the contention that the aristocratic ideals of the
slave-holders proves that they were not capitalists equates ideological and
structural positions too simply. The
critiques by slaveowners of industrial slums and the crass cash nexus of the
impersonal wage system (in contrast to their own paternal concern for their
slaves) may rather be interpreted as the fruit of political competition between
two groups of capitalists with
conflicting
interests. It would not have been the
first time (nor the last) that a group of capitalists with investments sunk in
a certain form or scale of production employed the ideology of paternalism (or
nationalism) against a newer competing
group.
The tendency for states to interfere with trade, to regulate
production, the establishment of guilds and associations attempting to protect
the positions of certain producers, are not seen as characteristics unique to a
period of "mercantilism."
Rather, these attempts to use political organizations to guarantee
market advantages are understood as normally recurring moments of capitalist
development itself. Every capital is
anxious to obtain political protection of its own markets and resources. Arguments in favor of free trade are employed
by those who fear the political controls of others. In such a light the similarities of
"mercantilism" with "monopoly capitalism" are most
apparent. Braudel (1977:95) makes this
point, but argues that the creation of a protected home market only became
crucial in the nineteenth century.
Previous to this, according to Braudel, it was city-states rather than
nation-states that performed the role of hegemonic core power. It may rather be that certain processes
create a fairly constant ratio between the size of the home market of the
hegemonic core power and the total size of the
system. Thus
The argument has
been made by some of the proponents of the "stages of capitalism"
view of periodization that the form and consequences of the core/periphery
relationship have changed from period to period. Plunder was followed by trade, and that by
direct production for export, and then trans-
national corporate control of production for the peripheral market
itself. While it is clear that the forms
of labor exploitation in the periphery have
changed, the basic underlying process of "primary
accumulation" (in the sense of politically coercive labor control) have
been a reproduced feature of the core/periphery relationship clear up to the
present.2
Clearly, the form
of primary accumulation has changed from epoch to epoch, and in some sense it has become less
"primitive" ─ that is, closer in form to the "free"
labor of the core. Thus the direct
plunder of treasure and slaves was
replaced by the establishment of "coerced cash crop" labor
(capitalist slavery and serfdom) ─ which was then later replaced by
colonial trade economies in which labor
was either paid a below-subsistence wage or else small commodity producers were
exploited through a politically controlled unequal market exchange. Neo-colonial forms of labor exploitation are
undoubtedly closer in form to the "free" labor of the core, but they
are still more determined by direct political coercion.
Giovanni Arrighi
(1978) argues that the different forms of core/periphery relations are used at
all times, with varying weights depending on the period and the historical
situation of the particular core power.
It is most probably the case that "economic" penetration of
the periphery by merchants, settlers, and direct investment is more prevalent
during periods of rapid growth in the world-economy, while more politically
coercive forms of penetration (formal
colonialism, monopoly trade, and forced labor) are employed more frequently
when the world-economy is in contraction and competing producers rely more on
state power to protect or extend their share of world surplus value.
It may be true
that the gap between the degree of coercion of core labor versus peripheral
labor has narrowed, but it has most certainly not been eliminated. What is clear from a long-run perspective is
that the juridical form of labor control, that is whether or not labor is
formally paid a wage, is not the only determinant of the commodity nature of
labor. Nor is it the only dimension of
the core/periphery hierarchy. Even if
the entire world-economy were to be composed of capitalists and wage laborers,
the interstate system would tend to reproduce the core/periphery
hierarchy. The size of the relative gap
between the incomes of core and peripheral workers has greatly increased over
the long run, while absolute levels of living have risen, albeit unevenly, in
the periphery. Thus immiseration has
tended to be relative rather than absolute (see chapter 12), while exploitation
(in the formal Marxist sense) has everywhere increased.
The
Wallersteinian world-system perspective implies that there has always been a
core/periphery wage differential, contrary to the arguments of Emmanuel, Amin,
and Mandel that this developed only in the late nineteenth century. This differential has certainly increased
over the long run operation of the system, but core workers and middle strata
have always had political advantages as well as an involvement in more
profitable and more skilled types of production, and these factors resulted in
higher average incomes for core workers from the sixteenth century on. This problem is considered further in chapter
10.
Brenner (1977)
has made an important critique of Wallerstein's contention that the
"second serfdom" in
Contrary to what
Brenner implies, Wallerstein has never said that coerced labor is the most
efficient or productive form from the point of view of the whole system. Serf labor, although not very productive, was
partially commodified in the sense that it was subjected to exploitation for
the purpose of profitable commodity production, and its "price"
(costs of political coercion) influenced the costs of production and the
profitability of the operation. Brenner
claims that the juridical nature of serfdom prevented it from responding to
market forces, but the enserfment of Polish peasants in the sixteenth century
was driven forward in response to the Western European demand for grain. And the further coercive exploitation of East
European serfs in the seventeenth century was a response to declining world
wheat prices (Wallerstein, 1980a:129-44).
Another point of
contention between the two types of periodization is whether or not the consequences of
core penetration of the periphery have changed with the reorganizations of
dominant capital and core/periphery relations.
As stated above, Kay (1975), and others (Warren, 1980; Szymanski, 1981)
have argued that contemporary transnational investment in peripheral countries
creates the possibility for autonomous capitalist economic development in
these countries. On the other hand, many
dependency theorists have argued that the consequence of transnational
investment is to continue to underdevelop peripheral countries relative to
those other peripheral countries which are less dependent on foreign investment. Cross-national comparative research on the
post-World War II period reveals that countries with higher levels of
penetration by foreign capital grow more slowly (other things equal) than
countries with lower levels of transnational corporate penetration (Bornschier
and Chase-Dunn, 1985). In addition, it
has been shown that, contrary to Szymanski's argument, it is transnational
investment in
manufacturing that has the largest negative effects on aggregate
national economic growth. These studies
reveal that foreign capital investment inflows have immediate positive effects
on GNP growth, but long-run negative consequences as the effects of structural
distortions and repatriation of profits spread to the national economy as a
whole. The implication of these findings
is that core capital continues to have the effect of recreating the gap between
core and periphery. Changes in the form and structure of
core/periphery relations which have occurred in the twentieth century have not
operated to eliminate the development of
underdevelopment.
It may be the
case that, just as the terms of trade between core and periphery are known to
alternate cyclically in favor of, or against, the periphery (Barrat-Brown, 1974), the wage gap
also exhibits a cyclical as well
as a secular
increase. If this is true the increase
noted by Mandel (1975) and Amin (1975) in the 1880s may have been part of a
cyclical shift rather than a completely new trend in core/periphery relations
as they allege. Research on recent
trends in the core/periphery wage differential is reported in chapter 12.
One explanation
for uneven development is the effect which class struggle has on the
concentration of capital in a particular area.
Successful accumulation is usually accompanied by rising wages due to
demand for skilled labor, exhaustion of the reserve army of the unemployed, and
the growth of trade unions which allow workers to influence access to jobs and
the level of wages. This process of class
struggle eventually causes capital to flow to areas where labor is cheaper to
obtain a competitive advantage. This is
part of the explanation for the flow of capital to the periphery. But the continued cheapening of constant
capital inputs by technological development can also cause cheap labor in the
periphery to lose its competitive advantage.
Thus, as Mandel points out, capital-intensive agricultural production
within core countries in the twentieth century has replaced much
labor-intensive peripheral agricultural
production. This "see-sawing"
process of uneven development, the alternating concentration and spread of
capitalist production techniques and investment, creates a cycle in which the
gap between core wages and peripheral wages may become alternately larger or
smaller. Though the gap has definitely
increased over the last hundred years (chapter 12), it remains to be empirically
demonstrated whether or not this trend also contains a cycle.
In discussing
differences between the twentieth century and earlier epochs it is important to
distinguish that which is superficially different from that which reveals a
fundamental crisis of capitalism. One
reason to carefully consider the problem of periodization within the
capitalist mode of production is so that we may be able to distinguish those
new forms which reproduce capitalism from those forms which might transform
it. The mere scale of phenomena in the
twentieth century, the rate of change, the size of organizations, the global
nature of production and consumption, are often seen as qualitative differences
from earlier epochs. The world cycles
approach emphasizes the continuity with the earlier epochs in terms of the
underlying processes at work, but also alerts us to those key features which
should be watched to detect signs of structural transformation.
The discussion of the growth of monopoly or oligopoly power, or
of the emergence of a new stage of
"monopoly capitalism," often assumes that the use of politically
organized price distortions alters the underlying rules of capitalist
development. On this basis some Marxists
have argued that volume 1 of Marx's Capital is no longer relevant for
the analysis of development because it assumes equilibrium prices and
competition among capitals. I argue in
chapter 4 that this is not the case. The
world-economy remains a very competitive arena in which there are no long-run
monopolies. The profit rate which
equalizes over the long run is the "surplus" profit rate, that which
includes profits due to the exercise of monopoly power. In this light Capital, with some
revisions (the addition of the interstate system and class struggle) remains
relevant for understanding contemporary capitalist development.
Also, in the
twentieth century we have "state capitalism" in which states themselves act like firms which are
competing in the world market. Here the
use of political mobilization, coercion, and productive advantage are combined in a very direct way. This is not, however, a completely new
feature. Political power has always been
used to distort market processes in
favor of certain
groups. What remains the case in this
epoch is that there is no single overarching political authority which can
control the whole arena of economic competition, and so the process of
capitalist accumulation continues. Let
us now look more closely at the period since
World War II.
Chapter 4: The
World-System Since 1945: What Has Changed?
To answer
completely the question posed in the title of this chapter we would need a
clear formulation of the deep structural logic of the capitalist
world-economy, and a way of determining the extent to which new developments
have altered its logic. The level of
specification obtained regarding the nature of the capitalist mode of
production in chapter 1 is somewhat crude.
Yet we can employ it and the schema presented in chapter 2 to make some
guesses about the extent of change. Of
interest in such an exercise is the problem of when quantitative change becomes
qualitative change.
A New Stage of
Capitalism?
Several recent
analysts claim that capitalism has entered a new stage since World War II. This contention utilizes the notion that
capitalism goes through stages which differ from one another in systemically
important ways.1 In chapter 3
I have argued that a theory of cycles and trends can account for most of the
changes across epochs which are claimed to be stages of capitalism. Here I want to focus on the changes alleged
to have occurred since World War II, or to be occurring now.
Albert Szymanski
(1981:95) contends that a transition from "early monopoly capitalist
imperialism" to "late monopoly capitalist imperialism" occurred
around 1960. John Borrego (1982) speaks
of the recent transition from national to metanational capitalist
accumulation. Robert Ross and Kent
Trachte (forthcoming) speak of the transition from monopoly capitalism to
global capitalism.
The substantive
observations on which claims of a qualitative
transition are
based differ among the different authors.
Szymanski argues that the most recent stage is based on the
decolonization and industrialization of the periphery. Borrego, and Ross and Trachte focus on the
increased importance of globally operating firms. These contentions and others will be analyzed
below.
This discussion
of the various claims about qualitative change will be organized into the
following topics:
1 the
transnationalization of capital;
2 technological
rents;
3 the new
international division of labor;
4 world classes
and world state formation; and
5 a socialist
world-system.
The
Transnationalization of Capital
Some social
scientists, on first perceiving the reality of the world-
economy, have
assumed that its importance as a systemic logic which has major effects on the
development of subunits is of quite recent origin (e.g. Michalet, 1976), or
that it has just recently become transnational2 (Hymer, 1979). These claims may be broken down into their
constituent arguments, which involve the logic of investment decisions,
monopolization, effects on states, and effects on prices and value. Arguments about the effects of the growth and
reorganization of transnational firms on the peripheral countries will be
discussed below in the section on the "new international division of labor."
It is undeniable
that transnational firms have grown in size and importance since 1945. Their expansion and operations in particular
countries have been the subject of many excellent studies (Biersteker, 1978;
Evans, 1979; Gereffi, 1983; Bennett and Sharpe, 1985) as have various aspects
of economic and political institutions within core states that have affected
the growth of transnational investment (Krasner, 1978; Hawley, 1983; Lipson,
1985). The fact of this growth is
presented by some analysts as evidence that the logic of the world-system must
have changed. This connection needs to
be examined.
The great
chartered companies were the first transnational corporations, engaging in both
merchant capitalism (buying cheap and selling dear) and productive capitalism
(the direct organization of commodity production). They were joint stock companies which were
allocated monopoly rights and political─ military protection by the
individual core states that chartered them.
These "monopolies," however, were usually incomplete and
often short-lived because competition among the chartered companies of
different core states was rife.
Some authors have
alleged that the contemporary transnational corporations are controlled by
international groups of capitalists not aligned with any particular core state
(e.g. M. Dixon, 1982). In terms of the
ownership of stock, it has been shown that almost all of the modern
transnational firms are, in fact, owned and controlled by capitalists from a
single core state (Mandel, 1975).
Nevertheless there is some co-participation in ownership across national
boundaries. This feature, however, is
not new. Barbour (1963) reports that
disgruntled seventeenth-century
The fact that
some capital was transnational in the seventeenth century does not contradict
the contention that it is more transnational now. The question is, what difference does that
make for the logic of investment decisions and capital accumulation? The great chartered companies, along
with peripheral
plantations and mines, were primarily operated according to the logic of
productive capital rather than merely merchant capital (Barr, 1981). But it is undoubtedly the case that the
direct organization of production by capital has become much more firmly
entrenched since the seventeenth century.
Merchant capital, buying products from independent producers, has been
increasingly replaced by productive capital directly controlling the production
of commodities. Transport and
communications costs have declined in a geometric fashion, facilitating the
expansion of the spatial extent of investment strategies. Thus the world-economy is more integrated by
global investment decisions and international sourcing than ever before. But does this constitute a change in logic or
merely a change in scale?
The schema
presented in chapter 2 accounts rather well for most of the recent changes,
especially as it designates trends toward transnationalization, capital
intensity, and the increasing size of firms.
It could be argued however that these quantitative trends have led to
qualitative changes in the nature of the game.
One undeniable
consequence of the increasing integration of the world-
economy by
transnational corporations and the shift away from merchant
capitalism is to
increase the systemness of the system.
Merchant capitalism trades commodities between regions that have not
become fully integrated as systems of production. Marx (1967b) describes how merchant capital
eventually creates "abstract labor" by subjecting qualitatively
different kinds of production to an equivalent standard ─ "socially
necessary labor time." Unintegrated
systems have price structures which vary according to their social structural
uniquenesses, differences in natural endowments, and techniques of production.
Merchant capital moves goods from areas where they are cheap to
areas where they are dear in the "exchange of unequals" (Amin,
1980a). But the long-run consequence of
such exchanges is to alter the allocation of labor time in both areas such that
they move toward the formation of a single equilibrated system in terms of the
"efficient" allocation of labor and other scarce resources.
No market system
is ever in perfect equilibrium in the above sense, and indeed a certain
inequality of labor values is part of the institutional nature of the
capitalist world-economy ─ the unequal exchange between the core and the
periphery (Emmanuel, 1972). The long-run
consequence of the action of market exchange is to produce a single
interactional set of prices which reflect the competitive rationality of a
market system. Most of the remaining
structural barriers to the equalization of wages and other prices are generated
by the structure of capitalism itself.
They are generally not "survivals" from the past, but are
themselves produced by competition and conflict among classes and states in the
context of the capitalist mode of production.
The trends toward
the transnationalization of capital, the further integration of the
world-economy, and the growing importance of production decisions on a global
scale, reduce the importance of the remnants of pre-
capitalist
systems. The capitalist mode of production became dominant in the European
world-economy of the long sixteenth century, but it still interacted with
precapitalist modes of production which continued to have some influence on the
historical development of the system. As
it expanded it incorporated other socio-economic systems into itself and these
too have left some institutional remnants which have influenced the particular
configurations of develop-ment in different areas (Wolf, 1982; Nolan and
Lenski, 1985). Some
institutional
aspects of these precapitalist modes of production undoubtedly remain, but
their importance has certainly decreased with the growing
integration of
the system.
Some analysts
have argued that the increasing importance of the trans-
national
corporations has altered the logic of capitalism toward a less competitive,
more monopolized and monolithic system.
It is true that a large and growing component of international trade is
made up of intrafirm transfers.
The affect of this, it is alleged, is to decrease the overall amount of
competition in the system.
This argument is
analogous to the discussion of the transition from competitive to monopoly
capitalism (see chapter 3). It is
alleged that there was once a stage of capitalism in which the state did not
interfere in production decisions or markets, but merely provided the
institutional support for the operation of free markets in land, labor, and
capital. Firms were small, start-up
costs were low, and thus the competitive market system forced firms to produce
as cheaply as possible and to sell their products at the lowest possible
prices.
This version is
alleged to describe
Once we focus on
the world-economy rather than national economies a number of things become
clear. First, most states most of the
time have attempted to influence production and market forces in favor of some
group of capitalists. The laissez-faire
state is merely a special case, in which one set of comparatively advantaged
capitalists has succeeded in reducing the political favoritism formerly offered
to another set. Second, although
monopolies are granted by states and enforced within municipalities and by
other political organizations, cartels, guilds, unions, etc., there are no
long run monopolies in the capitalist world-economy. The political organizations which grant
monopolies are themselves in competition with one another, and, since no one
can really escape interaction in the larger arena for long, protectionist
measures and monopoly rights are themselves subjected to a logic of
competition.
These
observations are no less true in the 1980s than they were in previous centuries,
except that the size of the largest firms has increased relative to the size of
states. It is this last development
which has caused some authors to argue that competitive capitalism has changed
into monopoly capitalism.
Monopoly pricing
allows firms to pass on costs to those consumers over whom they have some
direct or indirect political influence.
If the world-
system had a
single overarching state apparatus, true and complete monopolies could be
maintained. But in a world-economy with
a competitive interstate system, monopolies are partial and temporary.
At the global
level there are no industries that could be described as uncompetitive, despite
the growth of transnational firms. The
recent glut of steel, cars, oil, ships, and other world commodities reminds us
of the continuing "anarchy of production decisions" which has always
been a feature of capitalism (Strange and Tooze, 1981). Ross and Trachte
(forthcoming) argue the opposite: that the emergence of global capitalism increases
the competitiveness of the system as market shares are no longer guaranteed
within national boundaries. The relevant
market for leading edge, core industries has always been the world market (both
national and international markets). The rise of oligopoly within national
markets enabled core firms to compete in terms of product development instead
of price competition within their "own" national markets, while the
international markets have been and continue to be more price competitive. The product cycle, in which new products are
developed in the core countries and older products move on to the price
competition of second tier core and semiperipheral countries, is not new. Joint ventures between Japanese and American
auto and electronics firms may be somewhat novel, but they do not constitute
oligopoly at the global level.
Ross and Trachte
also claim that capital flight has taken on new significance as a key lever of
domination in the relationship between capital and labor. It is clearly true
that the spatial scale of production location has expanded beyond the
organizational capacity of contemporary trade unions, but this also happened in
the late nineteenth and early twentieth centuries as textile production moved
out of New England to the Southern states of the
Several analysts
have suggested that the increased importance of trans-
national
corporations has diminished the power of nation-states. Raymond Vernon stated this thesis most
strongly in his Sovereignty at Bay (1971). A modified version has been discussed by
Marlene Dixon (1982) in her essay on "dual power." It is clearly the case that transnational
corporations have increased their power vis-à-vis small peripheral
states. And, simply as a function of
their size, the largest firms may have increased their influence over core
states as well. It should be remembered,
however, that states have also increased their powers. The question is, whether or not the changing
relationship between the size of firms and the power of states has altered the
logic of the game.
Above it was
pointed out that, contrary to the contentions of some authors, most of the
world's largest firms continue to be primarily controlled by capitalists from
one or another of the core states. Thus
there are no truly multinational firms from the point of view of
ownership. Among the 50 largest transnational
firms in the world only Unilever and Royal Dutch Shell could be considered
"binational" in terms of ownership (Bergesen and Sahoo, 1985). The extent to which these firms may
constitute an integrated world bourgeoisie is discussed below in the section on
world class formation. Here I wish to
address the relationship between firms and states.
It is obvious
that transnational firms do not control their own armies, nor do they have
powers of taxation. The usual
distinction between "private" firms and public organizations becomes
very problematic when we consider the case of state capitalism. But, even considering the growth of direct
state control of production, there remains an important differentiation between
economic and political sources of power in the capitalist world-economy.
Firms continue to
rely on states for the provision of "order."
Transnational
corporations have contradictory interests vis-à-vis states. On the one hand they need world order,
not merely order within national boundaries, and this requires a fairly stable
set of alliances among the strongest core states. On the other hand, they make great profits
from their ability to play off states against one another. States compete to offer the best deals to
attract the capital investments of the transnationals. And transnationals desire to maintain the
maneuverability which the multistate system guarantees.
The power of the
transnational firms should not be overestimated, however. Their dependence on individual states is
still very great. They cannot suppress
strikes, political challenges, or nationalizations without being able to
mobilize the police forces and armies controlled by states, and so they must
maintain influence and control over states.
This cannot be done solely by threat of capital flight. It must also be done by supporting friendly
politicians, paying taxes, and demonstrating "good citizenship" shown
by public affairs campaigns and "social" activities. This is most true of their relations with
core states, of course. But even in
peripheral states they must coopt some support, even if this only means bribing
a few generals.
The
"sovereignty at bay" thesis was most believable when it was first put
forth in the 1960s. At that time the
world economy was still growing.
Stagnation had not yet raised the banner of protectionism and the use of
political power to maintain access to markets and profit-making
opportunities. When the pie is shrinking
the world-system turns toward a much more statecentric system, which provides
the basis for neo-mercantilist "realist" interpretations (e.g.
Krasner, 1978). M. Dixon's (1982) dual
power thesis is correct. The complicated
game of competition in the capitalist world-economy is a combination of
profitable commodity production with efficient use of geopolitical power. But this is not a new development. Rather we have experienced in recent years a
shift which has occurred many times before, from capitalist profit-making in an
expanding market to equally capitalist geo-
political
competition involving mercantilism, austerity, and the threat of world war.
"Creative
financing" and the casino quality of international financial transactions
(Strange, 1986) have certainly become more salient in the current period partly
because of the geometric leap in the turnover time of finance capital which is
associated with faster communications and the further trans- nationalization of
capital and commodity and money markets.
But it was in the 1920's that Hilferding (1981) first proclaimed that
finance capital had created a new stage of capitalism because the banks had
taken control of capitalist production.
The struggle between money capital and production capital has see-sawed
back and forth for centuries. The
seemingly magic quality of funny money has always eventually returned to some
more stable relationship to use values, and I expect that the current wild
period will lead to a period of
"devalorization"
by one means or another.
Technological
Rents
Mandel (1975) and
Habermas (1970) argue that post World War II capitalism is different in
important ways because the largest firms compete with one another for
technological rents derived from the application of science and engineering to
production. Rather than competing for markets
for the same product by cutting costs and prices, the largest firms compete by
developing new products. A version of
this argument has been used by Arrighi (1982) in his description of the
important differences between the last decades and earlier periods of
world-system development. Arrighi argues
that the emphasis on competition through the engineering of new products is the
major explanation for
"stagflation,"
the allegely peculiar combination of slower economic growth and higher
unemployment with price inflation. The
uniqueness of stagflation is in some doubt however. Goldstein (1986) argues that there has always
been a lag between the long cycle of production and the long cycle of
prices. This would imply a
"normal" interval of stagflation within the K-wave (see Goldstein,
1988: chapter 10). Nevertheless Arrighi
may be right that the operation of trends toward the larger size of firms,
transnationalization, and competition through product differentiation have made
stagflation a more salient feature of the most recent downturn.
The New
International Division of Labor
Several versions
of the "new international division of labor thesis" have been
offered. The most extreme version of the
thesis claims that the core/periphery territorial division of labor has been
eliminated, with metanational capitalist accumulation taking place globally
irrespective of territorial location.
Another version contends that peripheral capitalism (based on
"primary" accumulation using coerced labor to produce cheap,
labor-intensive raw material inputs) has been eliminated as a consequence of
the industrialization of peripheral countries.
Yet another version emphasizes the political autonomy of former
peripheral areas following decolonization and the demise of the colonial
empires. We shall examine these
arguments in turn.
First let us
describe the core/periphery hierarchy as it has been conceptualized in the
world-system perspective. The underlying
analytic basis of this territorial hierarchy is the distinction between core
production and peripheral production.
Core production is relatively capital intensive and employs skilled,
high wage labor; peripheral production is labor intensive and employs cheap,
often politically coerced labor. In core
areas there is a predominance of core production, and the obverse condition
exists in peripheral areas. This means
that there may be backwaters of peripheral production within core states. Wallerstein defines semiperipheral states as
areas containing a relatively equal mix of core and peripheral types of production.
One of the main
structural features which reproduce this territorial hierarchy is the exercise
of political─military power by core states. It is not simply a matter of original
differences among areas in terms of wage levels and "historical"
standards of living, as Emmanuel (1972) implies. The wage differential between core and
peripheral workers is a dynamic and reproduced feature of the system. Core states (the most powerful political
organizations in the system) are induced to provide some protection for the
wages of their citizens, as well as supplemental benefits composing the social
wage. The core/periphery wage
differential is greater than that which would be due to differences in
productivity alone, and this differential is maintained by restrictions on
international labor migration from the periphery to the core. The great differences in capital-intensity
between the core and the periphery also account for a good portion of the wage
differential.
This territorial
division of labor is not static. It has
expanded along with the expansion of the whole system, and there has been some
upward and downward mobility within the structure. The whole system moves toward greater capital
intensity, so production processes which were core activities in the past have
become peripheral activities at a later time.
The international
division of labor has been reorganized several times before in the history of
the capitalist world-economy (Walton, 1985).
The original plunder by core states of external arenas (extremely
primitive accumulation) was replaced by the production of raw materials using
coerced labor. Core investments in
plantations and mines were followed by investments in utilities, communications
and transportation infrastructure. As
domestic markets in the periphery developed, local and core capital took up
profitable opportunities in manufacturing, and, in the most recent phase,
industrial production for export has emerged in the periphery. Throughout these reorganizations the whole
world-economy has developed more capital-intensive production, but the gap
between the core and the periphery has been reproduced.
Ulrich Pfister
and Christian Suter (1987) have demonstrated that, despite the particular forms
of reorganization which the core/periphery hierarchy experiences, there are
recurring cycles in the core/periphery relationship as well. Their study, and another study by Suter
(1987), demonstrate the existence of waves of capital exports from the core to
the periphery followed by periods of debt crisis and default by peripheral
borrowers. The experience of the 1970s
and 1980s is similar in many ways to three earlier periods of massive debt
crisis which have occurred in the world economy since 1800.
The
core/periphery hierarchy has been reinforced by an unequal distribution of
political─military power among core states and peripheral areas. Historically this was organized as a system
of colonial empires in which core states exercised direct political domination
over peripheral areas. Chirot (1977:map
2) and Szymanski (1981) have argued that the nearly complete decolonization of
the periphery has reduced the power differential between core and peripheral
states. Contrary to most discussions of
neo-colonialism, Chirot claims that formal sovereignty has eliminated the
periphery, and that Asia, Africa and
Chirot's claim
that decolonization has created a world-system with no periphery is undoubtedly
a mistake. Was Latin America then
semiperipheral immediately upon attaining independence from
Clearly some
formerly peripheral countries have become semiperipheral, and the
Another
contention about the new international division of labor focuses on the growth
of industrial production in the Newly Industrializing Countries (NICS)
(Caporaso, 1981). Sometimes this
phenomenon is interpreted as the end of a core/periphery system. Deindustrialization in the core,
industrialization in the periphery, and a shift toward control by global
transnational corporations are portrayed as the beginning of a new era of
metanational capitalism (Borrego, 1982).
The notion of
upwardly mobile semiperipheral countries has been convincingly utilized to
understand the recent developmental paths of
Cardoso's (1973)
analysis of the shift from classical dependence (production of raw materials
for export to the core) to dependent development (Brazilian production of
manufactured goods for the domestic market by transnational corporations)
claimed that a change had occurred in the effects of dependence on overall
economic development. Cardoso argued
that the trans-
national firms
would now have an interest in the expansion of the domestic market and so they
would act economically and politically to foster the growth of the national
economy, albeit in a way which might exacerbate inequalities among classes, and
this was allegedly demonstrated by the Brazilian "miracle."
Cross-national research on the effects of dependence on foreign investments in
manufacturing does not support Cardoso's claim.
Bornschier and Chase-Dunn (1985:chapter 7) find that dependence on
manufacturing transnationals has a large long-run retardant effect on GNP
growth in a cross-national comparison, although short-run effects of inflows of
foreign capital are positive, accounting, at least in part, for the short-lived
miracle in
Fröbel,
Heinrichs, and Kreye (1980) have emphasized the importance of manufacturing in
the periphery for export to the core.
They document the growth of so-called free production zones, areas
juridically outside the tariff and labor regulations of peripheral countries
which allow transnational firms to have "export platforms" for the
utilization of cheap peripheral, often female, labor. The industrial exports of the Asian
"Gang of Four" (
production
overseas. Ross and Trachte (1983) have
argued that the recent growth of sweatshops employing undocumented immigrant
workers in
The problem is
whether or not these developments are the first stages of a shift toward
metanational capitalism, or are simply the continuation of uneven capitalist
development in a period of economic stagnation, with upward and downward
mobility occurring in a structural hierarchy which is still intact. The decline of the economic hegemony of the
of the sequence
of core competition, with uneven development occurring within countries (the
decline of the older industrial Northeast "rust bowl" and the rise of
the
No one could
seriously claim that the core/periphery hierarchy has already been eliminated.
Immense differences still exist in the level of living and the capital
intensity of production. The
transnational corporations have their headquarters in the great world cities of
the core countries. Industrial
production in the periphery has certainly grown, but it remains a very small
proportion of world industry (Petras et al., 1981:chapter 6). The heavy intermediate industries (e.g.
steel, chemicals) which have grown in the semiperiphery are no longer the
leading sectors of the world economy. In
chapter 12 I review studies which estimate recent trends in the magnitude of
core/periphery inequalities. There is no
evidence in favor of the notion that the core/periphery hierarchy is moving
toward greater equality.
Neither the
industrialization of the periphery nor the deindustrialization of the core have
reduced the magnitude of core/periphery inequalities. To claim that core countries have become
peripheralized because some areas within the core have experienced economic
decline is certainly an exaggeration.
Similarly, discussions of the arrival of "post-industrial"
society in the core are certainly premature.
The proportion of the work force in services and non-manual labor has undoubtedly
grown in core countries. But, at least
for the
consumer goods,
followed by the export of capital goods, and finally lived out the twilight or
their golden ages as centers of world finance and services. The
While it is true
that industrialization has occurred in some areas of the periphery and in most
of the semiperiphery, it should be remembered that industrialization is the
application of greater amounts of fixed capital, machinery, and non-human
energy to production. It is an increase
in capital intensity. This increase has
continued in the core at the same time that it has occurred in the periphery,
and thus the relative distribution of capital intensity may not have changed. Indeed, much of the peripheral industrialization
has been quite labor-intensive. The free
production zones exist primarily to exploit cheap labor. And even capital-intensive production in
semiperipheral areas generally uses technology which has become obsolete in
the core.
Very little
quantitative empirical work has been done on changes in the magnitude of
inequalities in the world-system as a whole.
In chapter 12 this matter is discussed in detail and new evidence is
presented. Here I would like to refer to
a table which calculated the distribution of world resources in 1950, 1960, and
1970 which was presented in a review of theories and trends of convergence and
divergence among national societies by Meyer, et al., (1975). An updated version of the table is presented
in chapter 12 but it does not contain data on educational enrollments, and it
is these which I wish to compare to other attributes and resources here.
Table 2 in Meyer,
et al. (1975:232) showed that the poorest countries did not increase their
share of world GNP between 1950 and 1970, while they did increase very slightly
their proportion of world electrical energy consumed.
Between 1950 and
1960 they increased their share of the world's nonagricultural work force from
7.3% to 9.5% These figures confirm the
impression that the economic structure of the peripheral countries has indeed
changed, but that they have not, as a result, increased their share of world
output.
The largest
increases for the least developed countries are in the areas of educational
enrollments and urbanization. These institutional features (which are often
associated with "modernization" of national societies) have grown
rapidly. These changes, however, are only superficially similar to the
educational expansion and urbanization processes which occurred during earlier
times in core countries.
Education does
not expand only as a function of the growth of domestic demand for skilled
labor in industry and services. Between
1950 and 1970 educational enrollments expanded everywhere in the periphery and
semi-periphery of the world-system (Meyer and Hannan, 1979) regardless of the
level or rate of economic development.
It is much easier for peripheral states to create the trappings of
modernization than to change their relative position in the core/periphery
division of labor.
Similarly, the
urbanization explosion in the periphery and semiperiphery has been dubbed
"overurbanization" by some observers because it has occurred in the
absence of a similar growth rate of industrial employment. The gigantic cities of the periphery most often
import capital-intensive technology from the core, which does not create a
large demand for workers in industry.
Squatter settlements and the teeming "informal sector" of
peddlers, domestic servants, and small commodity producers swell the urban population. This informal sector provides cheap inputs to
large-scale enterprises and government by subsidizing the costs of reproducing
labor power, and by producing products for sale which are cheap because of the
exploitation of unpaid family labor, or subminimum wage labor (Portes, 1981).
The urban informal sector, then, is the functional equivalent of rural labor
reserves, village economies, and the "domestic mode of production"
which cheapened the wage bill in classical dependent economies by reproducing
part-life-time proletarians (semiproletarians).
In addition, the
city systems which have grown up in
World Classes and
Although it has
been difficult to maintain in the recent period of international squabbles,
warmongering, and neo-mercantilism, some have made the argument that the world
bourgeoisie is becoming more integrated as a class (Sklar, 1980; Borrego,
1982). International organizations such
as the Trilateral Commission are alleged to form the core of an emergent
monolithic world bourgeoisie based on the global transnational
corporations. This is a new formulation
of the old debate which started at the end of the last century among members of
the Second International about superimperialism versus continued
imperimperialist rivalry. Many of the
issues discussed above in the section on the internationalization of capital
are relevant, but here we shall focus on changes in the interstate system.
There has been a
world bourgeoisie since the beginning of the modern world-system, but it has
been a very differentiated, competitive, and conflictive class. Peripheral capitalists employing coerced
labor have produced for export to the core.
Core capitalists, divided by nation-state, sector, and access to state
power, have made alliances and fought wars among themselves. Often these alliances have crossed the boundaries
of core states. It is undeniable that
the frequency and importance of intracore capitalist alliances has increased as
the scale of transnational firms has grown.
The question is,
does this lower the competitiveness of the system (addressed above) and does it
alter the operation of the interstate system?
Recent attempts to forge a core-wide common policy against OPEC
(spearheaded by the internationally-oriented portion of the
Some may discern
a trend toward international political integration in the emergence of the
United Nations. Indeed, international
organizations have proliferated as the world-economy has become more
integrated. The Concert of Europe fell apart to be reorganized as the
The effect of the
spread of nuclear weapons must be discussed here. Do these weapons make
continued competition among subgroups of world capital by means of war
obsolete? Clearly a world war involving
nuclear weapons would disrupt the operations of the capitalist system, hardly
the most tragic of its consequences.
Such a war would lead to social devolution, if not the end of our
species. This outcome is a real
possibility because the separate contenders who are risking nuclear holocaust
are not in control of the outcomes of their combined interaction. Is the world-system a headless horseman, or
rather a horse with many heads, galloping toward the edge of an abyss?
State managers,
world bankers, and transnational firms create war machines as a mechanism to
provide investment opportunities, to be sure, but the weapons also have a
potential "use value" as threats to maintain or extend political
hegemony. These threats involve the risk of a holocaust even though none of the
major actors desire this outcome.
The presence of
nuclear weapons does not, in itself, change the basic logic of system
interaction. Commodity production and geopolitical competition remain the main
forms of competition. But the existence
of these weapons does imply that the normal operation of the system, with the
usual sequence of hegemonic decline followed by world war, followed by the
emergence of a new hegemonic core power, cannot continue. A real world war among core powers would
undoubtedly bring the final holocaust.
Thus the mechanism which has formerly resolved the contradictions of
uneven development can no longer operate.
Or rather, if it does operate, the game is over.
The combined
effects of increases in the ability of international organizations to mediate
conflicts and the increasing destructiveness of weaponry have decreased the
probability of the outbreak of war among core states to some extent. But it should be remembered that both
international organization and increasing destructiveness are not unique to the
post-World War Two era. Many turn-of-the
century observers believed that the "Great War" was unthinkable
because of the brutal destructiveness of industrial warfare technology. My point is that, while the probability of
war among core states may have decreased somewhat, the basic logic of
competition and conflict in the world-system has not changed and there are no
existing institutions which are strong enough to guarantee the peace.
In a purely
mechanistic system doom would be the most certain prediction. But we are dealing with a human system, a
somewhat intelligent set of actors who can surely see the way out of such a
dilemma. Perhaps a sufficiently large,
but not entirely devastating, nuclear conflagration will jar the peoples and
leaders of the world into consciousness and create the political will necessary
to outlaw warfare. The existence of such
a threat to the survival of the human species could provide the motivation for
the mobilization of a movement to create a real basis for collective security,
a world federation capable of preventing warfare among nation-states. The current manifestations of this potential
are, however, a long way from that goal.
A socialist
world-system.
Another version
of the claim that the current world-system has undergone or is now experiencing
transformation focuses on the emergence of the socialist states. I have elsewhere presented an interpretation
of these states as territories in which intentionally socialist movements have
come to state power, but have not yet successfully introduced a
self-reproducing socialist mode of production (Chase-Dunn, ed., 1982b).
Polanyi (1944)
discussed the dialectical interaction between the market principle and the
needs of society for protections against certain of the consequences of market
rationality, but his analysis focused primarily on national societies. At the level of the world-system and its
anti-systemic movements we see that the attempts to create non-commodified
relations of cooperation become encapsulated politically within organizations:
co-ops, unions, socialist parties, and socialist states. The market principle has, so far, been able
to expand its scale to reincorporate these collectivities into the logic of
competition within the larger world-system.
Thus the
contemporary socialist states are important experiments in the construction of
socialist institutions which have been perverted to some extent by the
necessities of survival and development in the context of the capitalist world
market and the interstate system.
The large
proportion of the world population now living in avowedly socialist states, and
victories within recent decades of socialist national liberation movements in
Africa, Asia, Latin America, and the
My own
interpretation disputes this contention.
I see the socialist states (including
Frank (1980:chapter
4) draws the same conclusion from recent trends in which the socialist states
have increased their exports for sale on the world market, imports from the
avowedly capitalist countries, and made deals with transnational firms for
investments within their borders.
National economic
planning, which is most highly developed in the socialist states, may be simply
the most complete expression of the trend toward state capitalism which is
occurring in most core and peripheral countries. And while distribution is more equal within
socialist states, this does not change the competitive logic with which they
interact with other states. Thus one
possible world is composed of states which are internally socialist but which
compete with one another in international markets and geopolitics.
The increasing
number of socialist states does not seem to have weakened the logic of world
capitalism. Rather the political constraints on the free mobility of capital
which these states have created push the logic of
capitalist
organization to expand its scale. States
become firms, and transnational corporations deal with all players in a
competitive world which remains subject to the anarchy of investment decisions.
Discussion and
Conclusions
To argue that the
world-system's logic has not fundamentally altered does not imply that this is
impossible or even unlikely. Nor is it
to argue that the massive expansions, emergent institutions, and shifts of
capital from place to place have not had drastic effects on the lives of
people. I would like to revisit some of
the questions raised in the earlier sections to speculate about the possible
consequences of changes which have not yet occurred, but which might occur in
the future.
What if it were
true that recent trends were the beginning of the end of the core/periphery
hierarchy. World-system theory has
claimed that peripheral capitalism is a normal and necessary part of the
capitalist mode of production, and that the reproduction of expanded
accumulation in the core requires the existence of primary accumulation in the
periphery.
This idea is not
based on the claim that peripheral production creates the bulk of surplus value
in the system, but rather on the insight that the relative harmony of classes
in the core, the somewhat peaceful accommodation between capital and labor
which exists as social democracy, corporatism, or "business
unionism," is based on the ability of core capital to emphasize
nationalist bases of solidarity.
Nationalism in the core is sustained by competition among core states
and by the ability of core capital to pay off an important segment of core
workers with higher wages, better working conditions, more welfare provisions,
and greater access to political power through democratic processes. This is possible, at least in part, because
core exploitation of the periphery provides a measure of additional surplus
value through unequal exchange (e.g. cheap bananas for core workers), profits
derived from investments in the periphery, and the status-based affects of
comparison with "less developed" countries.
In a pregnant
sentence Wallerstein predicts that "When labor is everywhere free we shall
have socialism" (Wallerstein, 1974:127).
This implies that, if the core/periphery division of labor disappears,
capitalism will no longer be able to overcome its own contradictions, and the
political structures maintained by the core/periphery hierarchy will
crumble. Socialist transformation, which
Marx predicted would occur first in the core nations, will finally visit them.
If the above
analysis is correct the imminent approach of metanational capitalism would be
good news for the world socialist movement as Borrego (1982) contends. But the formal proletarianization of the
world work force (the end of coerced labor and the decreasing availability of
alternatives to dependence on the world market) does not necessarily mean the
end of "segmented labor markets."
Political and ethnic stratification have proven effective in maintaining
wage differentials among formally "free" proletarians. The core/ periphery hierarchy could become
increasingly based on inequality between politically protected labor and
"free" labor.
If a contraction
in the magnitude of the core/periphery hierarchy were to occur, this would
exacerbate the contradictions of capitalism, which have been softened in the
past by what David Harvey (1982) has called the "spatial fix," the
ability of capital to find fresh room for accumulation by moving to where
opposition and constraints are fewer.
The expansion of the capitalist world-system has been driven by the
search for new markets, cheap inputs, and profitable investment
opportunities. Lenin (1965) pointed out
that by the end of the nineteenth century the core states could no longer find
new worlds to conquer and were forced to divide and redivide the
already-conquered world. This extensive
expansion has been supplemented by intensive expansion, the conversion of more
and more aspects of life to the commodity form, and thus expansion of
profit-making opportunities in the provision of fast food breakfasts, etc. The potential for further commodification is
great, especially in the periphery, where a substantial terrain of production
and consumption for use remains.
Increases in the
extent and depth of commodification must eventually reach limits. Only so much of human activity can be
commodified and the ability of markets and capital to expand beyond political
regulation must decline as the density and scale of political regulation begins
to catch up. Antisystemic movements
create obstacles to the manueverability of capital and place claims on
profits. The see-sawing back and forth
motion of capital must eventually generate consciousness and coordination among
groups who have an interest in collective rationality at the world-system
level. Capital flight has pitted workers
against one another for 500 years, but the growing scale and density of
political claims must eventually decrease the incentive to move and the level
of profit. The systemic crisis of
capitalism will involve the creation of democratic and collectively rational
control over investment decisions in a context in which "private"
wealth no longer has the power or the motivation to continue directing the
production process.
The growth of welfare
states, decolonization of the periphery, and the emergence of states in which
socialist parties control state power should be understood in this light. Despite the failure of these to change the
logic of the world-system so far, developments of this kind which manage to
coordinate their efforts on a global scale can indeed transform that
logic. These questions are addressed in
more detail in chapter 16.
Chapter 5 : World Culture, Normative
Integration and Community
A structural approach to the modern
world-system necessarily raises questions about the role of culture and of
normative integration. This chapter
discusses the extent to which world culture exists, and the role that
ideologies,consciousness, and collective solidarities play in the reproduction
and transformation of the contemporary world-system. Critics
of the world-system perspective have complained that cultural factors are
treated as epiphenomenal, and a number of recent efforts have been made to
remedy that situation (e.g. Wallerstein, 1984b:chapters 15 and 16; Robertson
and Lechner, 1985). My own theorizing
rests on historical
material¬ism, but I do not accept a general model in which the "base"
mechanically determines the "super-structure." This chapter argues that world culture and
normative integration play secondary rather than primary roles in the
reproduction of contemporary world order because of the nature of the
capitalist mode of production as it
operates on a global scale. But I am not
arguing that culture and normative integration are secondary in all systems, nor
that they will always be so in the future.
In this discussion of the structure,
functions, and content of contemporary world culture I will argue that
consciousness, ideology, and consensual definitions of reality and the good are
not the primary institutions which integrate the modern world-system. Rather, in comparison with earlier whole socio-economic
systems and with contemporary subunits such as families and national societies,
the capitalist world-economy is integrated more by
political/military power and market interdependence
than by normative consen¬sus. This is
not to contend that there is no such thing as world culture, but rather to
argue that its nature and structure depend on more potent economic and
political institutions. After consideration of the role played by normative
integration in the contemporary world-system I will discuss
the structure of the emerging consensual
world culture and the resistance to core cultural domination by peripheral
groups. A culture is defined broadly by
many anthropologists as a constellation of socially constructed practices (e.g.
Fox, 1985). I will use the term more
narrowly to refer to consciousness and
symbolic systems of belief and knowledge.
In this sense culture is composed of collectively held definitions of reality
and understandings of what constitutes good and evil. Thus we are talking about ideology, although
this is understood broadly to include all belief systems and religion, but also
science and secular economic and political ideas. The construction and sharing of complex symbolic systems is the most
important feature which distinguishes
human beings from other forms of life.
Symbolic systems allow us to collectively
accumulate knowledge. The relatively great
capacity for learned behavior (as opposed to instinctual behavior), which is
based on the unpreprogrammed cortex of the human brain, allows individuals to
internalize some of the accumulated cultural heritage of past social development. As Marx says in ”The Eighteenth
Brumaire• (Marx, 1978:9), this happy possibility has a dark side as well:
"the tradition of all the generations of the dead weighs like a nightmare
on the brain of the living." The institutions
and ideologies we inherit not only empower us, they also constrain us and
reinforce structures of domination.
Structures of power mold the cultural heritages which individuals
receive.
This is not the whole story of
course. If individuals only received
heritages nothing would ever change. The very process by which individuals
are confronted with competing and
contradictory cultural elements, and the
decisions between alternative
possibilities, builds an element of freedom into
the daily reconstitution of
consciousness. Large-scale processes may
carry on
according to a systemic logic which seems
to steam roll across individuals, but
individuals and groups of people
nevertheless choose their actions.
Structural
change, especially the transformation of
systemic logic, occurs at least in
part because people become aware of
alternatives and struggle in their behalf.Solidarity, the conscious
identification of individuals with one another and with larger collectivities,
is an important aspect of all social systems. Solidarity is based on
identification and the sharing of cultural agreements,of definitions, of values
and norms. It is this aspect of culture,
its
integrational function, which is the main
focus of the following discussion.
How does normative solidarity, which
unites people into families, communities,and nations, function in the modern
world-system? A recurring issue which
differentiates theories of social systems is the causal weight they attach to
ideological versus infrastructural institutions.
The shift in focus from national societies
to the whole world-system has been accompanied by a new set of debates about
this issue. Wallerstein (1979a) and Braudel
(1984) have emphasized the importance of political and economic institutions at
the world-system level which produce and reproduce material life, while others
have stressed the cultural and normative bases of world-
sys¬tem integration (e.g. Parsons, 1971; Heintz, 1973; Inkeles, 1975;
Meyer,
1987). This chapter will place the
characterization of the importance of cultural institutions in a comparative
context which employs a broad typology of whole socio-economic systems
developed by Polanyi (1977).
Talcott Parsons was perhaps the most
well-known exponent of the idea that
there exists a global social system which
is normatively integrated. In his
1961 article, "Order and community in
the international system," he argued that international law, shared
assumptions about the desirability of economic development, rational
bureaucratic organization, and political democracy form the basis of a world
normative order. Parsons did not
directly address the question of the role of the normative order in the larger
system,
because for Parsons all social systems
are, by definition, normatively based.
Social structures are defined in terms of
shared assumptions about proper rules
of behavior, and the value system is the
most fundamental feature of a social
system.
In his short book, ”The System of Modern Societies• (1971),
Parsons
applies his AGIL schema to the
"international system," arguing that core states
are differentiated in terms of their
functions within the larger international
system.
The French specialize in fine arts and diplomacy, the British in democracy,
and the
Types of Integration
But other students of comparative society
differ with Parsons's emphasis on the
role of normative integration in all
social systems. As discussed in the introduction to Part I above, Karl Polanyi
(1977) and Eric Wolf (1982) distinguish between social systems in terms of the
main type of glue which integrates them:
”Normative social systems”: Polanyi discussed reciprocal systems in
which
production and exchange are based on
consensually held norms of reciprocal obligation [Wolf's (1982) kin-based mode
of production; Wallerstein's (1979a:155) mini-systems]. While most contemporary families are
reciprocal subunits within a larger socio-economic entity,
that larger entity, the modern
world-system is, I will argue below,held together by a different kind of glue.
”Politically coercive social systems: Polanyi also discussed “redistributive"
systems which are integrated by political institutions which gather goods by
means of taxation or tribute-payment.¼2
States which organize threats and coercion are the main institutions in
such systems.
”Market-based social systems: A socio-economic system is integrated by markets
when a considerable share of its interactions, and the direction and nature of
its development, are conditioned by the buying and selling of commodities on
price-setting markets. A price-setting market
is one in which the exchange rates áÄá prices áÄá are determined by competitive
buying and selling by a large number of independent agents seeking to maximize
their individual resources.ÞZÞ
In reciprocal systems a consensual moral
order defines reciprocal role obligations and mobilizes social labor. Distribution is controlled by a set of rights
and obligations and social integration is guaranteed by socializing individuals
into the belief systems which define their roles and regulate interaction. Internalized definitions of self and the peer
pressures of a
face-to-face community are the main glue
which guarantees social order.
Redistributional activity in such systems may confer prestige on benefactors, but
authority structures are not very hierarchical.
These societies are classless in the sense that kinship groups are not
ranked and superior position is usually not inherited. And they are stateless in the sense that they
do not
contain a differentiated organization which exercises a monopoly of legitimate violence. Cultural integration based on consensus about
what is real and what is good is the main basis of social order. In politically coercive systems there are
classes and states. Surplus product is
gathered and redistributed by a political/military ruling class which utilizes
institutionalized authority backed up by force to accomplish appropriation. There are many institutional means used to
reinforce the appropriation of surplus product in these "tributary modes
of production"
(Amin, 1980a:46-70) and they vary in the
degree to which they are centralized (Wolf, 1982). European or Japanese feudalism represent a
rather decentralized, so-called "stateless," type in which each manor
is, in fact, a mini-state.
More typically we think of the great
agrarian empires in which a centralized state extracts resources from a wide
territorial division of labor. In such
socio-economic systems the political organization of force is the main
determinant of societal dynamics, but normative integration continues to play
a role. Religion is used to legitimate
the rule of the temple and the palace.
Codified law is created to centrally
define correct behavior and deviance across local communities which have
formerly relied on unwritten moral tradition.
Such "world-empires" (Wallerstein, 1979a:156) are primarily integrated
by the political competition which is articulated in a single state apparatus. The three types of integration -- normative,
political and market -- exist
in most world-systems, but world-systems
differ in the extent to which these are dominant and in the particular ways in
which they are combined. Normative integration
has changed its locus and function over the long history of human social
development. In the classless, pre-state
societies, clans and tribal groupings based on notions of kinship used
normative integration (Durkheim's mechanical solidarity) to produce social
order. Conformity was ensured by
informal social censure; the deviant was
often shamed into self-punishment (Malinowski, 1961).
When states emerged, societies became
integrated predominantly by the coercive power of political institutions of
various kinds. Military ruling classes
monopolized the means of violence and "protected" peasants from other
competing military centers in exchange for obedience and peasant produced surplus
product. The first urbanized
civilizations may have been theocracies, integrated largely on the basis of
normative legitimacy, but soon states,
competing with one another in warfare, became
integrated by political/military
institutions. Normative integration through religious and
ideological institutions continued as an important, but not central, adjunct to
military organization. Subunits such as
families, lineages, and peripheral communities remained integrated by normative
means, but these were articulated within a dominant tributary mode of
production. It was the imposition of
rule across normatively-defined communities which created the need for the law
codified rules. Normative integration in
a consensual culture does not require codification because everyone knows what is
right and what is wrong, just as the incest taboo in modern society does not
primarily rely on legal enforcement. But when an imperial state seeks to impose a
uniform set of rules across a set of normative communities which it has
conquered, the written law becomes necessary.
In market systems integration stems
primarily from the specialization of
independent producers and their resulting
objective reliance on one another
(interdepen¬dence). Polanyi (1944) argued that pure market
systems are unstable
and tend to generate new political institutions to protect societal desiderata
from the negative consequences of
unbridled competition and narrow market rationality. In the Wallersteinian version
"world-economies" are systems in which there are multiple cultures,
multiple polities, and a single integrated division of labor. The ”capitalist• world-economy
combines commodity production
for markets with political-military
competition among unequally powerful nation-states. Integration is due to interdependence and the
dynamic interaction between uneven capitalist accumulation and the balance of
power mechanism of the interstate system.
The institutions of international capital investments and the interstate
system are reciprocally interdependent on one
another and provide the structural center
of what is an extremely expansionist and intensifying socio-economic system
(see chapter 7). Thus the glue which
holds together the modern world-system is a combination of market-generated
interdependence in the world division of labor and the
political-military power of core
states. Our global political economy is called
a capitalist one because the market-based logic of accumulation constitutes a
greater part of the dynamics of the whole system than in earlier (precapitalist)
world-empires and world-economies. We
know that markets and wage labor existed in the
determine, or even much influence, the dynamic of imperial expansion and contraction. Rather it was political competition within
the arena of a single state apparatus and the extraction of resources (tribute,
slaves) by conquest which was the structural kernel of the Roman system (T.
Hopkins, 1978).
The above characterization of the modern
world-system does not imply that
normative solidarities play no role in
integrating our global system.
Normatively integrated subunits such as
families are important components of
capitalist production because they operate as income-pooling units which allow
people to combine different sorts of
resources, and thus they play a crucial
role in the reproduction of the labor force
(Smith et al., 1984). Families,
neighborhoods, village communities, and
other face-to-face primary groups
continue to play an important part in the
generation and maintenance of
individual personalities. Social psychological research shows that the
cognitive relationships of individuals to
larger solidarities such as ethnic
groups or nations are crucially mediated
by such primary groups (Shils and
Janowitz, 1948).
Thus it is incorrect to see contemporary
solidarities like the family as
merely vestigial remnants of earlier modes
of integration. Their functional
importance in the modern world-system does
not, however, mean that we ought to
understand the larger system to be
normatively integrated in the same sense
that earlier socio-economic systems were. The system as a whole is not held
together by consensual
understandings. Most exchange in the
capitalist
world-economy is not specified by
agreed-upon reciprocal obligations.
Rather
exchange is primarily organized as market
trade of commodities and as inter-organizational and interstate political
bargaining. Normative Solidarities in
the World-system
Normative integration plays an important but subservient role in the modern world-system. Normatively integrated subunits such as the
family, the local community, ethnicities, and nations (national communities)
serve important functions for the operation of the capitalist mode of
production.¼4 I already
mentioned the importance of the household
as an income©pooling group which
bears the cost of child-bearing,
child-rearing, care of the elderly, and psychological sustenance of the
self. Probably
the most important resource-pooling solidarity in the modern
world-system is the nation, not because of
the direct effects of such pooling,
but because of its linkage with the state.
Most modern states are nation-states in
which the functions of formal political organization,
authority,interest-aggregation, etc. are blended with the ideology of national
interests.
Successful competition within the world
market requires a strong and integrated
state standing behind a strategy of
competitive commodity production. Such a
state must be able and willing to use
force to guarantee the market share interests of its national investors (which
may be state bureaucrats or "private" capitalists). The use of international force requires a
certain degree of legitimacy, as does the maintenance of peace and order at
home.
Class conflicts and domestic resistance to
the use of military power are often softened by the ideology of nationalism. The viability of nationalism as a solidarity
is, of course, reinforced by its frequent correspondence with the
nation-state. But nationalism is also
reinforced by the core/periphery
hierarchy, the structured inequality between
"developed" and
"underdeveloped" nations.
Unlike class solidarity, which is
usually crosscut by the core/periphery hierarchy, nationalism in both the core and
the periphery is reinforced by the inequalities among "advanced" and "developing"
countries.¼5 Class solidarity within
core countries is reinforced by the perception by core workers that they share
national interests with their "own" capitalists in exploiting
peripheral countries and competing with other core countries. An example of
this is presented in Sidney Mintz's (1985) recent study of the place of sugar
consumption in the emergence of industrial capitalism in
Mintz's anthropology of food consumption emphasizes the social symbolism of
what and how we eat, and its connections
with power. He notes that cheap sugar
provided fast calories for core workers at
the same time that it enabled them
to consume an item which, along with tea,
had in the past been an imperial
luxury available only to the rich.
Cultural Presuppositions of the World
Market
As the above indicates, there are various
forms of community (solidarity)which are institutionally embedded in the
operation of the modern world-system. Normative integration is important for
families, neighborhoods, villages,cities, subnational regions, ethnicities,
tribal groupings, clans, class and occupational groupings, political parties,
and most importantly, nations.
These various forms of community are not,
however, the main type of glue which holds the contemporary world-system
together. Neither is political
organization, although this plays an important role in mediating
competition. The main glue of our global
system is the interdependence produced by a market-mediated network of economic
differentiation, a division of labor which Durkheim called "organic
solidarity." This form of
integration is not a solidarity in the normative sense, because it does not
require identification with a larger collective interest or the sharing of
consensual definitions of proper behavior or the good. Markets are constituted by the buying and selling activities of
large numbers of individuals (or firms) operating on their own account to
maximize returns. Normative evaluations
of different persons are ideal-typically irrelevant. But, as Durkheim (1964) pointed out,
contractual market exchange
does require a certain level of normative
consensus, as well as a certain kind
of law.
Actors must come to agreement on prices and on a generalized medium of
exchange (money) and they should agree that one exchange partner is as good as
another, or rather is as good as his/her
ability to pay. The institutionalization
of a price-setting market is problematic within a normatively integrated group,
and it is even more problematic between different culturally-¬bounded groups. Phillip Curtin (1984) has studied the world
historical development of market exchange across cultural boundaries. Curtin observes that trade between culturally
distinct groups is most often carried out by a separate
normatively integrated group which
specializes in cross-cultural trade.
Curtin terms these specialized trading
ethnicities "trade diasporas"
because they usually set up enclaves within the boundaries of the cultural
groups that they link through trade.
Curtin's work on sub-Saharan Africa led him to the idea of a trade
diaspora, which he then used in a broader investigation of cross-cultural trade
ranging from ancient
of the European colonial powers. Without trust and credit long-term,
long-distance trade is difficult to sustain, and trade between different
cultural groups requires consensus about
notions of fairness and equivalent
values. A specialized trading group
which is integrated on a kinship or ethnic basis provides the social
organization which can sustain cross-cultural market exchange. Curtin observes that trade diasporas decline
as the separate cultures whose exchanges they mediate develop sufficient common
understandings to allow direct trade, a situation he
terms the emergence of a "trade
ecumene." A certain level of
cultural understanding facilitates the linkage of different cultural regions by
market interdependence.
Of course, the emergence of a trade
ecumene does not imply that a "perfect" market mechanism operates in
international exchange. For one thing, much
international exchange (even within a capitalist world-economy) is conditioned
by political agreements similar in form to the "state-administered trade"
which was typical in world-systems dominated by the tributary mode of
production. And the "market universalism"
assumption of the equivalence of all buyers and sellers is also violated by the
fact that the capitalist world-system remains divided into separate normatively
integrated groups -- primarily nations.
Also the universal equivalent, money, is imperfectly formed in international
trade, as the history of international monetary institutions
attests (Vilar, 1976). Nevertheless, the depth of the trade ecumene
and the relatively strong operation of price-setting international markets is
undoubtedly greater in the modern world-system than in earlier world-systems
in which capitalism and commodified relations were less prevalent. The normative relations which emerge among
nation-states and transnational
actors in a market-integrated
world-economy are primarily composed of expectations about fair trade and the
protocols of diplomatic interaction. The
analysis of these normative structures has been carried out under the rubric of
"international regimes" (Keohane and Nye, 1977). Charles Lipson's study of the protection of
private property in the periphery by core states describes well the limits of
such international normative structures:
“...stable property rights and contractual relations are exceedingly
difficult to establish across national boundaries. Although collective evaluations and expectations
are an important feature of international relations, these normative properties
are weaker because political, social, and cultural communities are constituted
primarily at the domestic level, where they typically overlap and reinforce one
another (as the term ”nation-state” suggests). Thus, while it is difficult
to establish the meaning and value of property rights domestically, it is far
harder
internationally.” (Lipson, 1985:4,
emphasis in the original)
Despite the limitations of
international normative structures, the study
of international regimes has revealed important
things about the contemporary
world order and the ideological premises
and disputes which accompany world
politics.
The studies by Stephen Krasner (1985) and Craig Murphy (1984)
demonstrate how arguments over world
distributive justice in many respects
parallel ideologies of inequality and
inequity at the level of national societies.Ideological support for socially
structured inequalities also plays a role in justifying and rationalizing
core/periphery differences. Some of the non-market barriers to wage
equalization between core and peripheral workers are ideological, as is also
the case with wage differences among men and
women, or blacks and whites. Michael
Hechter (1975) has shown that
cultural¬ly-based definitions of in-group and out-group (e.g. English versus Irish)
played an important role in maintaining the structure of "internal colonial¬ism"
within the
a supportive role in structures of
domination, these are largely reinforced by
economic and political institutions. And this is even more the case when the
object of analysis is international
inequalities. The political structure of
migration control is a complex function of
popular attitudes, the economic
demand for labor, and the political power
of competing organizations and states
(Portes and Bach, 1985). And migration control is certainly one of the
most
potent institutions which reproduce core/periphery
wage differentials.The Structure of Contemporary World Culture
Both the content and the level of consensus among participants vary across cultures. Consensus is never complete and it tends to
decrease with complexity. As discussed
above, Durkheim (1964) argued that there is a normative basis to contractual
exchange in market society, although it differs greatly from the customary
obligations prevalent in societies with a less complex division of labor. In societies with a complex division of labor
in which
individual producers exchange their
products for the products of impersonal
others through the medium of money,
consensus still exists on some matters.
But an elaborate division of labor tends
to create an objective dependence of
individuals and groups of producers on
market exchange such that ”broad
consensual agreement is no longer required
for” social order”. People
may
believe anything they want to about
religion or aesthetics. Market exchange only
requires certain social agreements about the formally equal status of buyers
and sellers, the legal obligations of contract, and the terms and content of
exchange value. This necessary consensus is somewhat minimal in comparison with the degree
of consensus about ontology and sacredness which exists in simple societies,
and this minimal consensus is backed up by law and formal sanctions.
Complex societies are characterized by
relatively individuated and voluntary
forms of consciousness in the sense that
individuals exercise many options in
choosing among culturally available
identities and ideologies. Marx's
(1967a:71-83) discussion of the "fetishism of commodities" implies that
the institutionalized opaqueness by which market relations obscure the concrete
relationships among producers is functional for the operation of
capitalism because producers and consumers
are alienated from the knowledge of their real interdependencies. More recent critics (Zaretsky 1976; Bellah et
al, 1986) argue that the consumerism and privatism which are associated with individuated
identity in capitalist society alienate the individual from meaningful social
action. It is clear however that modern
national societies continue to be integrated, at least in part, through
collective consciousness. Durkheimian processes
of moral boundary maintenance, as well as strong sentiments of
solidarity are evident with regard to the
"national community." The
nation is
probably the most important collective solidarity
in the modern world-system. Ideological
hegemony, as analyzed by Antonio Gramsci (1971), certainly plays an important
part in legitimating political and economic hierarchies within national
societies. To make this point, however,
is also to recognize that consensus and cultural agreement are primarily
organized along national
lines in the contemporary world-system, and ”not• at the level of
the whole
system. The
organization of consciousness and identity along national lines is,
in fact, an important characteristic of
the capitalist world-economy. It is
not only the multicentric character of the
interstate system which allows
capital to maintain its mobility and the
capability of outmaneuvering oppositional movements. In addition, the tendency of world culture to
be fragmented into national cultures, and for collective solidarities to be
organized nationally, reinforces the structural mobility of capitalist
accumulation. The poor record of
proletarian internationalism and the bloody conflict among contemporary
socialist states have resulted primarily from the institutional
structuring of political power in the
interstate system, but they have also
resulted from virulent nationalism.
The point here is that nation-building,
the formation of national solidarities out of formerly separate collective identities,
is itself
product of the long-run operation of the interstate system and the commodity
economy.
That nations are better integrated in terms of collective identity
in the core than in the periphery is
largely a consequence of colonialism and
economic exploitation which developed the
core and underdeveloped the periphery.
Colonialism often employed a divide and conquer policy which pitted
peripheral ethnic groups against one another, while effective nation-building
in core countries was facilitated by exploitation of the periphery. The most salient feature of world culture is
its multinational character.
Natural language is the most important
bearer of cultural meaning, and ”there
is no global language•. Language remains differentiated at the level
of the
world-system. And thus collective
identity as expressed through intimately understood symbolic systems remains
multinational. This is not to claim that
consensual symbolic systems are not emerging at the global level. Cultural imperialism and the ideological
hegemony of European religion, politics,
economics, and science have produced
obvious isomorphisms among the national cultures of the world-system. Robert Wuthnow (1980) has convincingly argued
that the institutionalization of science and the emergence of different types of
religious movements have been strongly conditioned by world-system
proces¬ses. Economic development,
political equality, and nationally delimited
collective rationality compose an
underlying set of consensual themes for world
culture (Heintz, 1973). Each nation expresses its own identity in
terms of
some "uniqueness" which is
nevertheless consistent with one or another version
of these basic themes.
On the other hand the civilizational
values which constitute European
culture, and thus the culture of
domination in the Euro-centered world-
system, are by no means universally accepted despite the trend toward morphological
similarities among national cultures.
Important aspects of non-Western civilizational traditions with very
different presuppositions
about the universe are still strongly held
by large numbers of the world's peoples (Galtung, 1981; Wallerstein,
1984b:chapter 16).
The study of international regimes [i.e.
the evolution of agreements about specific issues such as aid, debt, foreign
investment, the law of the
sea, etc. (Krasner, 1985; Lipson, 1985;
Wood, 1986)], as well as discussions of international distributive justice
confronted in the debate over a "new international economic order"
(Murphy, 1984), have certainly illuminated the process of consensus formation
and disputation in the emergence of a world normative order. Stephen Krasner's (1985) study of debates
between core and
peripheral states reveals the feature,
well-known from studies of other cases
of distributive justice, that peripheral
states are much more likely to favor
an international regime which controls
resources on the basis of globallydefined interests, whereas core states are
more likely to favor the
distribu¬tion of resources according to ability of individual states to pay. Systems of linguistic equivalence have become
institutionalized in the practices of international translators, and artificial
world languages such as
mathematics are accepted everywhere. A single method of time-reckoning is
nearly universal, and pressure is exerted
to standardize other measure¬ments.
Communications protocols, traffic signs,
medical terminology, national economic
accounting, social indicators, and even
aesthetic and literary judgements are
increasingly consensualized. World
literature, world history, and even theory
and research on something called the
world-system, are perhaps expressions of
the emergence of a unitary world culture
(see also King, 1984).
Identification with the human species
as a whole is not a very important
solidarity in the integration of the
contemporary world-system. The
boundaries
of human solidarities have been expanding
their scope for a long time. The
development of "world" religions
such as Christianity and Islam separated
kinship and blood ties from the definition
of membership in the moral order.
The actual content of the idea of
universalism has expanded to include the
whole human species in the
"brotherhood of man," the "species-being" of nineteenth century
socialists. Contemporary discussions of
the global village, or spaceship earth, stress the extent to which we share a
common fate as a species. And most of
contemporary social science assumes the oneness of the human species. Of course ideas may be expressed or held by
a minority without being institutionalized into social structures. When we discuss the nature of
contemporary solidarity, the global
community is revealed to be only weakly
institutionalized. There are many competing world religions,
including the
variants of socialism. And international law, although stressed by
Parsons
(1961) in his discussion of normative
order at the international level, is yet
poorly institutionalized. Even the most central actors simply disregard
the
World Court when it suits them.
Durkheim used changes in legal systems
as a measure of the type of solidarity found in a society. In a similar fashion, many people have
examined
the world legal system to see if global
normative regulation is emerging. The
function¬ing of the
criticism (e.g. Falk, 1982). The importance of normative regulation at the
global level undoubtedly oscillates with
the level of conflict in the worldªsystem.
In addition it seems likely that international law has indeed increased
its importance compared with earlier centuries.
But this trend has not significantly shifted the overall logic of the
world©system toward that of a normatively regulated system. That is to say, though normative regulation
may have increased to some extent at the
global level, it remains a very weak
force. Complex
cultures are never very homogeneous.
Parsons (1971) speaks of
differentiated cultures in which subunits
specialize in separate but interdependent forms of consciousness and
meaning. Thus lawyers think differently from
doctors (have a subculture) which is nevertheless part of a larger integrated
culture. Perhaps the emerging world
culture is such a differentiated civilizational whole. Many have argued that certain underlying
cultural
themes have penetrated everywhere in the
modern world-system, or at least are
shared by national elites
everywhe¬re. Variously called
"modernism" or
"Westernism," this world culture
has been spread by colonialism and market
relations to every corner of our
earth. Its content has been described as
focusing on economic develop¬ment,
rational bureaucracy, secular humanism and
science, and political democracy, although
the global status of the latter is
questionable because of oscillation between
democratic and authoritarian
regimes in the periphery and semiperiphery. Nevertheless, it must be significant that
there are few true monarchies left in the world. Legitimation of the state from
"below," that is, as an organization operating in the interest of the
"people," has become nearly universal. This is very different from the ideologies of
tributary states and empires, which are agents of the
gods (see chapter 6). If there is something like a world culture
then English is obviously one
of its major languages. This has resulted from the far©flung expanse
of the
British Empire, but also from the fortunate (or unfortunate for Anglophobes) circumstance
that the British hegemony was followed by that of a hegemon carrying a similar
linguistic gene, the United States. A
hegemonic core power promotes its own language and many speakers find it a
necessary ”lingua franca•.
Certainly the number of spoken languages
has decreased over the last 500
years as a result of the European conquest
of earth. And bilingualism has
expanded to increase the size proportions
of the world's population speaking one or another of the largest
languages. But synthetic languages (such
as mathematics) and methods of time-reckoning, methods of measuring space, etc.
have spread somewhat independently of natural languages. Linguists
such as Sapir (1949) and Whorf (1956) have argued that meanings
are non-trans¬lateable across language
groups employing incompatible assumptions
about the nature of reality. We have all heard of the difficul¬ties of
translating certain German or French words
into English. How much more
difficult to express the Eskimo notions of
snow or the Navaho attribu¬tion of
modified action to what we think of as
inanimate objects (e.g. the log bumps,
instead of a bump on the log). Problems of this kind are resolved (or
obliterated) when translation equivalences
become institutionalized, as when
professional translators at the United
Nations develop standard solutions to
the problem of equivalences. The original nuances are lost, but this is
nevertheless a form of global consensus creation.
Cultural Imperialism
Culture usually reflects socially
structured inequalities. That is,
culture is
itself hierarchical and comes to reflect,
and to reinforce, hierarchy. Marx
said that the dominant ideas of an age are
the ideas of its rulers, and Gramsci
analyzed ideological hegemony in terms of
the ability of a ruling class to
legitimate itself by propagating a dominant (but not totally exclusive)
world-view. Much of what has been called world culture by
Parsons (1961, 1971)
and the modernization school, has been
called Western cultural imperialism by
others (e.g. Galtung, 1971). William Meyer (1987) has performed an
operationalization of "the
struc¬tural thesis of cultural imperialism" which examines certain key
propositions with data on 24 developing countries. He uses cross-sectional multiple regression
analysis to examine the relationships between two indicators of the penetration
of a country by Western news and information flows and three indicators of
Westernization of the economy, education system,
and consumption. Meyer does not find support for the
hypotheses of cultural
imperialism, but his study can hardly be
cited as firm evidence. The small
number of cases (small for regression
analysis) and problematic operationalizations make the results
untrustworthy. The same can be said of
other comparative studies which have sought to examine the effects of media imperialism
(see Stevenson and Shaw, 1984). These
questions are important and relatively unplowed ground for careful comparative
research. While no one can deny the existence of hierarchical aspects of
world culture, I will argue that this form of imperialism is not very central
to the functioning and reproduction of inequality in the modern world-sys¬tem.
Cultural imperialism certainly propagates
core popular culture and the "preference structures" which create
demand for the consumption of core commodities in every part of the globe. This is accomplished in part through the
centralization of communications and information-providing systems in core countries
(Schiller, 1969). Coke, Pepsi, rock-and-roll,
and American television programs are everywhere.But just as oppressed groups
within nations have often found it possible
to redefine themselves, to throw off the
styles and identities provided by dominant groups and to assert their own
"traditional" or "unique" definitions of self or group,
this kind of resistance also operates in the core/periphery hierarchy. The psychology of national liberation is
essentially the creation of new national identities in reaction to the colonial
ideologies of the past.
This form of resistance has been fairly
successful, although not unproblematically so.
The secular trend toward unifying cultural understandings remains
subject to important resistance. Groups
and nations which feel short-changed by the accretion of allegedly
"universal" culture may often redefine themselves (Wuthnow,
1980). The traditionalism of the
Khomeini regime in
of core countries, on the other hand,
appears as an atavistic denial of the best universalist tendencies of modern
society. Yet both kinds are the productof
a world political economy which divides peoples from one another and promotes
conflict over resources. Albert
Szymanski's (1981:257-88) chapter on "ideological hegemony as a mechanism
of imperial domination" shows the important extent to which core culture
has been promoted in an effort to legitimate core/periphery exploitation. Szymanski also reveals the ease with which
nationalist regimes in the
periphery may produce their own television
programs, etc. The falling cost of
communications technology has provided the
means to counter the centralization
of the provision of information and
entertain¬ment. In the modern
world-system
it is much easier for oppressed people to
redefine themselves, to adopt a more
positive self-definition even in the face
of "hegemonic" cultures, than it is
to change the position of a nation in the
economic and political/military
hierarchy.
Thus we have the phenomenon of
"overmodernization." Many
developing
countries adopt the trappings of
development: state planning, mass
education,
national monuments, etc. without being
able to create the material bases of
economic development. Meyer and Hannan (1979) show that all
countries,
regardless of the rate of economic growth,
expanded their education systems in
the period between 1950 and 1970. The worldwide education explosion, however,
is unmatched by worldwide economic
development because it is much easier to
create students, schools and teachers than
it is to institute productive and/or
profitable economic enterprises in the
context of a competitive world market.Of course some countries have rejected
modernization as Western imperialism and sought to recreate
"indigenous" institutions and ideologies. The Iranian revolution is
an obvious example, but many other peripheral and semiperipheral countries have
similar elements in their national ideologies.
It is fairly easy to accomplish such a
redefinition and rejection of the
dominant world culture because ”the world culture itself is pluralistic.
Participation in the world market, or even
in the international system of
diplomacy, does not demand much in the way
of cultural uniformity. The above points
support the contention that world culture is not the main way in which order is
sustained in the contemporary world-system.
But a counter-argument could contend that the world-system is not well©
integrated,
and that this is due to a lack of
normative consensus. Many have perceived
international relations not as a system,
but rather as an "anarchy of nations"
each out to get as much as it can. It is clearly the case that warfare is an
institutionalized part of the competition
among states. Warfare is nearly
continuous on the earth, and war among
core states is periodically produced by
the process of uneven capitalist economic
development (see chapters 7 and 8). While
warfare indicates that competition regularly breaks into conflict, it is
misleading to characterize the world-system as an "anarchy of
nations."
The system reveals many regular patterns
of interaction despite the fact that it is not strongly integrated by consensual
culture. The market interdepen¬dence and
the related political-military balance of power mechanisms operate to produce
certain features which contradict the hypothesis of a disordered Hobbesian war
of all against all. First, the
core/periphery
hierarchy is fairly stable. Despite a certain amount of upward and
downward
mobility, any area is most likely to
remain in the position in which it has
long been.
And the number of sovereign states has increased rather than
decreased.
In a system based purely on conquest we would expect the number of
sovereign states to decrease as
empire-formation takes place.
One of the most persuasive arguments in
favor of a strong world culture
is that by John W. Meyer. Meyer claims that certain values, primarily
those of
economic progress and rationality, are institutionalized as normative rules in
the world polity. As Meyer (1987:50) puts it, “Explanations of the world-wide state
system that stress cultural factors are on the right track. Yet their commitment to a narrow
conceptualization of culture caused them to miss the awareness that modern
world culture is more
than a simple set of ideals or values
diffusing and operating separately in individual sentiments in each society. The power of modern culture -- like that of
medieval Christendom -- lies in the fact that it is a shared and binding set of
rules exogenous to any given society, and located not only in individual or
elite sentiments, but also in many world institutions (interstate
relations, lending agencies, world
cultural elite definitions and organization, transnational bodies, and so on).
The United Nations, although a weak body
organizationally, symbolically represents many of the rules of the modern world
polity....It symbolizes the rules of a political system in which national©
states are constitutive citizens. Meyer
contends that a strong set of institutionalized norms accounts for the
stability of the interstate system, supports the sovereignty of existing
states, and legitimates the expansion of
state regulation within national
societies in both the core and the
periphery. Meyer notes that peripheral
states expand their internal jurisdiction
and adopt the trappings of "modernity" (welfare systems, educational
systems, etc.) even in the absence of much domestic economic development, and
there is considerable empirical support for this contention (see Meyer and
Hannan, 1979).
I agree with Meyer that the normative
rules institutionalized in the United Nations and the protocols of diplomacy
support the interstate system and the sovereignty of states. But I disagree that these norms are the main source
of support for these important central structures of the modern
world-system. Meyer's assumption that
the norms are "shared and binding" is simply incorrect. When norms are binding the cost of deviance
is opprobrium from a valued other (shame) or self-punishment motivated by
guilt. Deviance from the norms of
diplomacy or the UN Charter is not sanctioned in these ways in the contemporary
world-system. My own explanation of the
expansion,
stability, and reproduction of the
interstate system also refers to institutions, but not to rules and values
embodied in the world culture. In
chapters 6, 7 and 8 I argue that the features of states and the interstate
system to which Meyer refers (and others) are produced by an institutional¬ized
capitalist mode of production and by the efforts of groups to protect
themselves from market forces and exploitation by core powers. I agree that normative and value-based
consensus has increased at the global level, and that world culture is evolving
as a differentiated complex system of institutionalized values. But I also contend that these emergent
features do not yet play a strong
integrative role in the dynamics of the contemporary world-system. System integration is primarily mediated by markets,
and this is backed up by the functioning of the interstate system, a political/military
balance of coercive power which regularly (but not randomly) employs warfare as
a means of competition. World culture
operates to legitim¬ate commodity production and the interstate system, but is
not an important determinant of the dynamics of our world-system. Consciousness may, however, play an important
role in the transformation of the current system to one based more on consensus
and normative integration.
The Future of World Culture and Community
The trend toward world cultural
integration which can be discerned in an
increasing convergence around basic themes
and the isomorphism of national
cultures has not yet reached the point at
which normative processes have a
central role in world-system
processes. The capitalist world-economy
is an
historical system with contradictory
tendencies which will eventually lead to
its transformation into a qualitatively different kind of system. My argument
above should not be interpreted to mean
that ideas will have no importance in
the transformation of this system. On the contrary, it is my hope that a
scientific analysis of the deep structural
tendencies of the system will be
useful in transforming it. As Polanyi (1944) suggested for national
societies,
the enormously productive forces of
capitalist develop¬ment are also enormously
destructive of certain human values which
do not easily enter into the calculus
of "private" (or partial)
profit-making. The normative assertion
of these
collective values is an important part of
constructing a world polity which can
democratically and rationally plan world
production, distribution, and development.
Thus the "universalism" generated by capitalist culture needs
to be carried forth to a new level of socialist meaning, albeit with a
sensitivity to the virtues of ethnic and national pluralism (see Chase-Dunn,
ed., 1982b:chap¬ter 14). The claims made
by Parsons (1971) and the other culturalist
theorists that normative universalism is a
central feature of the existing
capitalist world-system must be debunked,
but the possibility of such a world
society in the future should be
recognized. Now we turn to a consideration
of the role of states and the interstate system in the capitalist
world-economy. A summary of the
conclusions reached
in Part I is contained on pages 8 to 11 in
the Introduction.
PART II : States and the Interstate System
Economics and politics are not really separate phenomena despite
the needs of academic disciplines to maintain their boundaries. Political economy is the study of the
interaction and interdependence between economic and political activities. We cannot understand any social system
without knowing how both power and production are organized. A debate has emerged about the nature of the
modern world-system and the relationship between its economic and political
aspects. Some authors stress the
autonomous nature of geopolitics (Skocpol, 1977; Zolberg, 1981) and others,
like Modelski and Thompson (1988) analyze long cycles of power concentration in
which the long-range military capabilities of the "great powers" are
the main focus. The world-system theory
developed here looks at the specific ways in which economic and political
action are intertwined within the capitalist world-economy. It is argued that the interstate system of
unequally powerful and competing states is the political body of capitalism,
and that capitalist institutions are central to the maintenance and
reproduction of the interstate system, as well as vice versa.
According to Levy's
(1983:table 3.1) compilation there have been 64 wars involving two or more
states classified as "great powers" since 1495. Several different definitions and lists of
wars are reviewed and critiqued by Levy (1985).
Here I am defining world wars as major conflicts among core powers for
control of core areas of the world-system. Rather than simply counting wars,
students of international conflict have devised measures of the intensity of
warfare. Goldstein (1985, 1988) uses a
measure based on the number of battle deaths per year to study cycles of core
war severity.
Though individual
conflicts may have many causes, the cycles of world war among core states are
understood here as a normal part of the restructuring of political relations
which follows the uneven development of capitalist accumulation. The rise and
fall of hegemonic core powers is also seen as a consequence of uneven
development and associated processes of class conflict and core/periphery
interaction.
Contrary to the
arguments of some analysts (e.g. Schumpeter, 1955), capitalism is not a pacific
mode of production. Rather warfare
appears as a normal and periodic form of competition within the capitalist
world-economy. In the modern
world-system warfare is in large part an adjunct to strategies of trade and
investment, whereas in the precapitalist world-empires it was itself the
primary form of competition. For
capitalism the interstate system of sovereign nation-states, which presumes the
legitimacy of warfare, is a crucial institution. It is contended here that the emergence of a
system-wide world state, even a confederation limited solely to the prevention
of warfare among nation-states, would be the beginning of the end of capitalism
as a dominant mode of production.
Chapter 6 reviews
recent neo-Marxist and neo-Weberian studies of states as organizations. The process of state-formation, the size of
states, and differences between core and peripheral states are considered. Chapter 7
examines the arguments that claim that geopolitics is a separate
game from economic competition. I
contend that these two forms of competition
constitute a single integrated logic in the modern
world-system. The ways in which the
interstate system and the world economy are intertwined with one another in the
modern world-system are examined.
Chapter 8 continues this
discussion by focusing more explicitly on the institutions of
capitalism andtheir role in reproducing the interstate system; it compares the
modern interstate system with earlier world-empires, and considers the patterns
and linkages between different kinds of wars.
Chapter 9 outlines an explanation of the rise and decline of hegemonic
core states, the most successful states within the capitalist world-system.
Chapter 6: States and Capitalism
It has been implied that the world-system perspective is a
vulgar "economistic" approach which contends that political action is
determined by economic structures. It is
undeniable that the relationship between political action and socio-economic
structures is somewhat loose. This is
quite evident when we consider the extremely complicated connections between
class interests and political action.
They are by no means as simple and direct as Marx and many Marxists have
assumed (and wished). Analogously, if we
examine the link between the world-system position of states and the policies,
organizational forms, and regime structures of those states there is not a
simple and complete fit. To be sure, we
can not explain everything about political action and state structures by
knowing how and where a country is inserted into the world hierarchical
division of labor. As John Willoughby's
(1986:43) excellent discussion of core and peripheral state formation in the
context of the internationalization of capital concludes:
Neither imperial
nor subordinate state behavior can be explained without reference to more
specific national and international historical processes. No narrow focus on the tendencies of capital
can by itself explain state behavior.
The method does account for certain general structural trends in the
evolution of the global polity, but these findings only provide a basis for
understanding the subject of imperialism proper. It is still
necessary to develop a framework that can model the interactions among
core and peripheral nation-states and international organizations. Otherwise, it will not be possible to
anticipate the shifting contours of political-economic subordination and
conflict so basic to the capitalist world.
This said we can, however, observe certain general regularities
which may be helpful in understanding the world-system as a whole, and also the
constraints and possibilities of particular states.1 Those who wish to take a less deterministic, more voluntarist,
approach to the state and politics often take the Weberian methodological route
which emphasizes variability, and seeks to explain why it is different in such
and such a place. Charles Ragin and
David Zaret (1983) have recently clarified the distinction between Durkheimian
variable-based and Weberian case-based explanation. The latter emphasizes explaining the genesis
of diversity, while the former focuses on explicating the general model rather
than the deviant case. Both strategies
are useful and ought to be combined, as Ragin and Zaret contend. What they do not point out, and what is often
missing in those analyses of particular states which emphasize historical
contingency, is that explanation of variability or diversity assumes the
adequacy of the general model with which the particular case is being
contrasted. What I wish to do here is to
focus on the formulation of a general model.
This chapter considers
recent studies which compare core, peripheral, and semiperipheral states, and
which discuss generalizations about the connection between the core/periphery
hierarchy and features of states. It
also discusses the question of the nature of the state within a capitalist mode
of production, this in the context of the world-system perspective. The interconnection between the interstate
system, geopolitics, and capitalist institutions are examined in two following
chapters. Here we will focus on
individual states.
Is it true that the
typical capitalist state is one which only provides social order and does not
intervene in markets or production decisions?
Or, to ask the question another way, is capitalism a system which is
best conceptualized as operating within the context of such a minimalist
state? What is the real relationship
between states and markets within the capitalist mode of production? Are core states typically stronger than
peripheral states, anddoes this hold for both internal power and power vis--vis
other states? Are core states typically
more democratic, less centralized, and less authoritarian than peripheral and
semiperipheral states? If there is such
a correspondence between regime form and world-system position, what explains
this correspondence? These and related
questions are discussed in this chapter.
State Strength: Internal and External
Immanuel Wallerstein (1974) contends that core states tend to be
strong, both internally and vis--vis other states, while peripheral
states tend to be weak. And he argues
that these tendencies are reinforced by certain structural features of the
world-system, by ongoing processes of exploitation and oppression which
reproduce the core/periphery hierarchy (see also Rubinson, 1976; Kick,
1980). The strong state/weak state
formulation has been criticized by neo-Weberians, as has the alleged
"economism" of Wallerstein's approach (e.g. Skocpol, 1977). It is contended by critics that some non-core
states are very strong vis--vis internal oppositional forces, and that
some core states look rather weak internally.
It is generally agreed that in external stateto-state relations
Wallerstein's generalization holds. Core
states are always more powerful than peripheral states vis--vis other
states in terms of military power and economic power deriving from position in
the hierarchical international division of labor. These different types of power may not be
perfectly correlated, as some states emphasize one or the other, and attention
must be paid to the fact that some states change their relative position,
moving up or down in the core/periphery hierarchy. The thing which
distinguishes a capitalist world-economy from earlier
world-systems is the extent to which states in the core rely on comparative
advantage in production for the world market instead of politicalmilitary
power. This does not imply,however, that
the normal or typical capitalist state is one which does not interfere with
market exchange. The laissezfaire
state is, in fact, rather atypical, corresponding to those hegemonic core
states who are big winners in the world market without needing to resort to
strong mercantile interference, or small states greatly dependent on
international exchange who do not have the option of exercising effective
politicalmilitary influence.
The rough
correspondence between external state strength and relative status in the
core/periphery hierarchy is a matter of definition for those who understand
geopolitics to be the main arena of competition in the modern
world-system. It must be admitted that
politicalmilitary competition is important, but if we seek to understand how
the modern world-system differs from earlier world-systems we must examine the
interaction between states and capitalist commodity production.
The question of
internal power is admittedly more complex than the question of external
power. I will argue that the form of the
government, (whether it is constitutionally democratic, monarchical, or one or
another form of centralized authoritarianism) is not simply related to the
question of internal state power. A
democratically constituted state may be weak or strong vis--vis internal
opposition groups, as may an authoritarian state. And state strength vis--vis potential
and/or actual internal opposition varies over time (relative to itself) in both
core and the peripheral states, as do the formal constitutional forms taken by
states and the composition of class alliances backing particular regimes.
The question of the
internal power of states is, like all discussions of power, both theoretically
and empirically problematic. Much of the
recent literature focusing on states analyzes the "capacities" of the
state toimplement policy decisions in specific realms of social, political, and
economic activity (Skocpol, 1985). This
is a useful conceptualization of internal state power, but it needs to be
clarified in several ways. The quantity
of various kinds of resources directly controlled by governmental agencies
(assuming these can be quantitatively measured) needs to be compared with the
resources available to internal groups that are prone to resist state
policy. And both the state itself and
its contending oppositional groups need to be analyzed, not as monolithic, but
in terms of the degree of united action versus competing or even conflicting
subsections. One important determinant
of the power of a state visvis internal opposition is the degree to
which state managers and state agencies support each others' actions. Indeed, Arthur Stinchcombe has made this his
definition of legitimacy (Stinchcombe, 1968:
chapter 4).
It should be noted
that it is a characteristic of the most successful hegemonic core states to be
relatively decentralized in form. Thus
the
have a nationally directed education system or a serious public
central institution for national economic planning. The
Presiding over an
economy in which transnational capital is the dominant fraction of the
"local" bourgeoisie inhibits the expansion of the state's domestic
economic role in capital-exporting countries.
The interests of trans
national capital
coalesce with the geopolitical concerns of state elites around an
"externally strong, internally weak" state apparatus. The
Evans' usage applies
best when we are considering processes such as those that occur in declining
hegemonic core powers (see chapter 9).
It is then that capital exports become relatively great as domestic
opportunities for profit contract. In
these states formerly convergent interests among different types of capital,
and between capital and significantly large groups of core workers, show signs
of increasing divergence. No doubt the
state becomes weaker (relative to itself at an earlier time) in such a
situation as the coalition of interests behind the state becomes increasingly
problematic. But it would be inaccurate
to characterize core states as typically internally weak relative to
semiperipheral or peripheral states. The
power of a state as an organization really comes down to the amount of
resources it can mobilize relative to the amount of resources which can be
mobilized against it. In core countries
there are more total resources to be mobilized, and thus a state may need to
mobilize great resources against an actual or potential internal
challenge. But decentralization and
democratic political forms are not direct indications of state weakness. In fact, these forms may help create legitimacy
and consensus among significant supporters of a state, and thus undermine
challenges and resistance to state authority.
As implied by the
above discussion, legitimacy is an important component of internal state
strength. Robert Philip Weber (1981) has
demonstrated that there is an association over time in changes in the content
of political rhetoric and the Kondratieff cycle. Weber performed a content analysis of British
Speeches from the Throne from 1795 to 1972 which reveals a 52year thematic cycle,
and this closely corresponds to the Kwave.
Weber argues that this shows that problems of political legitimacy are
linked to the contradictions produced by capitalist development. Regardless of how this empirical relationship
is explained, the finding confirms the existence of a connection between
"internal" legitimacy processes and the long economic cycle of the
worldsystem.
We normally think of
autonomy and "sovereignty" as definitionally involved with state
strength. Sovereignty is a very problematic
thing when we turn to a consideration of peripheral states. The most extreme and obvious example of lack
of sovereignty in the periphery is the formal colony, an extension of the state
apparatus of a core power. Although such an apparatus may be very powerful vis--vis
oppositional groups in the periphery, we would not consider it to be, in
itself, an internally strong state. It
is likely to lack legitimacy, and it always lacks autonomy. External and internal state strength are thus
not completely independent from one another even at the level of
definition. Sovereignty vis--vis
other states is a requirement for internal
strength as well as external strength because it would be
nonsensical to characterize a comprador state or formal colony (whose very
existence is guaranteed primarily by the forces of a core state) as being an
internallystrong state. Internal state
strength must be defined in terms of those resources which are autonomously
controlled by the particular state under consideration. In practice this is a difficult distinction
to make, but we must make it in order to distinguish colonies and comprador
regimes from strong states.
A related point, first
outlined by Richard Tardanico (1978), is suggested by the conventional use in
comparative research of government revenues or expenditures as measures of
state strength. While these are
obviously direct measures of some of the resources commanded by a state, a
better measure would be the amount of resources commanded by the state during
a period in which state power is challenged. Though the seventeenthcentury
state to mobilize greater resources when they were needed
means that the state as an organization was probably internally stronger in
A strong state, then,
is strongly supported by an alliance of capitalists which is itself unified and
has relatively convergent interests, and is thesource of important
resources. There are analytically two
elements here: the magnitude of resources, and the relative unity within and
among classes. Richard Rubinson (1978)
has analyzed the political processes by which powerful coalitions among
capitalists and landed property owners were forged to form a strong bloc behind
upwardly mobile semiperipheral states in
An important
difference between a capitalist state (which mainly provides order and other
conditions for profitable commodity production and commerce) and precapitalist
states (which directly engaged in the tributary mode of production in which
political-military power was itself the main source of surplus appropriation)
suggests reasons why a lowbudget state may be, at the same time, very powerful vis--vis
both internal and external opposition.
Capitalists want effective and efficient states that is, states which supply sufficient
protection for successful capitalist accumulation at cost. A state which does this will be strongly
supported by groups with large resources, and yet these states may have
relatively small bureaucracies and sparse budgets.
The above arguments
notwithstanding, most crossnational studies of internal state strength use some
measure of the resources available to the government and compare this magnitude
to some measure of the overall resources available in the country. The most common measure is the ratio of
government revenues to the GNP. Most of
this research has examined the causes of the relative growth of states and the
effect of state strength on other variables(e.g. Rubinson, 1977b). A recent
Ph.D. dissertation by SuHoon Lee (1986) demonstrates that growth in the extractive
capacity of peripheral and semiperipheral states (revenues), their coercive
capacity (the military), and their integrative capacity (mass
education) are primarily the consequence of international interactions (such as
involvement in interstate wars and the international market) rather than
internal factors. Cameron (1978) finds
that the extractive capacity of core states is highly related to their degree
of involvement with the international market.
Here is some
additional evidence which bears on the question of the internal strength of
core and peripheral states. Table 6.1 is
taken from information contained in the World Bank's World Tables (1983). This table shows the average levels of
government consumption as a percentage of GDP for groups of countries from 1960
to 1981. The country groups are composed
by the World Bank. In our terms, the 21
so-called "industrial market economies" correspond almost exactly
with the core. "Lowincome
developing economies" (those 43
countries with less than $405 GDP per capita in 1981) are all peripheral, while
"middle income developing economies" includes 106 countries, both
peripheral and semiperipheral. So-called
"East European non-market economies" and "high income oil
exporters" are excluded from the above groups.2
--Table 6.1 About Here--
The item, general
government consumption, is defined as follows:
General
government consumption comprises all current expenditure for purchases of
goods and services by government bodies:
that is, central, regional, and local governments; separately operated
social security funds; and international authorities that exercise tax or
governmental expenditure functions within the national territory. It excludes outlays of public non-financial
enterprises and public financial institutions.
The current expenditure of general government covers outlays for
compensation of employees, purchases of goods (excluding the acquisition ofland
and depreciable assets) and services from other sectors of the economy,
military equipment, and other purchases from abroad. Capital expenditure on national defense (except
for civil defense) is treated as
consumption,
whereas all expenditure on capital formation (including civil defense) is
included in gross domestic investment.
(World Bank, 1983 I:xi)
Government consumption
as a percentage of GDP is by no means the ideal measure of internal state
strength. It does not take account of
the resources which would become available to a state in the face of an
emergency, an important component of state strength suggested in the argument
above. Neither does it at all capture
the dimension of unity (or disunity) among state agencies nor potential
opposition groups, nor does it include certain capital expenditures (e.g. land
purchases) or the expenditures of state-owned firms, which probably are
indications of internal state strength.
Undoubtedly this measure is affected by differences in accounting
procedures and other things which are unrelated to state strength. It does, however, take roughly into account
the magnitude of the economic resources normally available to states and
weights this by the total value of final economic transactions in the society
(GDP) As such this ought to be a rough
proxy (in cross-national comparison) for the normal extractive power of a state
vis--vis its national society.
Table 6.1 shows that
the percentage of GDP attributable to government consumption has risen since
1960 in all groups. This supports other
research (Boli, 1980; Lee, 1986) which has demonstrated increases in
state-formation in both the core and the periphery. Of more relevance to the question of
differential state strength, however, is the indication in table 6.1 of considerable differences between zones of the
world-system with regard to the extractive capacity of states. The level of average extractive capacity
amongcore states is significantly higher than that found in either middle or
low income developing countries, and this difference continues over time
despite the rise of each group.
The differences
between the low and middleincome groups indicate, with one exception, that
semiperipheral states may be internally stronger than peripheral states. This conclusion is uncertain however, because
the group of middleincome countries includes both semiperipheral and peripheral
countries.
A State-centric View of Exploitation
The recent literature which emphasizes the relative autonomy of
state managers, and their inclination to expand the state bureaucracy and to
organize stable and protected access to resources to expand state power and control as far as
they can needs to be considered within a
context of different modes of production. Charles Tilly (1985) has
characterized modern nation-states as predatory accumulators on their own
account, as legalized protection rackets operating to obtain the largest
possible share of resources. It is
important to remember that, although there has been a trend toward the growth
of states in both core and periphery, there still remain important differences
between
the operation of states within a capitalist world-system and the
operation of states within systems in which tribute-gathering is the main form
of accumulation. States within the
contemporary world-system certainly have a tendency to expand their resources. State managers often and regularly attempt to
extend their power by finding or creating constituencies which allegedly need
their services, and by expanding state access to resources. But the objects of state policy and the
continuing limitations on the use of state power need to be considered in the
context of a world capitalist system.
The expansion of core
states has been analyzed by certain Marxists in terms of necessary correctives
to contradictions produced by "monopoly capitalism." Baran and Sweezy (1966) noted that the
post-Korean war expansion of the
But O'Conner's work
also contends that there are important limitations on the further expansion of
state expenditures, and this idea of constraints is often neglected by
state-centric analysts. Contrary to the
tone one often finds in the literature on the predatory expansiveness of state
managers, there remain powerful forces which limit the tendency to state
expansion. Taxation has been resisted in
all historical systems, but in a capitalist
system the ruling class itself lives primarily on profits from
commodity production rather than tax revenues.
Tax payments constrain profit-making in many ways. Taxes on capitalist firms raise the cost of
products and reduce competitiveness, and thus profits. Taxes paid by consumers lower the effective
demand for commodities. Thus, as Fred
Block (1978) and many other analysts have pointed out, the very structure of
capitalist accumulation limits states to activities which promote an atmosphere
of "business confidence."
States themselves are
often major purchasers of products from the private sector, of course. And, increasingly, modern states are
themselves directlyentering into the production of commodities. Bennett and Sharpe (1985:71) usefully
summarize the differences between private and publicly owned firms. Public firms may operate at a loss if there
is sufficient political support to subsidize them. But even state capitalism is eventually
subjected to cost considerations emanating from competitive markets. If states produce for export they must
compete with foreign producers, and so cost considerations are important. And, even when they produce only for their
own monopolized domestic market, there are important constraints. If they produce for a protected internal
market the political costs of maintaining an internal monopoly vary with the
disparity between the internal price and the world price. Above a certain differential the costs of
preventing smuggling, and/or illegal internal production, become exorbitant. As long as the interstate system remains a
competitive arena there is a tendency toward the "equalization of surplus
profits" in which the political conditions for the maintenance of
monopolies are subjected to a logic of cost efficiency.
The historian of
action between state power and the economic growth of firms in
terms of the notion of "protection rent." States, which he calls
"violencecontrolling enterprises," are differentially successful in providing
effective and efficient protection to merchants and commodity producers. In the context of a competitive interstate
system and an international pricesetting market, protection rent is an
important component of the profits which are gained by firms. Lane explains,
An essential
charge on any economic enterprise is the cost of its protection from disruption
by violence. Different enterprises
competing in the same market often pay different costs of protection, perhaps
as tariffs, or bribes, perhaps in some other form. The difference between the protection costs
forms one element in the income of theenterprise enjoying the lower protection
cost. This element in income I will call
protection rent. (Lane,
1979:1213, emphasis in the original).
Though Lane's notion has been applied primarily to
"mercantilist" states, protection rent, and competition among states
to provide relatively efficient protection for their international merchants
and producers, continues to be an important constraint on state expenditures in
the contemporary world-system because capital can migrate to where effective
protection costs are lower.
Of course, as William
H. McNeill (1982) has argued, there is a contextual effect by which the level
of expenditure of any single state will be justified in rising along with the
general level. McNeill contends that
market-based industrialism and rapid innovations in military technology have,
in the context of the continuation of an extremely competitive interstate
system, created a virtual explosion (excuse the grim pun) of military
expenditures.
The expansion of
states everywhere, and their increasing tendency to become directly involved in
the process of economic development, may have somewhat weakened the constraints
on state expenditures which emanate from the world market and the competition
for investment capital, but these forces still remain important limitations on
the expansion of state appropriation of resources.
Raymond Duvall and
John R. Freeman (1981) present an excellent theoretical analysis of state
entrepreneurship in dependent capitalist states, mainly semiperipheral
ones. They make the important point that
there is no such thing as "the capitalist state" in general. The particular articulation of each state
within the larger world-system must be taken into account in any theory of
state economic policy. Peter Evans's
(1979) analysis of the Brazilian trip, the alliances, bargaining, and
competition among statemanagers, local capitalists and transnational firms operating
in
It is extremely
difficult for states to cut themselves off from the larger world-system and to
create a closed internal economy, although many have tried. The most successful in some respects are the
large semi-peripheral
states, especially
military threats that tend to undercut autarchic
industrialization.
Mercantilist state power (la Friedrich List) was also
used to protect infant industries from foreign competition in earlier
successful industrializers
England, the United States, Germany and Japan (Senghaas,
1985) and state power has been very
important in the recent industrialization of Brazil, Mexico, India, and of
course South Korea and Singapore.
Contrary to the
implications of Gershenkron's (1962) analysis, it is not only late
industrializers that employ mercantilist protection and state intervention in
order to foster capitalist accumulation.
Both
competition. In
Elizabethan England the government acted to constrain thebusinesses of foreign
merchants in
Similarly the
imperial revolution against
development, the granting of commercial monopolies, and the
regulation of land and water use.
Intervention at the federal level turned toward an increasingly
mercantilist "American System" policy in a series of struggles and
not a few setbacks, which were finally settled by the Civil War. My own study of tariff politics in the US
between 1812 and the Civil War shows, in this single policy terrain, how
shifting coalitions of core capitalists, peripheral capitalists, farmers, and
urban workers eventually led to the firm establishment of core capitalism in
the United States (Chase-Dunn, 1980).
It is obvious that
some states are able to move upward in the core/periphery hierarchy. These include Gershenkron's so-called late
industrializers. The world-system
analysis of "national development" views these cases of upward
mobility as exceptions against the background of the more frequent
"development of underdevelopment."
This is not just a matter of vocabulary.
Discussions of the state and national development which focus only on
the industrialization of national economies have difficulty accounting forthe
phenomenon of a reproduced core/periphery hierarchy based on uneven
development, with dependent industrialization occurring, but little or no
reduction in the overall magnitude of world-system inequality.
As Peter Evans and
John Stephens (1987) have persuasively argued, state control and market freedom
are not mutually incompatible alternatives. States increasingly act to both
control markets and create them. Indeed,
the policies of states that successfully promote capitalist development are
oriented, not
only to the efficient supply of social order, but also to the
creation of structures which promote profitable enterprises. State capitalism does not simply wait for
entrepreneurs to succeed so that it can tax them. It acts to create opportunities for
entrepreneurs, and sometimes it takes on the entrepreneurial role itself. Evans (1986) has analyzed the Brazilian
creation of a domestic micro-computer industry by some state
"technicos" who managed to create a constituency for themselves by
spawning domestic computer-producing firms.
"
It is not simply a
question of intervention versus the free operation of markets, but rather the
goals and content of state policies. It
is undoubtedly true that economic intervention has increased and become much
more sophisticated than when states imposed import and export tariffs primarily
as a mechanism for raising revenues. But the larger point is that this kind of
intervention is not precapitalist nor anti-capitalist, but is rather the normal
operation of states within a capitalist mode of production. The definition of capitalism as private
business conducted in the context of a minimalist state was a mistaken
representation produced by focusing on a core state (
Peripheral and Semiperipheral States
Rather than focusing on the exceptional success stories, the
upwardly mobile states, attention should be given to the more usual patterns
revealed by the relative stability of the core/periphery hierarchy and its
constraints on state action. As many have
argued, even core states are limited in terms of their possible actions by the
fact of their interdependence in the larger world-
system.
Political-military conflicts and economic competition restrict the range
of policies which a core state can adopt.
And of course, the longrun operation of the whole system conditions the
kind of class structure, political institutions, etc. which we find
"internal" to core states as well as
peripheral states (e.g. Walton, 1981). Nevertheless, the room for manuever is
considerably greater for core states than it is for peripheral states because
their access to resources is relatively greater and less dependent on external
forces or constrained by internal opposition.
The observation that peripheral capitalism relies more heavily on
political coercion to maintain class relations and to accomplish production and
distribution can be linked to an analysis of the organizational forms of states
and regimes in the periphery.
Marx's examination of
"primitive accumulation" (Marx, 1967a:part 8) discusses the plunder
and dispossession which occurred in the expansion of European hegemony. In a sense capitalist accumulation was
always, and still is, more "primitive" in the periphery. Its reliance on political coercion is more
than a passing phase which occurs during the creation of capitalist
institutions. The precapitalist
world-empires utilized political coercion directly in the maintenance of
master/slave or lord/serf relations.
Thegathering of tribute, taxes and rents was a relatively visible form
of surplus appropriation compared to the more opaque form of exploitation in
the
capitalist/proletarian relationship. So peripheral capitalism does bear a greater
similarity to precapitalist societies based on the tributary mode of production
than does core capitalism.
And yet it is
misleading to conceptualize the developmental process we observe in peripheral
areas as a separate precapitalist mode of production, or a period of transition
to full capitalism. These usages are
more than merely semantic differences because a mode of production ought to be
largely self-reproducing, and a period of transition should not last for
centuries. But peripheral capitalism,
though it changes its form and, in some ways, does take on aspects of core
capitalism (more proletarianization, more commodification, more state
formation, more nation-building, etc.) never arrives at the destination of
"advanced capitalism." That
is, the relative gap between the core and the periphery is reproduced, not
eliminated (see chapter 12).
Thus peripheral
states do develop, just as the economy of the periphery does. But they very rarely become core states. Peripheral areas experience state formation. Decolonization creates formal
sovereignty. The state arrogates greater
powers over other social and political organizations such as tribes, village
communities, ethnic and religious organizations, etc. The expansion of mass
education performs the magical rituals of nationbuilding by producing and
distributing national ideology and the political status of the citizen (Ramirez
and Rubinson, 1979).
And yet other features
of peripheral states seem more reticent to "develop." The literature on modernization and democracy
argues that economic development is necessary in order to institutionalize a
democratic polity. Yet, though there has been a good deal of industrialization
in many peripheral and semiperipheral countries, this has not resulted in
stable democratic government in most cases.
If we want to analyze
the possibilities and constraints of peripheral and semiperipheral states the
first thing we must notice is that many of these states depend on core states,
transnational banks, or core-based transnational firms for a significant share
of their resources, and these forms of support come with certain explicit or
implicit limitations on state action.
Bruce Moon's (1983) comparative study of the foreign policies of
peripheral states demonstrates that patterns of voting in the UN can best be
interpreted in terms of rather stable structures of core power domination based
on the dependency of periheral states rather than in terms of a more flexible
set of bargaining relationships. And
Nora Hamilton's (1982) study of Mexican social movements and the state in the
1930s shows the effects of changing core state policy on the limits of
peripheral state action. One of the
biggest factors allowing a significantly populist and nationalist state action
(the expropriation of USowned oil companies) by the
Maurice Zeitlin's
(1984) careful study of the class backgrounds of nineteenth century Chilean
statesmen and the class forces behind two civil wars in
factors") explain the failed "bourgeois
revolution" of Balmaceda, actually reveal a clash of interests between two
sets of peripheral capitalists: those vested in copper mines (who were
suffering from declining prices and stiff foreign competition for export
markets); and those rising producers of nitrate exports (who were enjoying
grand profits).
Balmaceda's intended
policy of state mobilization was supported primarily by the copper interests
who wanted to tax the nitrate exporters in order to invest in infrastructure
which would improve their position in the world market. Yes, this is a class struggle explanation,
but one in which the interests and actions of the important class fractions are
heavily influenced by their points of insertion and varying fortunes in the
world market. The "rounds of
accumulation" notion, which is useful in understanding regional uneven
development in core areas (e.g. Smith, 1984), can be seen operating in the rise
and fall of extractive and agriculture export products in peripheral
areas. These boom and bust sequences
account for a great part of the political changes which occur in the periphery
(Bunker, 1985).
The world-system
perspective also has implications for the nature of politics in semiperipheral
states. In addition to these states
being relatively intermediate in terms of their levels of internal and external
power, it is thought that the relative balance of peripheral and core types
of production within some semiperipheral states tends to create
combinations of class interests and regime form which, it is argued,
differentiate semi-
peripheral from both core and peripheral states. A recent volume edited by
Giovanni Arrighi (1985) presents a collection of studies which
examine the applicability of the semiperiphery concept for understanding
twentiethcentury political changes and patterns of economic development in most
of the countries of
These studies exhibit
the many problems of trying to understand the particular histories of countries
with a concept which has emerged from the attempt to describe and explain
features and processes which appear when we focus on the world-system as a
whole. When we use a telescope we see
different patterns than when we use a magnifying glass. Nevertheless the studies are illuminating,
not only because we learn a lot about the countries which are examined, but
because the exercise clarifies some of the confusing aspects of the
semiperiphery idea and some of the limits of its usefulness (see chapter 10).
Regime Form: Democracy and Authoritarianism
It is a common observation that, though both core and peripheral
states exhibit authoritarian regime forms, this feature is much more frequently
found in peripheral or semiperipheral states.
Kenneth Bollen (1983) presents a cross-national study of the
relationship between world-system position and political democracy which
demonstrates that both semiperipheral and peripheral states are less likely to
have democratic regime forms than core states.
Bollen's findings also show that peripheral states are even more likely
to be undemocratic than semiperipheral states (Bollen, 1983:table 2). This striking larger pattern requires
theoretical attention.
There has been much
recent study of peripheral and semiperipheral states, their organizational
characteristics, and the ways in which the state is linked to the local class
structure. Michael Timberlake and Kirk
Williams (1984) use cross-national data to examine the relationship between the
levelof penetration by transnational firms, exclusion of non-elite groups from
politics, and levels of repressiveness of peripheral and semiperipheral
governments. Their results show that
dependence on foreign capital does not have a direct effect on government repressiveness,
but it is associated with political exclusion, and affects repression
indirectly through its effects on exclusion.
Thus, among peripheral and semiperipheral countries, it is those most
dependent on foreign capital that are most likely to have authoritarian
states.
Guillermo O'Donnell's
(1978, 1979) analysis of bureaucratic
authoritarianism in the Southern Cone countries of
Clive Thomas
(1984) has examined the authoritarianism of more recently decolonized
peripheral states in Africa and the
articulated through the state apparatus are virtually the only game
around, and so fierce competition over control of the state undercuts the
emergence and maintenance of democratic institutions. Regimes tend to form around an authoritarian
one-party state.
When we compare the
analysis by O'Donnell with that of Thomas the first things that strike the eye
are the important differences between the semiperipheral Southern Cone
countries with their relatively developed national economies and significant
middle classes, and the more peripheral
While these
explanations may seem to be at odds, they are easily reconciled if we employ
two explanations suggested by the literature which compares regime types. A recent review of this literature by Peter
Evans and John Stephens (1988) suggests a useful synthesis of the approach
suggested by
Barrington Moore's (1966) study of democracy and dictatorship
and recentMarxist analyses which focus on the organizational power of the urban
working class.3
The class power
approach is complementary with
class struggle, building strong autonomous political
organizations (unions and parties) which exert power on the state to extend
citizenship rights and welfare rights.
This is, of course, a restatement in Marxist terms of
T. H. Marshall's (1965) thesis.
The relative size of
the urban working class and middle stratum is known to vary with the level of
national industrialization and the position of each country in the
core/periphery hierarchy. When we add
this to
Evans and Stephens
(1988) add two other factors. They argue
that late industrialization is more capitalintensive and thus creates a
relatively smaller industrial working class, and this weakens the association
between industrialization and democracy.
Also they argue that state strength and democracy are inversely related,
contrary to the argument I have made above.
In support of this last contention they interpret Mouzelis's (1986)
study of semiperipheral oscillation between populism and authoritarianism (see
below) as the consequence of a precociously overdeveloped state apparatus. Similarly they attribute the authoritarian
nature of the East Asian NICs to their relatively strong and interventionist
state apparati. And conversely, they
characterize the formerly British
The contention that
state strength is inversely related to the stability of democratic regimes is
another instance of the confusion (discussed above) of state strength with
authoritarianism. I argued above that
democratic states are more likely to be internally strong than authoritarian
ones. I know of no cross-national
comparative research which has directly addressed this question. The easiest
empirical test is the relationship between government revenues per capita (or
per GNP) and regime form. This would not
completely answer the question because of the problem mentioned above there is a potential discrepancy between the
actual amount of resources a stateobtains through taxation and the amount it
might obtain from supporters if its power were challenged. Even so I would guess that the correlation
between this somewhat faulty proxy for internal state strength and authoritarianism
would be negative in cross-national comparison.
Some support for this
surmise is had from the study by Thomas, Ramirez, Meyer and Gobalet (1973:table
11.3). They demonstrate that the level
of economic development has a negative effect over time on the centralization
of the party system and the likelihood of having a military regime. If internal state strength is associated with
the level of economic development (as indicated by my table 6.1), then it is likely
that authoritarian regimes are associated with (and perhaps caused by) weaker
states, not stronger ones.
To the agrarian class
relations and working class power explanations we can also add some other
features which come from our analysis of the capitalist world-system. We have already mentioned the world-wide long
business cycle (Kwave) which affects political developments in all states. There are several major forces operating
which, in combination, account for the overall core/periphery differences in
regime form. All states are more likely
to take an authoritarian form when they are significantly threatened by
internal or external opposition. Thus authoritarianism is a sign of weakness,
not of strength. This applies to both
core and peripheral states. Most
analyses of fascism understand it as a reactive response to strong opposition
from the Left. Socialist states are
thought to move toward authoritarianism as a result of both internal opposition
and threats from foreign capitalist states.
Political and economic crises in the core and in the periphery
tend to evoke a rise in centralization and authoritarian actions by
states. Even the strongly
institutionalized core democracies do this during time of war, and
recentpolitical commentary from the Trilateral Commission has suggested that
democracy in the core may have to be tightened up as the increasing
"cacophony of equity demands" produce an "ungovernable"
situation (Wolfe, 1980).
Against this tendency
for states to become more authoritarian in the face of increasing opposition is
an opposing tendency which affects all states in the capitalist world-economy
(but not equally). This is the tendency
for capitalism to structurally sustain a political ideology of egalitarianism. Commodity production and markets assume that
buyers and sellers have equal political standing, and the process of the
commodification of labor-power supports an ideology in which both capitalists
and workers are defined as equal citizens freely exchanging labor for
wages. These institutional supports for
democratic ideology exert pressure on states to adopt a form of government
based on legitimation from below, from the "people" or citizens. Rather than states as agents of God, most
modern states are constitutionally defined as agents of the people. This institutional push on modern states
stems from the process of commodification.
It has resulted in the nearly worldwide shift from legitimation based on
the divine right of kings to legitimation based on the consent of the
governed. Authoritarianism is, of course,
contrary to democratic ideology, but most contemporary forms of
authoritarianism are defined as temporary and necessary exercises, rather than
the restoration of legitimation from above.4
The ideological
pressure toward democratization affects the core and the periphery to different
extents precisely because commodification is more complete in the core than in
the periphery. This partly explains
core/
periphery differences in both the degree and the form of
authoritarianism. Not only are authoritarian
regimes more common in the periphery, but some of theseare legitimated by
traditional hierarchical ideologies. The
world's few really powerful remaining monarchs are in
Recent studies demonstrate that peripheral
states in which the majority of citizens are freeholding peasants have
difficulty enforcing policies and appropriating resources when peasants are
opposed (Hyden, 1980). Stephen Bunker
(1983) has shown that it is not really land tenure as such, but rather the
combination of local effective control of land use with the production of
export crops on which state development plans depend. This is a crucial
combination that allows local oppositional groups to effectively resist the
peripheral state. These and many other
possible combinations which create openings for either state autonomy or
effective local resistance may account for a good deal of the variation in
state strength, and thus affect the tendency to form an authoritarian
regime.
In addition to the
ideological effect which commodification has on politics, there is another
connection between core capitalism and democracy. It is an interesting fact that all the
hegemonic core powers, both citystates and nation-states, which have been
analyzed as hegemonic centers of the European capitalist world-economy by
Braudel (1984) have been republics or
federations, or both. This could
be explained by the ideological connection with commodity production, but it
may also be due to the need for a hegemonic core state to have a responsive
state apparatus which can adjust quickly to changing needs for appropriate
state policy in a rapidly changing world marketand geopolitical environment. If this sounds too functionalist, run it the
other way. Those core states that happen
to have such a democratic, pluralist regime are better able, in combination
with other necessary ingredients, to take on the role of a hegemon in a worldsystem
in which commodity production is an important form of competition. Thus
This factor does not
contradict
advantage of leaving primary production to others.
The matter of working
class power must be reinterpreted, but not discarded. It is not so much a matter of the industrial
revolution la the factory system creating an industrial proletariat as
usually conceived. Many
of these core powers did contain a significant industrial and
productive sector which facilitated their centrality in the expanding world market,
but in addition to this, the fact of being at the center of a world-system,
with higher rates of profit and more opportunities for "clean" work,
made class struggle a less contentious matter than it was in other areas (see
chapters 10 and 11). This extension of
Lenin's aristocracy of labor thesis back in timeadds a consideration to the
explanation of regime form which is not suggested by analyses which focus only
on national development and ignore the core/
periphery hierarchy.
Wallerstein suggests
that the internal strength and spatial scope of state power is an important
difference between semiperipheral and peripheral areas. Evans (1979) analyzes the strengthening of
the semiperipheral Brazilian state after the military coup of 1964, and the way
in which the state
effectively mediated bargaining between transnational firms and
national capitalists. Evans notes that
the effectiveness and autonomy of the authoritarian Brazilian state varied
across different sectors of the economy depending upon the production and
market characteristics of different
industries. Bunker
(1985:chapter 4) shows that the Brazilian state's autonomy was quite limited
when it attempted to regulate an internal extractive periphery, the Amazon
region.
O'Donnell's (1978, 1979)
work on bureaucratic authoritarianism contends that the semiperipheral states
in South America have been quite effective in regulating their national
economies, and have demonstrated a high degree of autonomy vis--vis both
internal private power groups and transnational firms. The recent reversion of these same states
back toward populist and democratic
constitutional forms supports the idea that they are relatively
strong states, at least if my argument about the relationship between regime
form and state strength is correct.
Portes and Kincaid (1985) have argued that the recent shift back toward
democratic forms in Argentina and Uruguay resulted from a "crisis of
authoritarianism" in which the economic policies of the authoritarian
regimes were exhausted and largely ineffective, and the level of opposition had
risen high enough to promote a return to democratic government, albeit under
difficult economic and political conditions.5
Nicos Mouzelis (1986)
has explicated a theory of semiperipheral politics which accounts for an
apparent long-run cyclical swing between authoritarian and populist regimes
based on the contradictory tendencies of dependent capitalist development. Mouzelis applies his theory to interpret the
political history of
authoritarian regimes depending on the shortrun successes or
failures of particular development schemes, and ups and downs in the world
market.
This explanation
differs from that of O'Donnell in two ways.
First it tries to explain semiperipheral politics in longrun structural
terms rather than focusing on the details of particular conjunctures. Secondly, Mouzelis and O'Donnell differ in
their portrayals of the strength or weakness of "civil society." O'Donnell claims that it is the middleclass
groups demanding more from the state which precipitate authoritarianism, while
for Mouzelis it is the lack of strong and supportive middleclass organizations
which underlies the instability of democracy.
Of course both could be correct if there is a nonlinear relationship
between middleclass strength and authoritarian regimes. A middle class may be strong enough to press
political demands, but not sufficiently strong to resist the opposition which
such demands engender.
The growing scale of
economic production has stimulated greater involvement of states in the economy
and the increasingly frequent use of the ideology of corporatism. As local economic circuits become more
completely integrated into national and international networks, states come to
be the only organizations which are large enough to exert leverage in the
economy. Corporatist ideology, the
notion of an interclass organic unity of interest which is mediated by the
state, takes many forms. In semiperipheral
countries both the bureaucratic-authoritarian and populist-democratic regimes
have utilized corporatist ideology.
Some of the apparently
contradictory implications of case studies of particular states with the
analysis of states located in the core/periphery hierarchy may result from
confusion between different types of comparison. Many case studies compare a state to itself
at an earlier point in time, whereas world-system studies most often compare
states to one another. Thus a state may
indeed gain in internal strength relative to itself, as Evans and O'Donnell
have claimed for
To summarize, I have
argued that core states are stronger internally and externally than peripheral
states, and they are more democratic.
These features of states are thought to result from a combination of
several world-system processes interacting with nation-building, state
formation, and class struggles. Only
further comparative research can place weights on
these various factors and settle matters of controversy. The matter of internal state strength needs
further conceptual clarification and empirical operationalization, as does the
relationship between state strength and regime form. For now we have established for certain that
important characteristics of states are associated with their position in the
larger world-system. Let us now turn to
the analysis of a larger structure, the interstate system which is composed of
these contending and unequally powerful states.
Chapter 7: Geopolitics and Capitalism: One Logic or Two?
As we have seen in previous chapters, the focus on the
world-system has raised anew the issue of the relationship between economic and
political processes in the capitalist mode of production. This has coincided with a new emphasis on the
autonomy of political processes by neo-Marxists seeking to correct the
overemphasis on economic determinism in earlier Marxist analyses.1 While focusing most directly on the
capitalist state and class relations within the core of the world-system, Nicos
Poulantzas (1973) and Perry Anderson (1974) have stressed the autonomy of
political processes and the "relative autonomy" of state managers
from determination by capitalist class interests.2 This emphasis on the autonomy of politics,
long central among political scientists, has been extended to a critique of the
alleged "economism" of the world-system perspective. At the international level this critique
argues that geopolitics is an autonomous game in its own right which can be
understood separately from an analysis of world economic structures. Theda Skocpol (1977, 1979) is the neo-Weberian
sociologist who has most explicitly made this argument, but it has also been
made by several political scientists who share a state-centric approach to
social science, e.g. George Modelski (1978), Aristide Zolberg (1981), and
Kenneth Waltz (1979).
All these authors
claim that Immanuel Wallerstein has reduced the operation of the
"international" system to a consequence of the process of capitalist
accumulation. Indeed some have contended
that geopolitics and
state-building are themselves the main motors of the modern
world-system (e.g. Winckler, 1979; Gilpin, 1981). Here I will argue that the capitalist mode of
production exhibits a single logic in which both political-military power and
the appropriation of surplus value through production of commodities for sale
on the world market play an integrated role.
This chapter discusses a metatheoretical issue, and presents an argument
about the interdependence of the interstate system and the capital accumulation
process.
First I will present a
case for a change in terminology. The
world-system scholars of the
A Metatheoretical Issue
In this chapter, rather than arguing at a metatheoretical level
about economics, politics, and political economy in general, I shall ground the
discussion in the particular processes which have been operating
in the capitalist world-economy since the sixteenth century. But before I advance arguments for my
contention that the interstate system and the capitalist accumulation process
are part of the same interactive socio-economic logic I would like to briefly
discuss a metatheoretical problem raised by this issue.
In order to know
whether it is most elegant to conceive of capitalism as a singular process
which incorporates both economic and political dynamics, or, on the other hand,
if it is more powerful to emphasize the autonomy of these processes, we should
be able to specify formally and compare a unified theory with a theory which
posits spearate economic and political subsystems.
Ideally these two
theories should have different implications for concrete social change, and for
our understanding of the dialectical transformation of capitalism into a
qualitatively different system.
Unfortunately my argument here does not proceed at this level of
theoretical clarity. Rather I only
adduce a case for the superiority of a unified theory. But it is important to cast this argument in
the context of the attempt to develop the world-system perspective into a
formalized theory of capitalist development.
Why have most of the
theorists who focus on politics tended to adopt a narrowly historicist approach
to capitalist development? Marx made a
broad distinction between the growth of the forces of production (technology)
which occurs in the capital accumulation process, and the reorganization of
social relations of production (class relations, forms of property, and other
institutions which structure exploitation and the accumulation process). Samir
Amin (1980a) has applied this broad distinction to the world-system. The widening of the world market and the
deepening of commodity production to moreand more spheres of life has occurred
in conjunction with a series of 40 to 60 year business cycles, the Kwave. The Kwave is associated with
"noneconomic" political events such as wars, revolutions, etc. This has caused some economists (e.g.
Adelman, 1965) to argue that long waves are not really economic cycles
at all, but are set off by "exogenous" political events.
The causal links
between wars, revolutions and long business cycles are not precisely understood
despite a vast literature on Kwaves (see Barr, 1979), but Amin (1980a) and
Mandel (1980) have made the insightful argument that the accumulation process
expands within a certain political framework to the point where that framework
is no longer adequate to the scale of world commodity production and
distribution. Thus world wars and the
rise and fall of hegemonic core powers can be understood as the violent
reorganization of production relations on a world scale which allows the
accumulation process to adjust to its own contradictions and to begin again on
a reorganized political foundation.
Political relations among core powers and the colonial empires which are
the formal political structure of core/periphery relations are reorganized in a
way which allows the increasing internationalization of capitalist production
and the spatial shifts which accompany uneven development. The observation that capitalism has always
been "international" (and transnational) does not contradict the
existence of a longrun increase in the proportion of all production decisions
and commodity chains which cross state boundaries the upward secular trend of the
transnationalization of capital.
The above discussion
does not establish causal priority between accumulation and political
reorganization. But it implies that
these are truly interdependent processes.
The tendency to a narrowly historicist approach on the part of those who
focus on political events may be due to the low predictability of politics and
the apparently more direct involvement of human collective rationality in
political action. On the other hand, the
over emphasis on determinism and mechanical models on the part of those who
focus exclusively on economic processes may be due to the greater regularity of
these phenomena and their law-like aggregation of many individual wills
seemingly independent of collective intentions.
These perceptions are
correct to a considerable extent precisely because capitalism as a system
mystifies the social nature of investment decisions by separating the
calculation of profit to the enterprise from the calculation of more general
social needs. Anti-capitalist movements
have tried to reintegrate economics and politics in practice, but up to now,
the expanding scale of the commodity economy has evaded them. The interaction of the world economy and the
interstate system is fundamental to an understanding of capitalist development
and also to its potential transformation into a more collectively rational
system. Neither mechanical determinism
nor narrow historicism is useful in this project.
States as Production Relations
The critiques of Wallerstein's work mentioned above contain
implicit assumptions about the nature of capitalism which tend to conceptualize
it as an exclusively "economic" process. Skocpol (1979:22) formulates the issue by
arguing that Wallerstein "assumes that individual
nation-states are instruments used by economically dominant groups to pursue
world-market oriented development at home and international economic advantages
abroad." She continues, explaining
her own position:
but a different
perspective is adopted here, one which holds that nation-states are, more
fundamentally, organizations geared to maintain control of home territories and
populations and to undertake actual or potential militarycompetition with other
states in the international system. The
international states system as a transnational structure of military
competition was not originally created by capitalism. Throughout modern world history, it
represents an analytically autonomous level of transnational reality interdependent in its structure and
dynamics with world capitalism, but not reducible to it, (Emphasis in the
original).
Modelski (1978) and
Zolberg (1981) argue even more strongly for the autonomy of the interstate
system in opposition to what they see as Wallerstein's economic
reductionism. These authors raise the
important question about the extent to which it is theoretically valuable to
conceptualize economic and political processes as independent subsystems, but
in so doing they over-simplify Wallerstein's perspective.
Wallerstein's work
suggests a reconceptualization of the capitalist mode of production itself such
that references to capitalism do not point simply to market-oriented strategies
for accumulating surplus value.
According to Wallerstein the capitalist mode of production is a system
in which groups pursue both political-military goals and profit-making
strategies, and the winners are those who effectively combine the two. Thus the interstate system, state-building,
and geopolitics are the political side of the capitalist mode of production.
As discussed in
chapter 1, Wallerstein argues that a mode of production is a feature of a whole
world-system, not of parts or subunits.
His distinction between world-economies and world-empires as different
types of world-systems emphasizes important structural differences in formal
political organization across economic networks. In Wallerstein's view it is very important
that modern capitalism became dominant in the context of an interstate system
of competing states. This view is shared
by many other analysts of the rise ofthe West, who focus on the decentralized
features of European feudalism which were conducive to the emergence of a
strong commodity-producing economy. In
the more centralized world-empires the logic of the tributary mode of
production was able to fend off the emergence of capitalism.
Max Weber was most
explicit about the connection betwen capitalism and the competitive interstate
system. Inspired by Leopold von Ranke's
study of early European states4 (von Ranke, 1887; see Weber, 1978:354) Weber
added the interstate system to his list of necessary structural conditions for
the emergence and reproduction of modern capitalism (see Collins, 1986:chapter
2). After mentioning in his General
Economic History (Weber, 1981:337) that the European states were
"competing national states in a condition of perpetual struggle for power
in peace or war," Weber continues:
This competitive
struggle created the largest opportunities for modern western capitalism. The separate states had to compete for mobile
capital, which dictated to them the conditions under which it would assist them
to power. Out of this alliance of the
state with capital, dictated by necessity, arose the national citizen class,
the bourgeoisie in the modern sense of the word. Hence it is the closed national state which
afforded to capitalism its chance for development and as long as the national state does not
give place to a world empire capitalism also will endure.
In Economy and Society Weber elaborates:
"Finally, at
the beginning of modern history, the various countries engaged in the struggle
for power needed ever more capital for political reasons and because of the
expanding money economy. This resulted
in that memorable alliance between the rising states and the sought-after and
privileged capitalist powers that was a major factor in creating modern
capitalism and fully justifies the designation "mercantilist" for the
policies of that epoch. ....At any rate,
from that time dates that European competitive struggle between large,
approximately equal and purely political structures which has had such a global
impact. It is well known that this
political competition has remained one of the most important motives of the
capitalist protectionism that emerged then and todaycontinues in different
forms. Neither the trade nor the
monetary policies of the modern states
those policies most closely linked to the central interests of the
present economic system can be
understood without this peculiar political competition and "equilibrium"
among the European states during the last five hundred years a phenomenon which Ranke recognized in his
first work as the world-
historical
distinctiveness of this era. (1978:3534)
To this I can only add that the neoWeberians ought to pay more attention
to Weber.
Some Marxists, such as
Colin Barker (1978) also recognize that the political basis of capitalism is
not the state but the interstate system.
Particular states vary in their emphasis on politicomilitary
aggrandizement or free market accumulating depending, in part, on their
position in the larger system. And the
system as a whole alternates between periods in which there is greater emphasis
on competition based on state power versus periods in which a relatively freer
world market of price competition comes to the fore (see chapter 13).
Core states with a
clear competitive advantage in production are usually the most enthusiastic
advocates of free trade. And, similarly,
peripheral states under the control of peripheral capitalist producers of low
wage goods
for export to the core usually support the "open
economy" of free international exchange.
As Stephen Krasner (1976) points out, smaller core states heavily
dependent on international trade also tend to support a liberal economic order. Semiperipheral states and larger second tier
core states contending for hegemony utilize tariff protectionism and
mercantilist monopoly to protect and expand their access to world surplus
value. Periods of rapid world-wide
economic growth are generally characterized by a relatively unobstructed world
market of commodity exchange as the interests of consumers in low prices come
to outweigh the interests of producers in protection (ChaseDunn, 1980). Inperiods of stagnation protectionism is more
frequently utilized to protect shares of the diminishing pie.
According to the model
proposed in chapter 1, the capitalist mode of production includes both
commodity producers employing wage labor in the core areas and coerced labor in
the peripheral areas. Peripheral areas
are not seen as "precapitalist" but rather as integrated, exploited
and essential parts of the larger system.
Capitalist production relations, in this view, are not limited to wage
labor (which is nevertheless understood to be very important to the expanded
reproduction of the core areas) but rather production relations are composed of
the articulation of wage labor with coerced labor in the periphery. This articulation is accomplished not only by
the world market exchange of commodities, but also by the forms of political
coercion which the core powers often exercise over peripheral areas. The direct and indirect use of
political-military power by core states is emphasized by James Petras (1981) as
the most central way in which imperialism operates to constrain political
action in peripheral areas. Petras's
research clearly reveals the operation of this kind of coercive power, and its
importance is without doubt. Albert
Bergesen (1983) in combating charges that the world-system perspective is
"circulationist" (i.e. a theory based on relations of exchange rather
than class relations of production) has emphasized the importance of
colonialism, core ownership, and other direct forms of control.
The states, and the
system of competing states, which compose the
world polity, constitute the basic structural support for
capitalist production relations. Marx
saw that the state stood behind the opaque exploitation of wage labor by
capital in nineteenthcentury
forces the commodified capital/wage labor relationship in the
core, the coerced labor extraction in the periphery, and the extra-economic
forms of exploitation between the core and the periphery. This constitutes the basis of production
relations for the capitalist system.
States are the
organizations which are often utilized by the classes that control them to help
appropriate shares of the world surplus value.
Market forces are either reinforced or regulated depending on the world
market position of the classes controlling a particular state. When I say "classes that control the
state" I am including state managers.
I am not a vulgar instrumentalist arguing that the state is simply the
executive committee of the bourgeoisie.
The extent to which business interests directly control a state
apparatus versus a situation in which state managers
successfully achieve a certain autonomy by balancing off different economic
interests is an important variable characteristic of states.
Richard Rubinson
(1978) has made the important point that state managers are most capable of
effectively pursing a policy of national development and upward mobility in the
world-system when there is a considerable convergence of political interests
within the dominant class in a nation.
This clarifies an issue which is posed by the "relative autonomy"
theorists, who ask whether or not the state represents the "general
interests" of capital. As Barker
(1978) reminds us, the world capitalist class exhibits a high degree of
interclass competition and conflict.
There is no single world capitalist state torepresent the interests of
capital as a whole, so the various national states represent the interests of
subgroups of capital. The extent to
which they do this effectively depends on the degree to which the interests of
the subgroups within a state converge or diverge as a consequence of their market
position and options within the larger world economy. Fred Block (1978) reminds us that state
managers are often able to expand the capabilities of the state in response to
the demands of workers and peasants, and thus states not only come to institutionalize
the interests of capitalists, but also, especially in the core, they take on
redistributive functions which benefit workers.
Both political
organizations and economic producers are subjected to a longrun "competing
down" process in the capitalist world-economy, whereas in the ancient
empires the monopoly of violence held by a single center minimized both market
and political competition between different organizational forms. This accounts for the much more rapid
transformation of both production
technology and political organization by capitalism. State structures themselves are submitted to
a political version of the "competing down" process which subjects
firms to price competition in the realm of the market. Inefficient state structures, ones that tax
their citizens too heavily or do not spend their revenues in ways which
facilitate political-economic competition in the world-economy, lose the
struggle for domination. In Marxist
theoretical terms, the interstate system produces an equalization of surplus
profits, the profits which return due to the use of political power to enforce
local monopolies. There are no corewide
monopolies. Even the largest organizations
(both states and firms) are subjected to the pressures of political-economic
competition.
The Emergence of Capitalism and the Interstate System
It has been pointed out by Zolberg (1981) and many others
(Ekholm and Friedman, 1982) that not all precapitalist world-systems were
world-empires.
Wallerstein's discussion implies that earlier world-economies
were short-
lived, tending to either dissolve into economically delinked
local systems or to experience empire formation. But Ekholm and Friedman (1982) have
noted that many ancient world-systems had interstate systems
which were quite stable in the sense that a balance of power mechanism operated
to prevent empire formation for rather long periods. Their most important example is the Sumerian
world-economy of city-states, but others have described rather stable
interstate systems in ancient
The fact that there
have been long-lived interstate systems prior to the emergence of the European
world-economy raises the question of whether or not these were structurally or
behaviorally different. Clearly the
normative rules of diplomacy were different (see Modelski, 1964), but it is
unclear if these are important determinants of the dynamics of an interstate
system. A comparative study of
interstate systems which employs a world-system perspective could perhaps
answer this question, but such a study has not yet been done (see Chase-Dunn,
1986). My guess is that the most
important difference between ancient and modern interstate systems is the
nature of the competition among states, and therefore the substantive content
of state policies. The modern interstate
system is composed mostly of states which are significantly controlled by
capitalists, which means that the goals of market protection and expansion
constitute a larger proportion of state action than in precapitalist interstate
systems. This characteristic probably
also leads to other differences. It is likely that threatened hegemons in
ancient interstate systems engaged in a policy of empire-formation, while in
the capitalist world-economy this does not happen.
Feudalism is another
type of precapitalist system which is not a world-empire. Zolberg (1981) is correct to point out that
classical European feudalism (i.e. around the ninth century) was not a
world-empire, but additionally it was a very strange kind of world-system. As a devolved residue of the Roman
world-empire classical European feudalism was characterized by a regional
political and cultural matrix organized across an economy which was almost
completely delinked into self-subsistent manors. The medieval states were so weak that in most
places most of the time the lord of each manor constituted a mini-state. Anderson (1974a) and many others have pointed
out that it was the "parcellization of sovereignty" within this very
decentralized system which allowed the capitalist mode of production to expand
in institutional interstices, and to begin to dominate exchange, production,
and politics.
The growth of
commodity production for both local, urban/rural, and long distance exchange
was stimulated by the limitations of the manorial economy and the opportunities
for profit-making presented by a system which had little regional political
ability to regulate production and exchange.
The constitution of cities as relatively autonomous elements within the
segmented matrix of manors enabled merchants and artisans to obtain "state
power" within a jurisdiction (the medieval city) which could then be used
to legitimate and militarily back capitalist exchange and production. The fact that successful cities soon tried to
protect their market advantages with politically-guaranteed monopolies simply
drove the market economy to expand elsewhere and to increase its spatial
dimensions. This process of capitalist
urban growthalso spurred the strengthening of the nation-state, as kings were
able to gain resources from capitalists to use against recalcitrant local
lords. Thus the nation-states and the
European interstate system came into existence.
It was the dynamic of mercantile and commodity production competition
between both state and private enterprises in the long sixteenth century,
together with the emergence of a core/periphery hierarchy, which led
Wallerstein to argue that the capitalist world-system was then born.
militarily threatening international state system emanating from
In the competitive
interstate system it has been impossible for any single state to monopolize the
entire world market, and to maintain hegemony indefinitely. Hegemonic core powers, such as
able commodity production; and the ability to produce
efficiently for the competitive world economy.
This is not the statecentric system which some analysts describe,
because states cannot escape, for long, the competitive forces of the world
economy. States that attempt to cut
themselves off or who overtax their domestic producers condemn themselves to
marginality. On the other hand, the
system is not simply a free world market of competing producers. The successful combination of political power
and competitive advantage in production is a delicate balance.
There have been
important differences among European states in terms of the strategies of
development that they have followed.
Some have relied more on continental military advantage and centralized
fiscal structures while others, the more successful ones, have employed a low
overhead policy of strategic protection of the vital business interests of
their national capitalists. Again, I
don't claim that all states equally employ a policy of support for their capitalists.
The most
successful core states have achieved their hegemony by having strong and
convergent business class interests which unified state policy behind a
sustained drive for successful commodity production and trade in the world
economy. Second-runners have often
achieved some centrality in the world economy by relying on a more directly
state-organized effort to catch up with the hegemonic state.
It could be argued
that the existence of states which successfully follow a more
political-military development path is evidence in favor of the thesis that
geopolitical and economic processes operate independently. The existence of such a development path is
unquestionable (e.g.
The Reproduction of Capitalist Accumulation
There are several ways in which the competitive interstate
system allows the capitalist accumulation process to temporarily overcome the
contradictions it creates, and to expand.
The balance of power in the interstate system prevents any single state
from controlling the world-economy, and from imposing a political monopoly over
accumulation. This means that
"factors of production" cannot be politically controlled to the
degree that they could be if there werean overarching world state. Capital is subjected to some controls by
states, but it can still flow from areas where profits are low to areas where
profits are higher. This allows capital
to escape most of the political claims which exploited classes attempt to
impose on it. If workers are successful
in creating unions which enable them to demand higher wages, or if communities
demand that corporations spend more money on pollution controls, capital can
usually escape these demands by moving to areas where opposition is weaker. This process of "capital flight"
can also be seen to operate inside of countries with federal states.
Class struggles are
most often oriented toward and constrained within particular territorial state
structures. Thus the interstate system
provides the political underpinning of the mobility of capital, and also the
institutional basis for the continuing expansion of capitalist
development. States which successfully
prevent domestic capital from emigrating do not necessarily solve this problem,
because foreign competitors are likely to take advantage of the less costly
production opportunities outside the national boundaries, and thus push the
domestic products out of the international market.5
The implication of the
above is that capitalism is not possible in the context of a single world
state, as Weber claimed. The
transformation of the interstate system into a world state would eventually
develop the political regulation of resource allocation. If this world state were socialist it would
more regularly and fully include social desiderata in the calculation of
investment decisions. The dynamic of the
present system, in which profit criteria and national power are the main
controllers of the use of resources, would eventually be transformed into a
system in which development combinesefficiency with a calculation of the
individual and collective use values of human society. Such a collectively rational system would not
constitute a utopia in which the problems of production and distribution would
be completely solved, but the political struggles for resources which would be
oriented toward a single overarching world government would exhibit a very
different longrun dynamic of political change and economic development than
that which has characterized the capitalist world-economy.
Of course this is an
optimistic assessment. It is also
possible that world state formation would bring about a transformation to a new
version of the tributary mode of production.
Both socialism and the tributary modes utilize primarily political means
of accumulation, but the tributary modes employ large amounts of coercion,
while socialism produces, distributes, and invests democratically. Either way though, capitalism would no longer
be the dominant mode of production.
Capitalist Reproduction of the Interstate System
Thus the interstate system is important for the continued
viability of the capitalist accumulation process. But is the accumulation process equally as
important for the generation and reproduction of the interstate system? First, what do I mean by reproduction of the
interstate system? I am not making fine
distinctions between types of interstate systems such as those introduced by
Partha Chatterjee (1975). By an
interstate system I mean a system of unequally powerful and competing states in
which no single state is capable of imposing control on all others. These states are in interaction with one
another through a set of shifting alliances and wars. Changes in the relative power of states
upsets any temporary set of alliances leading to a restructuring of the balance
of power. When is such a system not
reproduced? If an interstate system either:
1 disintegrates due to
the dissolution of the individual states;
2 dramatically reduces
to nearly zero the amount of material exchange and political-military interaction among
the states; or
3 becomes dominated by
a single overarching state,
the system can be said to have fundamentally changed (i.e. it
has been transformed, not reproduced).
In this definition the stages of classical, imperial, bipolar, and
"contemporary" interstate systems identified by Chatterjee are
subsumed into a single broad type which is nevertheless quite different from
the precapitalist agrarian empires or the economically self-
subsistent and "stateless" system which existed in
feudal
Which Came First?
Skocpol (1979) contends that the European interstate system
predates the emergence of capitalism6 and she implies that this is evidence of
its relative autonomy.7 No one denies that
states predate capitalism. At issue is
the genesis of a dynamic interstate system which is self-reproducing rather
than a transitional stage on the way toward empire-formation. It is clearly the case that multistate
systems exhibiting some of the characteristics of the European interstate
system existed prior to the emergence of the dominant capitalist mode of
production. The multicentric
"international system" which developed among the Italian city-states
and their trade partners in the East and West invented many of the institutions
of diplomacy and shifting alliance which were later adopted by the European
states. As Lane says of the sixteenth
century, "The Italian state system was being expanded into a European
state system" (1973:241). While this
constitutes prior development, it may not be evidence in favor of the autonomy
of the interstate system, as we shall see. Many
of the capitalistic financial and legal institutions later elaborated in the
European capitalist world-economy were invented in the Italian city-
states. The Christian
Mediterranean was part of an interstitial proto-capitalist regional
economy. Analogous to Marx's analysis of
merchant capitalism, the Mediterranean regional economy, though developing the
seeds of capitalist production with labor as a commodity, was primarily based
on the exchange of "unequals" between social systems which were not
integrated into a single commodity economy.8
Nevertheless this proto-capitalist regional economy succeeded in
developing several institutional features which were only later fully
elaborated in the capitalist world-economy which emerged in Europe and Latin
America in the long sixteenth century.
One of these was the interstate system, which as Zolberg (1981) agrees,
only became stably formed after its emergence in
But doesn't the
continuity of the Italian interstate system, and its failure to develop into a
world-empire, constitute a case for the independence of the interstate
system? Two factors militate against
this conclusion. The states of the
Italian system were already rather capitalistic, thus explaining the weakness
of attempts at empire-formation, and the Italian system became incorporated
into the larger European world-economy, which was already becoming dominated by
production capitalism in the sixteenth century.
I am not arguing that
capitalist institutions are the only factors which enable an interstate system
to resist empire formation. It is likely
that a common cultural matrix also works against empire-formation by
facilitating the diffusion of military and other technologies, and thus
maintaining a relatively equal distribution of power among contending core
states. The European system shared this
feature (an interstate common cultural matrix) with earlierlong-lived
interstate systems such as those in ancient
institutions such as international markets, money, banking, and
opportunities for investment further stabilize an interstate system by
inhibiting efforts at empire formation.
Skocpol's contention
about the prior emergence of the interstate system also receives support from
the "fully formed capitalist mode of production" as
becoming dominant only in the eighteenth century. Wallerstein's interpretation contends that
"agrarian" capitalism became dominant in the long sixteenth century.
Wallerstein's
interpretation implies that the capitalist mode of production became the most
important stimulus for change well before the "bourgeois revolutions"
in which explicitly capitalist interests came to power in nation-states.
An Outside
One clue to the dependence or independence of the interstate
system is its ability to reproduce itself, or to weather crises without
becoming transformed into either a world-empire or experiencing disintegration
of its network of international economic exchange. Wallerstein's analysis of the effort by the
Habsburgs to transform the still shaky sixteenthcentury capitalist
world-economy into a world-empire (1974:164-221) demonstrates the importance of
capitalism in reproducing the interstate system. I will discuss the later points at which
similar challenges to the interstate system were mounted (Louis XIV's, the
Napoleonic wars, and the twentiethcentury world wars) and the causes of
continuity of the interstate system, but first I want to consider another point
made by Zolberg (1981).
Zolberg argues that
the European interstate system occasionally incorporated powers that were
outside the capitalist world-economy into alliances which affected the outcome
of politico-military struggle. His main
example is the alliance between
Once again this shows
that the interstate system was important for the survival and growth of
international capitalism. On this there
is little disagreement. But what would
have happened to the European interstate system if international capitalism had
been encompassed by the Habsburg empire?
Obviously both international capitalism and the interstate system would
have been transformed into a world-empire, and probably one in which capitalism
as a mode of production was subordinated to the logic of imperial tribute and
taxation. Though I agree that capitalism
had become dominant over the logic of the tributary mode of production in the
long sixteenth century it is obvious that its domination in that first epoch
was somewhat shaky. The attempt to
convert the nascent capitalist world-system into a tributary world-empire was
stemmed, not by the institutional strength of capitalism alone, but in
conjunction with the somewhat fortuitous alliance between the French and an
"outside" power, the Ottoman Turks.
As we shall see below,
later challenges to the interstate system were undercut by the logic of
international capitalism alone. Zolberg
is right in pointing to the FrenchOttoman alliance as evidence of the
importance of the interstate system, but in later challenges it was the
strengthened
institutions of international capitalism by themselves that
prevented the interstate system from becoming a world-empire.
Another reason why
Zolberg argues for the existence of an autonomous logic of the interstate
system is his confusion over the difference between
colonial empires and world-empires. It is perfectly correct that core states
engage in imperialism in the sense of using military power to dominate parts of
the periphery. These colonial empires
expand cyclically with the growth of the modern world-system (Bergesen and
Schoenberg, 1980). But this phenomenon
is very different from the imposition of a single state over the whole system,
including other core states.
More Recent Challenges
The European world-system became a global world-system in a
series of waves of expansion which eventually incorporated all the territories
and peoples of the earth. Although
political-military alliances with states external to the system occurred after
the sixteenth century, they were never again so crucial to the survival and
development of capitalism as was the French/Ottoman alliance. But the capitalist world-economy continued to
face challenges of survival based on its own internal contradictions. Uneven economic development and the vast
expansion of productive forces outstripped the structure of political power,
causing violent reorganizations of the interstate system (world wars) to
accommodate new levels of economic development.
This process can be seen in the sequence of core competition, the rise
and fall of hegemonic core states, which has accompanied the expansion and
deepening of the
capitalist mode of production (see chapter 9).
After the failure of the
Habsburgs there have been three other efforts to impose a world-empire on the
capitalist world-economy: those of
Why Hegemons Don't Try Imperium
It may be argued that one or another of these did not really
constitute a serious effort at imperium.
There has been much dispute about German intentions in World War I (see
Fischer, 1967, and his critics) but the real issue is not intentions, but the
structural consequences which a German victory would have had for the
interstate system. If the balance of
power system, and thus the multicentric nature of the core, could have survived
such a victory, then these events did not represent real threats to the
interstate system as such, but merely a challenge to the extant balance of
power. If none of these efforts
presented a real possibility of world imperium (i.e. the formation of a core
state large enough to end the operation of the balance of power system) we must
ask why there have been no strong challenges to the interstate system since the
Habsburgs.
Some authors imply
that the size of the European states has been limited by the range of effective
territorial control, but this cannot explain the absence of empire formation in
It is the dynamic
uneven development of capitalism which systematically undercuts the
possibilities for empire formation, thus reproducing the interstate
system. One of the striking things about
these ineffectivechallenges to the interstate system is that they were not
perpetrated by the hegemonic core powers themselves, but rather by emerging
second runners among the competing core states.
This raises the question of why hegemonic core powers do not try to
impose imperium when it becomes obvious that their competitive advantage in
commodity production is waning.
Similarly we may ask, as Zolberg did of the sixteenth century, why
opposing forces were able to prevent the conversion of the system into a single
empire. To both of these questions I
would answer that it is the transnational structures associated with the
capitalist commodity economy which operated to tip the balance in favor of
preserving the interstate system.
Hegemonic core states
often use state power to enforce the interests of their "own"
producers, although typically they do not rely on it as heavily as other
competing core states. But, when a
hegemonic core power begins to lose its competitive edge in production because
of the spread of production techniques and differential labor costs, capital is
exported from the declining hegemonic core state to areas where profit rates
are higher. This reduces the level at
which the capitalists within the hegemonic core state will support the
"economic nationalism" of their home state. Their interests come to be spread across the
core. Another way to say this is that
hegemonic core states develop subgroups of their capitalist classes having
divergent interests; there comes to be a group of "international
capitalists" who support free trade, and a group of "national
capitalists" who seek tariff protection.
This explains the ambivalent, contradictory, and zig-zagging policies of
hegemonic core powers during the periods of their decline (Goldfrank, 1977; see
also chapter 9 below).
Schumpeter (1955)
pointed to the lack of patriotism which many capitalists evince9 as proof that
capitalism itself is a peace-loving system.
He claimed that modern warfare is caused by atavistic survivals of the
pre-capitalist era which periodically grip the world and lead to violent
destruction on a massive scale. It is
important to distinguish between capitalism as a system and the sentiments of
those who make investment decisions.
While some capitalists may be peace-loving, it is the export of
investment capital to other core states during hegemonic decline which is the
major factor which explains why hegemonic core states do not try to impose
imperium. And it is the reproduction of
the interstate system, which presumes the legitimacy of warfare, that
guarantees recurrent bouts of violent destruction.
Why Do Challenges Fail?
Why have the second-running core powers who have sought to
impose imperium on the world-economy failed?
Most theorists of the interstate system have not addressed this question
as such. The balance of power idea
explains why, in a multicentric system, alliances between the most powerful
actors weaken. Coalitions in a triad,
for example, balance the power by allying the two weakest actors against the
strongest. But this alliance falls apart
when the stronger of the partners gains enough to become the strongest single
actor (hegemon) because the weaker power can gain more by allying with the
declining former hegemon than by sticking to the original alliance. This simple game
theory is extended to the interstate system by the theorists of
power equi
librium, but it does not answer our question substantively. Again, in the modern world-system it is not
the most powerful actor that tries to impose imperium, but rather upwardly
mobile second runners with less than their "fair" share of political
influence over weaker areas of the globe.
Organski's (1968) theory explains why these second runners try, but not
why they fail.10
Of course one might
employ strictly historical explanations which make use of unique conjunctural
factors, a theoretical maneuver (or rather an atheoretical maneuver) which is
easy to accomplish when one is explaining only four "events." Here we seek an explanation of what seems to
be a regularity of the world-system based on our hypotheses about its deep
structural logic.
Morganthau (1952)
invokes the notion of a normatively organized liberal world culture which
successfully mobilizes counterforce against the threat to the balance of power
system. This conceptualization of a
normatively
integrated worldsystem has already been described and critiqued
in chapter 5. While I do not deny that
some normative patterns are generalized across the system, I emphasize the fact
that culture tends to follow state boundaries and that the larger system
remains significantly multicultural.
From this perspective it is far fetched to explain the failure of
empireformation in terms of commitment to internationally shared norms.
Craig Murphy (personal
communication) contends that another reason why hegemonic core powers do not
try to impose imperium on the whole system is the enlightened view of certain
core statesmen that the multistate system is necessary for the survival of
capitalism. Disraeli is suggested as an
example. This type of consciousness can
be understood as a response to the dispersion of investment capital and
consumer interests which accompanies the declining hegemony of the leading core
state. I doubt that liberal
internationalist ideology plays much of an independent role in the reproduction
of the interstate system.
On the other
hand, I have suggested above that interstate systems in which the states share
a consensual regional culture are more likely to resist empire formation
because new organizational and military technologies will rapidly diffuse and
maintain rough power equality among the contending core states. This explanation does not invoke normative
integration (the regulation of behavior by consensual belief in rules), nor is
it dependent on the specific content of cultural forms. It simply argues that information is more
likely to flow across state boundaries when the states have somewhat similar
ideological and cultural systems. Such a
condition existed among the core states of the European world-system and this
could have partly explained the failure of empire formation in
It is my argument,
however, that both the attempts and the failures of world imperium can be
primarily explained as reactive responses to the pressures of uneven
development in the world-economy. We
have already noted that the attempts were fomented, not by the most powerful
states in the system, but rather by emerging second-tier core powers contending
for hegemony. One striking thing about
all four cases is that they appear, in retrospect, to have been wildly
irrational. The countries who adopted
the strategy of aggrandizement reached far beyond their own capacities, and
failed to generate sufficient support from allied countries.
I agree with Modelski
(1978) that the predominantly land-oriented continental expansionism of the
French monarchy was not a strategy which could lead to hegemony in the
capitalist world-economy. It is notable
that the overhead costs of purely geopolitical expansionism (Oliver Cox's
[1959] "Florentine model" of domination) could not successfully
compete with the low overhead strategy of allowing a more decentralized
political system to bearthe costs of administration while surplus value
appropriation is accomplished by trade.
It was this "Venetian model" (Cox again) which was followed by
the states which became hegemonic core powers (
Why didn't the French
or German attempts at imperium receive more support? Probably in part because potential allies
doubted the extent to which their interests would be protected under the new
imperium, and because the path of capitalist growth in the context of the
multicentric system appeared preferable to the emerging bourgeoisies of
potential allied states.
If I am correct, the
interstate system is dependent on the institutions and opportunities presented
by the world market for its survival.
There are two main characteristics of the interstate system which need
to be sustained: the division of
sovereignty in the core (interimperial rivalry) and the maintenance of a
network of exchange among the states.
The commodified nature of the capitalist world-economy assures that
states will continue to exchange due to natural and socially created
comparative advantages in production.
Withdrawal from the world market can be accomplished for short periods
of time but it is costly and unstable.
Even the "socialist" states which have tried to establish a
separate mode of production have eventually returned to production for and
exchange with the larger commodity market.
The maintenance of
interimperial rivalry is facilitated by a number of institutional
processes. At any point in time national
sentiments, language and cultural differences make supernational integration
difficult, as is well illustrated by the EEC.
These "historical" factors may be traced back to the longrun
processes of state formation and nation building, and these processeshave
themselves been conditioned by the emergence of commodity economy over the past
500 years.
But the main
institutional feature of the world-economy which maintains interimperial
rivalry is the uneven nature of capitalist economic development. As discussed above, hegemonic core powers
lose their competitive advantage in production to other areas and this causes
the export of capital, which restrains the hegemon from attempting to impose
political imperium. Second-running
challengers, who may try to impose imperium, cannot gain sufficient support
from other core allies to win, or at least they have not historically been able
to do so. This is in part because the
potential for further expansion and deepening of the commodity economy, and
development in the context of a decentralized interstate system, appears
greater to potential allies than the potential for political and economic power
within the proposed imperium. Success
stories in the development history of the interstate system are frequent enough
to undermine empire formation.
Now let us further
consider the ways in which the transnational institutions of capitalism
interact with geopolitics to reproduce the interstate system.
Chapter 8: Warfare and World-Systems*
Part of this chapter is a response to William R. Thompson's
(1983c) valuable criticism of an earlier version of chapter 7.* The issues raised by Thompson's article are
addressed and a problem on which his analysis is conspicuously silent is
considered. Thompson's discussion fails
to address the argument that the reproduction of the interstate system is due
to the operation of specific institutions characteristic of a capitalist mode
of production. His comparison of
generally "political," as opposed to "economic," variables
ignores the role of historically specific economic institutions such as
commodity production, wage labor, commodified wealth, and capital in the
dynamics of the modern interstate system.
I shall make further comparisons of the modern capitalist world-economy
to pre-capitalist world-empires and world-economies in order to demonstrate the
importance of capitalist institutions for the reproduction of the modern
interstate system. In addition I will
examine the arguments and research on the relationship between the long
economic wave (Kwave) and world wars.
Thompson and George
Modelski (1978; see also Modelski and Thompson, 1988) have contributed
theorization and important research to the study of the modern
world-system. While their
conceptualization of that system is somewhat different from mine, they
nonetheless recognize it as a hierarchical structure in which unequally
powerful nation-states contend with one another for position. In this they have moved well beyond the still
widelyheld view that nationstates can be understood as either
"advanced" or "developing" without regard to the larger
context in which they are interacting.
___________________________________
*An earlier version of this chapter was coauthored with Joan
Sokolovsky.
On the other
hand both Modelski and Thompson proceed without any discussion of
capitalism. They instead focus on the
issue of the primacy of either "economic" or "political"
variables (Modelski, 1982; Thompson, 1983c).
While this may be a convenient short-hand, an understanding of the
underlying dynamics of the modern world-system requires comparison of its specific
institutional structures with those of other historical, large-scale social
systems. I do not claim that
"economic" variables are more important to the dynamics of the modern
system, but rather that several specifically capitalist features of the
political economy act to reproduce the interstate system. The notion of "historical systems,"
which differ from one another in fundamental ways, is absent from Thompson's
analysis, but central to mine.
Commodities and Markets
Thompson reduces the discussion of capitalist institutions to
"economic growth" and "uneven development." This may be, in part, due to a lack of clarity
in the work he was criticizing (Chase-Dunn, 1981), which uses such concepts as commodity
production without benefit of explanation.
In Marxist theory commodity production refers to the production for sale
in a price-setting market of somewhat standardized commodities (including
services). Everything is not a
commodity. Some items are produced, not
for exchange in a market, but for direct consumption by the producer. Other items are produced for exchange in a
normative reciprocal system or in a politically administered redistributive
system, but not for sale in a market.
Other things are produced for sale, but are sufficiently unique that
their conditions of production are not regularly and systematically subjected
to price competition, e.g. art objects.
Other things are sold but are not produced for sale, such as
untransformed land or other resources appropriated directly from nature.
Commodification is a process by which social relations become mediated by
markets. A price-setting market is one
in which the competitive interaction of a large number of independent buyers
and sellers determines the ratios (prices) at which particular commodities
exchange.
Capitalist commodity
production exists when wealth, land, and labor have become largely (but not
completely) commodified. Capitalist
production implies a hierarchical division of labor between controllers of
commodified means of production (capital) and the workers whose labor produces
commodities for profitable sale.
Empirically there are
no perfect price-setting markets. Prices
are always influenced to some extent by normative and political factors. And there are no completely capitalist social
systems, even our own. Capitalist
commodity production and markets have been present to some extent in most of
the historical systems since at least the emergence of cities some 5000 years
ago in
We should note that
commodities and markets are, in the final analysis, human institutional
artifacts. As such, they are
historically variable. Most human
societies in the history of social development have been predominantly based on
either normative reciprocity or politically determined production and exchange
in which markets have played only a very limited part.
Capitalists in Power
Weber's analysis of capitalism (1978) stresses the importance of
political control by capitalist merchants and producers. The medieval European cities which, by
revolution or accretion, came under the autonomous control of merchants, were
important early examples of institutionalized capitalist control of legal
systems and coercive power. The
capitalist cities of northwestern Europe provided political and economic
resources which enabled kings to gain power over feudal lords and to construct
the absolutist monarchies which first formed the European interstate system (
Capitalist merchants
and producers have gained power within most of the nation-states of the modern
world-system, but capitalist control of states has never been complete. Some states have been directly dominated by
capitalists, while others are in the hands of state managers who aggregate
contradictory class interests, but all states include other classes in their
ruling coalitions to some extent. Both
the class content of the ruling coalition and the constraints of the
world-system are important determinants of state policy.
Precapitalist World-systems
When I say that commodity production dominates world-system
dynamics I do not mean that it is all-inclusive. We must look at the way in which commodity
production is systematically interrelated with normative and political
institutions. All social systems are
built on a specific articulation of normative, coercive, and economic
institutions. In the modern worldsystem
capitalist commodity production is integrated with a set of political
processes: state formation, nation building, and the geopolitical game which we
know as the interstate system. Like
markets and money, these political institutions are not natural or
ahistorical. They are products of human
invention, and their stability, change, and/or transformation must therefore be
understood historically.
We know that the
process of state formation has occurred in various forms since the emergence of
agrarian civilizations 5000 years ago, and that for most of the history of
civilization this has been the most important process by which societies have
increased in complexity, hierarchy, and size.
Kinship systems were elaborated vertically into class systems in which
various forms of taxes or tribute were appropriated to support a non-producing
ruling class.
precapitalist world-systems.
Most systems based on the tributary mode of production tended toward
empire formation in which a single state apparatus came to exercise domination
over the core area. This facilitated
politically organized extraction of surplus product. The wage labor that existed in some of the
tributary empires was limited to certain sectors and was a very small
proportion of the work force. Slave
labor, a partially commodified form of labor control which is directly
dependent on political coercion, was important not only in the peripheral
regions but also in the core of many tributary empires.
The relatively large
role played by private property, markets, wage labor, and commodity production
in the Graeco-Roman world led to the invention of legal institutions
particularly appropriate for a capitalist society. This law later became the basis for municipal
legal systems in many of the capitalist cities of the European
Renaissance. But in classical
The Modern World-system
In the capitalist world-economy system dynamics are produced by
a single logic in which capitalist commodity production interacts with the
processes of geopolitics, state formation, class formation, and nation
building. Relative to the precapitalist
world-systems, the weight of commodity production is much greater. The business cycles, rapid technological
change, and the relatively rapid changes of fortune of different regions are
due to the operation of a world market which is much less constrained by
political-military structures than were the trade networks of earlier systems.
This is not to say
that political structures are unimportant, but rather that a particular kind of
political structure is most suitable for the continued operation of this
competitive economic system. A
centralized system tends to articulate political and normative constraints on
resource allocation and to enhance the possibility for longrun monopolies of
advantagein which production cost considerations are not important determinants
of income. Decentralized and competitive
political systems allow market
efficiency to be a much greater determinant of prices and the
distribution of returns.2 The
multicentric structure of the interstate system allows competitive advantage in
production to be an important continuing determinant of success.
Some states succeed in
becoming more central in the capitalist world-economy through a mode of
operation which bears some resemblance to that of the
Capitalist Institutions Support the Interstate System
More problematic is my contention that capitalist institutions reproduce
the contemporary interstate system.
I need to define clearly what the
interstate system is so that we may tell when it has been reproduced or
fundamentally changed, and I need to specify clearly my claims about attempts
to transform
that system.
Thompson (1983c) correctly points out that my definition of the
interstate system overlaps somewhat with my definition of a world-economy.
A world-economy is any
territorial division of labor (network of exchange of fundamental goods) in
which there are multiple political entities and multiple cultural systems. Thus world-economies always have interstate
systems.
An interstate system,
by contrast, is defined as a system of three or more states in direct or
indirect competition with one another in which none of the states is powerful
enough to dominate the whole system. An
inter-
state system could exist in which political interaction occurred
without exchange of material goods. In a
sense the European system of classical feudalism in the eighth and ninth
centuries was such a system. The
"states" were the manors, economically self-subsistent
units interacting
with one another in a shifting set of military alliances. While such a system is not a world-economy,
it is an interstate system.
The modern interstate
system is composed of nation-states, culturally integrated countries which
contain more than one city, whereas world-
economies of the past most usually contained interstate systems
composed of city-states and empires.3
Precapitalist world-economies have had interstate systems in which the
political and military competition among several states determined the nature
and extent of most material exchanges among states. This is Polanyi's (1977)
"state-administered" trade.
Precapitalist world-economies composed of citystates were most often
based on class relations in which each urban ruling class utilized political
institutions to extract surplus product from its own rural hinterland. Empires were formed when one of these city-
states succeeded by warfare in conquering others and extracting
tribute, and onforming a core/periphery system based on military
domination. Today warfare is less
central (though still important) to the workings of the modern world-
system.4
Modelski's (1978) and
Thompson's (1983c) explanation of the rise and fall of hegemonic core powers
invokes a systemic need for order.
International order is conceptualized as a "public good" which
is most effectively provided by a single hegemonic core state. This functionalist explanation has
difficulty accounting for the decline of core states or
decreases in the amount of systemic political order. In my model relative peace among states is
conditioned by a long period of economic growth in which a single hegemonic
core state supports both economic and geopolitical order. This stability has costs, however, and
periods of increased conflict emerge with the spread to other states of those
economic and technological advantages which enabled the 'great power' to become
hegemonic. World wars represent a
reversion to
military competition and a restructuring of the world political
order which, up to now, has subsequently allowed for the further expansion of
capitalist production.
My contention that
capitalist institutions acted to prevent the interstate system from evolving
into a world-empire led in the previous chapter to three questions:
1 Why have attempts to
create a world-empire been so few, and seemed, in Thompson's and Modelski's eyes (and the eyes
of the most central actors), so
unfeasible?
2 Why have hegemonic core
states never attempted imperium?
3 Why have
second-running challengers to hegemonic states been unwilling or incapable of creating a world-empire?Here
I will elaborate some further considerations which these questions raise.
An interstate system is
basically a game in which multiple players engage in shifting alliances in
order to gain advantages. In a simple
threestate version of such a game the two weakest states ally against the
stronger,but if their combined resources are fewer than those of the strongest
state it (the strongest) will eventually take over. Thus an interstate system will become a
world-empire any time a willing single state or coalition of states can
coordinate resources greater than the possible coalitions against it
(them). Why has this not happened during
the 500 year history of the capitalist world-economy?
Some would argue that
state formation on a sufficient scale is unfeasible due to technological or
cultural constraints. I have already
mentioned in the previous chapter that the
Of course, in the long
run they are crucial in all systems, as the demise of the
The more direct and
powerful constraints on state budgets in a capitalist world-economy are part of
the reason why there are so few attempts to take over the system and to create
a world state. The overhead costs of
such a state
would be greater than the costs of supporting competing
nation-states, especially to capitalist producers. And such a centralized world state, while it
holds the attractive possibility of world-wide monopoly, also holds the
potential for anti-capitalist social movements to coordinate their activities
and to concentrate political action toward a single center. Transnational corporations might enjoy some
advantages of centralized control, but they would have much less flexibility to
pit political organizations and states against one another.
Why don't declining
hegemonic core states change the rules of the game and attempt to retain their
hegemony by organizing a world imperium?
It is curious that no one, not even marginal political ideologues, ever
suggest such a project. The greatest
support for international organizations (of a limitedkind) comes from a hegemon
during its golden age of hegemony. Its
dominant ideology of free trade, firmly supported at first by its relative
advantage in commodity production for the world market, appears later to have a
life beyond its basis in competitive advantage.
There is an element of institutional and ideological momentum which
carries free trade ideology and liberal internationalism along as a guiding
policy in declining hegemons long after most other core states have shifted
back toward economic nationalism.
But the structural
basis behind this apparent ideological inertia is the power of
"international" capitalists within hegemonic core states. As a hegemon loses its comparative advantage
in production (because competitors abroad adopt and improve upon production
techniques, and because labor costs and taxes go up in a successful hegemon as
workers and other interest groups use political power to obtain a share of the
profits) a significant amount of capital from the hegemonic country gets
exported to where greater profits are
to be had. This gives the
"international capitalists," those who have invested abroad, a
continuing interest in free trade and a liberal international order. They oppose political thrusts toward
protectionism, and they also would be unsupportive of a policy of aggressive
political military expansion over other core powers if, indeed, anyone
suggested it. They do, however, support
maintenance of the interstate regime as currently structured against possible
challengers who want to create imperium or carve out a protected region. Some of the interests of workers, national
capitalists and international capitalists diverge during the period of
hegemonic decline, but because these groups are somewhat evenly balanced, the
result is a zig-zagging domestic and international economic policy in the
declining hegemonic core states. This
topic is explored further in chapter 9. But
why do the rising powers which militarily challenge the dominance of declining
hegemonic powers not go on to establish imperium? They may be ambivalent about the returns of
imperium.
periphery structures but they did not need to take on the
overhead costs of world imperium. As
mentioned in the previous chapter, if a challenger did propose the creation of
a world-empire it would be difficult to mobilize sufficient support because
most potential powerful allies would likely calculate their opportunities for
profitable commodity production in a world market to be greater than their
potential returns from participation in an imperium.
World Wars and the Restructuring of the World-economy
Jack Levy's (1985) discussion and critique of different
conceptualizations of world wars is a valuable review of recent scholarly
works, including the world-system perspective, the long cycle theory of
Modelski and Thompson, and Robert Gilpin's (1981) theory of hegemonic war. Levy compares different criteria for defining
world wars, and criticizes those approaches which define these conflicts in
terms of their consequences the
restructuring of the international system.
Levy claims that this builds a logical circularity into the definition
of world wars which makes it impossible to test important propositions about
the relationship between wars and system structure. His proposed criteria are
independent of outcomes and result in a list of ten "general" wars
since the sixteenth century. The
disputes over definitions and
lists of wars have generated a profuse literature. The following argument lays out a typology of
wars and their hypothesized roles in the world-system.
Without denying the
importance of the periodic emergence of hegemonic states in the development of
the modern worldsystem, I am not prepared to accept an explanation for the
dynamics of the system that is grounded solely in terms of the ascension to
military dominance of these leading states.
This necessitates a reinterpretation of the concept of world wars as
presented by Thompson and Modelski.
While I agree with Thompson that world wars represent attempts to
restructure political relations among states to correspond with changing
economic realities, I also think they may be seen as a way in which states try
to convert political-military strength into a greater share of world surplus
value. In part, this is a function of
the interdependence of political and economic factors in the capitalist mode of
production. In either case,
world wars, defined as those conflicts in which one state
attempts to take over and thus destroy the interstate system or struggles in
which leading power status is determined, are not the only important wars in
the struggle among states in the capitalist world-economy.
Extending our
consideration beyond the great powers to include the entire hierarchical
division of the world-economy since 1500, we can identify three structural
roles which wars have played. Most
fundamentally, they may represent struggles for control or dominance over the
entire interstate system. Although disagreements
clearly exist over the analysis of particular cases, the world wars identified
by Thompson all reflect struggles for preeminence to a greater or lesser
degree. The Napoleonic Wars and World
Wars I and II may be considered extreme examples of this type. There is no definitive answer to the
counter-factual question of what would have happened if the
unsuccessfulcoalition in one of these wars had managed to achieve its goals;
yet it is difficult to argue that the interstate system as we know it could
have continued to exist if either the Napoleonic or German attempts to conquer
most of the core zone had succeeded.
Secondly, wars may be
used to facilitate the upward or downward mobility of individual states and the
creation of a new structure of power that more accurately reflects the
strengths and weaknesses of key actors.
Examples of wars of this nature are legion.
During the eighteenth
century when economic and political power was quite evenly distributed
throughout the core of the system, wars of this second type were more
prevalent. The War of the Spanish
Succession resulted in the exhaustion of Dutch resources and the fall of the
status but did not lead to the immediate creation of a successor
state.
Finally, wars may be
used to restructure relations between core states and the periphery in keeping
with relative changes in power among actors.
The trade wars between
Conversely, when wars have resulted only in changes in relative
position within a more equal framework, victories have been accompanied by
changes in formal
control of particular peripheral areas to reflect the new
balance of power. Bergesen and
Schoenberg (1980) have documented the close relationship between periods of
multicentricity within the core and the development of formal colonial empires
as contrasted with periods of hegemony and the movement toward free trade and
decolonization.
Clearly, none of these
categories of wars are mutually exclusive, and historically world conflicts
have usually involved a combination of them.
Indeed the growth of exchange networks and interlocking alliances within
the world-system have insured the spread of conflicts to encompass states with
widely divergent motives. Thus the Seven
Years War linked the conflict between
In analyzing the
trajectory of military conflicts it is often nearly impossible to distinguish
between the real aims of participants, their statedaims, and the effect that
the victory of one or more actors is perceived as portending for the whole
system by other states. To make matters
more complicated, war aims regularly change with initial successes or
defeats. However it should be possible
to determine the impact of global conflict on the restructuring of the
interstate system.
When examined in these
terms wars have operated to reduce discrepancies between economic and military
power structures in the world-system.
Some states, like
Conversely, states
like
Overall, then, I share
basic agreement with Modelski and Thompson on the existence since the sixteenth
century of a multicentric world-system tied together by political relations
among states and economic exchange networks.
I concur as to the importance of the periodic emergence of a preponderant
economic and political power.
I differ from the
"power cycles" approach in my effort to go beyond a systemic need for
order to the construction of a causal model that will explain both the reasons
for the recurrent rise of these global powers and their failure to hold their
position within the structure of world power.
I believe that these causes lie in the institutions of capitalist
development. Upward mobility within the
system has been conceptualized as a two-sided process. While productive advantage has frequently
been converted to political strength by leading states in the system, other
states have been able to use the increased share of world surplus value gained
through military force to stimulate their own economic expansion.
The waning of dominant
status has also been linked to the operation of the capitalist
world-economy. Changes in competitive
advantage, costs of control and diverging interests among capitalists, workers,
and state bureau
crats within declining hegemons have been advanced as some of
the mechanisms through which this relationship is articulated.
The Kwave and Wars
One aspect of the link between the interstate system and the
world economy is revealed in the connection between the Kondratieff cycle
(Kwave) and warfare,first posited by Kondratieff himself. An impressive research literature has
blossomed in recent years investigating this linkage, and, although
considerable disagreement still exists about the causal connections involved,
quite a bit is now known.
An entire industrial
sector of international relations scholars have been working for scores of
years to code the timing, participants, territory, costs and destructiveness of
warfare in the interstate system. Levy's
(1983) book,
War in the Modern Great Power System, 14951975, presents a recent complete compilation of the data on
warfare. Although visual inspection of
the frequency of wars and other measures reveals an obvious sequence of periods
of more and less war, Levy reports that warfare does not exhibit any strictly
cyclical features. Joshua Goldstein
(1988:244) however, demonstrates that a statistical test (the AutoCorrelation
Function) applied to Levy's data on war severity (the number of battle deaths
per year) produces clear evidence of a 50 to 60 year cycle over the period from
1495 to 1975.
Thompson and Zuk
(1982) and Goldstein (1988) use time series analysis techniques to examine the
relationship between wars and the Kwave.
Goldstein uses several price and production series and the dates given
by four earlier scholars (Braudel, Frank, Kondratieff, and Mandel) to produce a
set of trough and peak dates for Kwaves between 1494 and 1975. He argues that strict periodicity is an
inappropriate standard for social cycles.
Thus he analyzes sequences of phases with unequal periods in his measure
of the Kwave. Goldstein's results reveal
a clear association between the Kwave and the severity cycle (battle deaths per
year) of war among core powers. He
argues that severe wars are more likely to occur during the upswing phase
of the Kwave and his empirical work finds support for this. This conclusion isdependent on the dates
Goldstein uses to distinguish between upswing and down
swing phases of the Kwave (see 1988: figure 11.3). Thompson (forthcoming) doubts the soundness
of Goldstein's periodization of the Kwave before 1790. Nevertheless Thompson finds support for the
relationship between war and Kwave price upswings in his own analysis of the
period between 1816 and 1914. What is
not in doubt is the finding that the Kwave and the war cycle are linked in some
systematic way.
The business cycle is
most often measured by price series.
Goldstein argues that there is a productionstagnation cycle which
precedes the price cycle by 10 to 15 years.
The war cycle peaks in between the peaks of the
production cycle and the price cycle in Goldstein's model. The data on long run series which indicate
real production and related indicators are rather scanty, so most empirical
work has focused on the relationship between price series and war. Some of that relationship is undoubtedly a
rather simple matter of the effects of warfare on inflation, a matter studied
by Thompson and Zuk (1982). They
conclude that most Kwave price downturns can be attributed to the ending of major
wars, but that Kwave price upswings regularly occur before the outbreak of
wars.
Goldstein's causal
model of the connection between war and the Kwave production cycle posits a
negative feedback loop. According to
Goldstein wars occur during Kwave production upswings because, though states
always desire to go to war, warfare is expensive and so states do it when
economic growth is providing them with more resources. Goldstein argues that warfare, on the other
hand, has a negative effect on economic growth through nonproductive
expenditure and destruction of people and property. Thus the two cycles spur one another on. Like Thompson, Goldstein does not consider
the capitalist nature of the institutions and processes which link warfare and
economic growth in the modern worldsystem.
For him states are simply war machines that go after one another when
they have the resources to do it, and since capitalist economic growth provides
great resources, warfare is endemic.
Goldstein's study
reveals empirical details of the relationship among business cycles, wars, and
the rise and fall of hegemons which must be
taken into account in any theory of the world-system. He shows that Kondratieff waves and peaks in
the severity of wars among core powers are closely
associated in time, with nine out of ten peaks in world war
since 1500 occurring near the end of an upward phase of the price
cycle. The somewhat surprising finding
here is that, according to Goldstein, world wars regularly occur during a
period of economic expansion. This is
surprising for two reasons. Most people
think first about World War II, which is the single exception among the ten
core war peaks since 1500. And many
theories of core
war are based on the idea that war is caused by increasing
competition among core states, [socalled "lateral pressure" (Choucri
and North, 1975)] which has been assumed to be most severe during periods of
economic stagnation.
Examining data series
on prices and several indicators of production, innovation and investment,
Goldstein concludes that the production cycle and the price cycle are somewhat
out of phase with one another, with the production cycle lagged about 10 to 15
years behind the price cycle. In
Goldstein's model (1988:259) the war cycle peaks just between the production
cycle and the price cycle. This model
is, of course, an idealized depiction of the exact features which are only
generally supported by Goldstein's analysis of actual data. Nevertheless, if Goldstein's representation
is correct we may be able toexplain why there has been considerable
disagreement about the timing of the relationship between warfare and the
Kwave. Goldstein follows Kondratieff and
many other theorists in arguing that warfare is most likely to be severe during
the upswing phase, but his own model implies that warfare actually peaks in
between the peak of the price cycle and the production cycle. If we believe that the psychology and logic
of production decisions as well as statecraft are involved in the relationship
between warfare and the Kwave, as the
Wallersteinian worldsystem perspective would emphasize, then it
is most interesting that the warfare peak is alleged to occur after the
peak of the investment cycle, in other words during the beginning of the investment
and production downswing. This is
simultaneously a period in which states have a lot of resources available for
war and capitalist investors have begun to slacken investments, presumably
because they perceive limitations on profit
taking. Increasing
competition for markets and investment opportunities is
due to overproduction by producers of core goods for the world
market relative to effective demand, and this kind of competition leads to
pressure for the use of extraeconomic power, that is state power, to protect
and/or expand market shares and investment opportunities.
Goldstein does not
distinguish the prices of different kinds of commodities, but other researchers
have shown that price cycles vary across different kinds of commodities. Michael Barrat-Brown's (1974) study of the
terms of trade between core and peripheral areas demonstrated that there are
differences in the degree of price changes such that terms of trade of
peripheral goods vis--vis core goods rise and fall over time, and this finding is confirmed by Paul Bairoch
(1986:205-8). This supports
Wallerstein's (1984b) argument that core commodities are overproduced in some
periodsrelative to effective demand, that is relative to the politically
structured distribution of resources in the world-system. This causes military conflict among core
states, and sometimes results in the restructuring of the inter
national order under a new hegemon.
Goldstein's
demonstration of the link between war cycles and the Kwave is extremely important
evidence in support of the contention that geopolitics and the world economy
are interdependent processes. The
connections between the war cycle, the Kwave, and the hegemonic cycle are
examined in the next chapter. Here I
would like to quote Goldstein's observations of four trends based on his
analysis of warfare over the last 500 years:
First, the
incidence of great power war is declining
more and more "peace" years separate the great power
wars. Second, and related, the great
power wars are becoming shorter. Third,
however, those wars are becoming more severe
annual fatalities during war increasing more than a hundredfold over the
five centuries. Fourth (and more
tentatively), the
war cycle may be gradually lenthening in each successive era, from about 40
years in the first era to about 60 years in the third. The presence of nuclear weapons has continued
these trends in great power war from the past five centuries any great power wars in this era will likely
be fewer, shorter and much more deadly.
(1985:432)
Now let us turn our
attention to the hegemonic sequence.
Chapter 9: The
Rise and Decline of Hegemonic Core Powers
Here it is argued
that three states have been hegemonic in the capitalist world economy since its
consolidation in the long sixteenth century ─ the United Provinces of the
Netherlands, the United Kingdom of Great Britain, and the United States of
America. The periods in between these hegemonies were characterized, I
contend, by a relatively equal (multicentric) distribution of military
power and economic competitive advantage
among core states, and by relatively higher levels of conflict and competition
within the core. I also think that these
periods were characterized by more
bilateral and politically controlled relationships between the core and the
periphery in which each core state attempted to
monopolize exchange with its "own" colonial empire. Stephen Krasner (1976) has contended that
periods in which a single great power has been hegemonic have been characterized by relatively more
free trade among different areas of the world economy. The interaction between the hegemonic cycle,
with its rise and fall of hegemonic core
powers, and changes in the structure of the core/periphery hierarchy is
analyzed in chapter 13. This chapter
focuses on the core zone itself, and
processes which cause the rise and fall of hegemons, and it considers the
current situation of the
Rise and Fall in
Different Systems
In order to
understand the dynamics of the current
One important
difference between the Roman world-empire and the
capitalist
world-economy is the organization of the state.
In
The main
consequence of this structural difference is its effect on the dynamics of
competition, reproduction, and growth in these two types of systems. In the Roman world-empire, competition was
primarily mediated through a single state apparatus and, although
monetarization of the economy was extensive (
centralized in a
single state apparatus. The main type of
growth was extensive, through the addition of control over lands and
slaves. As Keith Hopkins (1978b:62)
shows, the dynamic of the Roman economy was fueled by conquest. Roman military organization and transport
technology, by far the most advanced in antiquity, eventually reached their
cost-effective spatial limits, and territorial expansion halted.
The slave mode of
production required new inputs as slaves were worked to death on the latifundia
of
By contrast, the
capitalist world-economy, with its more decentralized polity, allows for much
greater economic and political competition. The existence
of the price-setting world market (which includes both national and
international markets), while it is not a "perfect" market, does
regularly encourage investors to increase the efficiency of production (to
produce at lower cost) in order to gain a greater share of income. Thus technical development of productivity is
facilitated by the fact that there is no central state that can impose monopoly
control on the whole arena of economic competition. Similarly, the multicentric interstate system
encourages the "export of capital" because political opposition to
profitable investment and labor exploitation is mediated by the individual
nation-states and often can be escaped by crossing state boundaries.
Political
competition in the capitalist world-economy is also much more dynamic than in
world-empires. The interstate system
allows different paths to "success" in the competition among
states. Some emphasize political-military
expansion, while others emphasize a strategy of competitive commodity production
for the world market. In such a
decentralized political system new forms of political organization can emerge
unconstrained by any central state and are thereby free to compete with one another for dominance. What makes the modern world economy a capitalist
world-economy, however, is its unique combination of an interstate system with
the institutions of commodity production for profit on a market, and a high
level of the commodification of labor power.
These institutions are
interwoven, along with the "private" nature of investment decisions,
into a competitive interstate system; and it is the institutional combination
of all these elements that creates the qualitative uniqueness which differentiates
the capitalist world-system from earlier world-systems.
The modern system
has shown its flexibility over the 500-year period of its expansion and deepening. Unlike the Roman world-empire, it has changed
its center without going into devolution as a system. Thus the present period is most probably not
a demise of "Western imperialism" (àla Galtung et al. 1980)
but rather the demise of the hegemony of the
The main
difference between the modern world-system and earlier world-systems is that
commodity production has become the dominant logic of competition at the
center of the system. In the Roman
world-empire and many other pre-capitalist world-systems there was much
commodity production, but it flourished
mainly in the interstices. It was the
business of clients, freedmen, or the specialty of semiperipheral trading
states, while the "perspective of
the world" was a game played exclusively by men more interested in
expanding state power as the primary means to wealth, power, and status. It was the emergence of a different kind of
state in the core region of the European world-economy, the hegemonic Dutch
state employing its military capabilities primarily to provide protection rents
to its capitalists, which signaled the consolidation of a world-system in which
capitalism had become the dominant mode of production.
Definitions of
Hegemony
Wallerstein
(1984a) defines hegemony in terms of economic comparative advantage ─ the concentration of a
certain type of commodity production within the borders of a single core
state. Remember that "core
production" is commodity production which utilizes relatively
capital-intensive technology and skilled, highly paid labor. Hegemony in this sense is comparative
advantage due to a combination of product development ─ producing the
most sophisticated and desirable products ─ and competitive prices
─ the ability to price the exported core products at levels which make it
hard for competing national economies to avoid purchasing them, and yet, at the
same time, to make a profit while selling at such prices. Wallerstein observes that economic
comparative advantage enables one core country to penetrate the home markets of
other core countries with capital-intensive commodities. The production of these types of commodities
has, of course, relatively denser forward and backward linkages, and spinoffs
which multiply the growth effects of investments within the national economy.
Very different
conceptualizations of global power are utilized by political scientists studying the
"international system." George
Modelski and William R. Thompson (1988) propose a theory of a long cycle of
political-military power in which great powers rise and fall. Their theory is similar in some respects to the Wallersteinian
perspective. They claim to be studying
the world system (without the hyphen ─ see Thompson, ed, 1983) and they
agree that periods of concentrated power
correspond with relatively lower levels of conflict among core powers, while
periods in which there is a more equal
distribution of power among core states tend to be more
conflictive. But they have a very
different notion of what constitutes concentrated power (they do not use the
term "hegemony"). For them the
central matter is naval power, which is an indicator of the ability of a great
power to exercise "global reach."
Predominant naval power enables a great power to maintain order at the
level of long-distance international interactions. Modelski and Thompson contend that land
armies and general military expenditures are not useful indicators of
"global reach" because these resources are usable primarily in regional or continental warfare,
not for domination of the global power system.
This argument illustrates Modelski and Thompson's "layered"
conceptualization of the world system, in which the global level of interaction
is analyzed as importantly distinct from local and regional interactions.
Modelski has
defined "world powers" as "those units monopolizing (that is
controlling more than one half of) the market for (or the supply of) orderkeeping
in the global layer of interdependence" (1978:216). The global system since the sixteenth century
is said to have experienced the rise and fall of four of these "world
powers:"
Other political
scientists speak of hegemony in terms of "power capability," which
is defined broadly as the ability of one state to control or influence the
behavior of other states using rewards and punishments.
Robert Gilpin
(1981) explicitly conceptualizes hegemony in terms of relative overall military
power, although he analyzes the concentration of economic innovations which
provide the economic wherewithal behind superior military advantage.
For Wallerstein
hegemony comes to include productive, commercial, and financial dominance
within the world economy. But productive
efficiency must be accompanied by state power.
The creation and maintenance of economic preeminence requires the
political and military capacity to preserve a domestic class structure
favorable to capitalist accumulation, innovative production, and the
prevention of external restrictions on flows of capital or goods. In this sense political-military power is a
necessary but not a sufficient basis for the attainment of hegemony in a
capitalist world-economy.
The free market,
which initially favors the competitive advantages of a leading power,
eventually results in the flow of capital and technological innovations to
competing states. This results in the
loss of productive advantage by the hegemonic power. Further, the costs of maintaining the global
order are borne disproportionately by the leader, resulting in rising
production costs and an excessive expenditure of national resources in the
non-productive military sector. Recent
efforts by the
When they seek to
operationalize power capability many researchers have used what Thompson
(1983b) somewhat disparagingly calls an "omnibus" measure which combines
a number of different indicators of economic and military resources.
Organski and Kugler (1980:30-8) have argued strongly for "total
output," that is total GNP, as the best measure of power capability of the
great powers. They also argue (1980:
68-84) that total output should be combined with a measure of political
development, which they define as the ability of the state to mobilize the
resources of its own society. Another
study which has explicitly tried to measure variation over time in the degree
of concentration of power among the "great powers" is that by Singer,
Bremer, and Stuckey (1979). They
combined a number of economic and military indicators of the capacity of
states. A similar composite measure was
constructed by Ferris (1973). Curiously,
none of the indicators used are measures of economic development as usually
understood, but are rather indicators which combine size and development, as
does total GNP. Of the studies I have
reviewed, only that by Doran and Parsons (1980) utilizes measures of economic
development, and they combine these into a single composite index along with
size indicators.
Some surprises
emerge from my review of the studies which have tried to empirically measure
changes in the relative power capabilities of states. Although these have been done by political
scientists, only Modelski and Thompson focus exclusively on military power, and
they examine only a particular kind of military power ─ naval
power. Organski and Kugler argue that
total GNP is the best overall measure of power capability, although later they
combine it with a measure of internal state strength. None of the studies measures hegemony in
terms of total military power. None of
the studies examines measures of development separately from measures of
size. And none of the studies tries to
examine the relationship between economic and political types of power.
Causes and
Conditions of the Rise and Decline of Hegemonic Core States
In this section a
set of causes and conditions of the rise and decline of hegemonic core states
is hypothesized. Then I present a brief
overview of the three hegemonies. Later I describe the few quantitative studies
of the hegemonic sequence which have been conducted.
The rise and fall
of hegemonic core states can be understood in terms of the formation of leading sectors of core
production and the concentration of these sectors, temporarily, in the
territory of a single state, which hence becomes the most economically and
politically powerful of the core states.
Decline sets in when the hegemon loses its ability to develop lead
industries ahead of its competitors.
This process can be understood as a feature of the world-economy as a whole
insofar as it involves the interaction of systemic variables such as the
Kondratieff wave, the application of new technologies to production (Mandel,
1978; Rostow, 1978 and Bousquet, 1980); and the violent reorganization of the interstate system
through warfare. As noted in the
previous chapter the strong association in time between long business cycles
and wars among core powers has been empirically demonstrated.
The uniqueness of
the world-system perspective is that it examines the system-wide dynamics of these cycles as well
as the exclusively national processes involved.
The cycles that occur are the consequences of the relative
over-production in different periods of different types of commodities (core
commodities and peripheral commodities) and the limits on effective demand that
particular political structures impose on consumption. During periods of the expansion of core
production labor unions, guilds, and other politically organized interest
groups increase their demands for shares of income. The expansion of core production increases
the need for raw material inputs, many of which are produced in the
periphery. The terms of trade between
core and periphery shift in favor of the periphery, enabling peripheral
producers and the states they control to attain a relatively more favorable
market position and to make more effective demands for a larger share of
surplus value. This redounds on the
struggle for shares among core states, where wage increases (and the
"cacophony of equity demands") are less easily met by increased
exploitation of the periphery. This dynamic
leads to heightened class struggle within core countries and to increased
competition among core countries for shares of a no longer growing pool of
world surplus value. It is striking that
hegemony in the core is consolidated after wars in which potential contenders
have destroyed one another, leaving an opening for the emerging dominance of
the new hegemon. The pattern which is
documented by Thompson (forthcoming) and Goldstein (1988) is as follows: a
rising challenger (B) initiates war against the declining hegemon (A). (A) makes an alliance with another rising
core state (C) to combat the military challenge by (B). (A) and (C) win the war and (C) emerges as
the new hegemon.
Goldstein's study
examines the relationship between the war/growth cycles and the rise and fall of hegemonic
core powers. Utilizing the Wallersteinian
conceptualization of three hegemonies ─ the Dutch, the British, and the
American ─ he demonstrates that they are related, but not in a very
regular way. As Goldstein (1988:287)
puts it:
I find the connection between the causal dynamics of these two
cycles--long waves and hegemony cycles ─ to be weak. They are not synchronized, and there is no
exact number of long waves that makes up a hegemony cycle. Rather, I see the two cycles as playing out
over time, each according to its own inner dynamic but each conditioned by,
and interacting with, the other.
And further along he concludes:
Each hegemony cycle contains several long waves, but not a fixed number. Each of the long waves within the
hegemony cycle ends in a war peak that re-adjusts the international power
structure without bringing in a new
hegemony.
(Goldstein, 1988:288, emphasis in the original)
In a footnote
Goldstein continues: "All great power wars affect relative positions in the international 'pecking
order.' Hegemonic wars determine the top
position in that order" (1988:288).
Although
Goldstein does explicate a causal model which explains the relation-ship
between economic growth and periodic wars (see chapter 8), his explanation of
the linkage between these cycles and the three hegemonies is more historical. He argues that many contingencies link the
particular outcomes which are revealed by the hegemonic sequence and the
particular countries which become hegemons, challengers, and "also-rans." Goldstein does not, however, address the more
system-wide questions which were the focus of chapters 7 and 8 above: why is
the interstate system reproduced, rather than evolving into world-empire; why
do the successful hegemons rely more on accumulation through trade than
military expansion; and why don't hegemons facing decline opt for world-wide
imperium? Also, why are the direct
military challengers of hegemons never successful? As argued in previous chapters these
questions require attention to the peculiar nature of competition and accumulation
in a capitalist world-economy.
Though Goldstein
treats the hegemonic sequence historically, it is possible to try to delineate
the systematic conditions and processes which
contribute to the rise and demise of hegemonic core states. Let us discuss some similarities of the three
hegemons which suggest the kinds of processes
which can account for their rise and decline.
The Three Stages
of a Hegemony
Hegemonies have
three stages. The first is based on
competitive advantage in mass consumption goods that can penetrate the markets
of core producers in competing countries and also can expand the size of the
market by lowering the price of the product.
The second stage is based on the expansion of capital goods production,
and the third stage is based on the export of financial services and the
performance of central place functions for the world-economy (See Wallerstein,
1984a).
In terms of the cities that became hegemonic
world cities over the history of the modern world-systems we may compare
A number of
conditions may contribute to the determination of which country becomes a
hegemon. Geographical location would
appear to play some role in facilitating
hegemony. Hegemonic powers have been
centrally located
within the
economic networks they come to dominate.
This is obviously an advantage in
terms of transport costs, but may become less important as transport costs
decrease. Technology adequate to a
breakthrough in competitive core production must be available, either from
local inventors or through borrowing.
All three hegemonies have involved "industrial revolutions" in
the sense that new, more economically efficient technologies were applied that
allowed the production of mass consumption goods more cheaply than by
competitors. Another necessary condition
is the existence of investment capital sufficient to develop the new types of
production in the hands of those willing to risk it in entreprenuerial
ventures. Each rising hegemonic power has early on developed
diversified, capital-intensive agriculture for home consumption and for
export. They have eventually developed
access to cheap imports of some staple foods and raw materials, most often
produced in the periphery, which have been important inputs to industry. Human capital, that is, labor with skills
relevant to the new type of production, must be available. All these conditions contribute to the
ability of an emerging core state to form a leading sector of core production
that can serve as the basis for hegemony.
The political conditions of rise to hegemony are rather more
complicated. A hegemonic state must be
powerful vis-à-vis other states and must also have the strong support of
the class coalition that composes its regime.
The quality and unity of this class coalition are also important. It must strongly include classes with an
interest in pursuing a strategy of profitable production for the world
market. Although the ability of the
state apparatus to appropriate resources is undoubtedly important (Tilly,
1985), the conception of state power I am using is not reducable to the
extractive power of government. As
discussed in chapter 6, a state is powerful if the classes that support it will
grant it great support during emergencies (Tardanico, 1978). Its ability to extract surplus through
taxation does not automatically show that it is strong in the sense that I mean
here. The Dutch state could raise a navy
overnight by convincing the Amsterdam merchants that their interests were at
stake, whereas the French state, whose peacetime government revenues per capita
were much greater than those of the British throughout the nineteenth century,
could not raise so great a subscription during time of war.
The size of the
state is also important, and as Wallerstein is fond of pointing out, it is
possible to be too large as well as to be too small,2 especially if economic regions with
contradictory interests are part of the same state, as was the case with
What, then, are
the conditions that lead to the decline of a hegemonic core state? First it should be pointed out that core
states do not decline absolutely. The
entire world-economy continues to grow, albeit at different rates. What happens is that core states relatively
lose their hegemony, but they do not plunge into the periphery. The most important cause of relative decline is the spread of leading core
industries to other competing core countries, and to parts of the
semiperiphery. Hegemonic states attempt
to monopolize the new types of
production, but unsuccessfully because of their inability to politically
control the diffusion of techniques, skilled labor, and investment capital. Competing producers in other states attempt
first to win back their home markets, often employing political regulation of
trade (protectionism) as well as the adoption of the new production techniques
(Senghaas, 1985). Later some of them
will successfully compete with the hegemonic power in international markets.
Another factor
that contributes to the loss of hegemony we may term the turnover time of fixed
capital, especially investment (both private and public) in infrastructural inputs to the
production system. This obviously
operates at the
level of heavy investment in technologies such as plant construction and
expensive large-scale machinery.
Latecomers have an advantage in that they can adopt newer technical
innovations, while earlier investors must wait to recoup initial investment. Steel plants in the
Transportation
systems, urban structures, communications systems, and energy systems involve investments of
resources that, once made, tend to be relatively permanent or not easily
reorganized. The canal system of
It can be asked
why entrepreneurs within a declining hegemonic core state do not invest within their national
economies to revitalize material production and increase productivity. It may be that a particular steel company
must wait for its sunk capital to be depreciated before building a new plant
that uses more productive technology, but why don't other corporations make such investments? Here we may point out that the structure of
national tariffs plays a role in the determination of investment
locations. Tariff protection of national
industries increases in a period of slower growth as states seek to protect
their national markets against international competition. It may be in the strategic and national
interest to make new investments in steel, and indeed states often adopt
policies that subsidize such undertakings.
But the purely profit-oriented logic of investment is unlikely to help a
country which is losing its competitive position in the world market. Building a new steel plant next to the old
plant means that the national market will have to be shared, while buying steel
from more competitive producers abroad and investing in more immediately
profitable enterprises (often located in other countries) is the most
attractive strategy for private investors.
Thus economic
nationalism by itself might prevent the relocation of certain industries, but
declining hegemonic core states are usually ambivalent about the choice between
nationalism and internationalism. This
reflects the contradictory interests of their "national" and
"international" capitalists as well as the contradictory interests of
workers as consumers and job-holders (Hart, 1980). The road of state-sponsored
"revitalization" advocated by those in hegemonic core states most
concerned with the interests of national producers (both labor and capital) may
be taken, but other competing states will also employ this strategy, and they
are more likely to do it effectively
because the coalitions of classes that control these other states are less
dominated by those who have international investments (Evans, 1985). The international capitalists within a
declining hegemonic core state can often
convince
consumers that it is better to try to
hang on to centrality in world exchange and to benefit from low-cost imports
than to adopt an expensive (and risky) program of economic revitalization.
In addition,
organizational features tend to have a certain inertia (or momentum, if organization is a process). Once a national economy becomes organized in
a certain way there is a tendency to crystallization around patterns which are then not easy to
change. While organizational forms may
be more malleable than the infrastructural features discussed above
(because material sunk capital is less
malleable), social rigidities do crystalize around organizational forms.
A frequently
cited explanation for the British decline at the end of the nineteenth century was the reticence of
family-held firms to adopt the newly emerging corporate form (Crouzet,
1982). In many sectors family firms apparently preferred continued control to
additional profits, and thus British industry was late to adopt the expanded
scale and new organizational forms becoming widespread in
Another factor is
the opposition that successful capitalist accumulation creates. The very political pluralism and relative
egalitarianism that was earlier a competitive advantage allows the formation of
constraints on the
maneuverability
of capital and increases the costs of production. The most obvious example of this is the
formation of political organizations that protect and expand the interests of
workers. Wages, both direct and social,
tend to go up in a successfully hegemonic core state. Capitalists who are making big profits are
more likely to accept a higher wage bill accompanied by a stable labor
supply. This changes when competition
increases and profits decline.
Similarly, other
constraints on the continued revolution of production become politically
articulated. The state begins to respond
to the needs of core workers and other
groups (e.g. consumers, environmentalists, etc.), and these
"non-economic" demands on capital may reduce the relative profit-
ability of
production within the country, at least compared to offshore locations where
workers and other groups are less well organized.
Mancur Olson's
(1982) The Rise and Decline of Nations stresses rising wages as the most
important among the "social rigidities" which cause the decline of
formerly successful national economies.
Olson's analysis points to some interesting organizational features
which reduce relative efficiency of
production and increase the obstacles to economic growth and revitalization. Nations in which interest groups are
fragmented and specialized, such as the
and the ability
to compete effectively in world markets are the best measures of progress. Thus the policy implication of his analysis
is that states and firms should remain as independent as possible of the
demands of workers, or other interest groups.
A more balanced definition of progress would suggest that the logic of
growth ought to include the needs of workers, consumers and the
environment. "Efficiency"
across the world-system should take these needs into account without pitting
workers in different countries against one another. This can be accomplished only by
democratically regulating major investment decisions at the world level.
Another
condition, one that is often related to increasing politically-articulated
constraints on capital, is the export of investment capital. Capitalists respond to differentials in
profit rates, and so increasing costs of operating in the home economy produce
the incentive to invest elsewhere, and thus the phenomenon of the export of
capital or "capital flight."
This means that fewer new investments in material production are made in
the home economy, although new lead sectors do continue to emerge, especially
in the provision of financial services to the larger world economy. Thus the world cities located in hegemonic
core states typically become more important to the economy of the country in
the latter days of the hegemony. This is
because the centrality in exchange that developed from the earlier centrality
in production is an important resource for the national economy and for the
functioning of the larger world economy.
It may be the case that, although benefiting from centrality to
a certain extent, the hegemonic core
state comes to bear too great a proportion of the costs of maintaining order
in the larger world-economy. The smooth
operation of the world-system requires the repression of deviance and the
maintenance of order, as does any social system. In the contemporary system an important
degree of order is maintained through political-military
expenditures. Military expenditures may serve some economic
functions (e.g. Baran and Sweezy, 1966),
but several studies demonstrate that they do not contribute to national
economic growth (e.g. Szymanski, 1973; Väyrynen, 1988). The small or non-existent military expenditures
of
Comparison of
Hegemonies
Let us now assess
and qualify the generalizations of the previous section by reviewing the
characteristics of the three powers that have been hegemonic in the modern
world-economy (the United Provinces of the Netherlands, the United Kingdom of
Great Britain, and the United States of
America) and by comparing them with the prior hegemony of Venice and with
the powers that contended for hegemony but did not attain it.
The Habsburg
Empire (which included the core "dorsal spine" of the European
world-economy in the first half of the long sixteenth century) (Bousquet, 1980) was based primarily on
political-military, rather than economic, centrality. The mercantile aggressiveness of the
Portuguese (Modelski, 1978) served as
the first wave of European expansion, but like the later centrality of Seville,
Portugal did not develop centrality in production.4 The Portuguese expansion, and the
"primitive accumulation" of money-capital by the Spanish, had
important, although complicated, effects
on the emerging European world-economy (Wallerstein, 1974:67-84), but
they did not lead to the development of core production in Lisbon or
Seville. Somewhat like the case of
The United
Provinces of the
Wallersteinian
conception of hegemony in a capitalist world-economy. The Dutch Revolution created a republican
federation in which the seafaring capitalists of
The Dutch state
is often seen as small, but in terms of the notion of state strength employed
in chapter 5, it was estimable (Braudel, 1984:193-5). Johan DeWitt, the Stadtholder of Amsterdam,
could raise sufficient funds in a day on the
The ideology of free trade and the rights of all nations to use
of the seas were propagated by the Dutch intelligentsia during the period in
which economic competitive advantage
enabled the
Barbour (1963)
contends that in many respects
Lane (1973)
observes that the Venetian ruling classes became land oriented during the
period of their decline, and he interprets this as an attempt to form a nation-state that could
compete with the larger states of the European world-economy in formation. The United Provinces seems rather small in
terms of land area and population size compared to the other states of Europe,
but it nevertheless played the role of
hegemonic core state rather effectively during the seventeenth century.5 The United Provinces may be seen as a kind of
midpoint between
The Dutch decline
exhibited the tendencies mentioned in my description of hegemonic stages: the shift toward financial services, the
export of capital, and the transformation of the capitalists from entrepreneurs
to rentiers (Burke, 1974; Riley,
1980).
The United
Kingdom fits the stages of hegemony best.6 Eric Hobsbawm's (1968) study, Industry and
Empire, depicts the three stages of hegemony in the rise of English cotton
textile production, its replacement in the middle of the nineteenth
century by the production and export of
machinery, railroads, and steamships, and the increasing importance of London
in the later nineteenth century as a center
of world financial services. The
Dutch hegemony, however, matches the three stages formulation rather well,
although the middle period of export of
ships, arms, and land reclamation projects fits the notion of
"capital goods" somewhat loosely.
The English
Revolution, like the Dutch, exhibits relative egalitarianism, pluralism, and
the firm incorporation of diverse capitalist interests into a flexible state
capable of mobilizing immense resources for international war while maintaining
a somewhat sparse and inexpensive peacetime bureaucracy. It should be repeated here that we are
describing features of the core of a larger capitalist system, not features of
capitalism as a whole. We do not want to
repeat the mistake of identifying capitalism as a system with the laissez-faire
state.
The unity of the
center coalition in the
The export of
capital from
Sometime after
1850 the average incomes of workers in
In general the
Southern peripheral capitalists, exporting raw materials to core industry in
The slave South
was the most successful peripheral economy the world has ever known, and it
sought to extend its political control over the West and over the Federal
state. After a series of confrontations
and compromises the "irrepressible conflict" was settled by the
Civil War, which resulted in the consolidation of control by core capital in
alliance with core labor and Western farmers.
In the 1880s the
The further rise
to hegemony by the
The success of
core production in the industrial North during the nineteenth century led to
the early export of cotton textiles, and not long after, to the export of
machinery. Electrical appliances and
automobiles became important mass consumption exports, along with other
industrial products. As discussed above,
the home market seems to have played a relatively larger role in the early
development of the
Let us compare
the
The
resulting
reorganization of the world political structure that would allow a new
hegemonic core power ─ not the
Soviet Union but perhaps
The apparently
shortening period of world-system cycles may reveal imminent changes in the
dynamics of the system which could alter the above scenario.
Since the late nineteenth century the system as a whole has begun to experience
certain natural and social "ceiling effects." Previously the contradictory aspects of the logic of
capitalist development have led to conjunctural reorganizations of the
political structure of the system through intracore war. This is what has pushed the decline of old
hegemonic powers and the rise of new ones able to operate on scales more
appropriate to the expanded size and intensified nature of the system. These reorganizations have allowed the
capitalist accumulation process to begin again on a new basis, that is, to
adapt to the problems created by its contradictory nature, and to continue
expansion and intensification. In the
twentieth century the ceiling effects have resulted in much deeper structural
problems for the system than previously (Chase-Dunn and Rubinson, 1979). The inclusion of virtually all global territory and population into the capitalist
world-economy in the late nine- teenth century eliminated the possibility of
expansion to previously unintegrated areas.
And the formal decolonization of the periphery, even though it has not
eliminated (or even reduced) the hierarchical core/periphery division of labor,
has increased the costs of exploiting the periphery. This reduces the amount of surplus value from
the periphery available for resolving class conflicts within core countries.
How do these
ceiling effects create a situation which is different from the one faced by the
Dutch or the British in similarly late phases of their hegemonies? One difference is a consequence of the
increased density of political
regulation of the capital accumulation process across the system. While most of this regulation is nationally
controlled, and thus is more an increase in state capitalism than a change in
the logic of the competitive accumulation process (Chase-Dunn, ed., 1982b),
there are incipient forms of supranational economic regulation, and this
provides a greater chance than ever before for the hegemonic core state to
engineer a political solution to the
trend toward increasingly bloody confrontation.
Ulrich Pfister and Christian Suter (1987), in their excellent study of
recurrent international financial
crises, argue that the ability of the contemporary world debt structure to
prevent (or postpone) collapse is due to the increased level of international
coordination among banks provided by such institutions as the International
Monetary Fund and the World Bank. The
process of world state formation may be unlikely to realize an effective
monopoly of legitimate violence in the next 40 years, but this outcome has a
greater probability of occurrence than ever before. The hegemonic core state, and especially that
sector of its ruling-class with the greatest dispersion of investments across
the globe, has the greatest interest in maintaining both the present order and
the global peace (Goldfrank, 1977).
From the point of
view of this cycle of core competition, where is the
matured in the
1945-70 period of the Pax Americana.
Economic hegemony began to
decline from about 1950 on. In 1950 the
Albert Bergesen
and Chintamani Sahoo (1985) show that the dominant position of US-based firms
has declined relative to the position of European and Japanese firms in various
world industries since the 1950s.
Financial centrality did not begin to slip until 1971, although signs of
unrest due to balance of payments deficits were visible earlier, as were
pressures to engage in trade protection (Block, 1977).
Examining trends
in their network measure of the structure of world trade of commodities at
different levels of processing, Smith and White (1986) show that the core is
becoming increasingly multicentric (less hierarchical) between 1965 and
1980. This finding supports the notion
that the decline of
The defeat in
Thus the golden
age of
It would seem
probable that the
Two trends may
portend a reorganization of US politics.
The
Cross-cutting
sectional differences are being evened out by the development of the sunbelt, a trend which
undercuts the domination of the South by "Dixiecrats." This makes the South more similar to the
North, and may reduce the regional cross-cutting which has been an important reason
that the major parties have resisted organization along class lines. Another factor may also increase the salience
of class interests in US politics ─ the trend toward a "shrinking
middle class" discovered in the income
distribution statistics since 1978 (Rose, 1986:table 5).
These trends may
well change the vocabulary of US politics toward serious consideration of
working-class interests and issues of democratic control of the economy, and in
this regard the
The Marxist theory of proletarianization is a theory about the
trajectory of changes in class structures in capitalism as such, not in
national units of capitalism. In a
period of rapid internationalization
of capital, therefore, national statistics are likely to give a distorted image
of transformations of capitalist class structures. If these arguments are correct, then one
would expect that changes in the class structure of world capitalism would be
unevenly distributed globally. In particular, there should be at least some
tendency for managerial class locations to expand more rapidly in the core
capitalist countries and proletarian
positions to expand more rapidly in the
This continued shift
toward a more core-like class structure may work against the other trends noted
above with regard to possible changes in US politics. While the decline of hegemony and the
increasingly unequal distribution of income may stimulate some polarization along
class lines and alter the focus of politics, it is unlikely that the
Conflicting class
interests over international economic policy will continue to make it difficult for the
The tendency for
core states to solidify relations with particular peripheral areas may undercut the tendency,
observed since World War II, for violent conflicts among core powers to be
fought out in the periphery. The export
of violent confrontation to the less powerful areas of the globe has been the
rule during the period of the Pax Americana.
But in a period of multicentricity, wars that break out in the periphery
may more easily become world wars.
Measuring the
Hegemonic Sequence
Most of the
studies of the hegemonic sequence have been of the type carried out above, a
narrative which tends to confirm the theoretical views of the narrator. The classic example is The Precarious
Balance by Ludwig Dehio (1962), a diplomatic history of the clashes among
the great powers of
There have,
however, been a few studies which have attempted to measure the sequence of concentration and dispersion
of power among core states. These have
been carried out primarily by political scientists studying the inter-
national system,
and they have employed very different ideas about what constitutes hegemonic
power, as reviewed above.
It would be
important, given that there are very different theoretical claims made by
students of the hegemonic sequence, to compare different measures, to study the timing and the
magnitude of hegemonic rise and decline, and to discover the causal
interactions between economic and political-military types of power. It should be noted here that only one of the
existing studies (Thompson, 1986) has employed a measure which could be
understood as even a rough proxy for the Wallersteinian notion of hegemony (see
below). It is, of course, difficult to
measure comparative advantages in core
production and the ability to penetrate foreign markets, especially if
we want to quantitatively compare our measures over long periods of time. I will
contend, however, that certain measures of national economic development are
good proxies for the idea of comparative advantage in core production.
After all, GNP per capita is highly correlated with national labor
productivity (GNP per worker), especially when we compare a large number of countries. Other measures known to cross- national
researchers to be highly correlated with GNP per capita may also serve fairly
well as estimates of comparative advantage in capital-intensive
production. It is well known that the
distribution of the national product across economic sectors ─ especially
the percentage of the total product in agriculture ─ is highly negatively
correlated with other measures of economic development in cross-national comparisons with large numbers
of countries, as is the percentage of the work force in agriculture.
I will contend
that these measures of economic development can serve as rough proxies for
relative core status in the Wallersteinian sense, at least until we can find
the resources to unearth data more directly related to the notion of
comparative advantage in core production.
Some results of the analysis of the distribution of economic development
measures among core countries for the
nineteenth and twentieth centuries are presented below.
But first let me
report what other researchers have found.
George Modelski and William R.
Thompson (Modelski and Thompson, 1988) have completed a monumental coding
project which estimates the amount of
naval power controlled by each of the great powers of
Because Modelski
and Thompson are focusing on naval power and not land-based armies, their list
of hegemons and the timing of hegemonic cycles comes out being fairly similar
to the sequence posed by Wallerstein.
They are focusing on the highest level of global control, which is the
arena of ships of the line, capital ships, and later, aircraft carriers. The reason that the
Modelski/Thompson
list of hegemons is similar to that of Wallerstein is that naval power and
global reach are especially important forms of control in a capitalist
world-economy in which international markets and core/periphery trade are
important in the determination of who wins and who loses.
There are,
however, some important differences between the "long cycle" and the
Wallersteinian narrative of hegemons.
Modelski (1978) claims that
century and a
second in the nineteenth century, following the Napoleonic Wars. Several smaller European states, which became
"developed" core countries in the nineteenth century, never make it
to the list of great powers.
The
Modelski/Thompson approach is similar in some respects to the
"transition" theory of Organski and Kugler (1980). They argue that important core wars occur
when a rising dissatisfied great power challenges the international regime
(from which it has been largely excluded) by making war against a declining or
stagnating formerly most-powerful state.
Organski and Kugler do not formulate a world-system level hegemonic
sequence, but they do see uneven rates of economic growth as the main cause of
changes in the relative power among core states. As mentioned above, they operationalize this by measuring total GNP.
I have always
thought of hegemony (in the sense of comparative advantage in core production)
as exhibiting a wave-like pattern in which
troughs are periods of relatively equal distribution of comparative
advantage among core powers and peaks are the golden age of a hegemon's economic
and military power. This picture is
typified, following Eric Hobsbawm's (1968) Industry and Empire, by the
rise of British productive advantage during the late eighteenth-century
industrial revolution, consolidated after the victory over
The picture
painted by Modelski and Thompson is very different. For them the peak of the cycle is right after
a core war in which the new victorious
global power is at the zenith of its naval advantage over contenders. Slow deterioration begins right away (in 1815
in the above example) and core war breaks out again when a new power is able to
imagine defeating the old hegemon.
Modelski and Thompson note that the military challenger never wins, but that the conflict
results in a reorganized international power structure, usually centered on a
country which was allied with the former
hegemon.
Thompson's
(1983a) analysis of his own data leads, however, to a modification of the
above hypothesized transition model. He
notes that the pattern of an increasingly powerful naval challenger is not
regularly revealed in his analysis of the relative power of core states,
although there is a tendency for the advantage of the hegemon to
deteriorate. He posits, instead, a
"two-step transition model" in which a secondary core power tries to
expand regionally through land based military adventures, and this regional war
becomes a global war when the old hegemon and another allied core power
perceive the regional challenge as a global challenge.9 As Thompson puts it:
It is most unclear, moreover, whether the primary challengers
realize, at the outset, the full extent to
which their regional activities will be viewed as threatening by either the 'reigning' world
power or its
eventual successor. How
else are we to account for the repeated surprise with which the primary
challengers confront the intervention of English, British and American
military forces? How else are we to
explain, especially in the twentieth century, war breaking out before the primary challengers have achieved
the capability base that they themselves
have projected as necessary for global competition? (Thompson, 1983a:112)
Before commenting
more on the Thompson/Modelski project I will describe another study of the
concentration of power among core states, that by Singer, Bremer, and Stuckey
(1979). This is one of the studies which
devises what Thompson (1983b) calls an "omnibus" measure of power
concentration which combines six separate indicators. The indicators used are:
"demographic"
─ the number of people living in cities having population sizes larger
than 20,000; and the nation's total population;
"industrial"
─ total energy consumption; and total iron or steel
production; and
"military"
─ military expenditures and the number of persons in the armed forces
(Singer, 1979:273).
It should be
noted here that none of the variables used in the composite measure is an
indicator of the level of development.
They all combine aspects of a
nation's size with its level of development.
Thus simply being a large country can give a high score on any of the
indicators.
The resulting
scores were used to calculate a measure of the relative concentration or
dispersion of power among the "great powers." This was done by summing the values of each
indicator for all the nations thought to compose the circle of great powers,
and then calculating the percentage of that sum held by each country, and then
averaging the percentages across the six indicators to determine an individual
country's final score. These country
scores were then used to calculate a statistic invented by Ray and Singer
(1973) to estimate the concentration of power in an international system. The resulting concentration scores are
presented for five-year time points from 1820 to 1976 (Singer et al.,
1979:277).
Thompson (1983b)
uses the same concentration formula (the Ray─Singer index) to compare his
data on naval power with the Singer (et al., 1979) measure. He calls the Singer measure the C.O.W.
measure after the Correlates of War Project, a usage I shall adopt. Figure 9.1 is taken from Thompson (1983b:152)
and shows the relationship between the concentration index as calculated using
Thompson's naval data and the C.O.W. indicator from 1816 to 1960. Inspection of this graph reveals that the
concentration indices go way up during periods of core war. Thompson's
index drops dramatically following the Napoleonic War, contrary to the
notion of a slow deterioration, and both measures seem to rise slowly,
"mounding" rather than peaking during the nineteenth century before
they zoom up during World Wars I and II.
(Figure 9.1 about
here)
Although the
above raises questions about the pattern of concentration hypothesized by
Modelski and Thompson it does not really tell us much about the Wallersteinian version of hegemony. This is because none of the indicators used
in the above studies examines the distribution of development among the core
powers. A paper by Kugler and Organski
(1986) presents data showing the relative distribution of total GNP across
the circle of "great powers"
from 1870 to 1980. Kugler and Organski
utilize part of the C.O.W. methodology, summing the total GNP for all the
countries and then calculating each
country's percentage of the total. They
use the results of this table to dispute the contentions of many authors who
have written about hegemony. They show,
for example, that
The problem with
this "economic" measure is that it confounds size with development.
By this same measure
(Table 9.1 about
here)
Table 9.1, being
a transformed version of the Kugler and Organski table, contains some countries
which were not core countries in 1870, such
as
Table 9.1 also
indicates that, in terms of GNP per capita, British predominance peaked in the 1890s. This is well after the slowdown of British
growth studied by Crouzet (1982: chapter 12) which ostensibly began in the
1870s. Crouzet (1982:377) shows that
British growth rates between 1870 and 1913 in total output, output per head and
output per man-hour were lower than in Germany, Sweden, and the US, and even
France increased its output per head and output per man-hour slightly faster
than Britain in this period. But the
British growth rates are not significantly different than the average figures
for ten European countries during this period.
The depression of the 1870s was hard on all the countries, and it may be
that the British relative decline did not gain real momentum until the 1890s.
Another rough
proxy for the development of core production is provided by the figures on the
distribution of product across economic sectors (Mitchell, 1975). Table 9.2 contains information on
(Table 9.2 about
here)
As we can see,
the
In 1860 we have
data for
formation
immediately. The
manufacturing
production, and also by Thompson's (1986) study of the concentration of lead industries (see
below). On the other hand Maddison
(1982:table C5) indicates that fifty percent of the
I have examined
the relative concentration of several specific types of economic production and
other indicators (Chase-Dunn, 1976).
Several of these indicators were presented by Eric Hobsbawm in Industry
and Empire 1968: diagrams 23-25c).
Hobsbawm calculated his figures in terms
of ratios, the percentage of various things such as world trade,
industrial output, coal production, pig iron production, steel production and
cotton comsumption which occurred in, or were attributable to,
When we examine
the same distributions utilizing difference scores we indeed find peaks during the nineteenth
century, or more usually "mounds."
The British advantage over the sum of French, German and US quantities
generally rises to some point between 1870 and 1890 and then declines. For example, installed railway line per
kilometer of land area peaks around 1870.
Crude steel output peaks in 1890.
Raw cotton consumption peaks twice, once in 1880 and again in 1890. Pig iron production peaks in about 1880. Coal
output peaks in about 1885.
National income per capita peaks between 1890 and 1900, but then rises
again between 1910 and 1913. Foreign
trade doesn't really peak in the nineteenth century but rather rises and
continues to rise, as Hobsbawm (1968:diagram 26) shows, until about 1930. Similarly, steam engine power peaks in 1880
and then rises even higher after 1890.
Two things can be
inferred from the above. The
generalization about an overall peaking
of British hegemony in the 1870s must be considered a sort of average. When we look at particular sectors or
particular types of production the peaks
come at different times. And in some
respects the British hegemony continued on into the twentieth century. This was
especially true in terms of British centrality in international
trade. The
A study which
focuses on leading industries as the measure of economic advantage in the
hegemonic sequence has been presented by Thompson (1986). He uses Rostow's (1978) description of
leading sectors in national development to construct a list and periodization
of leading sectors in the development of the world economy (Thompson,
1986:table 3). He then computes national
proportions for several great powers beginning in 1790 for each of the lead
industries, and then averages the proportions across industries for each core
country to produce a measure of the concentration/dispersion of core
production in lead industries (Thompson, 1986:table 5). Thompson's results produce a rise and decline
for both the
Thompson's
measure of leading sector concentration is undoubtedly the best effort so far
to operationalize the rise and fall of economic competitive advantage in the
hegemonic sequence. And, like most good
empirical studies,
it produces some
surprises. Contrary to the description
found in Hobsbawm (1968) and suggested by my analysis above, Thompson's measure
shows the British hegemony peaking earlier, between 1810 and 1830, and his
measure also indicates that the
figure 6).
Even though his
study is much better than any earlier study because of more complete data for earlier time points
and a better operationalization of core productive advantage, Thompson's study
still cannot be considered the final word.
His choice of leading sectors and his periodization of them is
controversial. One of his sectors, pig
iron production, was an old industry in 1790.
A better choice would have been steam engine power (Landes, 1968:
221). Also, the countries which Thompson
includes in his comparison to
compute a
concentration index are controversial.
He includes
Since our
definitions of coreness are partly what is in dispute we need to have data on
as many countries as possible as far back in time as we can get it.
This will enable us to compare different methods of measuring the
concentration of leading core production and produce a more certain under-
standing of the
concentration/dispersion sequence. It
will also enable us to study the causal relations between economic competitive
advantage and political/military power, a matter which is central for all the
theories of hegemony but which has yet only been scratched empirically.
Thompson's study is an important step forward, but much more needs to be done.
The above
comparison of hegemonies reveals that most features of the three cases examined fit our general
description of the causes and conditions of rise and demise. But here more research which compares the
successful hegemons with less successful core contenders also needs to be
done. If the hegemonic sequence is
indeed a structurally-based phenomenon produced by the dynamics of the larger
world-system, we may wonder how the future operation of system processes will
affect the rise and demise of hegemons.
The
nationalism
longer than other core states, just as the British refused to adopt protection
long after all other core states had. It
should also be the biggest proponent of corewide cooperation because it has the
most to lose from
conflict. The recent international policy of the
The main
conclusions of Part II on states, the interstate system, and the hegemonic
sequence are summarized on pages 11 to 14 of the introduction. Part III focuses more closely on the
core/periphery hierarchy, and chapter 13 discusses possible causal connections
between world-system cycles, including the hegemonic sequence, and periodic
changes in the structure of core/periphery relationships.
Part III :Zones of the World-System
The following four chapters examine the notion that one of the
most important structures of the modern world-system is the core/periphery
hierarchy, a socially structured spatial system of power/dependence relations.
In chapter 10 we
discuss how related terminologies have been employed by other scholars and
consider various possible analytic definitions of the core/periphery
hierarchy. I propose a definition of the
semiperiphery and a usage of the imagery of zones which avoid the search for
empirical boundaries among categories.
The nested nature of the core/periphery hierarchy is considered, as are
the reorganizations which have been characterized as several "new
international divisions of labor."
The problem of the homogeneity of the periphery is discussed, and
research is reviewed which reveals consequential differences among peripheral
areas depending on the type of indigenous society which was present before
incorporation into the Europe-centered world-system. The idea that the form of incorporation
varied with changes in the organizational nature of the core at the time of
incorporation is also discussed. Then we
review efforts which have been made to measure the position of countries in the
core/periphery hierarchy, the problems involved, and some proposed solutions.
In chapter 11 the question
of the function of the core/periphery hierarchy for the reproduction of
capitalism is confronted. Have
colonialism and imperialism been necessary aspects of capitalism in the core,
or have these been only unfortunate by-products resulting from misunderstandings
or vestigial atavism? Was primitive
accumulation simply a stage by which capitalism was brought to the periphery,
or has it been a form of primary accumulation which was and is necessary to the
reproduction of capitalism in the core?
Rather than solving this problem by definition, an explanation based on
the importance of peripheral exploitation in reproducing capitalist class
relations in the core and the multicentric interstate system is proposed. Various mechanisms which are alleged to
reproduce the core/periphery hierarchy are reviewed, as is the crossnational
research which examines the effects of those mechanisms on national
development.
In chapter 12 we
examine recent changes in social structural characteristics which have occurred
in core and peripheral countries. Such
phenomena as dependent industrialization, overurbanization and the burgeoning
urban informal sector are considered.
Then we review the evidence regarding trends in the magnitude of
core/periphery inequalities. Has there
been an absolute
immiseration of the periphery, as some authors claim, or have
both core and periphery developed, but at different rates, resulting in a
growing relative gap?
Chapter 13 examines
hypotheses which link cycles in the whole world-system with cyclical changes in
the core/periphery relationship. The
Kondratieff wave, waves of core war severity, and the hegemonic sequence are
discussed in connection with alleged oscillations in the core/periphery structure
such as alternate tightening and loosening of regional trade net-
works, periods of trade protection versus periods of relatively
more free world market exchange, waves of colonial expansion, waves of capital
exports from the core to the periphery, and cyclical international financial crises
which are triggered by defaults on peripheral debts.
Chapter 10: Core and Periphery
This chapter discusses the analytic meaning of core and
periphery. These conceptual categories
are unpacked into their underlying dimensions, and the controversy over
different usages is discussed. The
notion of the semiperiphery is also defined, and the nested quality of the
core/periphery hierarchy is considered. The problem of boundaries between the core,
peripheral, and semiperipheral zones is described, and various
approaches to operationalizing the
core/periphery relationship are reviewed.
Then we shall consider the problem of peripheralization as a process,
and discuss literature which considers the incorporation of different local
modes of production into the capitalist world-economy. Recent research by economic historians
disputing the importance of exploitation of the periphery for core
industrialization is critiqued. And we
review the literature about
reorganization of the form of the core/periphery relationship which has
been characterized as the "new international division of labor."
Terminologies
Regional differences have long been of interest to sociologists,
anthropologists, geographers, political scientists, and economists. The relationship between
"civilization" and "barbarism" is perhaps the oldest form
of the comparison of developed and less developed regions. The modernization theorists utilized the
distinction between modernity and tradition to compare societies thought to be
at different levels of development.
It has been discovered
that the so-called developed and undeveloped regions are often in interaction
with one another, and that this interaction
often importantly alters the structures of both partners (e.g. Lattimore,
1940). The idea of regional hierarchy
has been applied to the contemporary international system by dependency
theorists who discuss dominance and dependency, by political scientists who
study the interactions between the countries of the "North" and those
of the "South" (e.g. Doran, Modelski, and
The terms I shall
employ have been proposed by Immanuel Wallerstein. These will be defined below,
but here I will simply list the words.
We speak of a hegemonic core power, other core powers, and second-tier
core powers. We speak of semiperipheral
countries, including those which have moved down from the core and those which
are moving up from the periphery. We
speak of peripheral areas, of extreme peripheries and of external arenas those areas outside the boundaries of the
world-system. Core is preferred to
center because it suggests an area rather than a point. Periphery means a large category, not just an
outside edge. The scheme refers to a
socially structured stratification system which is spatially differentiated
because of the territorial forms taken by important organizations, especially
states. As implied by the above,
vertical mobility is possible within this structure of inequality,
although the overall hierarchy is
understood to be reproduced by several processes which operate in the
world-system. These processes are examined
in chapter 11. Here we will focus on the
nature of the spatial inequality which is produced by a capitalist
world-economy.
Core and Periphery in General
The contemporary world-economy is not the only world-system
which has had a core/periphery hierarchy.
Indeed, Kasja Ekholm and Jonathan Friedman (1982) have argued that all
world-systems, past and present, have core/periphery hierarchies and are based
on a similar general "capital imperialist" mode of production. They point out that all world-systems seem to
go through periods of concentration and dispersion of power and wealth, to
experience uneven development, and to exhibit the domination and exploitation of peripheral areas by core
areas. Much of the literature about the
evolution of states and empires supports the notion that core/periphery systems
have been important dimensions of organization in the ancient world-systems
(e.g. Rowlands, Larsen, and Kristiansen, 1987).
It is typical for a core area to extract surplus product from peripheral areas. It seems likely however, that there have been
important qualitative differences in the ways in which these core/periphery
hierarchies have been organized. One
interesting difference between the modern worldsystem and most precapitalist
worldsystems is the relative degree of inequality within core and peripheral
societies. Core countries in the modern
system have relatively less inequality, while in precapitalist systems the core
societies tended to be more stratified.
I have recently outlined the beginning of a comparison of core/
periphery hierarchies in different types of world-
systems which seeks to determine their similarities and
systematic differences, but this work is only at the stage of a prospectus
(Chase-Dunn, 1986). Here we will define
core and periphery in a way which is specific to our contemporary worldsystem,
although a more general definition will be needed for the comparative study of
different types of world-systems.
Core and Periphery in a Capitalist World-system
The analytic question which must be discussed is the essential
nature of the core/periphery hierarchy in a world-system in which capitalism is
the dominant mode of production. It is
relatively easy to compile a list of social structural features which
distinguish core areas from peripheral ones.
Thus, core states are internally and externally strong, contain
relatively integrated nations, and have articulated national economies in which
production is relatively capital intensive and wages are relatively high. Core states have relatively less internal
economic and political inequality than do peripheral states.
These generalizations
are true but they do not tell us what the essential quality of the
core/periphery hierarchy is. Of course
it may be that there is no single, most important, underlying dimension.
Nevertheless, some authors have specifically postulated a central analytic
definition of coreness and peripherality, and it would be theoretically
valuable to have a clearly specified definition. Here we shall discuss the various previous
formulations and present a new synthesis.
In general the
core/periphery hierarchy is a structure of domination and exploitation, of
course. As in all class-based
socio-economic systems, surplus product
the material product of direct producers beyond that which is required to reproduce those producers is appropriated by a class of
nonproducers. But how is this
appropriation organized and accomplished in
the modern world-system?
It is easiest to begin
with the ways in which it is not accomplished.
First, as argued in chapter 5, it is not organized primarily through
normative mechanisms. This is not to say
that these have no importance but, in comparison with political coercion and
market exchange, normative mechanisms of control and appropriation play only a
supportive role. Secondly, political
coercion, though still important, is much less central to the appropriation
process than in earlier world-empires and worldeconomies in which various forms
of the tributary modes of production were dominant. Taxation is certainly important as the source
of resources for states, but taxation and tribute are not the most central
forms of appropriation in the modern world-
system.
I disagree with those
who focus on politicalmilitary power as the main dimension of the
"North/South" relationship.
Political coercion is more directly operant in peripheral areas than it
is in core areas, and political coercion (including the rule of law and police
forces within nation-states and the use and threat of military force in
relations between states) certainly plays an important role in maintaining the
power relations which are necessary conditions for the operation of capitalist
commodity production. But, in comparison
with historically previous world-systems, this world-system is much less
reliant on direct politicalmilitary coercion, and more reliant on economic
exploitation which is organized through the production and sale of
commodities. The problem becomes: how
can we analytically define the typical
mix of political coercion and economic remuneration which allows
core areas to dominate and exploit peripheral areas in the contemporary
world-system? And once we have defined
coreness and peripherality, how can we
measure this dimension?
Several problems
emerge from a review of the definitions and
discussions we find in the literature on core/periphery relations. One concerns the units or actors which are
the nodes of core/periphery relationships.
Another is, ofcourse, the nature of the relational qualities alleged to
be central. And another is the
hypothetical shape of the distribution of core and peripheral activities.
Many authors have claimed or implied that the
main economic dimension of the core/periphery hierarchy is the division of
labor between industrial production of
processed goods versus the extractive production of raw materials or
agricultural commodities. Several
important pieces of research have been
done which examine differences in the "level of processing" of
commodities. These often imply that the level of processing distinction is the
main dimension of the core/periphery
hierarchy. This argument stems from a
1945 study by Albert Hirschman (1980) which examined the effects of the level
of processing of exports and imports on
national economic power. Hirschman, and
later Galtung (1971), reasoned that a national economy which produces mainly
highly processed goods will have a higher rate of growth because of more integrated
forward and backward linkages within the constellation of economic activities
in the national economy, and thus greater spinoffs and multiplier effects
caused by new investment. Regional
economies in which primarily
extractive raw materials or agricultural products are produced
are likely to be less internally differentiated, less internally linked, and
thus new investments are unlikely to stimulate much local growth.
Another argument in
support of the level of processing contends that it is easier to apply innovations and new
technology to the industrial sector than to the raw materials and/or
agricultural sector because the industrial sector is less dependent on
"natural" factors, and is thus more amenable to
reorganization. Stephen
Bunker (1984) contends that extractive economies export large amounts of
environmentally derived energy, and thus rapidlydeplete the local ecosystem,
which in turn tends to undermine the possibility for more diversified economic
activities. These negative effects of
extractive activities can (in theory and sometimes in practice) be overcome by
large investments to guard or reconstruct the local ecosystem, but the owners
of extractive enterprises are unlikely to make such preservationist
investments. Thus extractive activities
tend to prevent the development of core-type diversified industrial activities
in the same region.
The
Core activities receive disproportionately high returns, while
peripheral activities receive low returns.
And this distinction is conceived as dichotomous, so that each activity
is either core or peripheral. A
core area is one in which a relatively high proportion of economic activities
are core
activities, and vice versa for a peripheral area. A semiperipheral area is defined as a region
containing a relatively equal mix of core and peripheral activities.
Giovanni Arrighi and
Jessica Drangel (1986) have made a valuable effort to clarify the analytic
definition of core and peripheral activities.
They agree with Wallerstein that
the core activity/peripheral activity distinctionshould be conceived as
dichotomous, but they disagree with Wallerstein's tendency to equate core
activity with capitalintensive (machine) production. Wallerstein and other theorists (e.g.
Chase-Dunn and Rubinson, 1977) have criticized the notion (discussed above)
that equates the core/periphery dimension with a division of labor between
processed manufactures and the production of raw materials or agricultural
commodities the level of
processing. Both raw material and
agricultural production may be carried out as core production if
capitalintensive technology is combined with skilled well-paid labor, it is
argued. Thus the distinction between
core agriculture and peripheral agriculture, and also core industry and
peripheral industry, is made possible, with the underlying differences having
to do with the level of profits and wages, and these are assumed to be
associated with the relative degree of capital intensity.
While Arrighi and
Drangel (1986) agree that it is a mistake to simply identify core activity with
the level of processing, they also question the identification of core activity
with relatively capitalintensive production.
Rather they define core activity as that economic activity (not
necessarily production) which receives relatively high returns regardless of
what the substantive nature of the activity is.
Arrighi and Drangel adopt a Schumpeterian definition of core activity
based on entrepreneurial innovation.
Schumpeter (1939) argued that the driving force behind capitalist
accumulation is the ability of organizational entrepreneurs to develop new
activities which enable them to capture a large share of the returns to
economic activity. This may occur in the
realm of product development and production, but it may also occur in financial
or commercial activities. Arrighi and
Drangel argue that core activity consists in the ability of some actors to
capture relativelygreater returns by protecting themselves to some extent from
the forces of competition. Peripheral
activity, on the other hand, is exposed to strong competition and thus the level
of returns (profit, rent, and wages) is low.
The Arrighi and
Drangel definition is provocative, and it bears similarities with other recent
work on the nature of core capitalism.
Braudel (1984) focuses on haute finance, a combination of finance
and commercial capital aided by state-abetted monopolies, as the essence of
core capitalism. This similarly
differentiates coreness from any particular type of production. We are also reminded of Raymond Vernon's
(1966) notion of the product cycle in which core innovators create new products
which they are able to sell for high prices (technological rents) until the
products are copied by other producers who cut the price and utilize cheaper
inputs, thereby shifting production of the product toward the competitive
(peripheral) sector. Also Arrighi and
Drangel's discussion is reminiscent of the distinction made by James O'Connor
(1973) between the monopoly sector and the competitive sector,
and the related distinction between the primary
labor market and the secondary labor market employed in labor market
segmentation theories.
While these approaches
may seem to be similar on the surface, they differ in terms of the underlying
mechanisms by which differential returns are generated. The product cycle idea implies that research
and development costs for developing new products are recouped through
technological rents the ability of an
innovator to receive a high price for a new product. There is no exploitation
here. Those capable of creating new
products for which potential markets exist receive a fair profit on their
research and development investments. On
the other hand, monopolized or oligopolized industries may receive an
exploitative return because of their ability to control prices, thusresulting in
"surplus profits." It is
implied that protection from competitive forces is not due so much to
innovation as to the ability of firms to gain political protection from states,
or because of the prohibitively high cost of entry into the production of goods
which have attained high capital intensity and great returns to the scale of
production.
Core Production
I will define core activity as a certain kind of production, the
production of relatively capitalintensive commodities (core commodities) which
employ relatively skilled, relatively highly paid labor. This is a relational idea because the level
of capital-intensity which constitutes core production during a specific period
is defined as relative to the average level of capital-intensity in the world-system
as a whole. Since average capital-intensity
is a rising trend, forms of production which once were core production may
become peripheral production at a later time.
Capital intensity involves the utilization of
techniques which facilitate high productivity per labor hour. Thus a large component of capitalintensive
production is the utilization of machinery, or capital goods, in the production process. Capital-intensity is similar to Marx's idea
of the organic composition of capital
the ratio of capital to labor which is employed in the production
process. It is also closely related to
the idea of labor productivity, although both capital-intensity and the speed
and skill of human effort are involved in labor productivity. Capitalintensity
and the usage of skilled labor are usually combined, at least when we consider
the overall production process. It may
not take skilled labor to operate the machines, but it takes skilled labor to
build them and to keep them running.
A core area is
an area in which relatively capital-intensive production is concentrated. Capital-intensive production is often in the
manufacturing or industrial sector of a national economy, but it may also be in
the service sector, the agricultural
sector or other sectors. The definition
of core production is not restricted to "industry" even though this
is often the most capital-intensive sector.
Agriculture in core areas is usually also capital
intensive relative to agriculture in other zones of the
world-system, and the same is true of services.
It makes no sense to
me to dichotomize the distinction between core and peripheral production. Dichtomization creates the false problem of
where to
place the cutting point.
Rather the core/periphery dimension is a continuous variable between constellations
of economic activities which vary in terms of their average relative levels
of capital intensity versus labor intensity.
What is the unit which
can be designated as engaging in core or peripheral activities? Wallerstein uses the term "node" to
designate the locus of core or peripheral activity on a commodity chain. This term is intentionally vague because all
of the more specific designations have problems. One possibility is the firm, the unit of
capital, the organization which appropriates profit. But some firms are transnational and combine
both core and peripheral types of production.
And some important aspects of core production, such as relatively dense
forward and backward linkages, may not be characteristics of single core
firms. Cities are a possibility because
they are also loci of accumulation, and they can have the crossfirm
characteristics often associated with core production. But not all regions have cities, and some
aspects of relations among cities are important to the core/periphery distinction.
A third possibility is
the nation-state. The trouble with
focusing on a national economy, defined as a juridical unit, is that the real
economic networks often do not follow state
boundaries. Indeed, in the sense of
completely self-contained economic systems, there are no national economies in
the world-system. But regions and nation-states do differ in
terms of their relative levels of economic integration, as pointed out by
dependency theorists and Marxist scholars such as Amin and DeJanvry. Amin (1974) and de Janvry (1981) define core
capitalism as self-reproducing, relatively integrated, capitalist accumulation,
whereas peripheral capitalism is understood as a disarticulated regional
economy which is highly dependent on imports from, and exports to, the
core. It is worthwhile to remember that
core states are also dependent on the existence of the larger worldeconomy, but
it is also important to recognize the very different extent and nature of this
dependence. The differential integration of regions and nation-states has
long been, and remains, an important feature of the core/periphery hierarchy,
and this aspect of the hierarchy requires that we focus on regions rather than
the activities of individual firms in defining core and peripheral formations.
The unit of coreness
or peripherality will be the "region" then. This is still vague but it is clearly not the
firm, nor the nationstate. Cities but
also systems of cities and rural areas can be regions of relatively capital
intensive or labor-intensive production.
Arrighi and Drangel
dispute the identification of core activities with capitalintensive production because
industrial production is becoming an activity which is located in the
semiperiphery and the periphery, and they also point out that
"non-productive" activities are often more profitable and yield
higher salaries than activities which are directly associated
with industrial production.
Arrighi and Drangel
(1986:54) show that the average proportion of the work force in the industrial
sector has diminished in core countries while it has risen in semiperipheral
and peripheral countries.
Industrialization has certainly occurred in the semiperiphery and much
of the periphery, but I contend that this has not diminished the level of
inequality between core and periphery in terms of the capitalintensity of
production. The proportion of
the work force employed in the industrial sector in core
countries has gone down precisely because capitalintensity has risen. Robot manufacturing does not require a large
work force in the factory. Because
capital intensity has continued to rise in the core during the
industrialization of the semiperihery the overall distribution has not much
changed. If the distribution of capital
intensity were to become more equal there would also be a decentralization of
economic power which would threaten the operation of capitalism.
As for the Arrighi and
Drangel observation that capitalintensive production does not always yield the
highest returns, this is true, but again
it is necessary to examine economic constellations of activities. Speculators on the stock exchange may enjoy
the highest rate of profit, and this activity may increase in a period of
uneven development in the core and stagnating growth throughout the
world-system, because money-capital cannot find profit
able productive investments in a world in which productive
capacity greatly exceeds effective demand.
But this should not lead us to the conclusion that these
"innovative entrepreneurs" have succeeded in moving on to a new form
of accumulation which will be successful in the long (or even medium) run. Rather this form of speculative activity is
more a sign of world-level crisis than a
new form of core accumulation (see chapter 4).
On the other hand, it
must be admitted that the association between capitalintensity and the level of
returns is not exact. Successful
accumulation may or may not lead to a diversified, relatively capitalintensive
regional economy. The rise and fall of
peripheral boomtowns and extractive enclaves are evidence of this. Successful development of a core economy
involves all sorts of investments which are not immediately
profitable, as well as the sort of political regulation and state policy which
facilitates long term development rather than simply shortterm
accumulation. The fact that Arrighi and
Drangel (1986:44) claim that Libya has moved into the core (based on their use
of GNP per capita as a measure of core status) reveals the weakness of their
identification of core activity with shortrun returns based on any kind of
activity.
periphery hierarchy as a nested multilevel structure.
Nesting
The core/periphery hierarchy is a system-wide dimension of
structured inequality, but at the same time it is also a regionally nested
hierarchy. The states of the interstate
system are obviously important units in this
hierarchy, but states are not internally homogenous. As suggested by Galtung's terms, "the
periphery of the center" and "the center of the periphery,"
there are important regional inequalities within countries. Many of the processes ofuneven development which
we study at the level of the world-system also occur within countries (Hechter,
1975; N. Smith, 1986) and these are not only analogous processes. They are often historically linked with one
another. In the
In addition, the world-system
is nested into international regions as
well as regions within countries.
An analysis of the world-system hierarchy must not ignore the nested
forms of inequality which occur in continental
subregions. The discussion of
sub-imperialism (e.g. Marini, 1972) suggests that regional powers, such as
periphery." Even
with substantial nesting the multi-level core/periphery structure forms a
highly stratified hierarchy of dominance and dependence. Evidence regarding the magnitude of this
inequality is discussed in chapter 12. The Semiperiphery
The idea of the
semiperiphery is one of the most fruitful concepts introduced by Immauel
Wallerstein. It has been widely used but
there are important disagreements about definitions and some scholars have
criticized Wallerstein for vagueness and contradictory usage (see Lange,
1985). The suggestions I have made above
regarding the core/periphery hierarchy imply a reexamination of the
semiperiphery concept.
Wallerstein employs
two elements in his definition of the semiperiphery the dichotomy between core and peripheral
activities, and the notion that a state boundary encompasses an approximately
equal balance of both core and peripheral activities. Thus, by this definition, there are no
semiperipheral activities as such.
Rather there are semiperipheral states which contain a balance of both
core and peripheral activities.
Wallerstein also
argues that the existence of semiperipheral states acts to depolarize the
core/periphery hierarchy by providing intermediate actors whose very presence
reduces the salience of potential conflict along the core/periphery dimension
of inequality. He has also discussed the
opportunities for upward mobility of semiperipheral states during the
stagnation phase of the Kondratieff cycle (Wallerstein, 1979a:chapter 5).
Since I have
reconceptualized core and peripheral activities as a continuum of relatively
capital intensive/labor intensive forms of
production, it is hypothetically possible, according to my usage, for a
semiperipheral areato contain a uniformly intermediate level of production with respect to the core/periphery
continuum. The semiperiphery idea is an
important one because it enables us to focus on how the existence of
intermediate regions affects core/periphery dynamics in the world-system as a
whole. It also encourages us to examine
the ways in which intermediate actors have different strategies, and
intermediate states have different developmental possibilities different in the sense of systematically
differentiated from either typical core or typical peripheral regions. Wallerstein does not claim that the
semiperiphery is a homogenous zone or
set of states. Rather he contends that
being in a semi peripheral location visvis the core/periphery hierarchy
is a condition which encourages certain kinds of behavior.
The idea of
semipheripheral states containing a balance of both core and peripheral
activities is useful because this condition is likely to produce contradictory
economic and political interests within the boundaries of a single state. Wallerstein argues that this was an important
reason why
The notion of a mix of
core and peripheral activities is useful, but there is another type of
semiperiphery, that which contains activities which are predominantly
intermediate in terms of the relative level of capital
intensity/laborintensity.
This does not have exactly the same consequences as the mixed form. Wallerstein's usage sometimes suggests that
some kinds of class relations are intermediate in form, and thus indicate
semiperipherality. In his discussion of the semiperipheral Christian
Mediterranean of the long sixteenth century Wallerstein (1979a:chapter 2)
contends that share-cropping was a form of rural class relations which was
intermediate between the yeoman agriculture of core areas and the serfdom and
slavery of peripheral areas. This idea
of an intermediate form of labor control is similar in some ways to my notion
of intermediate levels on the capitalintensity/laborintensity continuum. This suggests that there are indeed some
activities which are usefully conceptualized as semiperipheral, and that the
prevalence of these activities in a region or state can constitute a
semiperipheral area.2
I am arguing that
there are two analytic kinds of semiperipheries: Type one are those states in
which there is a balanced mix of core and peripheral activities, and type two
are those areas or states in which there is a predominance of activities which
are at intermediate levels with regard to the current world-system distribution
of capitalintensive/laborintensive production.
Obviously much is left
out in terms of the kind of specification we will need in order to
operationalize these definitions of the semiperiphery. For now I wish to look at the implications of
the above for our expectations about semiperipheral behavior.
I have argued in
previous chapters (and will elaborate in chapters that follow) that the
core/periphery hierarchy supports the reproduction of capitalist accumulation
by depolarizing class conflict within core states, and also within those
peripheral states in which the state has come under the control of politicians
who are willing to use anti-imperial rhetoric to smooth over domestic
conflicts. This general argument also
applies to competition within ruling classes.
We expect more national solidarity among differentgroups of capitalists
in the core, and anti-imperial rhetoric may have an integrating effect on
different types of peripheral capitalists within peripheral states.
These harmonizing
effects of the core/periphery hierarchy on intraclass and interclass relations are less likely to
operate in the semiperiphery. The
harmonizing effects work differently on different groups. Among core
capitalists national solidarity is easier to achieve because there are more
opportunities to go around, and thus competition is less intense. For core
workers wages are higher, working conditions are better, and core
capitalists are more likely to make economic and political concessions to
workers precisely because they are less pressed by competitive forces.
These generalizations
apply only grosso modo. Within
individual core countries, depending
upon the heritage of institutional forms left over from past struggles, the
current state of the world-economy, and the future prospects of the national economy, these
effects may vary considerably. The same
holds for the following generalizations about peripheral and semiperipheral
states. It is also important to know the
particular trajectory of an area in trying to understand its political and
economic behavior.
Downwardly mobile countries are likely to be quite different
from upwardly mobile ones even though they might be at the same level in the
core/periphery hierarchy.
Peripheries in which the state is
substantially controlled by core powers or dependent on core-based
transnational corporations experience
heightened levels of competition among contending groups of peripheral
capitalists, and exacerbated class conflicts, though these may be largely invisible most of the time because of
externally-supported repression. These
conditions explain the high levels of political instability and likelihood of
authoritarian regimes, as well as the internal weakness of the peripheral states. When peripheral politicians come to state
power who are willing to employ anti-imperial rhetoric this reduces these
domestic conflicts somewhat, depending on the degree of implementation of
antiimperial policies.
Peripheral states that implement radical anti-imperial policies
reduce the level of domestic class conflicts, but they face the grave peril of
intervention by an offended core power.
In semiperipheral
states the effects of the core/periphery hierarchy are different. Both types of semiperipheral states defined
above may have opportunities for upward
mobility in the core/periphery hierarchy, and this will affect national
politics and state policy. Type one, a
balance of core and peripheral
activities, will experience political conflict over state policy because of the
conflicting regional interests. Type
two, a relatively uniform but
intermediate level of semiperipheral activities, will be much less likely to
experience conflict among different kinds of capitalists. Type two may, however, experience exacerbated
class conflicts because workers in
intermediate industries may be able to organize nationally
because of the similarity of their working conditions, and they are unlikely to
enjoy good wages or social benefits, at least compared to core workers.
Class conflict in Type
one semiperipheries may also break out, especially during periods of state crisis, but there are
likely to be problems of solidarity between the workers in the core sector and
the workers in the peripheral
sector. We can note the similarity here
between Wallerstein's definition of semiperipheral states (Type one above) and
Trotsky's notion of
combined and uneven development, which he employed
masterfully in his history of the Russian Revolution (Trotsky, 1932: chapter
1).3
The most important
thing about semiperipheries is that interesting political movements are more
likely to emerge in them. Movements of
both the right and the left have often found fertile ground in semiperipheral
and second-tier core states (Goldfrank,
1978). The basis of this political
fertility stems from the contradictory location of semiperipheral areas in the larger world-system.
Semiperipherality
produces both challenges to the dominant mode of production from antisystemic
movements and is fertile ground for upwardly mobile countries who achieve
success within the system. All three
hegemonic core powers were formerly semiperipheral areas (ChaseDunn,
1988). One factor which may partly
determine the path taken by semiperipheral areas is the relative degree and
nature of internal stratification. More
stratified semiperipheries are likely to produce social revolutions which
challenge the logic of capitalism, while relatively less stratified and
political liberal semi
peripheries can achieve the degree of class harmony necessary
for upward mobility within the capitalist worldeconomy.
Anti-systemic
movements are also likely to emerge in peripheral areas, but they are less
likely to survive there because resources to resist core inter
vention are meager. The
most important experiments with socialism have emerged in semiperipheral
states, and there is reason to believe that semiperipheral areas will continue
to produce powerful challenges to the capitalist mode of production in the
future.
Zonal Boundaries
If the core/periphery hierarchy is really a set of discrete
zones we should beable to determine the boundaries between the zones and to
unambiguously know the zone in which each country or area is located. Arrighi and Drangel (1986) have discovered a
fairly regular trimodal distribution of countries based on GNP per capita as a
measure of position in the core/periphery hierarchy. Nemeth and Smith (1985) and Smith and White (1986) have used a network analysis of
the level of processing of imports as a measure of the core/periphery
structure, and this reveals a four-tier hierarchy with two distinct
semiperipheries, one characterized as "strong" and one as
"weak." For myself the
vocabulary of zones is simply a shorthand.
I don't see any advantage in spending a lot of time trying to define and
empirically locate the boundaries between zones because I understand the
core/periphery hierarchy as a complex continuum. Since there is upward and downward mobility
in the system there must be cases of countries or areas which are in between
zones, at least temporarily. For me it
doesn't matter whether there are "really" three zones, four zones or
twenty zones.
The vocabulary of
zones is simply a useful metaphor, as is the notion of the semiperiphery. I don't think we need to reify our words to
the extent that we waste time arguing over the exact boundaries of zones. There are many more important problems, such
as the magnitude of inequalities, the operation
alization of the core/periphery hierarchy, and the study of the
relationships among different dimensions of world-system inequality, which
should receive our efforts. For me the
designations of second-tier core states or extreme peripheries are sometimes
useful, but I would not want to try to locate subzones or claim that there are
really four zones instead of three.
Thomas Hall's (1986) work on the continuum of integration into the
world-system similarly disputes the value of a simple dichotomy between those
areas which are outside of, and those that are inside, a world-system.
A Multidimensional Hierarchy
The definition of core and periphery described above focuses on
relative levels of the capital intensity of commodity production. This is an indicator of the economic basis of
national power in a capitalist worldeconomy, but other theorists conceptualize
the core/periphery hierarchy more directly in terms of power relationships
among states. Thus James Petras et al.
(1981) emphasize the military power of imperialist countries their ability to use coercive force to
control the behavior of other countries.
We have discussed various attempts to measure the military power of core
states in chapter 9. Such direct
measures of military power could be
utilized in the study of peripheral and semiperipheral countries. After all, most analysts agree that external
state strength is an important component of the world-system hierarchy. The research which has been done on different
types of economic dependency (reviewed below) demonstrates that these are often
not highly correlated with one another in crossnational comparison. This implies that the core/periphery
hierarchy may be multi-dimensional.
Another dimension has
been studied by Singer and Small (1966; Small and Singer, 1973). They have coded a measure of international
status ordering based on the exchange of diplomats among countries. What is needed is a long run empirical,
crossnational study of the relationships among the different political,
military, and economic types of power/dependence relations to determine how these dimensions interact with
one another. Such a study might not
solve the question of the best way to analytically define the core/
periphery hierarchy, but it would be very helpful in
further research on the causes and effects of a country's position in the
hierarchy.
Measuring World-system Position
Now let us turn to the problem of quantitatively measuring the
position of areas in the core/periphery hierarchy. Students of the worldsystem are by no means
agreed upon what units ought to be compared.
Most empirical studies examine nation-states or colonies, employing
political boundaries to designate the units of analysis. This is problematic because these boundaries
are themselves a creation of the system we are studying and they also have
effects on the processes which we want to study. The reality we want to study is a multilevel,
multidimensional, nested hierarchy composed of individuals, households,
communities, cities, classes, unions, parties, firms, ethnic groups, states,
international regions, zones, and the emergent characteristics of the whole
world-system. However, the goal of
science is to simplify a complex reality in a way which helps explain patterns
and predict outcomes. Thus there is no
point in making a map which is as complicated as the
territory. We can
simplify our analysis as long as we take into account the problems of inference
which may result from our simplifications.
Arrighi and Drangel
(1986) argue that their conceptualization of the core/periphery hierarchy can
be adequately operationalized by using GNP per
capita. This same indicator could
be used for my conceptualization of core production as relatively
capitalintensive production. A better
indicator of my concept would be the
amount of product divided by the number of hours worked, because
capitalintensity is very nearly the same as labor productivity. But a fair proxy for the above is GNP per
capita because, when we consider a large number of countries, there is a high
correlation betweenthe number of hours worked and the size of the
population. A better proxy would be the
ratio of GNP to the size of the active work force.
GNP per capita is also
a relatively feasible measure for studies of world-system position because it
is available for a large number of countries over fairly long time
periods.4 And, since we know that a
number of other indicators are highly correlated with GNP per capita in
cross-national comparison in recent decades, we might use those measures as
proxies for GNP per capita (and thus world-system position) back into the
nineteenth century or perhaps further. I
have in mind the level of urbanization, the proportion of the work force in
agriculture, and energy consumption per capita.
The proportion of the national product in agriculture was used in this
way in table 9.2 in chapter 9.
It is somewhat ironic
that the very measure which is most often used by modernizationists as an
indicator of national "development"
GNP per capita is also arguably
an indicator of world-system position.
And both schools of thought would
have similar reservations.
designates high productivity per labor hour due to the
employment of capital intensive production techniques.
At the other end of
the spectrum there are also problems.
Nonmarketed production is difficult to value. Dudley Seers (1983) tells enough horror stories from his days as a development
economist in
A better measure which
successfully weeds out the extractive enclave booms is based on the idea of the
level of processing. Some fairly
sophisticated measures using information on world trade have been analyzed, as
discussed below. While these are fairly
highly correlated with GNP per capita when we compare all countries, they do
not lead to the erroneous conclusion that
To my knowledge very little has been done to
construct a quantitative measure of world-system position for countries and
colonies in the nineteenth century. For
more recent time periods we have much better data, of course, and recent
research has attempted to operationalize world-system position as well as
specific kinds of dominance/dependence relations. Because GNP per capita has long been thought
of as a measure of economic development, most analyses of cross-national data
have not employed it as a measure of
overall world-system position. Rather
several studies have examined specific types of international dependence. The first cross-national comparative analysis
of penetration of peripheral countries by direct foreign investment was
performed by Albert Szymanski (1971) in his doctoral dissertation at
The above studies, and
many others, operationalized international dependence as variable attributes of
each country, usually derived by
determining the extent of some international connection and weighting it
by some measure of national size. David
Snyder and Edward Kick (1979) applied network analysis to international
relations in order to develop a block model measure of world-system
position. They used a dichotomous form
of cluster analysis to determine the structural similarities among groups of
countries
based on four international interaction matrices. The four
matrices they analyzed were the value of trade between countries; military
interventions; the exchange of
diplomats; and the existence of treaties.
The block-modeling technique used by Snyder and
Kick wasted a lot of information for the metric measures such as trade because
it reduced each cell to a zero or a one.
And the interesting relationships between the economic and political
dimensions were unanalyzed, as the four dimensions were combined to produce a
single structure which grouped countries into categories which are positionally
similar on the combined four dimensions.
Nevertheless this network measure of world-system position has achieved
wide usage in cross-
national research, although later studies have modified it
to correct apparently mistaken
categorizations (e.g. Bollen, 1983).
Another network
measure has been developed by Roger Nemeth and David A. Smith (1985). Using the idea that the level of processing
is the key to the core/periphery hierarchy, Nemeth and Smith combine five trade
matrices containing different commodity
classifications of imports. The five
trade categories were determined by factor analyzing the trade matrices of 53
two-digit commodity classifications. This
produced clusters of commodities by empirically examining their associations in
world trade patterns rather than by trying to decide ahead of time which
commodities embody higher degrees of processing. This structural trade network measure
produces a core, two semiperipheries and a periphery.5 Nemeth and Smith also show that being in the
periphery, and what they call the "weak semiperiphery," is associated
with relatively slower growth rates of GNP per capita, whereas being in the core or the "strong
semiperiphery" is associated with higher levels of growth.
The question of the
effects of different kinds of dependency relations and overall world-system
position will be discussed in the following chapter. Here I want to point out that the measurement
of world-system position has been carried out only for very recent times, and
even for these there remains considerable controversy about how the
core/periphery hierarchy ought to best be operationalized. We still know little about the interactions
among different forms of economic and political dependency, but we do know that
dominance/dependence relations in the world-system are empirically
multidimensional. This is clear from the
fact that many of the measures of different types of dependency are not highly
correlated with one another in crossnational comparisons. Thus a peripheral country may be dependent in
one respect but not in others. Also we
do not have a good understanding of the
interaction among different types of dependency and overall world-system
position.
The Process of Peripheralization
Conceptualizing the spread of the European world-economy to the
whole globe has raised a number of thorny theoretical issues as well as many
difficult empirical ones. Amin (1980a)
argues that we should not assume that theperiphery which resulted from the
incorporation of external arenas into the European-centered capitalist
world-system is a uniform zone, and Eric Wolf (1982) has demonstrated this
clearly. There were a number of
important determinants of the particular local and regional social structures
which were created in the process of peripheralization, most important of which
was the nature of the existing social structure before the Europeans arrived.
The globe was already covered with world-systems
of various kinds when the Europeans began their expansion. The nature of the indigenous social systems,
their abilities to militarily resist, their susceptibilities to the trade goods
(and diseases) offered by the Europeans, and their possession of things valued
by the Europeans, all affected the nature and timing of incorporation. The most developed and powerful civilizations
were able to hold the Europeans at bay longer, despite the strong attraction
which their wealth exercised on the Europeans.
And regions with very little wealth were not early incorporated because
the Europeans were not interested, unless some strategic or transportation
utility existed in a particular location.
Because of the fact that European core powers were competing among
themselves some areas were formally colonized simply to prevent competing
powers from having them, a phenomenon which has been called
"anticipatory colonialism."
The continuing
importance of the pre-conquest social structures is indicated by the recent comparative research
of Gerhard Lenski and Patrick Nolan (1984; also Nolan and Lenski, 1985). They show that countries which had a technological heritage
of horticultural (digging-stick) methods of production in agriculture continue
to show lower levels of development in the 1960s and 1970s than countries which
had a heritage of more technologically advanced agriculture, employing the
plow, animal energy, and irrigation.
They also contrast, within the category of agrarian (more
technologically advanced) societies those which have made the transition from
horticulture to plow agriculture more recently and those which made that
transition long ago. Interestingly, they
show evidence in support of a version of the hypothesis of the "advantages
of backwardness" by demonstrating that the growth rates and levels of
development of the group of countries that have more recently developed plow
agriculture are higher than those agrarian countries which developed it long
ago (Nolan and Lenski, 1985). Although
other factors probably account for some of the differences found by Lenski and
Nolan, their findings support the contention that features of the
pre-integration social structures continue to have relevance for contemporary
development patterns.
Others have pointed
out that the nature of peripheral social structures is also affected by the
timing and particular circumstances of incorporation. It is held by some that the stage of
capitalism which exists in the core at the time of incorporation has important
effects on the kind of colonial structure which is created in the
periphery. While my analysis of the
stages of capitalism as cycles and trends of world-system development (chapter
3) would describe changes in the core somewhat differently, I would not dispute
that the particular organizational forms which existed in a period, as well as
the peculiarities of the conquering power and its situation visvis
competing core powers, have had important consequences for the kind of
peripheral social structures which emerged.
These, as well as the original differences among indigenous social
structures and the particular role performed by indigenous groups have recently
been demonstrated by Hall's (1986) excellent study of the incorporation of the
Another matter which should be kept in mind
when thinking about the periphery as a whole, and variations within it, is the
question of ebb and flow of influence from the core. In chapter 13 we shall discuss waves of
colonial expansion and the export of capital from the core to the
periphery.
It should also be noted that both the nature and intensity of
core influence over peripheral areas vary over time. This is seen in the sequence of peripheral
cash crop booms; the rise and fall of particular kinds of extractive economies;
and the emergence, during periods of low interest by core powers, of more
autonomous social and economic structures in the periphery. Andre Gunder Frank's (1979b) study of the
transformation of agriculture in colonial Mexico (New Spain) from 1521 to 1630
shows how market forces generated by the expansion of silver mining linked
many farmers and livestock raisers into
the commodity chains connected to the European core. When the silver in an area ran out, these
cash crop producers reverted to the hacienda economy, a largely self-sufficient
rural system containing many of the social attributes of feudalism. The argument about whether or not the resulting
peripheral social structure is "capitalist" needs to take account of
the ebb and flow of core penetration of
the periphery.
The assertion that the
core/periphery hierarchy is an important structural feature of the worldsystem
certainly does not imply that the core determines everything that happens in
the periphery or that peripheralized peoples are inert objects of exploitation
and domination. As with class
structures, the core/periphery structure is a dynamic tension of domination and
resistance. This has been demonstrated
by a number of studies, especially by anthropologists, who have sought to
understand the linkages between local and world-system processes and the
methods by which peripheralized peoples resist core domination.6
The above points are
important to keep in mind when we are doing a case study of a particular
situation of peripheralization. I would,
however, contend that despite the results of Lenski and Nolan cited above, it
is likely that the secular strengthening
of world-system level economic and
political processes has resulted in an increasing homogenization
of the periphery a kind of convergence
in which original differences have tended to be reduced. I realize that there seem to be contemporary
forces, such as cultural nationalism and dependent industrialization, which
might further differentiate the
periphery. Nevertheless, the extent of
commodification and the high level reached by the trend toward the
globalization of production by transnational firms increasingly subject
peripheral areas to forces which homogenize their social structures along the
lines of peripheral capitalism. This
also includes the consequences of the efforts of peripheral peoples to resist
core exploitation and domination. While
these resistance efforts may differ from one another in terms of the particular
cultural forms they adopt, their organizational forms are becoming
morphologically similar, and the overall relative homogeneity of social
structures in peripheral areas is increasing.
The Necessity of Imperialism
I have already claimed that the core/periphery hierarchy was and
is a necessary feature of the
world-system which allows the capitalist mode of production to adjust to its
own contradictions. A number of
contrary interpretations have been
presented elsewhere. Some Marxists (e.g.
Albert Szymanski [1981] and David Harvey [1982]), have argued that capitalist
imperialism (the export of capital to peripheral areas, the extraction of raw
materials, and the penetration of new markets) is simply an alternative which exists for core capital, but is not
necessary to the reproduction and expansion of capitalist social
relations. This supposes the
possible existence of a world in which
the core/periphery hierarchy has disappeared and yet capitalism remains the
dominant mode of production.
Others (e.g. Chirot,
1977) have claimed that imperialism was simply a mistake. The European states allegedly conquered and
subdued the periphery, not because it was profitable, but because they were
suffering from a false ideology the belief that colonialism would be
profitable and the contagion produced by the knowledge that other powers were
doing it. It was a kind of
self-fulfilling prophecy. Some activists
in the periphery apparently shared this delusion. We can recall the Irish revolutionary who, in
1916, declared that
My argument, discussed
in greater depth in the following chapter, is
that exploitation of the periphery by the core made an important and
necessary contribution to capitalist accumulation in the core in earlier
centuries and that peripheral exploitation continues to be important because of
its effects on politics within and among core countries. Briefly, exploitation of the periphery, and
the threat of capital flight to the periphery, has acted to prevent labor
unions and socialist parties within core countries from successfully
challenging core capital. Thus the core
areas of the world-system remain dominated by capitalists. Socialist revolutions have happened, not in
the core as Marx predicted, but rather in the semiperiphery and the
periphery. The problems they have
encountered with economic and political development in thecontext of a
world-system still dominated by capitalism have further bolstered the political
and ideological hegemony of capital. The
implication here is that a future decrease in the magnitude of core/periphery
inequalities will reduce the political dampening effects on class politics
within the core, and make transformation of the world-system toward socialism
more likely.
But let us return to
the question of the importance of the core/periphery hierarchy for capitalist
economic development. There are two recent studies published by economic
historians, Patrick O'Brien (1982) and Paul Bairoch (1986), which dispute the
importance of European colonialism for the industrialization of
authors claim that European colonialism before 1800 was not
economically important for the development of industrial production in
international trade as well.
He also claims that the items imported from the periphery were not
important to European economic development, consisting mainly of luxury goods
and bullion. And he argues that the
profits made on investments in peripheral trade and production were not
abnormally high once the risk factor is taken into account, and these profits
did not play a very large role in the capital formation which spurred the
industrial revolution. O'Brien also disputes
the claim that the core/periphery trade was important in the diversification of
core production in
Bairoch (1986) takes
the similar position that colonialism was not an important stimulus to European
economic development before 1800. He
argues that
the European powers did not contain a numerically large
population compared to the population of
The arguments of
Bairoch and O'Brien can be disputed on several grounds. O'Brien uses world-system concepts in order
to frame his argument but both he and Bairoch make several errors in their use
of statistical categories, at least if their evidence is supposed to be
relevant to worldsystem hypotheses.
O'Brien's discussion of the small percentage of international trade
composed of exchanges between Europe and the continents of Asia, Africa and
Latin America, and his characterization of the items exchanged as unimportant
for core development, ignore the sizeable Baltic trade,8 and the role of closer
peripheries such as
Bairoch claims that
"traditional" colonialism, a type which allegedly included European
colonialism before 1800, did not involve significant core exploitation of the periphery, and did not
result in very large differences in the level of living between core and peripheral
areas. And he intentionally ignores
city-states such as
unimportant for core development contradicts the claims of the
Ekholm and Friedman (1982) who find core/periphery exploitation playing a
central role in the uneven development which occurred in the first state-based
world-system, that in Sumer in the third millenium BC. The process of empire-formation
carried out by the Assyrians, the Persians, and the Romans is
thought by almost all historians of antiquity to have involved significant
degrees of core/
periphery exploitation.
Bairoch's claim that the magnitude of
inequalities resulting from "traditional" colonialism was small may
be due to comparing the wrong units.
Lenski and Lenski (1982) claim that the agrarian empires were the most
unequal social systems that have ever existed.
But even if it were true that core/
periphery inequalities were quantitatively smaller in the
ancient world-systems, it does not follow that they were thus unimportant in
the process of the rise and fall of empires.
We know, for example, that urbanization was at a very low level in
"traditional" societies compared to industrial ones, but it was
certainly significant that cities were built at all. Lumping all the"traditional"
societies together in a heap is not something we have come to expect from
historians. The geometric rates of
change and the gross
inequalities we see in the modern world-system should not
prevent us from seeing the importance of quantitatively smaller but nonetheless
significant differences among the precapitalist world-systems.
Bairoch looks only at
the colonies of European states, ignoring most of the non-colonial dependencies
such as
to paint with broad strokes when they approach the core. Bairoch includes Europe as a whole, including
least have limited himself to those European powers which held
overseas colonies. Instead he includes
peripheral
O'Brien points out
that his statistics indicating the small importance of core/periphery trade contain some deviant
cases, small countries like
peripheral powers, such as the Phoenicians and Venice, play such
a role, but the Dutch Republic was the first capitalist state to be hegemonic
in a core/periphery system.
Both Bairoch and
O'Brien seem to believe that the industrial revolution which began in
and economic growth such as the spread and deepening of
commodity production across Europe, and the concentration of competitive core
production in
European states, some of which continued to pursue largely
continental military expansion, while others were following the path of
commodity production for the world market.
Both O'Brien and
Bairoch claim that the tropical foods imported from colonies by the core were
not important for core development. This
might be disputed on two grounds. First,
the provision of lowcost food for mass consumption lowered the cost of labor
within core countries. Lowering the cost of living to workers made it possible
(in theory), for wages to be lowered without killing off the work force. On the other hand, evidence indicates that average wages went up in the
core. Tropical food imports contributed
early on to relatively harmonious class relations within core countries.
As Sidney Mintz's (1985) study of sugar consumption in
Bairoch admits that
the small percentages which peripheral markets added to the demand for core
exports "may have sizeable influence on the profit
ability of a particular industrial sector" (Bairoch,
1986:211). O'Brien
estimates that "commerce with the periphery generated a
flow of funds sufficient, or potentially available, to finance about 15 percent
of gross investment expenditures undertaken during the Industrial
Revolution" in
Many of these facts
could be easily reinterpreted to support the importance of the periphery for
core industrialization. Fifteen percent
of capital formation is not a small share.
And, as argued above,
Reorganizations of the Core/Periphery Hierarchy
The overall core/periphery hierarchy is a socially structured
system of economic and politicalmilitary
inequality. The forms which this hierarchy
has taken have changed while the hierarchy itself has been preserved. The first
forays which began the peripheralization process in many areas involved pillage
and plunder by the European powers.9
Marx described this "primitive
accumulation" thus:
The discovery of
gold and silver in
extirpation,
enslavement and entombment in mines of the
aboriginal population, the beginning of the conquest and looting of the
East Indies, the turning of
This direct use of
coercive force has moved slowly in the direction of institutionalized economic
power based on law and private property, although the element of coercion in
core/periphery relations and within the periphery is still greater than within
the core. Slavery and serfdom have been
largely abolished in the periphery. And
other forms of obviously coercive labor control, such as contract labor,
indentured servitude, debt peonage, village tribute workers (mita), etc.
have declined. Other groups have
increased: "semi-proletarians"
living in village enclaves and working for wages seasonallyor
"temporarily" during their most productive years, rural proletarians, share croppers, and formally
free peasants owning their own small plots but forced to sell their cash crops
to monopolies or state marketing boards (Paige, 1975:13). So has the
"formal sector" of proletarians working in large firms, often
protected in some way by state regulations guaranteeing minimum wages, etc.
But this has been accompanied by a burgeoning informal sector of
unprotected wage earners, often working for small firms in the
"competitive" sector (Portes, 1981).
The categories of free
labor and coerced labor do not fully capture the kinds of labor control which
occur in the core/periphery hierarchy.
Rather core labor is "protected" to a greater extent by state
regulation and welfare institutions, although this seems less evident during
the current period of attack on the wages of workers in both the core and the
periphery. It is nonetheless still the
case that core labor is more protected than peripheral labor. The uneven process of state-formation and
economic development allows core workers access to the strongest states and the
most profitable and capital-intensive sectors in the world-system. The process of proletarianization has
occurred in both the core and the
periphery, with a lag in the periphery.
And with the additional feature that proletarians in the core can gain
not only "free" status but also a certain amount of political
protection visvis capital and visvis competition from other
workers.
It must also be noted
that colonialism, the direct organization of
formal political/military control over peripheral areas by core states,
has largely passed away. This certainly
does not mean that politicalmilitary
forms of power have ceased to operate in the core/periphery
hierarchy. We still see "gunboat
diplomacy," the overt and covert use of force, and the useof military and
economic aid to friendly peripheral regimes which support the core/periphery
hierarchy. But perhaps the weight of
politicalmilitary versus economic-based power has shifted a bit. The growth of the transnational corporations
has shifted the use of core state power toward the support of property rights
in the periphery, as demonstrated by Charles Lipson's (1985) study of the
development of institutions protecting foreign capital.
Several authors have
suggested the importance of major reorganizations in the core/periphery
division of labor. John Walton (1985)
speaks of the third "new international division of labor," and a
similar discussion of stages of dependency is offered by Dale Johnson (1985:22-8). The first international division of labor
corresponded to what others have called "classical dependency" the export of raw materials from the
periphery to the core and the export of manufactured goods from the core to the
periphery. This involved the
"deindustrialization" of those peripheral areas which were producing
products which were in competition with core exports. This first international division of labor
evolved over a long period of time in which various rounds of cash crop booms,
and mining booms, visited peripheral areas unevenly, in cyclical waves, and in
ebb and flow.
The second
"new" international division of labor, according to Walton, began in
the 1930s when some peripheral countries began the process of dependent industrialization. Backed by state policies of import
substitution, national capitalists tried to capture the domestic market for
manufactured goods. Similar attempts had
been made in the nineteenth century by groups in Chile (Zeitlin, 1984) and
Mexico (Hale, 1968) without much success, but the political transformation of
the United States, Germany, and several other hitherto peripheral areas in
Europe, which occurred during the nineteenthcentury, made possible the
formation of new centers of core capitalist industrialization (Chase-Dunn,
1980; Senghaas, 1985). These instances
of upward mobility into the core were not the most typical pattern, however. Most areas continued to be
peripheralized. It is doubtful if many
of the newly industrializing countries (NICs) will be able to accomplish more
than dependent industrialization. Peter
Evans' (1979) analysis of the competition and bargaining among state
bureaucrats, domestic capitalists, and transnational corporations (TNCs) over
shares of the Brazilian national market shows that even in countries where
import substitution has been the most successful the TNCs have succeeded in
controlling many sectors of the domestic market.
John Walton's third
new international division of labor is the globalization of production by the
transnational firms. This has been
revealed in world-wide sourcing and the
production of industrial components in the periphery for export to the core. Volker Bornschier (1976) shows evidence that the bulk of TNC manufacturing investment
in the periphery remains focused on production for the domestic market, but
there has also been an expansion of transnational production in "free
enterprise zones" in the periphery.
This is the "new international division of labor" studied
by Frbel, Heinrichs, and Kreye (1980).
While these reorganizations of the
core/periphery hierarchy have altered some of the organizational forms and
institutional mechanisms operating in the world-system (as discussed in
chapters 3 and 4) the more basic dynamics of the system have not changed. Evidence in support of this view is presented
in the review of formal comparative research contained in the next chapter.
Chapter 11: Reproduction of the Core/Periphery Hierarchy
This chapter examines different arguments about how the
core/periphery hierarchy is reproduced as a structural feature of the
capitalist world-system. It has been
argued above that underdevelopment is more than simply a transitional stage on
the way to core capitalism. Many
students of imperialism have examined the forces which drive capitalism to
expand into the periphery. The forces of
expansion opportunities for cheap raw
materials, lower-cost labor, and expanded market demand tell us why the periphery becomes exploited,
but they do not tell us why the processes of underdevelopment which recreate
the core/periphery hierarchy are sustained.
The real question is: why is the periphery necessary for the
reproduction of core capitalism, or what would happen to core capitalism if the
core/periphery hierarchy were to disappear?
It is unsatisfactory to simply contend that capitalism and the
core/periphery hierarchy are inseparable by
definition. We need to examine
the question of the "necessity of imperialism" directly, and the
following discussion reviews several
theoretical perspectives with the purpose of answering that question.
It should be stated at
the outset that many, even most, of the theories reviewed do not hold to the
proposition that the existence of the periphery is a requisite for the survival
of core capitalism. Rather most
theorists hold that the periphery is simply a convenient way for the core to
ameliorate certain contradictions caused by its own logic, but that these
contradictions have other possible solutions, and so the periphery is
fortuitous (for the core), but not necessary (e.g. Harvey, 1982). I shall review the extensive recent
literature by loosely combining arguments according to their main emphasis, a
method which causes some difficulties because most arguments are, in fact,
multivariate. Nevertheless some order of
presentation is necessary. Another
difficulty is the different timeorientation of the various theories reviewed. Some are focusing on processes thought to be
occurring only since World War II, while others are attempting to formulate a
theory which applies to the core/periphery hierarchy over a much longer
period. The consideration of various
explanations cannot escape arguments found in chapter 3 about stages of
capitalism, and in chapter 10 about reorganizations of the form of the core/
periphery hierarchy. The intent,
however, is to construct an explanation which accounts for the long-run
reproduction of the core/periphery hierarchy.
The consideration of
the driving forces behind expansion to peripheral areas does not, by itself, explain the
consequences of that expansion.
Expansion to some areas, such as the United States, resulted in the
extension of core capitalism, a relatively self-generating process of
capitalist accumulation, while expansion to other areas produced the more usual
dependent peripheral capitalism and reproduced the world-level core/periphery
hierarchy. What are the systematic mechanisms which give "backwash
effects," the cumulative
underdeveloping properties of international inequalities (Myrdal, 1957) greater weight than "spread
effects," the more even development of core type capitalist accumulation
across space?
In addition to a discussion of the various
theoretical mechanisms which have been proposed to reproduce the periphery, I
will review the corpus of cross-national research which has been done on some
of these mechanisms. For various reasons
that research does not provide the final word, but it does shed some light on the larger
question.
A number of mechanisms
have been asserted to be the mainstays of the
core/periphery hierarchy. Neil
Smith (1984) claims that uneven development in the capitalist mode of
production stems from the basic contradiction
between use-value and exchange-value.
Arghiri Emmanuel (1972) propounds a theory of unequal exchange which is
said to account for the extraction of surplus value from the periphery by the
core, and several authors have elaborated upon and criticized his theory. Others (e.g. Petras et al., 1981) have
suggested that it is state power which is the main support of the
core/periphery hierarchy. Thus the
Wallerstein (1983a) argues that differences in the process of
class formation between the core and periphery, and the political consequences
of those differences, account for the reproduction of international
inequalities. Alain de Janvry (1981) contends that the peripheral reliance on
export markets reproduces a disarticulated economic structure in the periphery,
which remains necessarily dependent on the core sector for markets.
It is of course
possible that all the above mechanisms operate together, varying over time in
their relative importance. We have
already noted in chapter 10 that the form of organization taken by the
core/periphery hierarchy varies over time, and to some extent from place to
place. Here I want toreview some of the
arguments and discuss the problem of doing research to determine which are the
most important mechanisms of core/periphery inequality.
The Marxist geographer
Neil Smith (1984) has recently argued that uneven development at the urban,
regional, national, and international levels results from the contradictory
tendencies of differentiation and equalization inherent in the process of
capitalist accumulation. Smith's valiant
effort to ground a theory of uneven development in Marx's accumulation model
does not contain a systematic explanation for the reproduction of the
core/periphery hierarchy, however. He
notes that capital's need for cheap labor and cheap raw materials are
contradictory with the need to expand consumption in the periphery in order to
widen markets, but he does not explain why capital continues to maintain
international inequality rather than using the periphery as a new locus of core-type accumulation (Smith,
1984:141). It is implied in Smith's
approach that there is nothing necessary about the core/periphery hierarchy for
capitalism. As he says, "the
emphasis on accumulation over consumption is just that, however, an emphasis" (Smith, 1984:141). Core capital continues to reproduce
international inequality simply because it is stuck in its old ways. As Smith says, "Empirically, however,
and despite the dramatic industrialization that has taken hold in the 1970s in
select Third World economies, a general and sustained industrialization seems
unlikely. This kind of restructuring is,
so far, blocked by inherited patterns of capital accumulation" (Smith,
1984:158). While I agree with Smith's
prediction, the purely contingent basis he asserts for it is somewhat
disappointing, especially if we are searching for a systematic explanation for
the reproduction of the core/
periphery hierarchy.
Arghiri Emmanuel
(1972) claims that unequal exchange is the mainstay of uneven development. Unequal exchange, as Emmanuel defines it, is
asserted to be a main basis of the extraction of surplus value from the
periphery by the core, and to result in the underdevelopment of the
periphery. I shall summarize Emmanuel's
theory and the other approaches which focus on core/ periphery wage
differentials or differences in the processes of class formation between the
core and the periphery. I shall also
review arguments which focus on:
--the nature of the
core/periphery division of labor between raw material producers and
manufactures (trade composition);
--state-centric
explanations;
--the disarticulation
of peripheral economic structures; and
--the economic and
political activities of transnational firms.
Then I shall present my own synthetic explanation of
core/periphery reproduction. We shall
then discuss recent cross-national research which is relevant to the question
of core/periphery reproduction.
Wage Differentials
Emmanuel's theory of unequal exchange explains how surplus value
is transferred from the periphery to the core by the operation of prices in the
international market. Working from
Marx's theory of the organic composition of capital, Emmanuel distinguishes
between two kinds of unequal exchange:
that which occurs due to differences among sectors in the organic composition
of capital (the ratio of capital to labor in the production process); and that
which occurs due to differences among sectors in the average wages paid for equivalent
kinds of work. Emmanuel defines the
first type as a normal transfer across sectors with differences in the level of
organic composition. It does affect
core/periphery exchange because average core production has a higher organic
composition (more capital per wage hour) than does peripheral production. Amin (1977) and Mandel (1975) contend that
this alone is a significant component of core/ periphery exploitation which is
reproduced by the monopolization of more productive technologies in the core. But for Emmanuel the most important type of
unequal exchange is that which is due to wage differences returning to workers
who perform similar kinds of work. He
notes that core workers receive much higher wages on the average than
peripheral workers even when differences
in productivity due to technology are accounted for. For example, a carpenter in the
Amin summarizes his
estimates for 1976 as follows:
For industry,
while productivity is about half what it is in the center, wages are only one
seventh; in agriculture, while productivity is 10 percent of what it is in the
center, peasants' earnings are one twentieth of the center; and in other
activities productivity of close to one third of the center compares with wages
of one seventh. Since ratios of
productivities are uniformly less unfavorable in the periphery than the ratios
of remuneration, it follows that there is superexploitation of labor in the
periphery which totals as much as $300 billion and which is largely hidden in
the structure of prices. (Amin, 1980b:18-19)
De Janvry and Kramer
(1979) have criticized certain logical difficulties in Emmanuel's formulation
of the theory of unequal exchange. They
contend that the theory only holds when there is complete specialization among
countries in terms of the goods produced for export, and that
initial wage disparities will tend to disappear if capital and labor are
mobile. Bill Gibson (1980) shows that
Emmanuel's theory can be generalized to account fortrade in which specialization
of exports is not complete. And he
analyzes a 67sector matrix of world trade which provides evidence that there is
indeed a large transfer of value from the periphery to the core as a result of
wage differences beyond productivity differences.
There is substantial
agreement about the importance of wage differentials between core and periphery
in the exploitation which occurs at the world level. But disagreement is wide on the question of
the origins and processes which reproduce this wage differential. Emmanuel argues that wages are an exogenous
variable. Original differences exist
because nations have different "historical and moral" heritages, and
these differences are reproduced because labor is relatively constrained from
migrating across national boundaries. If
really free mobility of labor were allowed migration would eventually eliminate
core/periphery wage differentials.
Emmanuel suggests that, for various reasons, capital is more mobile than
labor, but his theory does not include a systematic analysis of the processes
which reproduce the core/periphery wage differential. Other theorists do address this question but
their explanations differ. We shall
review these below.
In addition to the
causes, there is some dispute about the consequences of unequal exchange. While it is widely agreed that it accounts
for a substantial transfer of value to the core from the periphery, the
estimates of the actual amount vary
considerably (see Gibson, 1980). And the
consequences for the relative rates of growth in peripheral
countries is also in dispute. Many
neo-classical economists claim that exports to the core should have positive
growth effects (export-led growth), while Emmanuel and others claim that specialization
in low-wage exports has a retardant effect on growth. Evidence on this is discussed in the next
section. Another explanation of
underdevelopment which focuses on wage differentials between the core and the
periphery is that formulated by Raul Prebisch (1949). His argument focuses on returns to rises in
productivity. He claims that core
workers are institutionally able to tie their wages to increases in
productivity because they have strong trade unions and important access to
core-state power, while in the periphery increases in productivity lead to
lower prices for peripheral products and increases in unemployment (and lower
wages) as workers are made redundant by more productive technology. Peripheral capitalists do not have the market
power to prevent price decreases, as core capitalists often do, and peripheral
workers do not have the institutional power to tie increases in productivity to
wage increases. Thus the core benefits
from cheaper peripheral imports, while the periphery is hurt by more expensive
core products, and peripheral workers experience continuing low wages and high
unemployment. While Prebisch's theory
focuses on wage differences, it can be
seen that the key variable is differential access to state power and strong
unions. Theories which explicitly focus
on that dimension are discussed below.
Class-Formation
Other theorists focus on the differential processes of class
formation between core and periphery.
Immanuel Wallerstein (1983a) sees the reproduction of non-wage forms of
exploitation in the periphery as the key to the reproduction of the
core/periphery hierarchy. Others focus
on the interaction between core domination and the economic status of women in
the periphery. Claudia von Wehrlof
(1984) contends that the most fundamental dimension of oppression and
exploitation in the capitalist world-economy (both core and periphery) is that
which exploits the unpaid labor of women and peasant
"semi-proletarians."
Thedefinitional blindness which fails to calculate the labor
contribution expended within households and within the peripheral non-monetized
sectors to the reproduction of the world work force is seen as fundamental to
the nature of the modern system, rather than as a vestigial backwater of a
former mode of production. Wallerstein
(1983a) argues that the systematic reproduction of non-wage forms of labor is
the main mechanism which prevents the equalization of core capitalist
accumulation. Other authors (e.g. Ward,
1984), working from the "women and development" perspective, contend
that it is the combination of Western health technology and the kind of
dependent economic development occurring in the periphery which stunts the
demographic transition and worsens the economic status of women. Research on these topics is discussed below.
Trade Composition
Several authors have argued that it is the particular form of
the international division of labor which accounts for the reproduction of the
core/periphery hierarchy. Johan Galtung
(1971) claims that it is the "level of processing" of exports relative
to imports which differentially effects rates of national economic growth. The classical form of the core/periphery
division of labor was the export of raw materials from the periphery to the
core and the export of manufactured goods from the core to the periphery.
Galtung and
others (e.g. Hirschman, 1980) argued that the spinoffs and
multipliers associated with diversified manufacturing are greater than those
associated with raw material production, and thus the core will experience
faster economic growth than the periphery, and the gap will be reproduced.
Stephen Bunker (1984)
characterizes extractive economies as the main
element reproducing the core/periphery hierarchy. Resource depletion and changing demand
account for the costly boom and bust cycles of peripheral extractive
economies. Because material resources
are located randomly in relation to urban agglomerations, the restriction of
spinoffs and multipliers in raw materials extraction is exacerbated by the high
costs of relocation of social and economic infrastructure as resources are
depleted or international demand changes.
Use values are lost to the region both through exports of the resources
and through the disruption of the eco-systems from which they are
extracted. Unequal exchange of labor is
accompanied by the unequal exchange of matter and energy. The dominant classes engendered by such
economies tend to invest available capital in infrastructure and organization
for transport and exchange rather than in industry. Political institutions are largely adapted to
control access to natural resources.
Neither are geared to the protection or rational exploitation of the
resources on which the economy is based.
The extractive region is ecologically impoverished by a process which
does not generate alternative productive economies. The core's demand for raw materials, and the
peripheral social organization which extracts them, combine to reproduce the
core/periphery hierarchy.
Another type of
trade-based mechanism which is argued to support the core/periphery hierarchy is the ability of
core areas to maintain themselves as leaders in the development of new
productive technology. This gives core
producers the edge in international competition with non-core
producers even when there is industrialization in the periphery. The product-cycle theory elaborated by
Raymond Vernon (1966) describes the concentration of research and development
in core areas, where new products are introduced. The technological monopolies which enable the
developers to sell new products at a high price (technological rent) are
eventually diminished as other producers introduce comparable competing
products. Price competition eventually
moves the production of the product toward the periphery, where labor costs are cheaper. Thus it is the greater returns to firms which
can develop and introduce new products, and the process by which these products
diffuse through the system which reproduce international inequalities. The division of labor between "monopoly sector"
research and development and the "competitive sector" production of
standard products becomes the key differential within the core/periphery
division of labor.
Disarticulation of Peripheral Economic Structures
Osvaldo Sunkel (1973) was perhaps the first from the dependency
school to focus on the disintegration and disarticulation of national economic
structures which is characteristic of peripheral capitalism. Criticizing the modernization school, which
focused on the "bottleneck" between modern and
traditional sectors, Sunkel argued that the extreme inequalities
and disarticulation of peripheral economies are functionally related to the
reproduction of international inequalities.
Sunkel and Andre Gunder Frank (1969) contended that the important social,
economic, and political linkages between the "modern" and
"traditional" sectors, rather than their separateness, were
responsible for reproducing the traditional sector and slowing the development
of peripheral countries.
Alain de Janvry (1981)
has reformulated this early emphasis on disarticulation into a well-specified
theory of the features of peripheral and core capitalism. Core capitalism is a self-reproducing process
of capitalist accumulation which is subject to certain contradictions, in particular,
the tendency to produce more commodities than can be sold given a certain level
of effective demand. The tendency toward
market saturation can be temporarily resolved by the development of new
products, the raising of wages and consumption levels, and/or the expansion to
new markets in the periphery. Core
capitalism therefore contains within it a tendency to resolve economic crises
by expansion of consumer demand and/or state expenditures, although these forms
of adjustment are limited by the competition among core states for shares of
the world market. Peripheral capitalism,
on the other hand, is primarily dependent on the production of products not
sold in the domestic market. Thus there
is no dynamic which encourages the expansion of domestic consumption. Rather the reproduction of low levels of
remuneration for most workers is sustained by a political coalition between
dependent peripheral capitalists and the transnational core firms and core
states who benefit from the relatively cheap exports. This encourages the reproduction of forms of
semi-proletarianized labor receiving extremely low wages, often below the level
needed to reproduce the worker over a lifetime.
These low wages are often subsidized by non-market forms of subsistence
production or an urban informal sector which operates at a low price level due
to the incorporation of subsistence
activities and unpaid family, or mutual aid, forms of labor.
De Janvry discusses
the relationship between his model of core capitalism and peripheral capitalism
in terms of features of necessity and possibility. For peripheral capitalism,
core capitalism is both necessary as a source of market demand and needed
imports, and it provides certain opportunities for profitable business
activities. For the core, on the other
hand, the existence of the periphery is not necessary. Rather its existence only provides certain
opportunities, but core capitalism could proceed to reproduce itself in the
absence of a sector of the world economy in which peripheral capitalism is
operating. This is because the
opportunities for low wage production, cheap raw materials and food, and the
expansion of markets are desirable but not essential. There are also possibilities to resolve these problems
within the core economy itself. Thus,
for de Janvry, "Peripheral capitalism and the associated pattern of
primitive accumulation is not a sui generis reality with its own
distinct laws of motion, but is only a phase however historically exceptional and prolonged
in the development of capitalism in particular areas of the world
system" (De Janvry, 1981:22 emphasis added).
De Janvry outlines
versions of his model of disarticulated peripheral capitalism for both the export-enclave
economy and import-substitution industrializing economies. In the second, certain circuits have been
internalized and there is a growing domestic market, but the process of import
substitution leads to new forms of dependence on capital goods imports from the
core, and the disarticulated traditional sector remains. This extension was earlier outlined by Sunkel
(1973) but de Janvry describes it dynamically as a struggle between comprador
capitalists allied with the core and national
capitalists who politically support the development of a home
market. This struggle is alleged to
account for the periodic oscillations
between populist and authoritarian regimes in semiperipheral countries.
Transnational Corporate Exploitation
Several explanations focus on the institutions of core-based
transnational private banks and production firms (TNCs) as the key
organizations reproducing the core/periphery hierarchy. Most of these focus on recent decades. It is argued that dependence on foreign
(core-based) equity capital and loans retards economic development in the
periphery. Core corporations
decapitalize peripheral countries by repatriating much more than they invest,
and they act politically to reinforce peripheral regimes which support the
pattern of uneven economic development which produces great intranational
inequalities in the periphery. The
classical form of exploitation was the production of extracted raw materials
and agricultural goods for export to the core.
It is obvious that such producers have no interest in raising the income
of the masses in the periphery because those masses are not a significant
source of demand for their products.
Though many transnational firms have begun producing manufactured goods for
the domestic market in peripheral and semiperipheral countries, these firms
still support regimes which perpetuate inequalities between the poor masses and
the well-off ruling groups and their relatively small coterie of middle class
subalterns. Contrary to the theories of
those who suppose that the growth of transnational manufacturing in the
periphery gives the foreign firms an incentive to support policies which would
increase the incomes of the masses (because they now are selling within the
domestic market), Bornschier and Chase-Dunn (1985:chapter 2) have argued that
the continuing near subsistence level of incomes of the great majority of
peripheral peoples does not present an opportunity for expanded markets for
core-based transnationals. These firms
continue to provide resources to support existing inequalities because most of
their products are either capital goods consumed by the state sector or other
large firms or else are relatively expensive goods (e.g. consumer durables) for
which the masses do not, and cannot in the foreseeable future, constitute a
source of effective demand. A quantity
of TNC resources concentrated on
maintaining the current pattern of uneven development receives a greater return
than if those resources were used to alter the
regime in favor of a more even distribution of income. Increasing the income of near-subsistence
households does not create demand for expensive goods, but rather leads to
greater consumption of food and other basics.
Thus, even though transnational corporations are producing for the
domestic market, they act politically to sustain a disarticulated peripheral
economic structure which leaves a large proportion of the population out of the
process of development.
State-centric Explanations
Several theorists claim that it is the imperial structure of
political/military power among states which reproduces the core/periphery
hierarchy. This argument is made
explicitly by James Petras et al. (1981).
Underdevelopment in the periphery and the economic exploitation of the
periphery by the core are supported by the political/military power of core
states. During the period of formal
colonialism core states exercised direct control of peripheral areas. In the contemporary world-system it is the
A more complicated
version of the state-centric approach is outlined in Chase-Dunn and Rubinson (1977). Here the process of unequal exchange is
theorized as linked to differential forms of state-building, power-block formation and class struggle which occur in
core and periphery. This perspective
combines the state-centric approach with elements of other explanations. It poses a question which is not asked by the
imperial power theorists described above: what reproduces politicalmilitary
inequalities among states? We know that
some states change their positions in the world politicalmilitary
hierarchy.
Richard Rubinson
(1978) also examines the processes which have made it possible for some states
to move up in the core/periphery hierarchy in his analysis of the political
transformations which occurred in
Power-block Formation
A power-block is a coalition of classes and subgroups of classes
which supports a state. Power-block formation is the process by which
the interests of a political economic coalition are institutionalized within
the state apparatus. Political means are
necessary to secure stable economic advantages, and thus groups of capitalists
constantly attempt to convert their economic power into political power. Consequently, a struggle over state power
develops among subgroups of the capitalist class within particular states. Historically, in those areas where peripheral
capitalists producing primary products for export controlled the state, this
type of production became prevalent and the position of indigenous
manufacturing and merchant classes tended to decline. In those areas where manufacturing and
commercial capitalists dominated the state, those economic interests tended to
be sustained. This is because the
possession of political authority is an important mechanism which reproduces
the international division of labor.
Once a set of economic interests becomes dominant within a state
apparatus those types of production become relatively
secure and resistant to challenges from capitalists with contradictory needs visavis state
economic policy. The political victory
of core producers in core states is complementary with and acts to sustain the
political victory of peripheral capitalists in peripheral areas. Consequently, the economic division of labor
becomes structured into a geopolitical division of labor.
During the era of the
"classical" division of labor between the core and the periphery,
based on manufacturing industries and raw material production, the process of
power-block formation reinforced this division of labor. In the core an alliance of commercial,
financial, and industrial capital favored relatively flexible state policies
which allowed for the continuous development of core capitalism. Powerful and flexible coalitions supported a
strong state, and conflicts over state economic policy were settled relatively
amicably. In the periphery the power
block was usually composed of an alliance among raw material producers who were
dependent on core markets. New producers
who would challenge the market shares of existing core capitalists tended to be
excluded from state power in the periphery, although challenges to this happy
arrangement between core and peripheral capitalists occurred often. In most peripheral countries the challengers
lost and the state continued to be in the hands of peripheral capitalists tied to the current core/periphery division
of labor.
This model can be
adapted to the more recent period of "dependent
industrialization." In many countries
the above characterization is still apt, but in others the level of
importsubstituting industrialization has risen such that we need to take it
into account in the negotiations among local capitalists, transnational
capitalists, and the peripheral and semiperipheral state managers. The work of Peter Evans (1986) makes it clear
that the state managers are themselves often important actors, sometimes
creating constituencies in the private sector to support programs which would
develop coretype production. The case of
semiperipheral states is considered more generally below.
State Formation
State formation is the process by which states increase or
decrease their strength, both in relation to their own populations and in
relation to foreign actors. The processes of power-block formation and
state formation are closely related.
Core areas have strong states, states in which political authority is quite extensive and
stable. Peripheral areas have relatively
weak states, in which political authority is less extensive and stable
(remember chapter 6). A strong state has
the ability to mobilize great resources whenever it needs to. State strength, in this sense, is partly a
function of the extent to which the support of a country's dominant classes is
institutionalized within the machinery of the state. The relationship between power-block
formation and state formation arises from the political demands and
requirements of the different kinds of economic interests institutionalized
within states.
Before explaining why different types of power blocks produce
different types of state formation, it is necessary to recall the importance of
state action in the competitive
capitalist world-economy as discussed in part II. Strong states can be effective mechanisms for
protecting economic actors from the
risks and uncertainties which inhere in competitive commodity production in the
world market. States as
violence-controlling enterprises can
provide important protection rents to their capitalists, and under some
conditions states can effectively intervene to support new types of production which can maintain or increase the
share of world surplus value which returns to domestic capitalists
(Rueschemeyer and Evans, 1985).
Because of the
importance of states in the process of capitalist accumulation, all capitalists
desire to utilize state power for regulating
the market to their own advantage.
Thus, dominant groups of capitalists attempt to institutionalize their
interests within the state. But the type
of state formation is dependent on the nature of the power block. The industrial
commercialfinancial block in core states produces strong states,
while the export-oriented block in peripheral states produces weaker states. This difference
derives from the different political requirements and demands of these types of
production.
The effects of the
nature of the power block on the degree of state formation can be separated into external and
internal causes. There are two external
mechanisms. First, where the interests
in a state are composed primarily of industrial and commercial capital, great
demands are put on the state to create an aggressive foreign policy. This is because such economic production
requires access to foreign markets both for raw materials and for the selling
of both capital and consumption goods.
Capitalists engaged in this type of production will be competing for
access to such markets. One of the most
effective means to compete is to employ an aggressive foreign commercial and
military policy. Thus, we can expect
that the demands for such policy will lead to an increase in the authority and
strength of the state, as commercial and industrial capitalists support
increasing its strength in order to pursue such policies. Countries in which the dominant capitalist
groups are producing primary products for export will experience many fewer
demands on foreign policy. Peripheral
capitalists producing raw material exports are unlikely to support an extensive
foreign policy because their gains are not usually affected by an aggressive
state economic policy. It is almost
impossible to affect the demand for such primary goods by state action.3 Thus, there will be fewer vectors of economic
interest pushing the state toward an aggressive foreign policy, and
consequently the authority and strength of the state will be less.
The second external
consideration affecting the relative strength of states arises from strong
states actively attempting to weaken peripheral
states. Core states employ this
process because weakening peripheral states is one way to further monopolize
markets and to insure a steady supply of raw materials.
Among the internal
reasons, strong states arise in the core because extensive political regulation
is needed to foster and protect core industrial and commercial activities. For example, a highly sophisticated trade and
commercial policy is required both to protect home industries and to provide
the infrastructure needed for industrial production. Since production in the core is much more
extensive and diversified, there will be many more political demands placed on
the state, and consequently a greater degree of state formation.
Once strong states
develop, they become a central mechanism for reproducing the core/periphery
division of labor. As the world market
changes, and new areas of profit are created, a strong state, which has
incorporated a large variety of interests into itself, is in a position both to
feel the pressures for shifting political advantage to some rising new area of
profit and to have the ability
because of its greater
authority to effect such a shift of
political advantage to new types of production.
Political coalitions
between classes and state formation may take on distinctive forms in the
semiperiphery. Because of the mix of
core and peripheral activities in some
semiperipheral states, different kinds of capitalists tend to have very
opposing interests. Some have alliances
with core powers based on their control
of peripheral activities, while others favor more independent policies which
would expand core-type activities. Thus
it is often the case that the state apparatus itself becomes the dominant
element in forming the power block and is able to shape the political coalitions among economic
groups. In semiperipheral countries with
potential for upward mobility, state mobilization of development has often been
an important feature. Those upwardly mobile
countries that rely on alliances with core powers tend to develop rightist
military regimes (e.g.
Peripheral countries
tend to have high levels of political instability and either right-wing regimes backed by core
powers or left-wing anticore regimes.
But in the periphery the opportunities for real upward mobility in the system (the expansion of core activities)
are much more limited, and so
the class forces backing coretype development tend to be
weak. Anti-imperialist movements may
take state power and try to mobilize for development (e.g.
Class Struggle
The fourth mechanism by which the core/periphery division of
labor is reproduced is the operation of
class conflict on a world scale. Classes
are conventionally understood in terms of their operation within national
societies, and class struggle is therefore seen as taking place primarily
within countries. Objective economic
classes cut across national boundaries
to form a structure that can only be understood in terms of the world-system. Amin (1980b) develops this type of analysis
in his discussion of the world proletariat and the world bourgeoisie.
Since the major political organizations in the
world economy are nation states, class struggles tend to be oriented toward
these state structures. Hence there is
contradiction between the economic basis of class formation and the political
basis of class struggle. Class struggle
takes the form of
competition for the control of particular state structures, and
this fragments objective classes and stabilizes the larger system. Proletarian inter
nationalism has not yet been an effective unifying force even
between core workers of different states, let alone between core and peripheral
workers. The interstate system tends to
confine class struggle within nation-states.
This has the effect of reproducing the core/periphery division of labor
by producing class alliances that politically stabilize the global mode of
production.
Exploitation takes
place along two main dimensions: between capital and labor and between core and
periphery. The exploitation of the periphery
by the core helps core capital to coopt
core labor into a national alliance.
Similarly, opposition to core exploitation sometimes produces alliances
between domestic capital and labor in
the periphery. In many semiperipheral
areas, on the other hand, interclass alliances are more difficult because the
core/periphery hierarchy presents contradictory alternatives. There are real simultaneous possibilities in
some semiperipheral countries for either an alliance with core powers or a
mobilization for auto-centric development.
The state is the key organization
and its control is hotly contested between groups with widely opposed
interests. This results either in
authoritarian regimes that strongly suppress class struggle or in class
struggles that result in leftist
movements taking state power. The
contradictions between capital and labor are muted in the core and sometimes in
the periphery by the operationof the core/periphery contradiction, but in the
semiperiphery the class struggle may be exacerbated. This is part of the explanation for the
emergence of social democratic class alliances and business unionism in the
core. It also partly explains why
socialist movements based on the power of workers and peasants have first come
to state power in the semiperiphery (
The coming to power of
socialist movements in the periphery does not contradict the above. These have most often been anti-imperialist
struggles that have taken state power on the basis of anticore class alliances. Of course, nationalist class alliances have
been important in the semiperiphery also, but the intensity of domestic class
struggle involved in the creation of
socialist regimes has been greater in the semiperiphery than in the
periphery. The point here is not to
completely explain the class basis of all states but to point out that one
consequence of core/periphery exploitation is to stabilize interclass alliances
in the core and in the periphery. This
reinforces the interstate system and helps reproduce the core/periphery
division of labor. The political
function of the semiperiphery, according to Wallerstein, is to stabilize the
system by concentrating deviant political forms in an intermediate position. This also tends to depolarize and stabilize the
core/periphery dimension of exploitation.
A Political Theory of the Necessity of Imperialism
Beyond the simpler question of the mechanisms which reproduce
the core/periphery hierarchy in the short run is the matter of the necessity of
the existence of a peripheral sector for the reproduction of capitalist social
relations in the core. We have already
mentioned that many Marxist scholars (e.g. Szymanski, 1981; de Janvry, 1981;
Harvey, 1982) contend that theperiphery, and peripheral capitalism, is simply a
phase on the road to the extension of core capitalism (fully developed
capitalism) throughout the globe.
Peripheral areas are alleged to present opportunities for profit-making,
but are not seen as necessary for core capitalism. This is the case, it is said, because though
peripheral areas allow the process of
capitalist accumulation in the core to adjust to its own contradictions,
these contradictions can also be periodically resolved within core capitalism
itself. Thus, hypothetically, the whole
world could become the realm of core capitalism without creating a systemic
crisis for capitalism. The great
inequalities among core and peripheral countries could become reduced, and the
same form of accumulation could operate in all areas.
My own position is
that the core/periphery hierarchy is necessary for capitalism because of the political effects
which exploitation of the periphery has in the core. These have already been alleged in the above
discussion of power blockformation, state formation and class formation. It is my contention that capitalism as a mode
of production must be understood in more than economic terms, and that
political processes are more than simply conjunctural and historical. Capitalism as an historical system creates
opposition to itself, and this opposition must be adjusted to and
accommodated. This occurs in the form of
slow adjustments, but also through economic and political "crises"
which restructure certain relationships, but which allow the logic of
capitalism to be reproduced. Thus the
rise and fall of core powers, and the uneven development (which occurs both
within the core, between the core and periphery, and within the periphery)
operate as an adjustment process which drives capitalism to expand to an even
greater spatial scale. The anti-systemic
movements which try to change the logic of the mode of production are
outflanked and reincorporated, and often end up providing new ways for
capitalism to adjust to its own contradictions.
The role of the
periphery and semiperiphery in this process of expansion, intensification,
crisis, and struggle is two-fold. The
exploitation of the periphery by the core provides an extra measure of surplus
value which can be used by core capitalists as a source of new capital formation,
or as a reward for core workers, or as a resource for sustaining powerful core
states. It can also help resolve
potential conflicts among different groups of core capitalists. The availability of cheap raw materials and,
especially, cheap food stuffs in the core is a benefit to core workers as well
as core capitalists. The hierarchical
division of labor which concentrates cleaner, more skilled jobs in core
countries creates a class structure there which is more amenable to nation-building and a strong state
apparatus. In a number of different ways
the class struggle within core countries is made less antagonistic because of the existence of a world periphery. This operates directly by allowing for higher
wages in the core, and indirectly by supporting nationalist ideology which
compares (favorably) the core nation to peripheral areas, although these
factors operate differently in different core countries (see Lipset, 1977,
1981).
Uneven development
occurs within the core as well as between the core and the periphery, and the class structures
within core countries are not homogeneous.
My theory does not suppose that all core workers have benefited from
imperialism, nor that class relations within the core have always been
pacific. The structure of the working
class within core countries is only partly composed of the labor aristocracy of
primary sector workers. These
havesuccessfully organized business unions in which wages are tied to
productivity in the context of an agreement for industrial peace. But this regime has been the outcome of a
long struggle in which some militant workers sought to challenge the logic of
capitalism and to build the logic of
socialism. The radicals lost those
battles to others who sought a less antagonistic relationship with core capital
in large part because of the material incentives which core capitalism could
provide to the labor aristocracy of primary sector workers.
On the other hand,
there is still a large proportion of workers in some core countries (especially
the
Uneven development
within the core is also evident with the long business cycle and the changing
comparative advantage in commodity production which occurs among core
states. It is not the case that the core
is a uniform region, or that the political structures which shape the
distribution of returns are unchanging.
High wages, and the "social wage" which is built into the
welfare state, are not invulnerable, as is obvious in the current period of
economic relocation. When competition
and high wages and taxes create a profit squeeze on core capital, the regime of
business unions and welfare nets isattacked by austerity policies, capital
disinvestment, and capital flight this
last often to the periphery or the
semiperiphery (Walton, 1981; Ross and Trachte, forthcoming). Thus exploitation of the periphery undermines
socialist challenges to capitalism in the core and allows core capital to limit
the wage bill in the core by threatening job blackmail, or by actually moving
production to the low wage periphery.
Several authors (Amin,
1980a; Braudel, 1984:614) have argued that the positive relationship between
wages and increases in productivity in the core only began operating in the
middle of the nineteenth century. If
this is true it might be argued that the political explanation for the
necessity of the periphery is not applicable before then. The evidence for this claim is based on a
study of the wages of English building craftsmen and laborers from the
thirteenth century to the twentieth (Phelps-Brown and Hopkins, 1955). Braudel shows that the wages of a stone mason
were inversely related to rises and declines in the level of consumer
prices until the mid-nineteenth century, at which time the relationship became
positive (see Braudel, 1984:616, figure 58).
Amin has contended that working class wages in the core were at a subsistence
level until well after the industrial revolution in the core, and he argues
that the core/periphery relationship changed as a result of the introduction
within the core of a regime which tied the wages of certain workers to
increases in productivity. Amin contends
that the political explanation of the function of the periphery for the
reproduction of core capitalism outlined above applies only to the period after
1850.
I would like to
suggest a different interpretation which extends the political mechanism further back in
time. First, the political explanation
does not focus only on the wages of workers in the core. The argument also extends to the shape of the
overall class structure within core states, and to relationships among
different subgroups of the capitalist class within core countries. Thus, even though stone masons, and perhaps
broad sections of the working class, were receiving subsistence incomes up
until the middle of the nineteenth
century, the shape of the class structure in core areas was less hierarchical
than in peripheral areas, and the quality of the relationship among different
types of capitalists was different. The
size of the middle class, the opportunity for larger amounts of skilled labor,
and the ability of relatively successful capitalists to ally with one another in support of a (therefore) strong
state, are conditions which predate the extension of a greater than subsistence
wages to a sizeable portion of the core working class.
The argument that
colonialism and exploitation of the periphery helps resolve conflicts among
different groups of capitalists within core states is supported by several different observations
which have been made about the connection between imperialism and core
politics. Volker Bornschier (1988)
contends that colonialism "fostered non-revolutionary political change
favorable to industrialization" by dampening the conflict between
older sections of the core ruling class
(whose incomes had been partly based on state-guaranteed privileges) and the
newly arising industrial bourgeoisie. It
has been observed by others (e.g. Schumpeter, 1955) that much of the British
colonial bureaucracy was staffed by the second sons of landed aristocrats.
But Cain and Hopkins (1986) show that the dominant party in the
British imperial bureaucracy was composed of agricultural capitalists before
1850 (so-called "gentlemanly capitalism") and that, when the old
colonial system was replaced by free trade, it was the commercial and financial
barons of the Cityof London who took over imperial policy, not the
manufacturers. In any case it seems
plausible that the availability of resources derived from imperial exploitation
can help smooth over conflicts between emerging sectors of capital and those
groups which experience a decline in their fortunes as a result of the dynamics
of uneven development within the core.
We do not have data on
the relative distribution of incomes in early
core and peripheral areas which might be used to detect the pattern now
familiar a diamond-shaped distribution
of income in the core versus a
pyramid-shaped distribution in the periphery. But Braudel does provide some documentary
indications that wages were often higher in core states. Of fifteen century
But the
peacefulness of the Venetian social scene is
nonetheless astonishing. It is
true that even the humblest toilers
fortunate enough to inhabit the heart of a world-economy might pick up scraps
from the capitalists' table. Was this
one of the reasons for the lack of trouble?
Wages in
It may be that a
version of Braudel's imagery of a layered world-system (capitalism at the top,
eventually extending itself downward), is useful here. Perhaps the most important effects of core
exploitation of the periphery were at first on relations among capitalists
themselves, creating unity and support for a strong state, and encouraging
state policies in the service of successful capitalist (as opposed to
tributary) accumulation. Later, as the
world-system became centered on core nation-states rather than core citystates,
the incorporation extended to a middle class of sub-alterns, and even later it
came to include core workers as well.
Braudel's discussion of the Dutch hegemony implies that both rural and
urban workers received relatively high incomes.
He quotes Pieter de la Court (1662) as saying:
Our peasants are
obliged to pay such high wages to their workers and farmhands that (the latter)
carry off a large share of the profits and live more comfortably than their
masters; the same inconvenience is experienced in the towns between artisans
and their servants, who are more insupportable and less obliging than anywhere
else in the world. (Braudel,
1984:179-180)
Another problem with
the extension of the political explanation for the necessity of the periphery
back in time is the lack of any socialist challenge to capitalism before the
existence of modern working class trade unions and parties. Eric Hobsbawm (1959) has contrasted the
modern organizations of the working class with the forms of action taken by
what he calls "primitive rebels."
It must be agreed that socialism was not a viable alternative to
capitalism before the middle of the nineteenth century for many reasons. Even if the Chartists could have succeeded in
establishing a democratic form of collective rationality in
It has been argued in
Part II that the capitalist mode of production is dependent on the existence of
a competitive interstate system in which the core remains politically
multicentric. The formation of a
core-wide world state would be likely to reduce the operation of those
capitalist institutions which limit and disorganize the establishment of political
control over investment decisions. Of
course, political control over investment decisions can be of several
kinds. Many have pointed out the
structural similarity between a tributary world-empire and a hypothetical
socialist world government. Both
establish political regulations over major investment decisions in the world
economy, although presumably one would operate at the behest of a democratic
majority of the world's citizens, while the other would be the instrument of a
new ruling class based on some modern form of the tributary mode of
production. In practice these two
situations are somewhat difficult to distinguish (because, as Marx and Engels
said, every new class represents its own interests as universal interests), but
that is not the problem I am addressing here.
Rather I am arguing
that the existence of a peripheral sector of the world-economy is a necessary
sustaining structure for the reproduction of
the interstate system, and thus capitalism in the core. This is the case because of the effect of
core/periphery exploitation on the nationalism of core workers and core
capitalists, and because the competition over exploitation of the periphery
intensifies inter-imperialist rivalries among core states. This last effect has certainly increased, as
Lenin (1965) observed, since the whole globe has been brought within the
capitalist world-economy because of the necessity of redividing territories
which have already been claimed by one or another core state. But this interstate competition over
peripheral areas and external arenas was already strong before the capitalist
world-economy had become completely global.
The attempts to
transform the interstate system into a core-wide world-empire were partly
motivated by competition for raw material extraction from peripheral areas, and the failure of these
attempts resulted in part from the superior access to peripheral raw materials
held by the hegemonic core statesand their allies, which enabled them to stave
off the attempts at world imperium.
Thus the existence of
a core/periphery hierarchy has allowed capitalism in the core to reproduce and
expand by undermining political challenges to capitalism within core states and
by reinforcing the multicentric structure of the interstate system, which is
itself a necessary structural basis of capitalism.
Crossnational Studies of the Effects of Dependence
While the above arguments are not all formulated at the same
level of generality, they all do have
implications for the way in which the core/periphery hierarchy is
reproduced. Most authors do not ask the
question in terms of the maintenance of a structural feature of the
world-system, but rather they argue in terms of the consequences for national
development. Here we must mention a
point to be discussed in more detail in chapter 15 on methods evidence which shows that something which
affects national development is not directly evidence about the world-system
level. This is the aggregation
problem the ecological fallacy in
reverse. It is possible that a certain
mechanism causes the underdevelopment of some national economies, but not the
reproduction of international inequality.
This could occur if the mechanism
under consideration slowed the development of some countries relative to others,
but nevertheless peripheral countries were, on the average, catching up with
the core. Thus, though we will review
below the crossnational research which examines the effects of various
mechanisms on national development, no firm conclusions about the reproduction
of the core/periphery hierarchy are possible from these results. We shall consider studies which examine
recent trends in the magnitude of core/periphery inequalities in thefollowing
chapter. Formal studies which directly
evaluate evidence about the causes of the reproduction of the core/periphery at
the worldsystem level have not yet been done, but this should not prevent
us from using the less formal evidence that we have to evaluate the
hypothesized mechanisms.
Here we shall review
crossnational comparative research which has been done on the effects of various kinds of
dependence on economic growth in the periphery.
Generally, dependency theory implies that having a high level of dependence causes a relatively slower rate of
economic development. There are several
types of dependence which have been studied by the method of cross
national comparison.
1 dependence on
foreign aid and loans;
2 dependence on
foreign equity capital;
3 the composition of
trade in terms of the levels of processing of exports and imports (trade
composition);
4 the concentration of
exports in terms of specialization in particular commodities (commodity
concentration); and the concentration of trade partners,
usually the percentage of exports going
to the largest export partner (partner concentration).
One study has examined the effects of the export of low wage
goods hypothesized as central in the unequal exchange theory.
Foreign Investment Dependence
Dependence on foreign equity capital, as measured by the extent
to which a national economy contains a
high proportion of capital stocks owned by foreign corporations, has been found
to have a retardant effect on economic growth, and also to be associated with a
relatively greater degree of national income inequality (Bornschier and
Chase-Dunn, 1985). Contrary to the
arguments ofsome theorists of dependent industrialization, this also holds true
for dependence on foreign capital invested in manufacturing (Bornschier and
Chase-Dunn, 1985:chapter 7).
Thus both classical
dependence (investment in extractive and agriculture exports) and dependent
industrialization probably reproduce the core/periphery hierarchy by slowing
the rate of economic development in dependent countries. Studies of the several mechanisms which have
been found to mediate the retardant effects of investment dependence on
development are summarized in Bornschier and Chase-Dunn (1985). Evidence indicates that decapitalization
resulting from repatriation of profits by TNCs is one important mechanism
accounting for the negative growth effect.
Trade Dependence
Trade dependence has several dimensions, as outlined by Rubinson
and Holtzman (1981). Galtung (1971) shows that export partner
concentration and commodity concentration are negatively correlated with GNP
per capita, while trade composition (the export of manufactured goods and the
import of raw materials) is positively correlated with GNP per capita. Thus these measures of trade dependence are
distributed as expected across the core/periphery hierarchy. This result is confirmed using a more
sophisticated measure of trade composition developed by Firebaugh and Bullock
(1986). But the question remains about
the effects of these various forms of trade dependence on economic development. Rubinson and Holtzman (1981:93) review 13
crossnational quantitative studies of the effects of these various forms of
trade dependence on the growth of GNP per capita. They conclude that commodity concentration
does not have systematic effects. They
also conclude that trade composition positively affects economic growth. Thus exporting manufactured goods and
importing raw materials, as expected by Galtung and others, has been found to
be associated with higher rates of economic development.
As described in the
previous chapter, Nemeth and Smith (1986) have constructed a network-based
measure of the position of countries in the world-system based on the
composition of imports. Their network
analysis produced four tiers of the world-system: a core, a strong and a weak
semiperiphery, and a periphery. Using
these categories in a cross-national panel analysis to determine the effects on
economic growth, Nemeth and Smith generally confirm the results of other studies
of trade composition. Both the weak
semiperiphery and the periphery show negative effects on the percentage growth
rate of GNP per capita, but the strong semiperiphery exhibits a positive effect
which is similar in size to the positive effect associated with being in the
core category (Nemeth and Smith, 1985:548-53).
Bornschier and
Hartlieb (1981), however, find that, when attribute measures of trade dependence are entered into
a multiple regression analysis together with a measure of foreign investment
dependence and other appropriate control variables, none of the trade
dependence variables have significant effects on the growth of GNP per capita
between 1965 and 1977 (Bornschier and Hartlieb, 1981:38). This suggests that the apparent effects of
trade composition indicated in other studies may have resulted from its
association with penetration by transnational firms.
Low Wage Exports and International Split Labor Markets
Kristen Williams (1985) reports the results of a crossnational
study which sets out to evaluate competing explanations of international uneven
development: neo-classical economics, Emmanuel's unequal exchange theory, and
Prebisch's hypothesis of unequal returns to productivity increases. Williamsexamines data from both core and
non-core countries for five, ten and fifteen year time lags between 1960 and
1975. She examines the causes of wage
increases, productivity increases, changes in the organic composition of
capital, and GNP per capita growth. Her
results are surprising for all of the theoretical positions examined.
She tests the
hypothesis that trade with high-wage countries slows the rate of GNP per capita growth for non-core
countries (derived from Emmanuel's unequal exchange theory) but finds no
support for it (Williams, 1958:57).
Neither is Prebisch's model of the periphery supported. Increases in productivity, and/or the organic
composition of capital, increase both wages and growth in the periphery, a
result which Williams interprets as support for the neo-classical model. She finds a negative effect of the growth of
peripheral exports on productivity and wages, and thus economic growth. This is not the unequal exchange effect
predicted by Emmanuel, but neither is it consistent with the neo-classical
theory of "export-led" growth.
Growth in the volume of peripheral exports lowers economic growth
because it lowers both productivity and
wages.
Williams finds that a
quite different model explains the dynamics of the core. "The neo-classical model does not seem
to fit. Increases in productivity do not
seem to lead to either growth in wages or economic growth... Instead, one of
the predictions of Emmanuel's model does seem to have some empirical
support. Increases in wages appear to
lead to growth. In fact, wage increases
seem to have the largest impact on growth" (Williams, 1985:65). She later argues that this is probably due
more to the Keynesian consequence of increasing demand than to anything
connected with Emmanuel's theory of trade imperialism. Williams suggests a version of split labor market theory to
account for her findings. "For the
periphery, low wages encourage labor-intensive development, which has low
productivity, so wages stagnate" (Williams, 1985:68). This is indirectly related to international
trade, rather than directly, as Emmanuel
theorized. Wage differentials between
core and periphery are maintained by immigration controls and by the ability of
core producers to avoid having to compete directly with peripheral producers. Specialization in the development of new
hightechnology products, and the product cycle which shifts these to the
periphery once price competition begins to operate, is consistent with the
"monopoly sector/competitive sector" imagery which finds support in
Williams' research.
Debt Dependence
A growing focus on aid dependence and dependence on foreign
loans has accompanied the international financial crisis which has become
evident in this decade. Excellent work has been done on long cycles
of international financial crises by Ulrich Pfister and Christian Suter
(Pfister and Suter, 1987; Suter, 1987; Suter and Pfister, 1986). This research is reviewed in chapter 13 on
cyclical changes in core/periphery relations.
Excellent research has also been done on the international aid regime
and its relationship to the debt crisis by Robert Wood (1986). Here I shall review the crossnational
evidence on the effects of aid and debt dependence on economic development.
Crude early attempts
to estimate the effect of debt dependence (e.g.
Chase-Dunn, 1975:734) provided some evidence that per capita external
public debt loans to the government and
government-guaranteed loans in 1965 had
a negative effect on the growth of electrical energy consumption per capita
between 1965 and 1970. The estimated
effect on per capita GNP growth was alsonegative, but not statistically
significant. Debt dependence was also
found to be cross-sectionally associated with higher levels of income
inequality among households. The reason
these findings were weak was the limited availability of data.4
Ulrich Pfister (1984)
more conclusively demonstrates that total accumulated foreign debt is
negatively correlated with GNP per capita growth between 1975 and 1981, in a
group of 77 developing countries. When
accumulated debts to private banks are studied, the negative
relationship is even larger. This
relationship holds when additional control variables are used in multivariate
analyses. Interestingly, Pfister finds that
inclusion of a measure of
penetration by transnational non-financial firms (TNCs) does not
lower the estimated effect of debt dependence, but the estimated effect of
investment dependence (penetration by TNCs) is much smaller than when debt
dependence is not included with it in
the same regression equation. This
implies that either the negative effect of equity investment dependence is
spurious due to its association with debt dependence or, as Pfister (1984:11)
suggests, it may be that bank credits are a "mediating variable in the
process of decapitalization induced by the operation of multinational
corporations."
Pfister also shows
that current flows of foreign aid between 1975 and 1977 have a positive effect
on GNP per capita growth and these effects are largest for a subsample of 31
lowincome developing countries. However, accumulated debt due to foreign aid does not
positively affect growth, suggesting that "returns on this kind of
financial flow are in fact rather low and that they do not set the economy of
the recipient country on a more rapid growth path in the long run"
(Pfister, 1984:16).
Pfister also examines
the processes which mediate the negative growth effect of debt to private
foreign banks. He presents evidence that
this effect is not due to high levels of
debt service because, though high debt
service is associated with high accumulated debts, it is
unrelated to
economic growth.
Sell and Kunitz
(198687) find that the level of indebtedness in the 1970s is associated with
slower declines of the mortality rate among a group of peripheral countries in
Asia and the
One of the
statecentric approaches (that by Petras et al., 1981) focuses on the power of
the hegemonic core state as a main mechanism behind the reproduction of the
core/periphery hierarchy. One way this
is alleged to operate is through the effects on peripheral states of military
aid programs supplied by the hegemon.
Hartman and Walters (1985) have examined the effects of national
dependence on
The network measures
of worldsystem position have been used in studies which examine the causes of
economic growth and other aspects of development. Snyder and Kick (1979) used their block model
measure which combined four
different international interaction matrices (described in the
previous chapter) in a crossnational analysis to determine its effects on
economic growth. Peripheral and
semiperipheral locations produced by their network measure are associated with
relatively slower rates of economic growth, while core countries show higher
rates of growth. Kukreja and Miley
(1988) have used Snyder and Kick's measure (as corrected by Bollen, 1983) to
demonstrate a crosssectional relationship with a measure of industrial
diversity of national economies. They
show that peripheral and semiperipheral national economies are less likely to
have a complex division of labor as measured by the distribution of the work
force across nine economic sectors.
Nemeth and Smith's (1985) network measure, which is based on
international trade matrices (see previous chapter) has also been shown to be
related to different rates of economic growth.
Nemeth and Smith show that their core and "strong"
semiperiphery categories are associated with relatively higher rates of
economic growth, while their "weak" semiperiphery and periphery
categories are associated with slower economic growth.
Other Effects of Dependence
A cross-sectional analysis by Volker Bornschier and Thanh-huyen
Ballmer-Cao (1979) presented comparative evidence which supports the
proposition that relative income inequality among households is associated with greater degrees of penetration by
transnational corporations. This has
been further confirmed in recent research by Charles Ragin and York Bradshaw
which analyses data over the period from 1938 to 1980. Ragin and Bradshaw (1986) show that a
multiple indicator measure of economic dependency stunts economic development,
depresses the rate of improvement in the physical quality of life, and increases distributional
inequalities within countries. Several
other studies confirm similar findings (see Bornschier and ChaseDunn, chapter
8).
Kathryn Ward's (1984)
research on the effects of investment dependence on the status of women and fertility rates
illuminates another way in which the core/periphery hierarchy may be
reproduced. Ward's review of the literature on women and development provides
several causal propositions about the effects of international economic
dependence on the position of women in developing societies and the fertility
rate. She postulates that the kind of
economic development which occurs in a context of high dependence on foreign
investment decreases the economic opportunities of women relative to men, and
also maintains the economic value of children for subsistence production. This latter obstructs fertility
reduction. Dependency, she predicts, will
be associated with fewer economic opportunities for women in the wage-labor
sector. Further, some of the potentially
liberating and anti-natalist cultural influences, which might otherwise diffuse
from the core to the periphery, are blocked by the consequences of
underdevelopment and the inferior economic status of women. Analyzing data from 126 countries Ward finds
that her hypotheses are generally
supported. Both investment dependence
and trade dependence are found to have negative direct and indirect effects on
women's share of the labor force and the
level of women's participation in the labor force. These findings also hold when agricultural
and industrial sectors are examined
separately. Fertility is shown to be
largely determined by economic development and family planning programs
(negative effects) and by infant mortality (positive effects). Lesser but significant causes of fertility
are income inequality, investment and trade dependence (positive), and the
educational and economic status of women (negative). Thus Ward's research demonstrates that
certain aspects of the core/periphery hierarchy are direct and indirect causes
of the population explosion in the periphery as well as perpetuation of the
patriarchal oppression of women. A more
recent study by Patrick Nolan (1980) finds only a weak relationship between Nemeth
and Smith's (1985) network measure of trade composition and fertility, however.
Peter Evans and
Michael Timberlake (1980) showed that investment dependence in peripheral
countries is associated with relatively large service sector employment, and Jeffrey
Kentor (1981) demonstrated that investment dependence has a positive effect on
measures of urbanization and over
urbanization, and that these effects are partly mediated by the
expansion of the service sector. The
effect on service sector growth was confirmed by Robert Fiala (1983) for the
decade of the 1950s but weaker effects were found for the 1960s. Moshe Semyonov and Noah LewinEpstein (1986)
have shown that investment dependence expands the part of the service sector
which is
associated with producer services, but not consumer services.
Conclusions
What can we conclude from the crossnational studies regarding
the operation of mechanisms which reproduce the core/periphery division of
labor? Unfortunately, due to the aggregation problem mentioned above, the
studies of national development cannot be used as conclusive evidence about
which are the most important processes producing international
inequalities. Even though investment
dependence has been shown to be an important cause of relatively slow economic
growth in peripheral countries, it does not necessarily follow that it accounts
for the reproduction of international inequalities. No one has studied the relationship between
changes in the level of worldwide investment dependence and changes in
international inequalities over time.
The crossnational
findings reviewed above, however, do provide us with our best guesses at this point about which
mechanisms reproduce the core/periphery hierarchy. Equity investment dependence and debt dependence
are probably more important than the various forms of trade dependence. There is little support for the hypothesis
(derived from Emmanuel's theory) that low wage exports directly retard national
development. These findings are based on
research on the last few decades, and thus ought not to be generalized to
earlier periods. In order to be more
certain we need studies over longer periods of time, and studies which examine
the core/periphery dimension at the world-system level rather than at the level
of national development. The few studies
which have been done are descriptive in the sense that they examine changes in
the magnitude of core/periphery inequalities but do not examine the causes of
these changes. These are reviewed in the
next two chapters.
Chapter 12: Recent Trends
This chapter reviews studies which have examined recent changes
in the distribution of social structural features among core, peripheral, and
semi-
peripheral countries.
Research has shown that great changes in certain social structural
features have occurred in peripheral countries in recent decades. As mentioned in chapter 10, several authors
have claimed that the international division of labor has moved into a new
stage. Industrialization has begun in
many peripheral and semiperipheral countries and some formerly peripheral
countries such as
urbanization;
the increasing
population primacy of the largest cities within developing countries;
the growth of mass
education;
the expansion of state
structures;
the rapid growth of
population; and
the shift of the
structure of the work force out of agriculture
toward industry and services.
All these signs of
"development" (in the modernization sense) are viewed with some
skepticism however, by those who notice that some kinds of development are
happening much faster than others, and the overall level of inequality between
core and periphery has apparently not diminished. We shall examine studies which look at the
magnitude of core/periphery inequalities.
But first let us review the recent organizational changes within zones
of the world-
system.
In chapter 6 we
discussed evidence about the internal power of states which revealed that the extractive power of
states vis--vis their societies is increasing in both the core and the
periphery, but core states continue to have greater extractive capacity than
peripheral states (see table 6.1).
Patrick Nolan (1984)
presents the mean levels of several social structural characteristics in the
core, periphery, and semiperiphery for 1960 and 1970, as well as average growth
rates for the countries in these zones for various periods up to 1970. His tables (Nolan, 1984:112-16) generally
reveal the expected differences among core, semiperipheral and peripheral
countries on measures such as the percentage of the labor force in agriculture,
industry, and tertiary employment, as well as the percentage of GDP in
agriculture and manufacturing. Nolan shows that peripheral and semiperipheral
countries are changing in the direction
of the core on all of these measures between 1960 and 1970. Using a slightly different categorization of
countries into world-system zones, Michael Timberlake and James Lunday (1985)
show similar changes in the country averages of national labor force structures
between 1950 and 1970 (Timberlake and Lunday, 1985:334). They also use national data to examine the
structure of the world work force as a whole, rather than averaging country
scores (Timberlake and Lunday, 1985:335).
When they break the world work force into zones the results are very
similar to the analysis produced by averaging country scores.
The Timberlake and
Lunday study, though it examines the same kinds of data as did Nolan (1984),
points to one feature not discussed by Nolan.
Timberlake and Lunday compare the percentage of the work force in the secondary
(primarily manufacturing) sector with the percentage of the work
force in the tertiary (primarily services) sector. They make a ratio of these two proportions
with tertiary in the numerator and secondary in the denominator, and refer to
this as the T/S ratio. Their table 15.l
(1985:335) shows that the T/S ratio for the core and semiperiphery varies
between 1.01 and 1.25 from 1950 to 1970.
But the T/S ratio for the periphery is much higher, about 1.8. This fact is discussed in terms of the notion
of "overdevelopment" and the "bloated tertiary," referring
to the existence of a relatively larger tertiary sector in peripheral
countries. Timberlake and Lunday
theorize that it is the centralized pattern of capitalist accumulation in the
world-system which creates this kind of economic structure in peripheral
countries. A large portion of the
peripheral tertiary sector mediates the relationship between extractive and
cash crop exports in exchange for secondary sector imports from the core. Though the percentage of the peripheral labor
force in industry grew from 10.6 percent in 1950 to 14.3 percent in 1970, this
is still a very small proportion
compared to core countries, where this percentage has apparently peaked out at
about 40 percent. Interestingly, the 26
countries categorized by Timberlake and
Lunday as being in the semiperiphery do not show higher levels of the T/S ratio
than core countries.
Timberlake and Lunday (1985:337) also examine
core T/S ratios from 1850 to 1970 and peripheral T/S ratios from 1900 to
1970. They show that the average T/S
ratios of the core countries rose from 0.68 in 1850 to 1.24 in 1970,
demonstrating the pattern described by the literature on "post-
industrial" society.
Both the service sector and the secondary (industrial) sector have
grown, but the service sector has grown faster and continues to grow, while the industrial sector has grown
relatively slower and seems to have reached a peak. For peripheral countries data are only
available
beginning in 1900, but these show that the average T/S ratio for
peripheral countries was already much higher in 1900 1.25 for peripheral versus 0.81 for core
countries. By 1970 the peripheral countries
had increased their average T/S ratio to 1.76, although this had not increased
since 1950, when it was slightly higher at 1.79.
A similar pattern is
revealed when we examine levels of urbanization across zones of the
world-system. Firebaugh (1985) shows that the level of urbanization (the percentage of the
population living in cities larger than 20,000 in population size) has
increased in both the core and the periphery
since 1920, the first year in which Firebaugh is able to compute average
urbanization levels for both groups of countries. Firebaugh's (1985:295) groups include 11 Western European core
countries and an unspecified number of peripheral countries. In 1920 the average urbanization level for
the core was 31 percent, while in the periphery it was 7 percent. In 1940 the core had increased its level of
urbanization to 36 percent while the periphery had only changed to 10 percent. By 1960 the average level of urbanization in
the core countries studied by Firebaugh had reached 44 percent while the level
among peripheral countries was 17 percent.
Firebaugh points out that the "urbanization gap" between core
and periphery increased between 1920 and 1960, but that this trend is bound to
be reversed as core countries reach ceiling levels of urbanization while
peripheral countries continue to urbanize.
Another feature of
national urban systems is the degree of urban primacy exhibited by the
city-size distribution in a country.
Some national urban systems are flat in the sense that the ranked
population sizes of the cities
do not form a size hierarchy because the cities are of similar
size. Manycountries have city-size
hierarchies which conform approximately to the "rank size rule." The ranksize rule specifies a city-size
distribution in which the largest city is twice as large as the second largest,
three times larger than the third largest, etc.
A city system has a high level of "urban primacy" when the
largest city is much larger than would be the case in a ranksize
distribution. It has long been observed
that many peripheral countries have greater urban primacy than many core
countries, and that urban primacy seems to be a frequent, but not universal,
feature of peripheral city systems.
Many authors have
sought to explain this difference in terms of patterns of colonialism or the
effects of dependency in the worldsystem. Recent research which examines the
city-size distributions for countries in the world-system since 1800 reveals
that, indeed, there are now significant differences between the core and the
periphery in terms of average levels of
urban primacy, but these differences are of relatively recent origin
(Chase-Dunn, 1984, 1985b). For Latin
American countries the difference emerged in the 1930s and 1940s. This suggests that structures of colonialism
are not responsible for the high levels of urban primacy since Latin American
colonialism ended in the 1820s. Whatever has caused this
core/periphery difference is something which has had its effects in relatively recent decades.
Another urban-related
feature of the core/periphery structure is the city-size distribution of the
world-system as a whole. Chase-Dunn
(1985a) has studied changes in the
city-size distribution of the ten largest cities
in the world-system beginning with the pre-modern Europe of AD
800 until 1975. This shows that the
level of urban primacy in the world city system varies cyclically and
corresponds generally with changes in the distribution of military power and
economic comparative advantage among core powers. However, there has been a trend in recent
decades for some of the world's
largest cities to be located in semiperipheral countries. By 1975 four of the ten largest cities in the
world were located in semiperipheral countries.
In addition the world city system has been flattening since 1875 and has
reached a degree of city-size decentralization not seen since the twelfth
century. In terms of population size,
the world city hierarchy is becoming much less hierarchical, and much of this
is due to the rapid growth of a few extremely large cities in semiperipheral
countries.
Changes in the Importance of Different Types of Dependence
There is evidence of a shift in the relative balance of
different forms of dependency over time.
Bornschier and Chase-Dunn (1985:52) show that from the 1960s to the
1970s three measures of average trade dependence of peripheral countries
declined (see also table 13.2 in the following
chapter),1 while dependence on private foreign investment by
transnational firms and three measures of non-equity debt dependence increased
greatly. The rising trends of investment
dependence and debt dependence are found to hold for several subcategories of
non-core countries. An examination of
the World Bank categories of Low Income Countries, Middle Income Countries, New
Industrializing Countries and OPEC members shows that both an
index of transnational corporate penetration (table 12.1) and the total debt
outstanding as a proportion of GNP2 (table 12.2) have increased substantially
in each of these country groups.
(Tables 12.1 and 12.2 About Here)
Trends in the Magnitude of World-system Inequality
There has been much discussion of the extent of world-level
inequality butvery little empirical
work. Authors argue over whether or not
there is a growing gap, and some claim that the capitalist world-economy has brought absolute as well as relative immiseration to
the periphery. Here I will review the
few studies which allow us to estimate changes in the magnitude of world-system
inequalities, and discuss the complicated question of the overall effects of
capitalist development on the quality of life in peripheral areas.
The first thing to
point out is that most of the quantitative studies focus on the post-World War II period,
especially the 1950s and 1960s, in which the world economy was generally
growing, albeit at different rates in different places. It is undeniable that many groups have
experienced absolute immiseration, both material and cultural, as a result of
their contact with the expanding European world-system. The demographic collapse of New World
populations shortly after the conquest by Spain was a result of both epidemic
diseases (Crosby, 1972) and socially structured exploitation (Frank,
1979b). And, since this first era of
pillage and enslavement, there have been many other times and places when
absolute disaster as well as relative deprivation has followed as a direct
consequence of the incorporation of people into the expanding
world-system. Cultural and material
exploitation and immiseration are very difficult things to quantify. The destruction of indigenous cultures, like
the extinction of genetic species, is unweighable.
We should evaluate
the destructiveness and creativity of a
mode of production by comparing it to other modes (Wallerstein, 1983a;
Wallerstein, 1984b:chapter 14). When we
compare the modern world-system with earlier
world-systems I tend to agree with Marx. It is not the best of all possible worlds,
but it does represent progress over earlier systems in many respects. Earlier systems were also exploitative, especially once states had
beeninvented. They also destroyed
indigenous cultures and spread diseases and killed people through
over-exploitation and recurrent warfare.
The quality of cultural and spiritual welfare is difficult to compare,
but even here the critics of modern culture
have not convinced me that the contemporary lowest common denominator is
below that of earlier civilizations.
In material terms the
problem is somewhat easier. We must
consider population growth and the
material level of living. Earlier modes
of production allowed the human species to increase its numbers, but only slowly (Coale, 1974). As Marx pointed out, the ability of
capitalism to revolutionize productive technology is its most progressive
feature, although this is more of a mixed blessing than Marx realized. The rapid
increase in the numbers of human beings in the twentieth century has
been largely due to a combination of more productive technology and public
health measures. While acknowledging
that rapid population growth creates problems, I believe that population growth
is a positive consequence. Certainly the
carrying capacity of the earth is limited, but we have not yet come close to
that limit. The main problems of
pollution and resource depletion are caused by a lack of collective rationality
with regard to long-term consequences. These are world-level political problems
which capitalism as a socio-political
system is probably unable to solve, but we should not allow these observations
to blind us to the basically progressive developments in productive and medical
technology. It is too easy for residents
of the core countries to undervalue basic food supplies and non-animal energy
sources. An appreciation of the overall
collective rationality needed to wisely expand productivity and energy use is
not enhanced by neo-Luddism.
When we examine the
quality of individual lives across the whole
periphery it is by no means clear that there has been absolute
immiseration over the long run. Recent
evidence (described below) suggests that the
average material level of living has risen slightly even in the
periphery of the world-system in recent decades. Earlier instances of absolute immiseration
probably involved larger portions of the world population, whereas more recent
episodes of absolute immiseration have become less frequent and less
widespread. If this is true the problem
has become more a matter of relative rather than absolute immiseration. Evidence about recent changes in the
distribution of certain resources is reviewed below.
There is, however, one
sense in which the notion of absolute immiseration is undoubtedly applicable to
the modern world-system, and that is the matter of the destructiveness and
scale of warfare. The percentage of the total population involved in warfare,
the severity of warfare in terms of the proportion of the total population
which is injured and the proportion of social resources destroyed, have
increased rapidly in the last centuries, and especially in this century (Galtung,
1980:7). The potential now, as everyone
knows, is the destruction of not only the accomplishments of human endeavors
but of three billion years of biological evolution as well. Any mode of production which brings about
such a consequence is guilty of much more than absolute immiseration. It is in connection with the production of
warfare that the historically progressive nature of capitalism is cast into
doubt.
Absolute Immiseration or Relative Deprivation?
What is the evidence about recent trends in world-level
inequalities? Every region on earth
experienced increases in production per capita during the post-World War II
period. Of course, average levels of GNP
per capita do nottell us about intranational distributions. It is well known that some groups in some
regions have experienced declines in material welfare, and the extent of these
absolute declines has certainly grown during the recent period of world-wide
economic slowdown (Frank, 1981:125-31).
But do these pockets of absolute decline outweigh the areas of growth
such that the average level of living in
the periphery has declined since World War II?
The question of
absolute and relative levels of poverty are examined in a study conducted by
Bourguignon, Berry, and Morrison (1983). The conclusions with regard to changes in the relative
distribution of income and consumption in this study are qualified by another
paper by the same authors (Berry, Bourguignon, and Morrison [1983]) which warns
that measurement errors in the determination of relative levels of world
inequality do not allow much certainty about trends. Nevertheless, the first paper is important
because it combines both intranational and international data on inequality of
household incomes to estimate changes in the level of world inequality from
1950 to 1977.
With regard to
absolute immiseration, Bourguignon et al. (1983) report that the total number
of poor people below the poverty line of $200 yearly income increased between
1950 and 1977. This might be understood
as an indicator of absolute immiseration, but other facts lead to a different
conclusion. While the absolute number of
people below the poverty line increased, this was due largely to population
growth in the poorest countries. When
Bourguignon et al. examine the proportion of the world population below
the poverty line, this decreased from 43 percent to 28 percent in the
period studied. This means that although
the number of poor people increased absolutely they decreased as a proportion
of the whole world population.
Other studies indicate
that immiseration has not increased.
Studies of average life expectancies in peripheral countries show
increases (Sell and Kunitz, 1986:87:table 1).
Studies of caloric and protein content of food consumption in the
periphery show stable levels, and an index of per capita consumption for the
"low-income economies" shows an increase from 1960 to 1981 (World
Bank, 1983, volume 1:548). Ragin and
Bradshaw (1986:table 1) show that an indicator of the physical quality of life
in poor countries increased substantially between 1938 and 1980. Thus there is no evidence for absolute
decline in the average material level of living when we examine the periphery
as a whole. This may not hold for
specific areas or groups, and it is possible that a more widespread decline has
occurred since 1981 as the world economic crisis has caused severe dislocation
in many countries. Sell and Kunitz
(198687) argue that the recent debt crises and stagnation of the world economy
has brought "the end of an era in mortality decline."
What can we conclude
about trends in the distribution of relative shares of world resources? The Bourguignon et al. (1983) study found
that the poorest 40 percent of the world's population received 4.9 percent of
world income in 1950, but only 4.2 percent in 1977, and the gap between the
poorest and the richest halves of the world population increased. Even though this is the only study to combine both intranational and
international data on inequality, the revealed trends must be viewed with some
uncertainty because of the measure
ment problems discussed by Berry et al. (1983). Nevertheless it is fairly certain that the
magnitude of world income inequality did not decrease in this period. Ragin and Bradshaw (1986:table 1) examine two
indicators of welfare, telephones per 1,000 population and a composite measure
of the physical quality
of life. These both show
an increase in the relative gap between core and peripheral countries between
1938 and 1980.
Let us now examine the
distribution of some other resources.
Most studies of global resource
distribution compare the means of country groups. Studies which compute averages for country
groups are informative, but they do not allow us much certainty about changes
in the overall magnitude of inequality because the countries in each group are
held constant over time. Since some
individual countries are likely to be changing their ranks, estimates of this
sort do not provide firm knowledge about changes in the magnitude of inequalities. Table 12.3 presents figures which show
changes in the relative concentration/dispersion of several kinds of resources or social structural features
across percentile groups of the world's population. The use of percentile groups in place of
country groups allows for changes in the
rank of countries to occur without disturbance to the estimates of percentile
shares.
Table 12.3 uses data from countries to compute the
proportion of world "resources" held by percentile groups of the
world population. For each resource the
countries included in the overall distributions are the same for all time
points so that the addition or deletion of countries does not affect the
proportions. The proportions are
calculated by ordering the countries in terms of the per capita or
"level" indicator, such as GNP per capita. All
countries are listed from high to low in terms of GNP per capita
and their total GNP and total population are also listed. Then the group of countries highest on GNP
per capita that include nearly 20 percent of the world's population are
examined to determine what proportion of world GNP they produce.3 (Table 12.3
About Here)
Table 12.3 shows the
concentration of economic production as measured by GNP and energy
consumption. It examines the world
economic structure as indicated by the proportion of the agricultural labor
force and the proportion of the
industrial labor force, and it also examines the shape of world urbanization. It shows changes in the distributions of
these features to the top and bottom
quintiles of the world's population, as well as the middle 60 percent, and the
shares accounted for by the United States.
The main conclusion indicated by the overall
pattern in table 12.3 is that the distribution of these features has not
changed very much in the 20year period
from 1960 to 1980. Certainly there has
been no decrease in the relative distribution of productivity as indicated by
GNP. Somewhat surprisingly the middle
group has not gained in this period, and neither has the top group lost,
despite the well-known phenomenon of the "newly industrializing countries." The United States has fallen considerably
from 32.1 percent of world GNP in 1960 to 26.9 percent in 1980. This is a continuation of the declining position of the US in world
commodity production which was demonstrated to have already been occurring
since 1950 by Meyer, et al. (1975:table 2).
The poorest quintile shows a decrease from 1.4 percent to .8 percent. Recall, however, that this is a relative share,
not an absolute decrease.4
Energy consumption
shows a similar distribution in 1960 to GNP, but in the following decades an
important changes occur. The top
quintile loses 13 percent of world
energy consumption to the middle group.
This might be due to the industrialization of semiperipheral
countries. When we examine which
countries grew the most or the least between 1970 and 1980 in per capita
energy consumption we discover two factors which account for the
change in the world distribution. First,
three high consumption energy producing countries in the top quintile
experienced large declines in their energy
consumption: Trinidad, Bahrain,
and Kuwait. The United Kingdom and
Luxembourg, both in the top quintile, also experienced large declines. On the
other end of things, large increases were registered by a number of countries
in the middle group: Libya, Austria,
Yugoslavia, Japan,5 Romania, South Korea, Hungary, Greece, Spain, Portugal, and
Ireland. These, along with the milder
growth of other semiperipheral countries in the middle group and the declining
consumption in the countries within the top quintile (mentioned above) were
great enough to overwhelm the increases experienced by other countries in the
top quintile such as Norway, Australia, Canada, the United States, Saudi
Arabia, Iceland, Finland, the Netherlands, Czechoslovakia, East Germany, and
Belgium. A combination of declines on
the part of some oil producers (plus the United Kingdom and Luxembourg), plus
growth among Eastern European and semiperipheral industrializing countries (as
well as Japan and Austria), accounts for the shift in the distribution of world energy consumption.
It is interesting that the shift in energy
consumption, which is at least partly due to the spreading of industrialization
to semiperipheral countries, is not
reflected in the distribution of GNP. A
sectoral analysis of GNP growth might help resolve this apparent inconsistency. It is possible that financial services and
high technology/low energy industries located primarily in core countries have
expanded enough in the period studied to offset the semiperipheral industrial
growth component of the world distribution of GNP. Much of semiperipheral industrialization has
been in heavy intermediate industries which consume great amounts of energy per
unit of GNP.
When we consider the
distribution of economic structural characteristics there are some interesting
trends. The world's agricultural work
force is shifting toward those countries which are already highest in
terms of the proportion of the work
force in agriculture. From 1960 to 1980
the proportion of the world's farmers and farm workers located in the quintile of countries with the lowest
proportion of the labor force in agriculture declined from 3.8 percent to 2.6
percent, and declines were also experienced by the middle group.
The countries in the quintile highest on the percentage of the labor
force in agriculture i.e. peripheral
countries increased their proportion of
the world's farmers and farm workers from 33.2 percent to 40.4 percent. This finding suggests a different twist to
the notion that national economies are developing by shifting workers out of
the agricultural sector. It is true that this is happening in most countries, but
it is happening much faster in the more developed countries, so that the
structure of the world work force is
shifting toward a greater concentration of farm workers in the least developed
countries.
When we examine the
relative global distribution of industrial workers the major feature is
stability in each of the groups, although the lowest quintile has increased from 6.2 percent to
7.2 percent. This stability is interesting
because the most developed countries are not increasing the percentage of their
work force in industry, and many are experiencing declines in this proportion
as the service sector continues to grow.
The United States, as shown, experienced a reduction in its proportion
of the world labor force in industry from 12.9 percent to 9.5 percent in the
period studied. This
"export of the
proletariat" is rather small, however, when we examine the top
quintile as a whole. Its proportion of
the world industrial labor force declined from
42.7 percent to 41.0 percent.
Urbanization, on the other hand, reveals
fairly substantial changes in its world distribution. The most urbanized countries declined in
their proportion of world city-dwellers
from 38.4 percent in 1960 to 34.1 percent in 1980. Both the middle group and the bottom quintile
experienced increases. This means that the well-known phenomenon of rapid
urbanization in the periphery is redistributing the world's city dwellers. Of course urbanizaton the proportion of a population residing in
cities is subject to ceiling limits such
that increases above 100 percent are impossible, and so concentration would be
likely to decline in a world in which some countries have become completely
urbanized while others are still on the road to urbanization. Similar ceiling effects, however, might be
expected to reduce core/periphery differences in the distribution of the
world's work force across economic
sectors, but for these the inequalities are either stable or becoming
greater. I would suggest that the
reduction in core/periphery differences in urbanization levels is indicative,
not of the periphery catching up with core with regard to a structural feature
of modernization, but rather the increasingly direct integration of the
populations of peripheral countries into the world labor market, including the
rapidly expanding informal sectors in
the large cities of the periphery (Portes and Walton, 1981). Thus this is not evidence of the reduction of
core/periphery inequality, but rather a more direct inclusion of peripheral
peoples into the structure of international dominance and dependence. As reported in the previous chapter Jeffrey
Kentor (1981) has shown that dependence on foreign investment causes increased
urbanization in the periphery and that this is mediated by the expansion of
tertiary sectoremployment. Peripheral
urbanization is, thus, a very different phenomenon from the urban-industrial pattern of
development which occurred in core countries.
Another recent trend
is the spread of the production of military weapons to the "
Conclusions
The most important conclusion from the evidence reviewed in this
chapter is that, despite the undeniable industrialization of many peripheral
and semiperipheral states, there is no evidence of a reduction in the magnitude
of core/periphery inequalities. Thus,
reports of the demise of the core/periphery hierarchy are certainly premature.
On the other hand,
there is little evidence of general absolute immiseration of the
periphery. Some studies show small
increases while others show little change in various indicators of the average
life chances and level of living in the periphery. Data are not yet available for the recent period since 1981, a period in which
there have been famines and economic crises in several regions of the
periphery. Periphery-wide data on this
time period might yield a different result.
My best guess is that it is relative rather than absolute immiseration
which will have significant political effects in the long run.
Chapter 13: World-System Fluctuations
This chapter examines empirical studies of worldsystem cycles
and arguments about their causes and effects.
Early theorizing about relationships among worldsystem processes posited
both temporal models showing the relations among different cycles over time and
models depicting causal relations among variables (e.g. ChaseDunn,
1978:1701). Subsequent empirical work
has confirmed some of the original hypotheses and exposed others as overly
simplified or completely wrong. This
chapter reviews the state of a field which is rapidly changing because so much
new empirical work is being done, and yet much more needs to be done before we
will have a clear picture of the causal nature of worldsystem processes.
As was discussed in
chapter 9, the distribution of competitive advantage in commodity production in
the core of the world-system varies from
a situation of hegemony, in which one core state has a clear advantage, to a
situation in which profitable production of core products for the world market
is more evenly distributed across the
core states. For analytic purposes we
can refer to the former situation as hegemonic and the latter as "multicentric." In chapter 9 we discussed the causes of the
rise and fall of hegemonic core powers
the national-level and world-system dynamics which result in the cycle
of core competition. We will now examine
how this fluctuation between hegemony and multicentricity affects the structure
of control and exchange between the core and the periphery.
One of the features of
the core/periphery relationship which is hypothesized over time is the
fluctuation between a more or less multilateral system of
"free" trade between the various core states and the various
peripheral areas on the one hand, and a more bilateral system of
colonial empires on the other. These two hypothetical structures are
graphically represented by figure 13.1.
(Figure 13.1 about here)
This figure
illustrates two different idealized structures of control and exchange between
core states and peripheral areas. In a
system of colonial empires each core state monopolizes exchange with its own
colonies and excludes other core states from this trade. In the multi-lateral structure trade is less
controlled by mercantilist state policies and the exchange comes closer to the
ideal of a free world market. These
schematic alternatives are, of course, only roughly approximated in complex
reality. Some core countries never have
formal colonial empires or have only small ones; truly price-setting
international markets in which commodities exchange between core and periphery
without regard to bilateral political considerations have never been completely
realized; and, of course, the composition of the core and the periphery changes
over time with the rise and fall of different regions and the incorporation of
new regions into the modern world-system.
Nevertheless, the above hypothesized
structural types may have analytic utility.
Two additional
features of the core/periphery relationship will be considered in this
discussion. The first is expansion. The European world-system expanded to
incorporate and peripheralize formerly external arenas in a series of waves
since the sixteenth century. The nature
of this incorporation varied depending upon the type of society being
incorporated, as well as the particular features of capitalist accumulation and
political organization which were current in the core at the time of
incorporation (chapter 10). Nevertheless
we may note that the rate of incorporation of new populations and territories
varied over time such that incorporation occurred in waves. And another feature of the core/periphery
relationship which has varied over time is the amount of resistance to core
domination from peripheral areas. These
variations, of course can and should be studied in particular contexts to
understand the conjunctural elements which nuance each situation; but here I
want to examine them as features of the world-system as a whole.
The following discussion
also incorporates two other cyclical variations which we have noted operating
in the world-system: Kondratieff waves (Kwaves); and the cycle of the severity
of war among core powers. These cycles
and variable features will each be discussed in turn below before we
consider the temporal and causal
relations among them.
As discussed in
earlier chapters, in my view the core/periphery hierarchy plays an important
role in the reproduction of capitalist accumulation. Let us now examine some of the ways in which
the structure of domination between the core and the periphery changes and is
changed by the process of capitalist accumulation in the core. Intensive and extensive development are
alternatives for capital, and their relative profitability is determined in
part by the level of resistance which is encountered. Thus, one reason that capital is exported to
the periphery, and that colonial expansion is undertaken, is that resistance to
capitalist exploitation is often weaker in the periphery than in the core. Also, expanded reproduction in the core
creates demand for food products and industrial raw materials which can be
cheaply produced in the periphery. And
core producers seek market outlets in colonial areas in order to realize
additional profits on product lines originally introduced in core markets.
According to Rosa
Luxemburg (1968) the motive of market expansion is the key reason for
imperialism, as capital seeks to escape the market saturation resulting from
cyclical overproduction of commodities.
All of the above processes desire
for expanded investment opportunities, desire for commodity markets, and the
rising demand for cheap food and raw material inputs stimulate capitalist core
states to use politicalmilitary power to sustain the property relations and
other institutional supports which facilitate core exploitation of the
periphery. This general tendency is
influenced by the cycles of economic growth, uneven development, periodic
warfare, and hegemonic rise and fall which emanate from the arena of
competition and conflict within and among core states. Economic and political competition among core
states often produces defensive expansion in the periphery, a kind of
"anticipatory colonialism" which appears to be unmotivated by existing
economic
opportunities, but which can be understood in terms of the
"strategic" interests of competing core powers.
A Hypothetical Model of Temporal Relations
This section focuses upon a specific world-system level process:
the way in which competition among core states and their national bourgeoisies
affects the relationship between the core and the periphery. The descriptive form of the hypothesis is
that a hegemonic distribution of productive advantage in the core, in which the
bourgeoisie of a single core state is hegemonic in the world economy, leads to
a relatively multilateral structure of exchange and control between the core
and the periphery, and the relaxation of
political controls over core/ periphery exchange. Conversely, a more multicentric distribution
of competitive advantage in the core leads to a bilateral structure of exchange
and control between core states and their colonial empires, and also to the
expansion of colonial empires into new territories.
The aggregate rate of
growth in investment and production which varies over time to produce the
Kondratieff wave is an average for the world-system as a whole, at least in
principle. But development is, of
course, uneven among core states, between the core and the periphery, and within
the periphery as well. So-called
"rounds of accumulation" based on the expansion of new types of
production are obviously not uniformly distributed across space. And the overall rates of economic growth, as
well as the levels of slowdown and stagnation, vary from country to country,
and also regionally within countries.
Thus capitalist development is spatially uneven, and it is this which
produces the hegemonic cycles and the upward and downward mobility of countries
in the core/periphery hierarchy. But let
us ignore these spatial variations for a moment to focus on the average levels
of growth and the overall system-wide variations in the severity of competition
and conflict among core states.
In my 1978 article I
argued that the Kwave and conflict among core states were inversely related
(ChaseDunn, 1978). This was before
Goldstein's (1988) research, which shows that the severity of wars among core
states varies not inversely but conjointly (with a small lag) with the
Kwave. My earlier reasoning was
influenced by my study of tariff politics.
I argued that in periods of economic expansion capitalists are less
likely to support state intervention, while during periods of stagnation they
are more likely to support the use of state power to protect or extend their
interests. A recent study of nineteenth
century tariff politics (McKeown, 1983) agrees with my supposition that Kwaves
and tariff protectionism are inversely related.
Goldstein's results show that warfare is most intense during Kwave upswings,while
McKeown (1983) concludes that protectionism increases during downswsings. Apparently some kinds of competition are
greater during upswings while other kinds increase during downswings. These leads to different sorts of state
intervention.
The concentration of
productive advantage in a single core state means that commodities are being
produced at a low cost and in enough volume to invade the markets of
competitors and to create new markets by affecting consumption patterns. When the price of these commodities is low
enough, and demand in other core countries is high enough, the political
barriers to trade across state boundaries are likely to be lowered, and a
period of relatively free trade ensues.
Stephen Krasner (1976) first described this consequence of
"hegemonic stability" in his study of tariff barriers and trade
patterns in the world economy during the nineteenth and twentieth centuries.
In the nineteenth
century the period between 1820 and 1870 was one in which tariff barriers among core states were
generally lowered. The Anti-Corn Law
league carried on an extensive program of propaganda and political mobilization to abolish
There is a constant
tendency for all states to have protective tariffs because the returns to the
protected producer are great, while the loss to
individual consumers is small.
Political policy is thus subjected to the influence of strongly
motivated interests seeking protection, while the interests of consumers are dispersed and
difficult to mobilize. But if the
gap between home market prices and international market prices
becomes too great either smuggling will occur, or political pressures will be
brought to bear to lower tariffs. An additional element is the overall rate of
economic growth. Producers are more
adamant about protection during periods when profits are falling and
alternative investments are difficult to find.
Just as tariff
barriers among core states were lowered, barriers constraining core/periphery trade within
colonial empires between an individual "mother" country and its
colonies tended to relax as the
advantages of buying cheap imports came to outweigh the forces
supporting colonial trade monopolies.
Consumers within core states wanted cheap sugar, for example, and the political support for
monopoly faced strong opposition as the difference between the price of
protected sugar and world market sugar
grew. Similarly, consumers in peripheral
areas had an interest in buying their imports from thecore power that sold them
most cheaply, and this tended to disrupt
bilateral colonial exchange monopolies.
Both colonial monopolies and home market monopolies were politically
attacked during a period of overall economic growth of the world economy.
The diffusion of
technological innovations from the hegemonic core state (Henderson, 1965) and
the stimulus to more efficient production resulting from core competition
leads, in combination with the right domestic political conditions, to the
expansion of industrial production in other core states and in some
semiperipheral states. In the nineteenth
century these were the
The expansion of
production of processed goods in the core caused the demand for raw materials
to rise. Raw material production was more dependent on "natural"
factors such as climate and the location of natural resources, so the
geographical distribution of extractive production was necessarily more
widespread than the production of manufactured goods. This dependence on natural conditions also
slowed technological improvement in the production of raw materials relative to
that which was possible in manufacturing
production, which is more easily amenable to reorganization. In addition, the increased accumulation of
capital in the core stimulated an
organized labor force
demanding higher wages and other amenities. This provided an incentive for capital to
utilize cheaper peripheral labor when
possible. All these factors
expanded and intensified the economic exploitation of the periphery.
The evening out of
competitive advantage across the core led to increased competition among
producers for access to markets and raw
materials. This wasmanifested
within the core by the reemergence of protective tariffs around home markets,
and between the core and the periphery
in the tightening of colonial monopolies and the expansion of colonial empires
to new areas. The core/periphery trade
network shifted back toward a more bilateral (colonial) structure. This "new" mercantilism and
"new" imperialism occurred because the overall growth rate of the
worldeconomy slowed down and so competition increased and groups intensified
their utilization of state power to maintain shares of a shrinking
"pie". Each core country
increased its ties with its "own"
colonial empire, and trade among core countries declined in importance
(Woodruff, 1967).
Colonial expansion and
economic penetration of the periphery mutually reinforced one another, although
formal colonization was fiscally expensive
and was often a defensive (or even preventive) result of core competition
over access to peripheral resources and markets. The scramble for sub-Saharan
A period of conflict
and disorganization of the world-economy brought about by increasing competition among core
states created room for the emergence of peripheral resistance. Similarly, but during a different phase of the world-wide Kondratieff, the increased
demand for raw materials improved the market position of producers in the
periphery and thus may have encouraged
peripheral resistance. Peripheral independence movements received support from
those core powers in a position to benefit from breaking down the colonial
monopolies of other core powers. British
support for Spanish American independence movements is a case in point.
Decolonization and
resistance from the periphery increased the costs of exploitation, and this
forced core capital to reconsider the possibilities for more intensive
exploitation at home. The creation of
formally independent states in the periphery, although their sovereignty was
compromised by neocolonial forms of core/periphery domination, nevertheless
increased the cost of exploitation. It
created barriers to further colonial expansion (or recolonization), and
multilateralized the structure of trade as peripheral states obtained some
latitude to play off core countries against one another. As in the core, however, opposition from one
area drove capital to where opposition was less, and this provided the motive
force for the continual expansion and deepening of capitalist exploitation.
Before reviewing the
empirical studies which have been done, let us describe another model, this
time in the form of a set of hypotheses about
directional causal effects. The
model below was presented in Chase-Dunn (1978) and a related version was
formulated by Pat McGowan (1985) based on the arguments contained in Bergesen
and Schoenberg's (1980) study of waves of colonialism.
A Causal Model
Let us decompose the process of uneven development in the core
into three parts: (1) the growth of production in the hegemonic core state
resulting from the concentration of productive advantages; (2) the growth of
production in competing core states resulting from the evening out of the
distribution of productive advantage; and (3) the aggregate rate of economic
development in the world economy.
Similarly, we can decompose the core/periphery
structure into four components: (1) economic penetration of peripheral areas by
core firms; (2) expansion of formal colonial control by core states; (3) the
oscillation between a bilateral and a multilateral network of exchange and
control
between the core and periphery; and (4) the increase of
peripheral resistance as manifested by decolonization movements and other forms
of resistance to core domination.
The relationships
among these variables are mediated in part by the level of conflict among core states. These variables and the hypothesized causal
relations among them are illustrated in figure 13.2.
(Figure 13.2 about here)
As specified, this
model is not testable with quantitative data even if the proposed variables
could be operationalized over a time period
sufficient to include meaningful variation in world-system structures
because it is underidentified too many
of the variables are endogenous in the sense that they are affected by other
variables in the model. Nevertheless each proposed effect is adduceable. The model is presented to clarify my
arguments. Further specification needs
to be done to make it testable, and, of course there are difficult problems of
data availability. For now let us simply
use this model as an heuristic tool.
A Review of Existing Evidence
There have been quite a number of interpretive studies which
assert dating schemes for various of the worldsystem fluctuations but only very
recently have researchers begun to actually measure worldsystem variables and
study their interaction over time. The
pathbreaking study of waves of colonialism by Bergesen and Schoenberg (1980)
has been followed by a number of other efforts.
These are reviewed below and the implications for the above
hypothetical arguments are discussed.Cycles in the Core
The findings from
Goldstein's (1988) study of cycles of core war, Kwaves, and the hegemonic
sequence indicate that some previous worldsystem hypotheses were
incorrect. Hopkins and Wallerstein
(1979) argued that each hegemonic sequence was composed of two Kwaves, but
Goldstein concludes that the Kwave is related only loosely to the hegemonic
sequence. Each hegemony contains a
different number of Kwaves and associated war cycles, but the hegemonies all
begin after an unusually large peak in core war severity. These matters are still in some doubt,
however, because of remaining controversies over the conceptualization and
measurement of the hegemonic sequence (see chapter 9).
Also, contrary to many
earlier worldsystem arguments, Goldstein finds that war is more severe during
Kwave upswings than during Kwave downswings, and so the Kwave and the war cycle
vary conjointly rather than inversely.
If it is true that tariff protectionism, on the other hand does vary
inversely with the Kwave then the variable "core conflict" in the
middle of figure 13.2 needs to be decomposed into different types of
conflict/competition among core states.
Colonial Expansion
I propose a slightly different approach to that employed by
Bergesen and Schoenberg (1980) in their study of waves of colonialism from 1415
to 1969. Measuring the expansion of the
modern world-system is not a simple matter.
Expansion is carried out by states through formal colonization, but also
by private entrepreneurs, and some areas become incorporated into the world-system
through trade or treaty which does not involve formal colonial subjugation
(e.g. Hall, 1986). As David Henige's
(1970) compilation shows, many colonies were settled first by private parties
and later received officialcolonial status.
Bergesen and Schoenberg's operationalization treats each newly
established colony equally, but surely some were more important than others in terms of the amount
of territory or number of people subjugated.
Unfortunately, only rough estimates of the territorial or population size of the colonial empires
are available for the earlier centuries (Bairoch, 1986; Taagepera, 1988). The use of the number of new colonies
established is, however rough, the best continuous measure we have available at
this time.1
Bergesen and
Schoenberg devote most of their analysis to the net number of colonies,
an indicator which shows the cumulative number of colonies established minus
the colonies which have been terminated.
This indicator reveals the two waves of colonialism, an earlier one in
the seventeenth and eighteenth centuries, and a later one in the nineteenth
century known to many students of colonialism as the "new
imperialism."2 But Bergesen and
Schoenberg should have paid more attention to the expansion of colonial empires
separately from their termination. These
are separate processes, one emanating mostly from the core and the other due
largely to resistance from the periphery to core domination. Bergesen and Schoenberg present graphs of
these phenonomena separately (1980:2345), confirming what has also been noted
by McGowan, that they have quite different patterns in time. Figure 13.4, from Bergesen and Schoenberg
(1980:234), shows an 11-year non-cumulative moving average of the number of new
colonies established.
(Figure 13.4 about here)
This figure, I will
argue, shows the periods in which the world-system is being territorially
expanded due to colonizing by core powers.
The figure reveals ten peaks of expansion since 1500, and visual
inspection suggests that these may be related in a lagged fashion to the
Kondratieff waves and war severity cycles demonstrated by Goldstein (see Figure
8.1 above).
Pat McGowan
(1985:table 4) analyzed the Bergesen and Schoenberg data using the method of
time series regression analysis. McGowan
finds no relationship between the
measure of new colonies created and the Bergesen and Schoenberg coding of the
existence of a war. McGowan tested for a
simultaneous correlation examining the
correlation between war and expansion within each single year. There are two problems with this
finding. The hypothesis that wars and
colonial expansion should occur in the same year sounds faulty to me. I would not expect expansion by core powers
in peripheralizing areas to occur simultaneously with war among core
powers. Bergesen and Schoenberg seem to
suggest that these variables are correlated within larger blocks of time with a
time lag between them. My guess is that
the making of war with other core powers does not leave resources free for
extensive adventures in the periphery.
We know that some colonies change
hands during or just after a core war, but these cases should be deducted from
a measure of system expansion (see note 1 above).
The three different
factors theorized to be behind system expansion are: demand for cheap raw
materials; need for new investment opportunities due to the declining rate of
profit on investments within the core; and the glut of core markets resulting
in the search for new effective demand in the periphery. These three factors are most likely to occur
at different points in the Kondratieff cycle.
Increased demand and rising prices for raw materials occur during an
upswing when production is expanding.
The need for new investment opportunities and markets occurs at the peak
and during the downswing. In addition
Walter Goldfrank (personal communication) points out that during downturns
there are pressures for new areas of settlement for the unemployed, thus
pushing toward expansion at the frontiers.
McGowan's finding of
no simultaneous relationship between the creation of new colonies and the presence/absence of
core war may be due to the contradictory operation of the above factors, or it
may be due to the crude measure of core
war used by Bergesen and Schoenberg.
Goldstein's (1988) findings are based on the severity of core
wars, the number of persons killed. This
measure shows the intensity of core conflict as well as its
presence/absence. My guess is that the simultaneous
time series correlation between core war severity and the expansion of colonial
empires will be shown to be negative or zero because of the trade-offs in the
costs of core war and colonial expansion.
I predict, however, a significant association when time lags are
analyzed.3
Terry Boswell (1986)
has performed a time series regression analysis of the relationship between
colonialism, warfare, the Kwave, and the hegemonic sequence. Using a measure of war intensity the
annual number of battle deaths as a proportion of the total European population
(different from Goldstein's measure of severity which is the unweighted
number of battle deaths) and Bergesen and Schoenberg's measure of net
cumulative colonies, Boswell finds no relationship between wars among core
states and colonialism. My main
reservation about this finding is the use of the net cumulative measure of
colonialism. As stated above, combining
the establishment of colonies with their termination confuses two very
different processes. Future research
should examine the lagged relationship between warfare and the number of
colonies established.
Boswell finds support
for the hypotheses in ChaseDunn and Rubinson (1977) regarding the relationship
between colonialism, the hegemonic sequence and Kwaves. Kwave upswings are associated with less
colonialism; downturns with more colonialism.
Periods of "hegemonic victory" are associated with less
colonialism. These findings are
encouraging but they suffer from the same defects mentioned above: the net
cumulative number of colonies was used and the "measure" of hegemony
is a set of dummy variables based on the dates asserted by Hopkins and
Wallerstein (1979).
It is important to
remember Goldstein's (1988:176) argument that
sequential "cycle time" rather than strictly stationary
periodicity is the appropriate form of analysis for many social phenomena. We would not expect, for example, a constant lag of exactly X number
of years to characterize the relations among world-system variables in each
period. A test for lagged relationships should allow for this by
utilizing at least fiveyear blocks of time.
The methods problems associated with time series analyses of
world-system processes are considered in chapter 15.
Resistance in the Periphery
Regarding the notion of resistance from the periphery we should
examine another figure produced by Bergesen and Schoenberg (1980:235) of the
number of colonies terminated. Remember
that resistance takes many forms from labor
slowdowns, flight to refuge regions, demonstrations, strikes, banditry, tax
evasion, protectionism, import substitution, expropriation, armed rebellion
which is quashed, and successful rebellions which expel the colonial power,
such as occurred in the US, Haiti, and the Spanish colonies of Latin America in
the late eighteenth and early nineteenth centuries. The conditions for resistance are always
present in the periphery, so why should we expect peripheral resistance to
cluster in time across many very different peripheral situations? Andre Gunder Frank (1972) has argued that
resistance in the periphery clusters during core wars and high levels of core
competition because the level of control by core powers is diminished. Also it may be easier for peripheral
countries to play core powers off against one another in such periods.
Another factor which
may produce the clustering of successful peripheral resistance in time is
related to the hegemonic sequence. A
rising hegemon often supports the liberation of the colonies of other core
powers. This situation produces the spectacle
of a policy of "liberation" which supports anti-colonial movements in
other colonial empires while simultaneously extending its own (in the name of
free trade) by force. Robinson and
Gallagher (1953) described this aspect of British policy in the first half of
the nineteenth century as the "imperialism of free trade." A milder, but functionally equivalent,
version was United States anti-colonialism after World War II.
Observing figure 13.5
we see that the termination of colonies does cluster over time, with the
biggest peaks being the decolonization of the
Americas in the late eighteenth and early nineteenth centuries, and the
decolonization of most of the rest of the African and Asian periphery after World War II.
(Figure 13.5 about here)
While the British fought hard to retain the
thirteen colonies which became the United States (and these new states were
aided by a contending core power, France) it was not much later that these same
British aided the national liberation movements of Latin America against Spain,
thus contributing to the first big wave of decolonization. After World War II it was the UnitedStates, the new hegemonic core
power, which, again in the name of free trade, supported the dismantling of the
empires of other core powers. The Dutch
hegemony did not produce such an upsurge, perhaps because, though the Dutch too
espoused a liberal international order, no one thought to apply this idea to
the pagans and Moslems of the East Indies.
Rather a policy of "armed trade" prevailed in which Portuguese
colonies were either taken by force or burned to the ground lest they compete
with nearby Dutch establishments. Much
of this action occurred in areas which, according to Wallerstein (1974), were still
external arenas outside the boundaries of the European world-economy. Plunder and pillage have become somewhat less
frequently employed by core powers over the centuries and this may partly
account for the rising success of peripheral resistance movements in obtaining formal state sovereignty.
A time series analysis
of peripheral resistance has been carried out by David Kowalewski (1987) who
has coded the existence and success of revolutionary movements in 34 peripheral
and semiperipheral countries between 1821 and
1985. Kowalewski demonstrates
the existence of an upward trend over time in revolutionary activity in the
periphery, and this is shown to be independent of the secular increase in the
amount of coverage of peripheral events in the news media from which activities
are coded. Kowalewski's results also
show that there has been no accompanying trend in the amount of success of
revolutionary movements over the same time period. He surmises that peripheral revolutionaries
are "making more revolution now but perhaps enjoying it less."
Kowalewski also
examines propositions which link the hegemonic sequence and the Kondratieff
wave to peripheral resistance with his data on revolutions. His indicator of the hegemonic sequence is a
series of periods designating phases of the world-leadership cycle as defined
by Modelski (1978).
Modelski'speriodization into phases of global power (1822-48),
delegitimation (1849-73), deconcentration (1874-1913), global war (1914-45),
global power again (1946-73) and delegitimation (1973-) is both conceptually
and operationally
controversial. In any
case Kowalewski does not find any striking empirical relationship between these
periods and changes in the level of revolutionary activity or success.
Kowalewski's analysis
does, however, demonstrate a relationship between revolutionary activity and
the Kondratieff wave. Revolutionary
activity is much more likely to occur during three upswings of the Kondratieff
wave than during four downswings.
Revolutionary success was not found to be regularly related to the
phases of the Kondratieff wave. Further
research needs to be done using better measures of resistance and better
measures of world-system cycles before we can know about the causalities
involved, but the surmise that periphery-wide waves of resistance exist and are
related to larger world-system processes is supported by Kowalewski's findings.
Protectionism and the Colonial Regulation of Trade
Another set of related hypotheses about cycles in core/periphery
relations refer to the political regulation of trade through colonial
monopolies and protectionism. We know
that Johan DeWitt, the seventeenthcentury mayor of Amsterdam, wrote pamphlets
about the universal benefits of international free trade; and Hugo Grotius, the
Dutch legal philosopher, formulated the doctrine of the free use of the seas
(which became the basis for the international law of the sea) during the Dutch
hegemony. At this same period the
English were becoming more nationalistic, expelling Jewish merchants and
attempting to protect their woolen textile industry against cheap Dutch
imports.
Later, during a period
in which competitive advantage begun to turn in
favor of the British, Adam Smith developed a theory of market-regulated
exchange in which the general welfare would be maximized by the abolition of
state interference in economic transactions.
Somewhat later, during the industrial boom based on cotton textile
production, Cobden and Bright
campaigned for British and international reduction of
tariffs. And the Americans followed
suit. After remaining protectionist
during nearly its whole period of upward mobility in the world-system, the US
finally became the advocate of free trade after World War II, establishing the
GATT agreements.
The above narrative
indicates that hegemonic core powers like free trade, while those with less
market power most often favor protection.
About this there is little disageement.
Stephen Krasner (1976) has added that smaller core powers that have
small home markets and are dependent on imports also favor free trade, although
Dieter Senghaas (1985: 23) points out that only Switzerland and the Netherlands
have maintained a continuous free trade policy.
It has likewise been noticed that peripheral capitalists, who are
exporting goods to the core and importing from the core, also favor free trade, although here
the matter of colonial monopoly sometimes cuts a different way. English sugar planters in the Caribbean were
often unable to beat the prices of competing operations in Brazil or on the less
soil-depleted Caribbean islands. Thus
they lobbied for colonial quotas and price supports. US cotton planters exporting to England were,
on the other hand, staunch defenders of free trade.
Again we see that
different groups have different interests, and the particular way in which each is inserted into
the complex mosaic which is the world-system is going to weigh heavily on the
decision to support protectionism or free trade. Stephen
Krasner (1976) examined three indicators of openness versus closure in the
world-system: protectionist taxation by
states; the ratio of the average size of the national domestic market to
foreign trade; and the degree of concentration of trade within regional areas
or colonial empires. On the basis of
these three features he concluded that the structure of trade and political
regulation of trade has alternated between periods of relative openness and
periods of relative closure. Examining
the nineteenth and twentieth centuries, he finds five periods (Krasner, 1976:330):
Period I (1820-1879): Increasing openness tariffs are generally lowered; trade
proportions increase. Data are not available for trade patterns.
However, it is important to note that this is not a universal
pattern. The United States is largely
unaffected: its tariff levels remain high (and are in fact increased during the
early 1860s) and American trade proportions remain almost constant.4
Period II (1879-1900): Modest closure tariffs are increased; trade proportions
decline modestly for most states. Data
are not available for trade patterns.
Period III (1900-1913): Greater openness tariff levels remain generally unchanged;
trade proportions increase for all major trading states except the United
States. Trading patterns become less
regional in three out of the four cases for which data are available.
Period IV (1918-1939): Closure
tariff levels are increased in the 1920's and again in the 1930's; trade
proportions decline. Trade becomes more
regionally encapsulated.
Period V (1945-c.1970): Great openness tariffs are lowered; trade proportions
increase, particularly after 1960.
Regional concentration decreased after 1960. However, these developments are limited to
non-communist areas of the world.
To this we may add a sixth period following 1970 in which
tariffs are again rising.
As mentioned above, a
more recent study by Tim McKeown (1983) disputes Krasner's argument that the
decline of protectionism among core states is due to the action of the
hegemonic state. McKeown shows that the
British state did not aggressively pursue a policy to induce other core powers
to lower their tariffs. Nevertheless the
hegemons adopt free trade earlier and keep it longer than competing core
states. Suzanne Frederick's (1987)
critique of Krasner and the other theorists of "hegemonic stability"
argues that free trade breaks out during a period of instability following the
beginning of the hegemon's decline. Certainly
the motives of the hegemon and the competing powers differ, and perhaps we
should pay more attention to the timing, location and degree of changes in
protectionism than Krasner has done. But
Frederick's conclusion about the "instability of free trade" is
wholly dependent on her use of Modelski's dating of the power (hegemonic)
sequence, and thus suffers from the same defect as studies described above.
The idea of a cycle of
the core/periphery exchange structure from bilateral to multilateral exchange
would predict a shift back toward a more bilateral pattern of trade in the
recent period since 1970. This should
show up in rising levels of export partner concentration (the percentage of
exports going to a single other nation).
Table 13.2 shows evidence to the contrary, however.
Except for the OPEC nations, every category of countries experienced a
decline in export partner concentration between 1970 and 1980, indicating that the international trade
matrix is continuing to become more multilateral in this period.
(Table 13.2 about here)
It may be that the increase in OPEC partner concentration
is due to heightened levels of
competition among core powers for access to oil exacerbated by theOPEC oil
cartel. This probably encourages a more
bilateral politically determined structure of exchange in the world oil market.
While some of
Krasner's periods correspond with well-known phases of the Kondratieff, others do not; for example
the long period between 1820 and 1879.
It is, of course, possible that variations of smaller amplitude occurred
within the long period in the middle of the nineteenth century, but the current
crude level of measurement is inadequate to detect them. A more quantitative approach to the
measurement of the average height of protectionist barriers is desirable, and
also feasible.
Core/Periphery Cycles in the Terms of Trade
Raul Prebisch's (1949) important work on the unequal returns to
gains in productivity in the core and
periphery emphasized the longterm deterioration of the terms of trade for
peripheral exports. However, Prebisch's
claim that the terms of trade for peripheral exports always decline has been
disproven. Michael Barratt-Brown (1974)
first presented evidence showing cyclical variations in the terms of trade
between core and periphery, and this has recently been further supported by
data analyzed by Paul Bairoch. Bairoch (1986:205-8) finds that "between
the 1870s and the 1926-29 period the terms of trade for primary products
relative to manufactured goods improved
by some 10-25 percent..." This is
contrary to the results of the League of Nations study by Folke Hilgerdt which
Prebisch and, much later, Rostow
(1978:98) used as evidence of changes in the terms of trade. Bairoch argues that the main problem with the
League of Nations study is that it employed only British data on import and
export prices and the import prices include a large component of transport
costs, which were falling throughout the period. Bairoch also notes that export prices of
British manufactures were rising more rapidly during thisperiod than those of
other competing core powers (due to falling productivity in Britain), which
also biases the terms of trade figures when only British prices are used. Bairoch goes on to observe:
The fact that the
terms of trade, or in more precise and 'technical' terms the net barter terms
of trade of less developed countries
have improved, does not mean that this is necessarily a positive
development. It would have been so if
this had been accompanied by a rise in wages and in other incomes as has
happened in the developed countries.
While the real wages of primary-goods producers in the Third World
remained stagnant between the 1870's and the 1920's the real wages of the
producers of manufactured goods in the developed world increased by some 100-160
percent in the same period. This implies
that in 1926-29 an average Third World worker could buy with his average wage
10-25 percent more manufactured goods than his grandparents could around 1875,
while an average worker in the developed world could buy with his average wage
80-130 percent more primary goods originating from the Third World than had
been possible for his grandparents. In
more technical terms, this means that
the factorial terms of trade for primary
goods from the Third World declined. (1986:206)
Bairoch presents a table (1986:207; table
13.6) which shows that the net barter terms of trade of the peripheral
countries improved from 1938 to 1950-54
and then deteriorated until 1960-64, then improved until 1970 and then
deteriorated again until 1983 when the data series ends. Excluding the major oil exporting countries,
the terms of trade of the peripheral countries tend to vary inversely with
those of the core countries until 1970.
After that the terms of trade of both core and peripheral countries
worsen together and inversely with those of the oil exporters.
Bairoch notes the
paradox that the post-World War II worsening of the peripheral terms of trade
coincides with the decolonization and political independence of many of the peripheral
countries. He attributes this to a number of causes: the slowdown in demand and
increase in supply of many primary products; the development of core-produced
synthetic substitutes; tariffs on the importation of some tropical goods; and technological progress that has reduced the
input
coefficients of
raw materials in manufacturing industry and, last but not least, what is called
the Singer-Prebisch thesis. This thesis
suggests that due to weaker organization, the unequal relationship between the
developed and the underdeveloped worlds leads to a situation where, in the case
of primary products, the gains in productivity are translated into a decline in
prices, while in the case of manufacturers, those gains are translated into
higher salaries and profits. The irony
is that, to a certain extent, independence could mean a freer hand for big
purchasing companies to press for lower prices since, in such a case, the local
social situation has no effect on the developed country. (1986:207)
Bairoch also mentions the possibility that part of the terms of
trade deterioration since the 1950's could have been due to productivity
increases in tropical agriculture which were greater than the increases in the
productivity of manufactured goods.
While evidence of
cyclical variations in the net barter terms of trade is convincing, the variations noted do not
seem to correspond in any simple way with other world-system cycles. The improvement noted in the late nineteenth century occurs during the upswings
and downswings of Kondratieffs and during the generally declining phase of
British hegemony, while the variations noted after World War II occur during
the golden age of US hegemony and the beginning of its decline.
Hopkins and
Wallerstein (1979:496) outline a more complex model which posits a set of
shifts in the demand for core (high wage) and peripheral (low wage) products
over a pair of Kwaves. The hypothesized
changes in core/
periphery terms of trade as applied to the two twentieth century
Kwaves do not correspond well with the changes found by Bairoch.
What is still at issue
is the matter of the factorial terms of trade, which takes into account changes
in wage differentials. Data are not
presentlyadequate for the detection of cycles.
But according to Bairoch the factorial terms of trade worsened during
the period around the turn of the century at the same time as the net factor
terms improved. It would appear that
this worsening probably continued after World War II when the net factor terms
declined. This is further indication of
a secular trend in the direction of an increasing relative gap in the
purchasing power of core and peripheral workers; both groups are increasing
their purchasing power, but at different rates.
Cycles of Capital Exports to the Periphery
Capital exports and colonization are very different kinds of
expansion toward the periphery, yet the model presented above hypothesizes that
both of these are related to the hegemonic sequences and the Kwave. Recent work by Ulrich Pfister and Christian
Suter (Suter, 1987; Suter and Pfister, 1986; Pfister and Suter, 1987) focuses
on the cyclical features of the international financial system which links the
core and periphery. Pfister and Suter (1987)
posit a theory which links the Kondratieff wave to capital exports from the
core, capital imports by peripheral countries, and international financial
crises which are triggered by defaults in the periphery. Suter (1987) also hypothesizes that international
financial crises will be more severe during multicentric periods of the
hegemonic sequence because no single hegemonic financial center is able to
perform the role of "lender of last resort." He examines data on core capital exports, peripheral
imports of capital, and financial crises to determine the cyclical nature of
these and their relationship to Kuznets cycles, Kondratieff cycles, and the
hegemonic sequence.
Summarizing the theoretical argument made by
Pfister and Suter (1987), Suter (1987:6) writes:
...capital flows
into the periphery occur in later stages of the long wave when markets of the
core are saturated and profit rates begin to decline due to the exhausted
innovative potential. Since returns on
equity investment are low, capital tends to flow into the more profitable
financial assets. This rising supply of
international liquidity meets a corresponding demand from peripheral countries,
which have not been fully integrated yet into the development process of the
long wave. Structural constraints, however, such as short-falls
in export earnings, low returns on external capital, and consumptive uses of
external resources due to legitimatory pressures on governments, cause low
income effects of imported capital. This
means that the profits from investments
financed by external resources do not match debt service
obligations linked with these capital flows.
Thus, peripheral borrowers tend to incur large debts towards the end of
a long wave. As a consquence, the
international financial system is over-extended and increasingly prone to
disruption and crisis.
Suter notes that the
above theory, along with the long wave explanations of Mandel (1980) and Mensch
(1978), predicts a clustering of international financial flows from the core to
the periphery towards the end of a Kondratieff upswing and during the beginning
of the downswing phase. In contrast,
Rostow's (1978) theory of Kondratieffs as driven by changes in the terms of
trade between raw materials and manufactured goods predicts that capital
imports of sovereign borrowers occur during the upswing of the Kondratieff
price cycle.
Suter and Pfister's
(1986) findings about cycles of capital exports from the core are based on
general surveys of the issuance of government-backed bonds. They conclude that there are seven boom
periods of capital exports since the beginning of the nineteenth century (Suter
and Pfister, 1986:table 1). These occur
in the early 1820s, the late 1830s, the 1860s and early 1870s, the 1880s, the
decade before World War I, the 1920s, and the 1970s.5
Suter (1987) also
presents two case studies of cycles of capital imports, one for
Suter argues that the
patterns of capital exports and capital imports that he observes can be
explained by understanding different periods as phases of the expansion process
of the capitalist world-system. The
first half of the nineteenth century does not reveal cyclical patterns of
international finance because, he argues, the mechanisms of international
finance produced by industrialization are as yet insufficiently
developed.6
During the second half
of the nineteenth century Suter finds both Kuznets and Kondratieff waves of
development affecting international financial cycles. He interprets this as involving two different
processes. The Kuznets cycle a 15 to 25 year business cycle allegedly operated only within the sphere of
the growing "Atlantic economy" (an inner circle of recent settlement
in which population migrations, housebuilding, and railroad building booms are
important). Brinley Thomas's (1954)
analysis of the inversely related British and US Kuznets cycles was the
original work which set off a tradition of research on inversely related
economic cycles. While Rostow (1978)
declared that the Kuznets cycle has no general validity, being a process which
is revealed only under special circumstances, Suter disagrees. He argues that the Kuznets cycle existed
within the expanding inner circle of core capitalistgrowth, and he claims that
evidence of it can be seen not only in the late nineteenth century but also in
the interwar period. In the postwar
period only the Kondratieff wave is evident in international financial
cycles. Amin (1974) and others have
argued that the shorter cycles have been dampened by the counter-cyclical
Keynesian economic policies employed by states, but that the longer cycle
remains because it flows from an international dynamic not controllable by any
state.
Cycles of Financial Crisis
Suter also studies debt crises, defaults, and international loan
reschedulings. His chart 6 (Suter,
1987:21) graphs the number of countries in default, the number of occurrences
of default from 1815 to 1950, and the number of loan reschedulings from 1950 to
1984. This graph reveals that debt
crises are highly correlated with Kondratieff waves, with high default levels
occurring during downswing periods, as the theoretical approach taken by
Pfister and Suter predicts. The most
interesting finding, however, is the combination of the analysis of the cycles
of debt crises with a regional specification of the countries experiencing the
financial difficulties (Suter, 1987:22, table 2). This shows that the areas accounting for the
major financial crises shifted from period to period while the cycle of debt crises
varied regularly with the Kondratieff wave of the world economy. In the crisis period from 1825 to 1840 the
areas most affected were nine US states, Portugal, Colombia, Venezuela,
Ecuador, Mexico, Spain, Brazil, and Greece.
Suter points out that not only do the regions involved vary, but the
particular combination of factors which contributed to the debt crises
varied. In the 1820s the decolonizing
Latin American countries employed their external financing to make
anti-imperial war on
In the debt crisis
period from 1875 to 1882 the major defaulting
countries were Turkey, Egypt, Peru, Mexico, eleven southern US states,
Tunisia, Uruguay, Costa Rica, Honduras, Venezuela, and Colombia. Some of these were the same countries which
defaulted before, but a number of others were new. In the period from 1932 to 1939 the
defaultors were
Suter also hypothesizes that international
financial crises will be more severe during a period of multicentricity in the
hegemonic sequence, and this is borne out by the numbers for the 1930s compared
to earlier crises and morerecent ones.
Pfister and Suter (1987) have also argued that the ability of
the contemporary international financial system to avoid (or
postpone) collapse is due to the increased ability of international financial
institutions to coordinate debt rescheduling.
Suter contends that the greater integration of international
organizations such as the IMF frees the international financial system to some
extent from dependence on a hegemonic core power to perform the role of
supplying world money as the "lender of last resort." He also points out that in earlier crises upwardly mobile
semiperipheral countries were often able to weather financial crises without
collapsing. Some of the current countries having debt problems (i.e.
Conclusions
What can we conclude about cycles in core/periphery
relations? Of course more work needs to
be done, as suggested several times above.
We still have much to do at the level of discovering the temporal
relations among variables. Measurement
of variables needs more work in order to specify cyclical
variations. It is much
easier to determine overall trends. The existing research confirms that there are
indeed cyclical variations but we are still uncertain about the true relations
in time among many of these features.
The core/periphery trade network, including trade partner concentration,
tariff barriers, and colonial trade monopolies is found to vary from period to
period, and is related in a general way (but not simply) to the hegemonic
sequence. The waves of decolonization,
one form of resistance from the periphery, occur during the upswing and peak of
the hegemonic sequence when the hegemon supports the break-up of the colonial
empires of the other core states. The
Kondratieffwave is less tightly related to the hegemonic sequence than we have
earlier supposed. Waves of colonization
are likely to be related to the Kondratieff/
core war severity cycle, but we don't know the exact timing of
the relationship. Flows of capital from
the core to the periphery are related to the Kondratieff wave, occurring
approximately at the peak of an upswing, and international financial crises
which are triggered by defaults in the periphery occur during Kondratieff
downswings. As noted in note 5, visual
comparison of the graphs of waves of capital exports with the waves of colonial
expansion suggests that these two types of expansion are not simply related to
one another in time.
Evaluation of the
causal model presented in figure 13.2 is impossible at this point. Our review of the few studies which have
tried to examine core/ periphery cycles does, however, suggest some
changes. Different sorts of expansion (capital
exports versus colonialism) may have somewhat different causes, as might the
three indicators of core/periphery trade structure used by Krasner.
Protectionism, trade partner concentration, and the average ratio of
domestic production to foreign trade probably have somewhat different
causes. And "core conflict"
must be differentiated to distinguish between economic protectionism and
warfare.
It is premature to
produce new causal models before we have done
additional data-gathering and made more progress on the temporal
relations among variables. A summary of
the main conclusions reached in Part III on the core/periphery hierarchy is
contained on pages 1416 of the Introduction.
We now turn to a consideration of the metatheoretical assumptions and
empirical methods used in world-system
research.
Part IV: Metatheory and Research Methods
The shift to a world-system frame of reference necessarily
raises questions of epistemology and the philosophy of science. Is there a stance vis--vis these problems
which is most appropriate for understanding world-systems? Are certain methodologies ruled out? What difficulties do we encounter as we
attempt to build theories about world-systems and to subject those theories to
evidence, and how can these be overcome?
Only a few authors
have directly addressed these questions in the terms just used, but there is,
of course, a wider literature which is relevant. Most of the controversies which have rocked
social science since its beginnings about how we should scientifically study
human societies are germane, and so are the particular twists and turns,
attacks and defenses, which have been so evident in recent years.
Simply because there
is now only one world-system the method of historical interpretation has seemed
preferable to most world-system scholars.
In the following chapters I will argue that the comparative method as it
has developed within modern empirical social research provides a valuable set
of tools for world-system studies, notwithstanding certain problems which
emerge when we utilize these tools. I
will discuss the claims of authors who have disputed the value of particular
kinds of comparison, e.g. quantitative cross-national studies, and will
describe solutions to the problems encountered when we use comparative methods
to study a single world-system.
There is now only one
world-system. Its spatial scale is grand
and variations within it are striking.
Thus many world-system studies have focused on the particular qualities
of periods or places, and have employed themethod of historical
interpretation. While these studies are
valuable additions to our knowledge and rich sources of hypotheses about the
world-
system, their theoretical contributions can only be evaluated by
the use of comparative research designs.
Of course the very
systemness of the now-global world-system is itself a problem, and a research
question. My project, only begun in this
book, is to formulate and test models of the modern world-system which can explain
its past and predict its future, and hopefully be of use in transforming it
into a more humane set of social relations.
This project assumes that the world-system is fairly systemic. You can not model something which is so
ephemeral that its structure and dynamic nature change very rapidly or are
infinitely complex.
Chapter 14 on theory
construction discusses the nature of world-system theory and criticisms of the
structural approach. It describes a
continuum of positions ranging from ideographic historicism to completely
ahistorical nomothetic theory, and it argues against the tide of recent attacks
on structuralism. The ontological status
of the world-system is compared with that of other units of analysis in the
social sciences. It is pointed out that,
contrary to the common assumption, levels of abstraction are not necessarily
related to the spatial scale of units of analysis. Three types of world-system models are
described and the problems of concept formation, dialectical modeling, and the
implications of structural theories for political action are discussed.
Chapter 15, on
methods, confronts the claim that we need a new method of research in order to
study world-systems. It discusses the
logic of comparative method, case studies, and quantitative analysis. It describes Galton's
problem the problem of
the independence of cases and its
solution, thetaking account of non-independence, including contextual and
relational variables, within the theoretical model. Then problems of applying the generalized
logic of time series analysis to the whole world-system are confronted. Cross-national, transnational and multilevel
research designs are considered, as are the prospects for comparing large
numbers of historical world-systems.
Chapter 14:Theory Construction
Recent
attacks upon structuralism and formal theorizing in the social sciences have
occurred within the context of a global ideological shift toward the
right. The renewal of nineteenth-century
glorifications of the free market and entrepreneurial profits by neo‑conservatives
has been surprising, but even more curious is the Left's drift to the
center. This has occurred in many forms
─ the emphasis on modernizing the state sector and increasing its
productivity, the shift toward markets and profit incentives in the
"socialist" states, and Eurocommunist support for nationalism, state
capitalism, and austerity. We know that
academia and world politics are not separate islands, and so it is tempting to
assert a connection between the renewed emphasis on methodological
individualism and the shift away from analyses of structural and institutional
forces in both political ideology and social science.
Exactly what this connection is I cannot
say. The Zeitgeist seems to
affect us all, and all of our activities.
I am not arguing that all those who have attacked Marxist structuralism
are tools of the capitalists. Rather I
want to make the case that the building of a formal structural theory of
capitalist accumulation remains an important goal for social science and a
potentially valuable contribution to the creation of a more humane, egalitarian,
and balanced world society, and one that is less likely to send our collective
experiment with life up in smoke.
In
my view the structuralism of Louis Althusser and his students has been attacked
on the wrong grounds. The effort to
formulate a structural theory of capitalist development, now understood at the
world‑system level, is a necessary effort if we are to collectively get
control of this headless horseman, this
driverless (or multidrivered) behemoth
which is almost certainly hurtling us all toward the precipice.
A Defense of Theory
E. P. Thompson (1978) has made perhaps
the most influential attack on
Althusserian structuralist theory in his essay "The Poverty of
Theory." Thompson argues that men
and women are more than simply occupants of structural positions, merely
agents of social forces. They are
historical actors who actively negotiate and struggle to create their own lives
and to change (or maintain) existing social institutions. A structuralist theory is condemned because
it implies a mechanical view of human beings hopelessly caught and scuttled
along by social forces beyond their control.
This debilitates purposive action and promotes fatalism. Structural theories are also themselves
ideological weapons by which power is legitimated, and this power is used to
oppress and exploit people. Stalinism
and the Third International are suggested as a pertinent example.
As a substitute for structuralism Thompson
puts forth his own method of historical analysis that emphasizes the authorship
of culture by individuals and classes engaging in struggles to survive and
create a better world. The historical
contingency of outcomes is emphasized, as are the elements of consciousness and
intentional organization.
Thompson
has been joined by a large number of other academic Marxists who have made
similar arguments with respect to particular subject areas, or in general
philosophical and methodological terms.
The critique of Marxist
structuralism has succeeded in
mobilizing a broad rejection in favor of historicist, voluntarist,
particularistic (area or local) studies.
The growing popularity of the deconstructionist critique, which
demolishes all theories as textual tools
of political power, is also quite evident.1
Here
I will formulate my own critique of Althusserian structuralism, and I will
argue for a renewed effort to build a formal theory of the deep structure of
capitalism, rather than a rejection of all theories.
The Historicist/Structuralist Continuum
First I want to pose the existence of a
continuum between two extreme metatheoretical positions in the philosophy of
social science and locate various intermediate positions on that
continuum. The end points are what have
been called nomothetic versus ideographic analysis. There is a long history of contention within
all the social science disciplines between these two very different
stances. Indeed many of the disciplines
exhibit a somewhat cyclical pattern of variation in terms of the popularity of
approaches along this continuum (e.g. Harris, 1968). Nomothetic analysis attempts to formulate
general laws which explain the regularities, patterns, and forms of change
exhibited by a phenomenon. In social
science the most completely nomothetic formulations assert that a single
ahistorical model can account for all human social systems, large or small,
primitive, ancient, or modern. Talcott
Parsons and his followers represent this position, contending that the idea of
social structure (composed of normatively defined statuses and relations) and a
list of systemic necessities can be usefully applied to two‑person dyads,
small groups, organizations, national societies, and global systems.
Ideographic
analysis, the other end of the continuum, focuses on what is unique about a
person, a time period, or a locale.
Rather than asking what cases have in common, it stresses their
differences. It paints in rich detail,
endeavoring to evince the mentality, both cognitive and affective,
of a historical setting. This is what I am calling historicism.2
While this effort constitutes a valid and
valuable exercise in the humanities, many of its proponents assert that a more
generalizing approach imposes a false philosophy of physical science on human
beings. Beings which can alter their own
behavior intelligently and in ways difficult to predict are not billiard balls,
it is contended. Social scientists
respond that even "thick description" implicitly makes comparisons
and utilizes generalizations in order to make its narrative statements
intelligible.3
Perry Anderson (1980) has formulated a
valuable response to E. P. Thompson's attack on structuralist theory which
rescues much of what is valuable in the Althusserian apparatus while acknowledging
the dialectical nature of structural
determination and voluntaristic action.
Anthony Giddens (1979) has addressed this issue in great detail and his
work is perhaps the most systematic effort to resolve the problem, but one
quickly tires of the discussion of structure, agency, and action at a purely
abstract level. I am not agreeing with
the historicists that every theory or piece of research must contain people,
places, and events, as Charles Tilly (1984) has argued, but the absence of a
specified context behind Giddens's discussion leaves one wondering. The discussion of structures, agency, power,
ideology, and social change at such a completely abstract and general level
sounds like so much talk. But perhaps I
am only revealing my own predilection for a stance near the middle of the
continuum.
In
between the extreme poles of total generalization and pure description are an
infinite number of possible combinations of the two, not only with the
quantitative mix varying, but also with the scope and nature of content
assigned in different ways. Many
sociologists have argued that "grand theory" is vacuous (e.g. C.
Wright Mills, 1959). On the other hand,
"theories of the middle range" applied to particular contexts are
alleged to be more scientifically valid and socially useful (Merton, 1957).
Fernando
Henrique Cardoso and Enzo Faletto (1979) introduce their important study of
dependency in Latin America with a description of their "historical─structural"
method, which emphasizes the differences among several qualitative types of
dependency situations and the possibilities for maneuver within the structural
constraints which emanate from the core (see also Bennett and Sharpe, 1985:9‑13).
Immanuel Wallerstein has characterized his
focal unit of analysis as "historical systems," a term which neatly
captures the dialectical antinomy
between ideographic and nomothetic analyses. Wallerstein's approach emphasizes the
interaction between the historicity of socio‑economic systems and their
deep structural or essential elements.
In this he is similar to Marx.
Marx criticized the ahistorical generalizations of classical political
economy which ignored the unique qualities of different modes of production. Assumptions about timeless human nature, he
pointed out, obscure the social origins of institutions and the qualitative
transformations which occur in the development of human societies. Rather than attempting to model all socio‑
economic systems, Marx focused on
capitalism, an historical mode of production with its own unique logic of
development and contradictory tendencies.
But whereas Marx attempted to formulate his deep structural model
explicitly in the volumes of Capital, Wallerstein's remarks about theory
and his somewhat casual approach to theoretical specification reveal a semi‑phenomenological
attitude. The statement of editorial
policy in the
Althusser and his followers have been
criticized for excessive structuralism and the allegedly associated sins listed in the discussion of
E. P. Thompson above. But I do not think
these are the main problems with Althusser's approach. A structural theory of the tendential laws of
capitalism, rather than implying a deterministic universe in which political
action is necessarily futile, is rather
a guide to the weak links and openings for positive socialist politics, or it
ought to be. There is no contradiction
here because no one claims that all social action is determined by
structural forces. A Marxist structural
theory, as Engels (1935) long ago pointed out, is an effort to say how
historical forces are moving and to indicate what is possible for
"scientific socialism" within that context. Both structural theories and historical
accounts can be used to legitimate political power. A good (or true) theory can be used for bad
ends, but this possibility is not an argument for ripping down all theories.
Althusser
and Balibar's approach (Balibar, 1970)is valuable precisely because they make
the distinction between:
1 the mode of production ─ an abstract level of analysis which
specifies essential structural tendencies of a
socio‑economic system, and
2 the level of the social formation ─ a concrete and directly
observable level of historical events and social institutions which may contain
more than one mode of production as well
as more purely conjunctural features which combine to determine historical
events.
The distinction between the deep
structure and the conjunctural level of historical events is contained within
the overall analysis. Althusser and his
students are more explicit about the content of the deep structural level, and
they put more emphasis on its causal importance than does Wallerstein or other
scholars who are closer to the historicist end of the continuum.
Of
course it is not only a matter of the relative emphasis of structure and
conjuncture, but also the substantive content of the distinction. Among the structures, Parsons makes norms and
values the master variables. Exactly how
one understands the distinction between "base" and "superstructure"
(or between essence and epiphenomena) is of great importance to the substantive
content of a theory. In chapter 1 I have
proposed a reformulation of Marx's model of capitalist accumulation which not
only changes the framework of analysis to the level of world‑systems, but
also argues that certain elements such as state‑formation, nation‑building,
and class formation (which Marx consigned to the conjunctural) should instead
be incorporated into the model of deep structure. Here I am not defending those substantive
theoretical decisions, but rather the prior decision to continue the project to
produce a structural theory.
My own criticism of Althusser et al. focuses,
not on the structuralist project but on the content of the theorization (see
chapter 1), and also on the failure of the Althusserians to concern themselves
with the confrontation between theoretical formulation and empirical
research. The philosophy of praxis
employed by the Althusserians defines the confrontation with the empirical
world in terms of political activity. I
do not agree that political practice is a substitute for systematic comparative
research designed to distinguish among contending theoretical
formulations. Rejecting systematic
comparative research as a bourgeois method, the Althusserians became mired in a
scholastic world of textual interpretation, rationalistic deductive analysis,
and political debate. Many of those
Marxists who were concerned with the real world of ongoing social movements
outside of the immediate context of French political debates turned away from
structuralism. Thus Manuel Castells,
himself an early proponent of Althusserian structuralism, embraced historicism
in the form of a kind of populist romanticism (see Molotch, 1984), throwing out
theory with the bath.
My
defense of comparative empirical analysis is contained in the following chapter. Here I only want to point out that deductive
reasoning, the logical exposition and critique of theoretical concepts and propositions, is only half of the process of
scientific theory production. The other
half is inductive empirical research.
The central tendency of American sociology has been the opposite error ─
research without theory. As Arthur
Stinchcombe (1968) and many others have pointed out, research can only
distinguish between theories which predict different things, and so it is
important to formulate contending theories in ways which make it clear what
they do and do not imply
about the empirical world.4 Admittedly this demanding ideal of how theory
construction ought to proceed is rarely followed, and I have not been able to
completely follow it in this book.
Nevertheless, archaic as it may seem, it is one of the main
justifications I offer for my theoretical effort, and for the review of
comparative research results.
Ontology
In many discussions with students and
colleagues in sociology I have discovered the curious assumption of a
connection between spatial scale and the level of abstraction. Many seem to assume that the world‑system
is abstract, whereas an individual person is concrete. I have also observed this error in the
published works of distinguished social scientists (e.g. Tilly, 1984:14). But on reflection everyone will admit that
there is no necessary connection between
size and abstraction. The sun is not
more abstract than the earth. I am not
more abstract than an ant. The world‑system
is not directly observable to the eye, and this may be part of the reason why
some people think of it as more abstract than smaller levels of analysis.
But neither is the earth visible as a whole to most eyes. And yet we would all agree that it, and many
other things such as atoms and the solar system, are concrete entities. The psychology of visual perception has
established that we employ ideas even in our perception of the immediately
visible. But this does not mean that
everything is equally abstract. Though I
need the concept "chair" to see the chair, I and most philosophers of
knowledge believe that a chair is more concrete than is (say) beauty.
In
social science most of us believe that individuals really exist and that they
are more concrete than classes or nations.
Since Lukács we have been cautioned to beware the reification of
society.5 But John W. Meyer,
a consumate structuralist, has pointed out that we more often reify the individual. As we all know from our undergraduate
courses, the self is socially constructed and biological individuals are
conceptualized very differently in different kinds of societies. But let us assume that individuals
exist. It does not follow that larger
levels of analysis are less concrete. At
the most concrete level the world‑system is composed of all (or nearly
all) of the people on earth, and the material interconnections (direct and
indirect) among them. It is large, but
it is not abstract, or not more abstract than other, smaller, objects of social
science analysis.
But
scale does affect the patterns that are observable. When we use a telescope we see very different
things than when we use the naked eye or a microscope, even when these are
pointed at the same "reality."
There is little point in arguing about which are the real concrete patterns
or objects. Rather, if we are writing
history we may choose to focus on one or another scale of analysis for its own
sake, as we choose between different aesthetic styles. But if we are making science we will wish to
understand the causal processes by which patterns change, and we will ask which
level of analysis accounts for more variation in a designated outcome variable
(explanandum). This last effort may
require the study of different levels of analysis simultaneously, a matter that
is discussed in the next chapter.
Models of the World‑system
World modeling is an enterprise which
has proceeded largely unconnected with the world‑system perspective. Here we will comment on the possible
relations between these two projects and designate three types of models which
are part of the effort to construct a theory of world‑systems.
World
modeling is a data‑gathering and simulations effort by different groups
of scholars. Much of this work is
valuable because it focuses on the world economy as a whole as the unit of
analysis and it uses empirical data to forecast trends. The theoretical models used to created the
simulations have varied greatly from project to
project, but none of the main projects has utilized the kind of theory
developed in this book, a theory which uses Marx's accumulation model of capitalist development
as its starting point. I would concur
with Patrick McGowan (1980) that, despite the very different paradigms employed
by world modelers and the world‑system perspective, these two projects
can be useful to one another.
Levels of Specification
I would like to distinguish between
three levels at which we may specify models of the modern world‑system. The first can be called a descriptive
model. This type of model specifies the
relationships in time between the several cycles and trends which are features
of the world‑system that vary over time.
Such a model is implied by the discussion of cycles and trends in
chapter 2 and several authors have presented such temporal models (e.g.
Chase-Dunn, 1978:170; Hopkins and Wallerstein, 1979:496-7). This sort of model does not specify causal
relations among variable features.
Rather it simply predicts regular
temporal relations among different
variables. This descriptive level of
analysis is rarely ever presented without some discussion of causal
relations. It is a valuable theoretical
effort in its own right to make explicit hypotheses about temporal
regularities because it facilitates the posing of empirical questions.
The
second type of model is a specification of the causal relations among variables
which are features of the whole world‑system or features of subunits,
such as zones, nation‑states, etc.
An example of such a model is given in figure 13.2 and a series of
similar examples are contained in McGowan's (1985) critique of Bergesen and
Schoenberg's (1980) study of cycles of colonialism. Procedures for testing these causal models
are discussed in the following chapter.
Their value is that they make explicit our arguments about what causes
what, and therefore they can be helpful in evaluating different theoretical
arguments about processes of the world‑system.
A
third type of theoretical specification is one which formulates a theory of the deep structure or the main engine
which drives the world‑system.
Marx's theory of the capitalist mode of
production is such a model. It posits
the existence of structural tendencies which can account for the long-run
dynamics of growth and reproduction of a socio‑economic system. The specification of such a deep structural
model may be formalized in a number of different ways. Part of Marx's model has been formalized as
an axiomatic theory of logically related statements by Nowak (1971). Morishima
(1973) has mathematically specified major aspects of Marx's accumulation
model. Models of the deep structure need
not be completely formalized, but formalization makes assumptions clearer and
makes it easier to see the empirical implications of a set of central
theoretical statements.
It
should be pointed out that the language of deep structure versus surface‑level
appearances does not require the nominalist philosophy of Hegelian
idealism. We need not assume that there
is "really" an unobservable essence beneath the complexities of
empirical appearances. The model of the
deep structure is like a map. It is a
simplification of the territory that helps us get where we want to go, to
explain and predict as much as possible.
It is desirable that the map be as simple as
it can be, while still remaining helpful.
Here again we find the continuum between historicism and completely
ahistorical theory. Historicism copies
the territory in rich detail, without an attempt to simplify. On the other hand, the map of completely
general ahistorical theory is so simple that only the most analytic features
are drawn, e.g. the grid of longitude and latitude. These features will provide a rough guide for
all locations, but will not provide sufficient information to be helpful for
most particular purposes.6
Thus it is not so much a question of the real existence of the deep
structure but rather the usefulness of the model for explanation, prediction,
and action. And in this sense there are
varying degrees of truthfulness rather than an absolute truth.
Concept Formation
Cardoso (1977) argues that concepts
should not be rigidly defined and/or converted into one‑dimensional
variables, because the social reality being studied is a dynamic, contradictory
reality which is oversimplified by such precision. This objection may be partly based on Marx's
notion that theoretical concepts in the social sciences should be reflective
of the relational and contradictory character of social processes themselves
(Marx, 1973). I think that this is an
important methodological idea, but it should not prevent us from making
tentative definitions clear in order to see how they may be useful in
explaining empirical reality.
Operationalizing a concept does not permanently commit us to either the
definition or the particular indicator we employ to measure it. Cardoso is right to point out that we should
be aware of the assumptions behind converting a concept like dependency into a
one‑dimensional variable. Indeed,
the evidence of cross‑national research confirms that dependency is a
multidimensional phenomenon (see chapter 9).
Much
of the concern about the implicitly static and mechanistic nature of formal
causal models (e.g. Bach, 1977) seems to be the result of a misunderstanding
of the logic of causal analysis. All non‑experimental
research is an attempt to infer underlying causal processes from data over
which we have little control. The type
of variables used (qualitative, metric, linear or curvilinear, multidimensional
or not) and the logic and nature of the causal relations that are implied in a
particular model are up to the theoretical imagination of the researcher. It is true that the most commonly tested
models assume one‑way causation and linear relations among variables, but
these assumptions are by no means necessary to causal modeling.
Dialectics and Contradiction
Cardoso (1977) and Bach (1977) argue
that conventional causal imagery should not be applied to dependency and world‑system
processes because these processes are
dialectical and contradictory. Presumably
these authors are making a claim about objective reality; if we can be clear
about what is meant by a dialectical process, there is no inherent reason why a
dialectical model cannot be specified and tested.
Many
students of social structure prefer to use dialectics as an heuristic aid to
thinking about processes of social change.
As such, the general notions of contradiction, opposition, and
qualitative transformation can be quite useful for interpreting complex
historical situations, and this heuristic aid is not at all incompatible with
causal propositions of a more conventional kind. When we assert that there is a negative
causal relationship between, for example, dependence on foreign investment and
economic growth (Bornschier and Chase‑Dunn, 1985), we do not deny the
interactive and reactive nature of the relationships between transnational
corporations, peripheral states, and peripheral workers. What we are asserting is that on the whole,
over many cases, in the long run, ceteris paribus, the more dependent a
country is on foreign capital the slower it is likely to develop
economically. The
possibility that there may be
exceptions, or that some countries may be able to successfully combine foreign
investment with a certain kind of growth, does not disprove the general
contention. Propositions of this kind
can easily be combined with a dialectical heuristic.
More problematic is the specification of
formal dialectical propositions within a model.
If we contend that a particular process or relationship is dialectical
it is possible to formally specify the meaning of this assertion. Although it is unpalatable to many
dialecticians, we can translate the notion of contradiction into conventional
causal logic. Thus contradiction may be
understood as:
─ causal vectors that affect a dependent variable
in opposite ways;
─ variables that affect one another (reciprocal
causation); or
─ a variable that negatively affects itself over
time (negative feedback).
It is true that most causal models in
social research employ only fairly
simple assumptions about feedback, reciprocal causation and interaction,
but much more complex forms of causation have been successfully modeled
with fairly standard mathematical
tools.
Most
serious dialecticians are not satisfied with such a simple trans- lation of the
central notions of dialects (contradiction, opposition, and qualitative transformation) into conventional
causal logic, however. They argue that
Aristotelian logic precludes the simultaneous existence of opposites, so a new logic and mathematics is
required. Work has begun to create a
formalized dialectical logic and mathematics (Rescher, 1977; Alker, 1982). This
type of work may eventually allow us to construct dialectical models that
clearly specify the meaning of opposition, contradiction,
synthesis, and qualitative
transformation in ways that make them empirically disconfirmable. This line of work was suggested by Erik
Wright (1978:chapter 1). He elaborated a number of "modes of
determination" which are particularly appropriate to the task of modeling
the causal structures of Marxian theory.
Critical Theory, Determinism and
Political Practice
Structuralist theory, formal modeling,
and quantitative comparative studies have been attacked by those who argue that
dependency and world‑system studies should expose and condemn the
structures of domination. Critical
theory and deconstructionism want to expose positive social science as
bourgeois ideology, and the argument is made that both the philosophy of
knowledge and the comparative methodology used in normal social science are
actually class‑based mystifications.
Cardoso (1977) contends that quantitative comparative research is
necessarily uncritical because it must wear the mantle of "value‑free"
objective science.
My
general defense of quantitative comparative methods is contained in the next chapter. Here I want to argue for the political value
of a renewed effort in the realm of structural theory. If we already had an adequate theory of capitalist accumulation and
sufficient understanding of the dynamics of world‑system processes and
the nature of the transformation of modes of production, then we could go on to
particular applications of this knowledge in social struggles. But the current "crisis of Marxism"
reveals that certain basic problems about the nature of capitalism and
socialism have not yet been resolved.
That is why it is politically sensible to exert effort toward the reformulation
of a positive theory of capitalist accumulation. If we could simply accept Marx's model and
apply it to the world of the twentieth century then we would not need to
produce a new (or revised) theory. A
structural theory, however, is more than a call to action. It is a model of how the socio‑economic
system actually works, its dynamics of growth and competition, and the inherent
contradictions which produce pressures for transformation.
Bach
(1977) and Cardoso (1977) argue that the construction of formal models implies
a deterministic view of development which ignores the voluntary efforts of
individuals and classes to resist and to alter the structures that are
exploiting and underdeveloping them. On
the contrary, models in social science are most usually probabilistic rather
than deterministic, thus to allow for the complexity and indeterminacy of human
behavior. Knowledge of the likelihood of
a social outcome given certain conditions is a potential contributor to human
freedom. We do not become more
determined by under-
standing the laws of nature or the
tendencies of social systems. On the
contrary, we become freer. For example,
models of dependency that show the size and nature of average effects on a
national economy caused by an increase (or decrease) in dependence on foreign
investment can be useful in helping policy‑makers in peripheral countries
avoid the negative consequences of exploitation by transnational corporations.
There
is also the question of the uses of causal models and quantitative research for
informing political practice. There are
many political organizations and movements that might make good use of the
results of comparative research on world‑system processes. But just as policy‑makers and planners
cannot apply models mechanically to every situation, so in politics (especially
in politics) practice is more an art than a science because of the complex and
conjunctural nature of the task. This
means that, as in medicine, the results of research must be used by informed and
artful practitioners.
As I see it the main use of a structural
theory of the capitalist world‑economy is its potential to help us
distinguish social changes and political
forces which reproduce capitalism from those which contribute to
its transformation, and to identify
"weak links" where political efforts at transformation can bear the most fruit. If the twentieth century is the beginning of
a period of transition to a socialist world‑system, but the extant "socialist" states are
functional parts of the capitalist world‑economy (Chase‑Dunn, ed.,
1982b), we need a clearly specified theory of
capitalism to help us distinguish newly emerging forms which contribute
to the transformation of the system from those which reproduce, further expand
and deepen it. And our new theory should
also have implications for the question of agency in the building of socialism.
Chapter 15: Research Methods
Many of the long‑standing
methodological feuds within the social sciences have been extended to the
question of the correct way to study
world‑systems. The
continuum of metatheoretical stances discussed in the last chapter generally
corresponds with a set of preferences regarding methodology.
I say generally because some theorists disapprove of empirical research
in any form, while some "raw empiricists" disdain theory. But among those who are willing to risk
contact with the messy empirical world there is a vast gap between proponents
of the case study and proponents of systematic
comparative analysis. As with
metatheory, there are many positions in between. Some researchers differentiate among
qualitative types. Some combine case
studies with quantitative analysis of a large
number of cases (e.g. Paige, 1975).
Charles Ragin and David Zaret (1983) contrast
Weber's individualizing approach with Durkheim's generalizing approach. Weber constructed ideal types in order to analyze the differences
between the Orient and the Occident.
Durkheim analyzed large numbers of cases in order to verify a general
model which was intended to apply to all cases.
Ragin and Zaret convincingly argue that both strategies are valuable
depending upon the goals of the study.
And they make a valiant effort to demonstrate the complementarity of these strategies. Adherents of these methodological stances
need not always engage in a confrontation in which the historians and area
specialists assert that it was different in
Either
of these strategies (individualizing or generalizing) can be usefully applied to the world‑system,
depending on what we are trying to accomplish.
A purely historicist approach would focus on the unique conjunctural qualities of the world‑system
during a particular period. Since there
is only one world‑system, the case study seems most appropriate. If it is
believed that structural models which seek to tease out fundamental dynamics
are inappropriate and misleading (a claim which reduces to the assertion that the reality is either so very
complex or so rapidly changing that its structure cannot be accurately
specified) then only a history is
possible. A history may be useful,
however, even when a more structural model is supposed. Structural models do not account for
everything. Conjunctural events still
need to be chronicled. The mentalities,
the struggles of individuals and classes, need to be described and interpreted.
On
the other hand, the same single case can also be studied structurally.
It is not true that a case study precludes a structural analysis. It is possible to formulate models of the
structural processes operating in a
single case and to test those models using a generalized version of the logic
of time series analysis. Time series
analysis measures changes in variables over time and enables us to test models
of the causal relations among these variables.
The problems which this approach encounters when we try to apply it to
the world‑system are considered below.
Here I simply want to make the point that having only one case does not
limit us to historical or qualitative methods.
Cross-national Studies
Contrary to what has been argued by
some authors (e.g. Bach, 1980), the study of other levels of analysis can be
quite relevant to world‑system theory.
Though I will devote much space below
to examining how we can study the world‑system as a whole, this does not
imply that studies of other units of analysis such as individuals,
organizations, classes, states, zones, etc. are irrelevant for our
understanding of the world‑system.
Indeed, our conception of the world‑system as a holistic structure
includes these levels. It is not merely
the largest layer of long‑distance interaction. It includes all the short‑range links
as well. This is reflected in our conceptualization of the
world economy as the whole which is composed of all the national economies
along with the international and transnational economic structures and networks
of exchange. Indeed some world‑system
processes must be studied by examining smaller units of analysis such as
nation‑states or transnational firms.
For example, studies of the effects of location in the core/periphery
hierarchy on national economic development must use countries as the unit of
analysis, and this is surely a research problem which emanates from world‑system
theory. The analysis of the world‑system
has implications for processes which operate at many levels, not just at the
level of the world‑system as a whole.1
Galton's Problem
The comparison of cases is a useful and
meaningful form of research for evaluating propositions which are implied by
world‑system theory. The strategy
of comparative analysis does not emanate, as some have implied, solely from the
practice of survey research in sociology.
Non‑experimental research designs have a long history in the
natural sciences, as well as in economics.
The fact that sociologists learned the logic of comparison through the
technique of survey analysis is largely a historical accident, although this
legacy does affect some of our
practices (e.g. methodological individualism) and our vocabulary. Similarly the use of probabilistic models has
been affected by the logic of sampling.
We still talk of "sample size" when we mean the number of
cases we are analyzing. But the logic of
probability has many useful applications beyond the sampling problem.
We
can draw valid inferences from comparing a number of cases only when those
cases can be treated as independent instances of the process under study. This is one of the basic tenets of
comparative logic (Zelditch, 1971). We
cannot increase our number of cases meaningfully by including the same case
twice because this is not an independent instance. This problem was first highlighted in social
science by Sir Francis Galton's critique of the comparative method proposed by
Edward Tylor. Tylor had proposed that
social structural and cultural processes could be studied by comparing
correlations of attributes across cultures.
Galton pointed out that some of the resulting correlations would stem
from cultural diffusion rather than from the functional interdependence of
institutions. Raoul Naroll (1968)
suggested that this problem could be solved by comparing cases in which
cultural diffusion could not have occurred.
This is one way of "controlling out" the effects of a
variable, in this case diffusion. A more
general solution is to measure non‑independence and to include the
hypothesized sources of non‑independence in the model, thus controlling
for them. This is done in a number of
different ways.
A
piece on the methods of studying national societies written by Terence Hopkins and Immanuel Wallerstein
(1967) (and since disavowed by them) suggested a set of distinctions which are
helpful here. Hopkins and Wallerstein contended that national societies
develop due to causes which are internal, relational, and contextual. Internal causes are variable characteristics
of the national society itself.
Relational causes are those variables which impinge on a national
society due to its relations with other national societies, for example the
dependence of a national economy on foreign investment. Contextual variables emanate from
characteristics of the world situation as a whole, for example the level of
world industrialization. Once we have
conceptualized the sources of non‑independence among a set of cases and
operationalized these we can use the method of quantitative comparative
analysis.2 It is not
necessary to find cases which are completely unconnected with each other.
Correct Methods
My stance is somewhat eclectic on the
question of correct methods. I think
that both historical interpretive studies and quantitative comparative studies
can be useful approaches to the world‑system. Historical studies are useful because some
things really are conjunctural and impossible to analyze structurally, and
also because they often generate hypotheses and conceptualizations which
stimulate the formulation of structural theories. Also, once we have successfully established
one or more structural theories, case studies which take a closer look at a
process on the ground can be helpful in telling us exactly how structural
forces operate. These cases or instances
are best selected after comparative analysis has already shown the outlines of
the structural process. We can
investigate the mechanisms which link causal variables by studying
"typical" cases, and we can examine contingencies by looking at
deviant cases.
Contrary to the liberal spirit of the above,
some world‑system militants contend that the correct study of the world‑system
requires a new methodology, or at least precludes the usefulness of some
methods, especially quantitative comparative analysis. Terence Hopkins (1978) and Robert Bach (1980)
argue that we should completely rethink our method-ological and epistemological
assumptions when we take the world‑system as the object of study. They call for the development of a new method
more appropriate for understanding the nature of the modern world‑system. The reasoning behind their concerns needs to
be examined in an atmosphere which is clear of the smoke of the sterile debate
between historicists and structuralists or between the adepts of quantitative
versus qualitative analysis. These old "ethnic" boundaries often
motivate intense debates, but they rarely produce new knowledge.
Robert
Bach (1980) organizes his argument about the desirability of a new method around two ideas: the notion of the "singular
process," and the idea of a "spatio‑temporal whole." The
world‑system is said to be characterized by a set of singular processes,
which are features of the whole system and cannot be understood by analyzing
sub‑parts or smaller units.
This claim is part of the stance
of theoretical holism, which holds that the logic of the socio‑economic
system Ä the capitalist mode of production Ä is a feature of the whole system,
and that all causal and developmental logics within an empirical world‑system
must be aspects of this logic of the whole.
This matter has been discussed at the theoretical level in chapter
1. It basically rules out the
possibility that the dominant mode of production is articulated with other
modes, and it equates the logical boundaries of a mode of production with the
spatial boundaries of a world‑system.
Bach seeks to extend this theoretical holism to a methodological
holism. His argument implies that only
study of the world‑system as a whole is relevant for understanding the
world‑system.
The Aggregation Problem
One reason Bach may be worried about
subunit comparisons is the aggregation problem. This problem is analyzed generally by Michael
Hannan (1971). It is the inferential
error which corresponds to the opposite of the ecological fallacy. The ecological fallacy occurs when an
association between two variables at the level of a set of larger units of
analysis (e.g. census tracts) is used as evidence in support of an assertion
about an analogous relationship at a smaller unit of analysis (e.g. households,
individuals). For example, it may be
observed that there is a correlation between the average level of education and
the average income of households when we compare census tracts, but this is not
evidence for this association at the level of house-
holds.
It is not impossible that poor Ph.D.s and rich high school dropouts live
in the same neighborhoods. Conversely,
evidence of an association at the level of a smaller unit is not evidence of
the analogous association at a larger level.
This is the aggregation problem which is mentioned in chapter 11 in the
context of the discussion of cross‑national research and its implications
for questions about the mechanisms which reproduce the core/periphery
hierarchy.
The
aggregation problem does not imply, however, that it is meaningless to
aggregate characteristics of subunits in order to construct a measure of a
feature of a larger unit. The world
economic product can be usefully estimated by summing the GNPs of all
countries.
Subunit Comparisons
Neither does the aggregation problem
imply that cross-national or other subunit comparisons are irrelevant for world‑system
theory. The rise and decline of
hegemons, the effects of international power‑dependence relations on
national development, the effects of location in the core/periphery hierarchy
on class formation and class struggles Ä these are all world‑system
processes which can be usefully studied by comparing national societies to one
another. It is simply not the case that
world‑system theory only has implications for processes which operate at
the level of the whole system.
Much of the reason for rejecting
cross-national comparisons of large numbers of countries apparently stems from
misunderstandings about the assumptions
behind such research. Critics have
contended that cross‑national quantitative analysis implies that all
nation‑states are the same, or that national development is a static,
mechanistic, deterministic process (e.g. Cardoso, 1977;
Space and Time
The second concept which Robert Bach
uses in his argument in favor of a new methodology for world‑system
scholarship is the idea of the "spatio‑temporal whole" (Bach,
1980). He claims that the capitalist
world‑system has a feature which makes it difficult to apply the standard
tools of comparative research. It is
alleged that the processes which operate at the level of the whole system are
not only singular, but they are processes in which time and space are
integral features rather than simply dimensions on which instances are
arrayed.
This
is a fascinating contention, and one that will surely stimulate much interest among social geographers, who
are always looking for new and theoretically interesting ways of
conceptualizing space (e.g. N. Smith, 1984).
But what does Bach really mean?
He says that time and space are integral features of world‑system
processes. Time and space are features
of all social processes in the sense that they enter into the determination of
outcomes. Bach is somewhat vague as to
how this claimed special feature of the world‑
system differs from social processes
operating at smaller levels. Possibly he
means that the social meaning of time and space is an important feature
of the cultural and economic structures of capitalism. This is undoubtedly true, but I fail to see
how this supports Bach's claim that conventional comparative methods are
inappropriate for world‑system studies.
Certainly
there are special problems relating to the temporal and spatial dimensions
which come up when we try to study the whole world‑system. Many of these are discussed below. The unreflective application of any
methodology, including historical interpretation, is unwise. But I am not convinced by Bach's argument
that we should throw cross-national analysis or survey research, or formal
comparative method in general, out of our tool box before we approach the world‑system.
World‑system Characteristics
In proposing the formal comparative
study of variable characteristics and processes of the world‑system as a
whole I am asserting, at the minimum,
that such a system exists and can be studied in its own right. The stronger claim that this larger system
has great causal importance for the development of smaller institutions and
entities can only be subjected to empirical research once world‑system
processes have been specified and operationalized. The focus of this section is on the problems
of studying the relationship between different variable characteristics of the
whole system. This is distinct from the
problem of studying the contextual effects of world‑system
characteristics on national development or the relational effects of a nation's
position in the larger system on its own development. For now I am focusing on the causal relations
between different characteristics of the whole world‑system.
There
is only one world‑system now, but historically there have been many and a
social scientist may make systematic comparisons across these cases in order to
bring evidence to bear on a hypothesis.
Wallerstein does this effectively with his comparison between the
emerging European world-economy and the Chinese world‑empire in the
sixteenth century (1974:52‑63).
This approach is discussed in the last section of this chapter. Another strategy, the one to be considered here,
is to employ methods which enable the comparison of a single system to itself
over time. We can use the generalized
logic of time
series analysis to test propositions
about relationships between system‑wide character- istics which vary over
time. Time series analysis is a statistical technique for testing causal models
in which time points are the units of comparison (Hibbs, 1974). Thus it compares a system to itself over time
in order to examine the relationships among variable characteristics. This requires that theoretically relevant
variables have been conceptualized and operationalized, and that comensurable
data are available over a long enough time period for processes of interest to
operate.
Operationalization and Measurement
The idea of employing time series
analysis to study the world‑system raises a number of critical empirical
questions. What are the spatial
boundaries of the system and how have they changed over time? Can data gathered on nation‑states be
used to study world‑system processes?
Can variables be measured over long enough periods to make valid
inferences possible? Does the logic of
causal analysis apply to world‑system events in a straightforward
way? Are long-run processes amenable to
comparative research and proposition‑testing? Complete answers to these questions depend on
the particular proposition being investigated, but I will give examples and
suggest possible approaches.
1 Spatio‑temporal
mapping
Any effort to apply the general logic
of time series analysis to the world‑
system as a whole must be able to
specify the spatial boundaries of the system.
In order to measure characteristics which vary over time we must know
which areas to include and which to exclude.
For example, my study of changes in the rank‑size distribution of
world cities (Chase‑Dunn, 1985a) required decisions about which cities to
include in the European‑centered world‑economy at various time
points. I employed the rough "spatio-temporal mapping" produced by
the
The problem of the spatial boundaries of a
world‑system is both
theoretical and operational. It is necessary to have a clear
theoretical conceptualization of spatial
boundaries before we can make headway on the measurement problem. But it would be disingenuous to imply that
these two problems are completely
separated in practice. A theoretical
definition which is completely impossible to operationalize is useless. Thus we must use our knowledge of potentially
available information in conjunction with the construction of a theoretical
definition.
I
argued in chapter 1 for a separation between the notions of logical and spatial
boundaries. The deep structural logic of
a system can change without changes in its spatial boundaries and vice
versa. Thus system logic ought not be
used to draw spatial boundaries between world‑systems. This separation makes it possible for there
to be more than one mode of production within an existing world‑system. This simplifies the problem of spatial
boundaries to one of establishing some (specified) form of connectedness. It is not necessary to decide which deep
structural logic is operating in any instance in order to determine the spatial
boundaries of a system.
The theoretical specification of spatial
boundaries is still contentious however, even after we have separated it from
the problem of the mode of production.
There are contending definitions, and most scholars agree that spatial
boundaries are "fuzzy."
Different aspects of a world-system (e.g. politicalÄmilitary and
economic) may have somewhat different spatial extents. For example, a region may have been formally
incorporated as a colony of a core power, but not yet linked into the world
economic network, or vice versa.
Fuzziness is also related to time.
Immanuel Wallerstein contends that the process of incorporation operates
over at least a fifty year period within which it is difficult to say whether
or not a particular region is in or out.
This implies that mapping which employs shorter intervals is necessarily
somewhat arbitrary. Thomas Hall (1986)
has argued that incorporation should be conceptualized as a continuum of
types. This approach introduces an
additional source of
"fuzziness."
Wallerstein's
simplest definition of the spatial boundaries of a world‑
system focusses on links in an
interdependent network of the exchange of "fundamental commodities,"
by which he means food and other necessities of everyday life. He excludes the exchange of
"preciosities" (luxuries) which are alleged not to have important
consequences for the exchanging parties or their societies. Charles Tilly (1984:62) argues that the
spatial boundaries of a world‑system should be understood generally in
terms of coherence and interdependence.
He correctly points out that allowing any connection at all to
constitute grounds for inclusion results in most areas of the globe having been
parts of a single "system" for millenia, a usage employed by Lenski and Lenski (1982). The notion that everything is somehow
connected to everything else, while undoubtedly true, is not very useful to a
science of human society.
Tilly proposes a "rule of thumb for
connectedness" based (not surpris-
ingly) on political control. He suggests that the boundaries of a world‑system
can be drawn as follows:
the actions of
powerholders in one region of a network rapidly (say within a year) and visibly
(say in changes
actually reported by
nearby observers) affect the welfare of at least a significant minority (say a
tenth) of the population in another region of the network. (1984:62)
Both
Wallerstein's and Tilly's definitions have conceptual and operational
problems. Tilly's is more precise, but
the precision is admittedly arbitrary.
Nevertheless, to operationalize we are often forced to make such
arbitrary dichotomies, and it is useful to recognize them as such because we
may want to alter them later when other empirical problems become more evident.
Tilly's
definition does not specify the boundaries of the "network," and it
seems to imply a system which has a single center. Both the modern and many ancient world‑systems
were (and are) multicentric and thus require that we conceptualize indirect as
well as direct economic and political connections.
A
third definition of interconnection based on interactive politicalÄ
military conflict is suggested by David
Wilkenson (1987). Wilkenson proposes
that polities which are engaged in sustained politicalÄmilitary competition with
one another should be considered parts of the same
"world-system/civilization."
Wallerstein's effort to rule out the exchange
of preciosities is problematic when we
look at primitive and ancient world‑systems, and even the early modern
world‑system, as convincingly argued by Jane Schneider (1977). Prestige goods have played an important
political role in most precapitalist world‑systems, and indeed it has
been argued that inter-societal prestige economies have constituted the most
important element in some premodern world‑systems (e.g. Blanton and
Feinman, 1984). Also as Schneider points
out, bullion was used as a means for hiring mercenaries in both modern and
ancient world‑systems, and this is hardly a matter which can be
considered epiphenomenal to the reproduction of power structures. The point here is that fundamental goods
networks constitute one network level because of the cost of transport, while
the range of preciosities is much greater, but these latter must be taken into
account in any theory which seeks to explain social change. Perhaps
we need to define sub‑ and super‑world‑systems based on this
distinction. If so the sixteenth-century
European world‑economy was a
subsystem within a larger multicentric Eurasian super‑world‑system. This kind of specification of boundaries
complicates things to some extent, but it allows us to continue with the
analysis of world‑systems without consigning clearly important
transactions to the basket of epiphenomena.
These theoretical issues notwithstanding, I
want to move on to some of the measurement problems. Assume we have a clear theoretical definition
of spatial boundaries. I already mentioned the problem of
fuzziness. This is
an instance of the problem of
measurement error, and we have developed a set of tools for dealing with
this. If measurement error is randomly
distributed it will not distort the outcome of studies of associations among
variables. If it is systematically
biased it may affect such outcomes, and is a bigger problem. In the context of world‑system boundaries
it is likely that we have better information and more information the closer we get to the core and the more
recent is the time period we are studying.
These systematic measurement errors should be kept in mind whenever we
seek to test any particular proposition.
Another point may be obvious, but is not
always appreciated. I learned this from
my experience with cross-national research.
The spatial and temporal scale of
research changes the significance of measurement error. In connection with the topic at hand, error
associated with the spatial boundary of a world‑system may be extremely
important if one is trying to explain what happened in a particular city but
much less consequential if one is trying to measure the level of a world‑system
characteristic such as the world city‑size distribution. In other words, whether Constantinople is in
or out is quite important if you are studying
2 Validity
and Reliability
As mentioned above, problems of
operationalizing variables differ greatly depending on the era being
studied. For earlier periods we must
usually rely on partial information and extrapolate to the system as a
whole. Thus Braudel and Spooner (1967)
estimate the growing integration of sixteenth-century
The
operationalization of abstract variables over long periods of time often
requires the use of different indicators in different periods. Thus, for instance, the timing of the Dutch
hegemony in economic competitive advantage might be measured in terms of the
extent to which Dutch products and
services replaced those of competitors in the sixteenth and seventeenth
centuries. The particular leading
sectors which were important in this
period are, of course, very different from the ones which indicated the
relative competitive advantage of the British economy in the late
eighteenth and nineteenth centuries, and
those which were important for the US hegemony of the twentieth century (see
Thompson, 1986 and chapter 9 above).
Thus the validity of these measures of the concentration of competitive
advantage in the core is dependent on the accuracy of judgments about which are
the leading edge industries over long periods of time.
3 Limitations
of Aggregating Data on Nation‑states
The very structure of data‑gathering
institutions may distort the available information in systematic ways which
create problems of inference (Wallerstein, 1983b). For example, many of the most comparable data
series are measures of characteristics of nation‑states. This institutional fact, which derives from
the importance of the nation‑states as tax‑gatherers and
controllers of flows, makes the study of certain world‑system processes
difficult. For example,
4 Transformation
of Nation‑state Data
It is possible, however, to transform
data based on nation‑states to make them more suitable for world‑system
research. For example, if we want to
know the distribution of world monetary income among households we can use GNP
data (national income) plus information from studies of intra‑national
income
distribution among households to
produce an estimate of the world distribution of income among households. This has been done in the research reported
by
Similarly
we can estimate the distribution of income among social classes in the world by
using occupational wage data from the International Labour Organization. Studies of social class using occupational
characteristics (e.g. Wright, 1976) could be matched to the occupational
categories used by the ILO. By this
method we could produce an estimate of world income distribution among social
classes.
Detailed
examination of national accounts, census information, and the use of special
studies to estimate parameters across the system can certainly provide
theoretically relevant measures which do not depend on the nation‑state
as the unit of analysis. The existence
of Gross National Product data mean that, at some point in the data‑gathering
process, there must be information (or at least estimates) on the monetized
transactions completed by individuals, households, firms, sectors, and the
state. This information can be
aggregated on other than a national basis if it can first be obtained in enough
detail across the system. Much of this
detailed information, however, is not
available for other than the most recent periods, and thus the study of long-
run processes must use less complete information. Although this reduces the level of certainty
we can have in results, these important processes should not be ignored for
this reason.
The
use of nation‑state data should not be dismissed as a tool for studying the world‑system.3 For some theoretical purposes nation‑states
are the relevant unit of analysis and the aggregation of national data,
or the
study of "international"
patterns of interaction, may shed light on hypotheses about the world‑system. For example, the causal model proposed in
chapter 13
hypothesizes that changes in the relative
distribution of productive advantage among core states affect the pattern of
exchange and control between core states
and peripheral areas. Although for both
of these variables it is not the states themselves that are the only actors,
the fact that actors are located within or associated with particular states or
colonial areas is of direct theoretical importance. And thus data which are organized by nation‑
state are appropriate to testing the
proposition.
5 Combination
of Data from Different Countries in the Same Indicator
The example given above of the attempt
to measure the relative distribution of competitive advantage among core
states involves the combination of data from different countries in the same
indicator. The problems of working with historical trade
data from a single nation are many (e.g. Schlote, 1952:3‑40) but they are
complicated immensely when one tries to combine such data from different
countries. For example, I have estimated
one aspect of
Another way in which data from different
countries are combined is in the production of the measures produced by network
analysis. These measures have been
discussed in the section on measurement of world‑system position in
chapter 10. The use of international
trade matrices can be a valuable way of studying the shape of the world‑system
and how it changes over time (e.g. Smith and White, 1986). And this kind of measure has also been used
to study the world‑system causes of national development. This is accomplished by using the location
of each country within a block produced by the network analysis as an attribute
of that country in cross-national analysis (e.g. Snyder and Kick, 1979; Nemeth
and Smith, 1985). Recall also the
contention that these network measures are superior because they are more
"relational" than other measures, and my demur in Note 5 of chapter
10.
6 Non‑continuous
Data Series
Another difficulty in attempting to
construct data series over long periods of time is that the methods of data
gathering and conventions of presentation often change radically over
time. This creates non-continuous series
which may place in jeopardy the task of determining the timing of a downturn or
upturn in a fluctuation. For instance,
in the above example we want to know when British textiles began to penetrate the
reversals. Such decisions must be made carefully. An excellent study which reflects
insightfully on these difficulties is Modelski and Thompson's (1988)
examination of world sea power.
7 The
Width of a Time Point: Measurement Error in Time
Here is another question about the
applicability of conventional causal logic to world‑system research. One of the basic canons of causal analysis
holds that a change in variable A causes a change in variable B if and only if
the associated change in variable A precedes the change in variable B in time. The future cannot cause the past. In practice we often use historical events
such as wars, regime changes, economic crises, etc. as indicators of underlying
structural changes. The problem is that
these events are often not direct simultaneous indicators of the underlying structural
changes which we wish them to measure, and thus the apparent time sequence of
indicators (events) may not reflect the true causality between variables.
The
exact timing of, say, the outbreak of a revolution or war may be determined by
factors which are conjunctural or only loosely related to underlying
world-system processes. And yet the
propensity for warfare or revolution to occur within a certain broad period may
indeed be connected with underlying structural processes. If we are using an event as an indicator of a
structural variable the exact timing of the event (e.g. the day and month) may
represent mainly measurement error, while the fact that the event occurred
within a broader (e.g. ten-year) period may be the relevant fact for our purposes
of measurement.
Another
way to think about this is that, when we are considering very long-run structural processes, the relevant
width of a time point may be very wide.
That is, the events which indicate changes in underlying variables may be "simultaneous"
even when they are years apart in historical time. Andre Gunder Frank (1978:20) asserts that simultaneity
of events disparate in space, while not proof of a causal relationship, is an
important clue for world‑system analysis.
My point is that simultaneity should be defined broadly enough to allow
for measurement errors due to lags between underlying structural changes and
the "events" we use to indicate them.4 And also the fact that event (A) occurs a few
months or years later than event (B) should not be taken as proof that (A) (or
rather the variable which it indicates) could not have caused (B). Their sequencing in historical time may not
reflect the direction of true structural causality.
8 Few
Instances of Long‑run Processes
Another difficulty which time series
analysis of world‑system data must confront is the problem of few
instances. If one is studying a very
long-run process, e.g. the K-wave, there have been only eleven of these cycles
since the beginning of the Europe‑centered world‑system. This means that a test of any theory about
the causes of these cycles will be necessarily somewhat inconclusive because
of the small number of "cases"
(instances). This is the intersection
between what is historically unique and what is scientifically proveable. The degrees of freedom rapidly run out as the
complexity of the explanation exceeds the number of independent instances we
can study. This does not mean we should
abandon the general logic of time series analysis, but rather that we must be
willing to accept low levels of certainty.
An example is provided by Goldstein's (1988) study of core wars and
K-waves. He finds that his model is
supported in ten out of the eleven instances.
This is very
unlikely to have occurred by chance,
but only a few more contrary cases would have cast doubt on his model as a
general explanation. With few cases it
is easy to disprove hypotheses, but hard to prove them.
9 Sequencing
and "Cycle Time"
Joshua Goldstein (1988:chapter 8), in
his discussion of methods problems encountered when studying variable
characteristics of the world-system, describes an approach to social cycles
which I find very helpful. Goldstein
points out that social cycles are rarely perfectly periodic, unlike some of the
cycles which occur in physical processes.
Thus attempts to decompose long economic cycles into their components
which use Fourier or spectral analysis (e.g. Adelman, 1965) are misleading
because they assume that social cycles should closely approximate sine waves. Goldstein proposes instead the notion of
"cycle time" in which social cycles are allowed to have variable
periods. It is generally recognized that
the Kondratieff wave varies in length from peak to peak from about 40 to about
60 years. Goldstein argues that we
should rely more on the sequencing of events and trend reversals rather than
their exact spacing along the time dimension.
This is a valuable suggestion, although we should be somewhat careful
about its application. We have already
mentioned above a related concern about the relationship between events and
underlying structural variables: the width of a time point. That discussion put into question the
complete reliance on temporal priority as an indicator of the direction of
causation.
Another
problem is the extent of periodic cycle variation which is allowable. If our data indicated, for example, that
K-waves varied in their period from (say) twenty to eighty years we would
probably reconsider the value of a "cyclical" model. Robert Philip Weber (1987) has explained
analytic differences between three kinds of "cycles." Weak cycles do not have "fairly
constant" periodicity and are termed fluctuations. Moderate cycles have "fairly
constant" periodicity, and strong cycles have, in addition, regular
amplitude and symmetry. What constitutes
"fairly constant" in world-system research will need to be determined
in practice. K-waves and war severity
waves are moderate cycles. The hegemonic
sequence is probably a fluctuation with peak curves being about the same length
and troughs of very different lengths.
Other Research Designs
Above we have discussed problems
associated with studying relationships among variable characteristics of the
world‑system as a whole. This is
but one of several possible research designs which have been or could be used
to study world‑system processes.
The following discussion compares six other broad types of non‑experimental
research designs. It proposes the
application of two new approaches: designs for testing multilevel models, and
cross‑world‑system comparisons.
The
six types of research design are:
1 timeseries studies of
single countries;
2 crossnational studies;
3 event histories;
4 a design which examines
the effects of world‑system contextual variables
on national development;
5 multi‑level models,
and two research designs for testing them; and
6 cross‑world‑system
studies.
Each of these categories is actually a
broad collection of possible approaches in terms of the way in which the
temporal and spatial aspects of comparison are utilized to study causal
processes.
First,
though, let us consider the problem of the choice of a unit of analysis.
This is a research decision which ought to be determined by theoretical
considerations. Ideally we would choose
units in which independent and comparable instances of the process of
theoretical interest are operating. But
often, and especially in world‑system studies, different units of analysis affect one another and
completely independent instances are difficult to find. Above we have outlined the solution to
Galton's problem Ä the inclusion of hypothesized non‑independence into
the theoretical model and into the comparative analysis. But the yardstick of relative independence is
still useful in the selection of a unit of analysis. Other considerations also affect the choice
of a unit of analysis such as the availability of comparable data and the
existence of policy instruments which make it possible to employ research
findings in political practice. These
last considerations ought not to overwhelm theoretical considerations,
however. If availability of hard data or
"policy relevance" were primary considerations we would never study
the world‑system as a whole.
If
the dependent variable is a characteristic of countries, such as national economic growth, then the best
choice of unit of analysis is the country.
It would be possible to make inferences about national growth on the
basis of a study of firms, regions, or cities but such inferences would be
subject to assumptions which could introduce errors. Traditionally, quantitative comparative
research has required a rectangular data set in which the "cases"
dimension contains instances of a single focal unit of analysis. Having chosen a focal unit of analysis we may
use information on other units, or on the focal unit's relationship to other
units or larger systems, but all these must be transformed into attributes of
the focal unit. These transformations
also involve assumptions which may be fraught with error, and thus we seek
below to develop a multilevel research design in which the same model can
contain data on different units of analysis.
But first let us review more typical research designs in which there is
a single unit of analysis.
1 Timeseries
Studies of Single Countries
Timeseries analysis is a method of
testing causal hypotheses using data collected over time (Hibbs, 1974). Thus, if we believe that dependency retards
economic growth, we can measure changes over time in the level of dependency
and economic growth for a single country and test the hypothesis. In this method time points are the unit of
analysis. Certain practical limitations
to this design are suggested by the results of cross-national research. First, multiple regression analysis, which is
a useful technique for estimating causal relations among variables, requires at
least 30 datum‑points for estimates to be stable. The time lapse between points must be a
function of the time over which meaningful change in the variables of interest
takes place. Most of the variables of
interest to dependency theory or world‑system analysis are structural
characteristics of large social units, and these change only slowly. This means that, for example, yearly time
points are not usually sufficiently independent instances of the process under
study. Five‑ or ten‑year
intervals between measurements may be necessary, and 30 five‑year
intervals span a period of 150 years.
For most variables it will be difficult to accurately measure
variation over such a long time period
because of the deficiencies of available data.
This practical limitation makes most studies of a single country impractical for purposes of formal hypothesis
testing. However, such a design may be
useful for studying variables which change meaningfully over shorter periods,
or for determining the particularities of the country under consideration. The modeling of world‑system processes
as they occur in a single country allows one to build in conditions which may
be specific to each country (e.g. Duvall and Freeman, 1981). Such models should presumably assuage some of
the more historicist critics of quantitative studies, such as Cardoso (1977).
2 Crossnational
Designs
Seven different cross-national designs are
employed in the 17 studies on the effects of international economic dependence
reviewed by Bornschier et al. (1978).
These designs all combine data from different countries, and they differ
chiefly in the way in which the time dimension is employed. They all use measures of dependency which
characterize a country's relational position in the larger world‑economy,
that is vis-à-vis other countries.
The designs which employ measures of change over time are superior to
those that don't because of the reciprocal causal relations between economic
growth and international economic dependence, and because of the time‑lagged
nature of dependence effects.
Rather than rehashing the description and
evaluation of these designs (which is contained in Bornschier et al. 1978) I
will focus on a design which is particularly useful for world‑system
studies. It is called "pooled
cross-sections," but more properly might be termed pooled panel analysis
(Hannan and Young, 1977). This design
employs data on countries measured at several time points and uses both
countries and time points as the unit of comparison (country‑times). It throws together in one analysis the cross‑sections
(or panels) from different periods of time. This design is particularly appropriate
for studying structural variables which change slowly, such as the city‑size
distribution of a country (Chase‑Dunn, 1979). The number of datum‑points are
increased considerably, thus overcoming one of the limitations of other
cross-national research designs: a rather small number of cases. There are only about 150 countries in
1970. If we study periods before the
post‑World War II decolonization period the number of sovereign countries
decreases considerably. If a process of theoretical interest changes
only slowly a short time lag contains mostly measurement error. Pooled panel analysis allows the use of more
widely‑spaced measurement points and the specification of longer time
lags, because the number of "cases" is increased when "country‑times"
are analyzed.
One
possible problem with pooled panel analysis is its assumption that the structure of causation remains constant
across different time periods. The
cycles and trends of the larger world‑system may alter the structure
of causal processes operating at the
national level. Volker Bornschier (1985:
table 1) has found that the causal structure of national economic growth which was shown to have operated during the
post‑World War II K-wave upswing
changed dramatically in the period of
downswing from 1976 to 1981. These
changes may themselves be periodic (cyclical), as when the relationship
between economic nationalism and
national growth alters according to whether the larger system is economically
expanding or stagnating, or transformational in the sense that the structure of
causation is altered in a completely new way.
3 Event
History Analysis
Of those research designs which analyze
more than one point in time, most use regular intervals of time for the
measurement of variables. Thus cross-national
panel designs usually measure variables
every so many years, at regular intervals.
Event histories record the actual exact time of the occurrence of events and use this additional
temporal information about timing and sequencing in testing causal models
(Allison, 1984; Tuma and Hannan, 1984).
The most obvious usage of this approach is for the analysis of
categorical variables in which the timing of changes between discrete states
can be determined exactly, but Tuma and Hannan (1984) have extended the use of
event history analysis to continuous variables as well.
Hannan
and Carroll (1981) have used event history analysis to reanalyze a
crossnational panel study of the causes of changes in the formal political
structures of regimes (Thomas, Ramirez,
Meyer, and Gobalet, 1979). The study
analyzes different national‑level and international relational causes of
changes in the level of regime centralization (see chapter 6). Their findings differ from those of the
earlier study and, since their analysis employs more information about timing
and sequencing, they conclude that event history analysis is a superior method
for macro‑comparative studies.
It
is difficult to argue against the use of more complete information, but the
reanalysis by Hannan and Carroll suggests a number of cautions. Above we have discussed the problem of the
width of a time point. If it is true
that largely conjunctural features determine the exact timing of events, or
rather alter (randomly perhaps) the length of the lag between a significant
change in a structural variable and the event we use to indicate this
structural variable, then the use of "exact" time and sequencing may
introduce noise rather than new information.
Also
the use of a more complicated research design may run into additional problems
of data availability. Hannan and Carroll
were forced to drop one of the two indicators of world‑system position
used in the prior study for this reason.
Investment dependence is not available over the whole time period they
study, and is not available in yearly estimates. Thus they dropped this measure from the
analysis. They found that the other
measure of world‑system position, export partner concentration, did not
operate as a cause of changes in regime form, contrary to the panel analysis
findings of Thomas et al. (1979). This
bolstered the conclusion presented by Tuma and Hannan (1984:321) that the world‑system
interpretation has been disconfirmed by the event history analysis.
This
raises the question of the costs and benefits of the development of more and
more sophisticated methods. Most new
methods increase the resource costs and narrow the feasible scope of
research. The gains in better inferences must be weighed against
these increased costs and constraints.
When the choice of methods determines which research problems will be
studied the tail is wagging the dog.
This is one reason why social scientists with an appreciation of theory
are wary of the rapid revolutionizing of methodological techniques. Event history analysis may indeed turn out to
be a better way to
study world‑system processes, but
further efforts to compare it with older methods should be more chary about
conclusions which are based on less information, rather than more.
4 World‑system
Effects on National Societies
Most of the cross-national world‑system
studies which have been done examine the effects of a country's position within
a hierarchical world structure. Hopkins and Wallerstein (1967) distinguish
between these relational characteristics (attributes of countries due
to their position in a larger structure) and contextual characteristics of the
world‑system as a whole. They point out that, for example, as the level
of world industrialization increases it may become harder for any nation to
industrialize. John Boli (1980:table 5.4) employs an ingenious method for
testing such propositions. He is concerned with the way in which changes
in the concentration of world economic power affect the dominance of states
over their own populations. He measures
the percentage of world trade controlled by the largest trading nation at seven
decennial time points and then estimates the effect of this world‑contextual
variable on the national characteristic (state dominance) by combining the two
variables in a single analysis. While
countries remain the focal unit of analysis, each country receives the score of
the world variable at each of the seven time points, and the seven temporal
cross‑sections are pooled. This
research design could fruitfully be employed to test other hypotheses about the
effects of changes in world‑system characteristics on national
development.
5 Multilevel
Analyses
We can imagine theoretical models in
which different kinds of units are simultaneously included (Teune, 1979). We implicitly include many levels of
analysis when we make a model of a
single unit. For example, when we
study countries or firms we imagine that
the behavior of and interaction among individuals or classes are systematically
related to the model, but we don't usually specify these relationships. In the discussion of contextual effects above
I gave an example of how Boli (1980) tests the hypothesis that change in one
level causes change in another. But a
more complete understanding of social reality would specify the interaction
between many different levels of overlapping units, some hierarchically nested,
some intersecting the
boundaries of others.
Multilevel
research designs will enable us to answer questions about the relative size of
interlevel effects. The
"internalÄexternal" debate has often wrongly assumed that world‑system
theory only addresses processes which operate at the international level,
whereas the world‑system is holistically conceived as containing
international, national and within-nation processes. In other words all these things are
"internal" to the world‑system.
But we may nevertheless want to ask about the relative importance of
international‑level versus national‑level processes in determining certain outcomes. Does the position of a country in the larger
world‑system or, alternatively, does the nature of its national class
structure have greater consequences for national development? Such a comparison would have to be specified
more exactly, but this is the kind of question which might be answered by
research designs which can test multilevel models.
Imagine, for example, a theoretical model in
which there are two focal units of analysis, such as cities and countries. Attributes of other units (firms, classes,
the world‑system) are hypothesized to affect both cities and countries in
a single model. This model can be
represented by figure 15.1.
(Figure
15.1 About Here)
Figure
15.1 only shows the units of analysis in this multilevel model. Any real theory would analyze attributes or
variable characteristics of these units.
Now let us consider what types of research design could be used to
estimate such a model and how data‑sets could be organized to allow this
estimation. The most common way in which
such models are estimated is the use of, in this case, two different data
matrices. In the first data-set (matrix
A) the case base will be countries because the dependent variable is an
attribute of countries and all other units will be variables which will
implicitly be treated as attributes of nation‑states. In the second data‑set (matrix B) the
case base will be cities, and likewise all the variables will be implicitly
treated as attributes of cities. These
two matrices may both contain the same information but this information will
have to be reorganized in order to estimate both equations.
Single Matrix Multilevel Time Series
There is, however, a research design
which will permit the estimation of both equations from a single data
matrix. This design can be described as time
series analysis with data on different units. It potentially can allow any of the units
which are included in the model to be analyzed as the dependent variable, and
thus multilevel models can be estimated from a single data matrix. A matrix of this type is illustrated in
figure 15.2.
(Figure
15.2 About Here)
In this design time points are the focal unit
of analysis. This differs from pooled
cross‑sections or the method used by Boli (see above) in that time points are not combined
with any other single unit. This allows
great
flexibility in the choice of dependent
variable, although once a dependent variable is chosen all independent
variables will be implicitly treated as attributes of the unit of the dependent
variable.
What
are the assumptions behind such a design?
When we do simple cross- national analysis (one time point, many
countries) we are assuming that the countries are instances in which the
process we want to study operate.
Commonly we think of the differences among countries as variation over
space because countries are territorially organized units and do not occupy
the same space. But this is simply a convenience. Countries are not completely independent
instances, and so we build their non‑independence into our models in the
form of relational or contextual variables which we treat as attributes of the
countries. The point here is that
spatial separation does not in itself tell us about relations (or nonrelations)
among countries (Naroll, 1968).
If
there is a hypothesized relationship between countries which affects the
dependent variable of interest we must include it in the model, and if we do not include a variable we are assuming
that it is unrelated to the other independent variables and can safely be
relegated to the error term. In the time
series designs discussed above, the convenient dimension on which we hang
observations of instances is temporal rather than spatial, and like variation
over space, variation over time may or may not reveal independent instances
(e.g. auto-correlation). Non‑independence
must be included in the model, and the failure to include it will result in
misspecification and errors of inference.
The advantage of using time rather than space
for multilevel analysis is that it is a dimension which is common to all units
regardless of their spatial
organization, and thus we can include units which are parts of other units, or
those which overlap others, as long as we specify their causal relations correctly in our model. Most discussions of multilevel analysis focus
on units which are hierarchically nested (e.g. Hannan and Young, 1976) but this
is not a necessary condition for the design proposed here.
This design also allows for a great deal of
flexibility in grouping units, so we can easily test alternative methods of
aggregating. Hypothetically such a design could include information on all
units of analysis and on each of the individual cases at all levels. These data could be aggregated in many
different ways or be used in disaggregated form. The limitations and tradeoffs involved with
the use of this single matrix multilevel design are examined by Chase-Dunn,
Pallas, and Kentor (1982). They also
explain how to use panel analysis to solve identification problems in
multilevel designs which examine different data matrices for each unit of
analysis.
My
hope is that the above discussion will encourage further consideration of the
methods for testing multilevel models.
We may be able to resolve some of the confusion and disagreement about
the importance of larger and smaller levels of analysis for determining outcomes if we include these
together in the same models.
6 Cross‑world‑system
Studies
Despite my contention that time series
analysis of the modern world‑system can be used in the process of world‑system
theory construction, it could be argued that a theory of the deep structure
cannot be tested by studying only one world‑system. Indeed it might be the case that if we want
to understand what is structurally distinctive about the contemporary world‑system
we need to compare it to systems which are really different. In any case, we may be able to gain
understanding of the processes by which modes of production become transformed
by studying earlier instances of transformation and comparing them to the
contemporary situation.
Was the rise and fall of world‑empires
directly related to the transformation of modes of production or was that
relationship a more complicated one? How
do the concepts which we have developed for studying the capitalist world‑economy
help us (or hinder us) when we begin to analyze really different world‑systems? Is it true that there have been many world‑empires
but few (or mostly short-lived) world‑economies? How many distinct world‑systems have
there been in the history of human social development? What is a useful typology for classifying
world-systems so that we can understand how they are systematically different
and how they are the same?
I have begun to work on some of these
questions (Chase‑Dunn, 1986) and so have other scholars (e.g. Ekholm and
Friedman, 1982; Blanton and Feinman, 1984; Mathien and McGuire, 1986; Kohl,
1987; Rowlands, et al, 1987). But here I
would like to discuss the feasibility of a cross-world-system research design,
one that would use world‑systems as the unit of analysis and enable us to
test causal propositions about world‑system processes by comparing large
numbers of world‑systems.
This idea brings a number of problems
immediately to mind. Many of the
difficulties we encounter when we do cross-national comparisons are encountered
again, only most of them are worse. Are
world‑systems sufficiently comparable to be considered a unit of formal
comparative analysis? What about the
independence of units (Galton's problem)?
Problems of data availability and comparability are certainly
stupendous. Is it possible to obtain a
represen-
tative sample of historical
world-systems?
Obviously
a lot of conceptual and theoretical work would need to precede any serious
attempt to construct a data‑set composed of world‑systems. The spatial boundary problems discussed above
would have to be resolved, or at least clarified sufficiently so that
contending specifications could be compared.
Contending theoretical approaches would have to be clearly specified so
that we could direct our attention and resources productively. The idea of the core/periphery hierarchy
would have to be reconceptualized in a way that facilitates cross‑world‑system
comparisons without evacuating its theoretical content. Do relevant data‑sets already
exist? Rein Taagepera (1978) has
organized a data-set which contains the territorial size and temporal duration
data of more than 100 historical empires from 3,000 BC to the present. There may be other such data‑sets.
There
are two main macrocomparative research traditions which should be examined for
suggestions about how to resolve these problems. Cross-national research is obvious. The other is the Human Relations Area File
(HRAF), a massive effort by anthropologists to create a comparative data set
for the analysis of cultures (Murdock, 1957).
The HRAF uses societies rather than world-systems as the unit of
analysis and most of the variable characteristics coded are not relevant to the
theoretical questions asked by the world-system perspective. Douglas White and Michael Burton (1987) are
engaged in an effort to add world-system variables to the Standard Cross
Cultural Sample, a representative subset of the societies contained in the
HRAF. This promises to be a valuable
contribution to comparative world-system research and it has the advantage of
building on a huge amount of previous work.
But, while this approach can potentially answer questions about the
effect of world-system variables on societal development, it cannot be used to
study dependent variables which are characteristics of world-systems
themselves. I suspect that the big
questions regarding the transformations of modes of production can only be
answered by using world-systems themselves as the unit of analysis.
I
am not an advocate of large science for its own sake. A project which studied large numbers of
world‑systems might easily consume the entire National Science Foundation
budget for the social sciences.
Assertions about the potential value of such a project should be
critiqued, and preliminary studies should be conducted to explore the value of
a strategy which compares world‑ systems. My guess, however, is that intersocietal
systems have long been the units in which the most important historical and
developmental processes have operated, and thus studies based on these units
will be found to provide a better understanding of social change.
The
main conclusions of part IV on metatheory and methods are summarized on page 16
to 18 of the Introduction. Now we turn
to a consideration of some major unresolved problems suggested by the foregoing
structural approach to the modern world‑system.
Chapter 16: Implications
This last chapter reflects on some unresolved problems, and then
discusses implications for new scholarly work and for political practice. We will take another look at the main
conclusions reached and examine the links and voids between them.
Despite the somewhat ex
cathedra tone of the theoretical conclusions presented in this book, they
are certainly not the last word.
Problems and gaps remain within my own formulation, and the process of
systematically comparing my theory with contending theories has not really
begun. The intended road of formal
theory construction and empirical research which evaluates contending theories
has been recommended, but only a few steps down that road have been taken. My own theory is insufficiently formalized to
allow certainty about the empirical propositions which may be deduced from
it. And, since we have only recently
extended our frame of reference to whole world-systems, the relevant phenomena
have been observed and theorized about for only a short time. I expect that other contending world-system
theories will emerge from the several theoretical perspectives which can be
seen entering the fray.
In order to compare
theories we must be able to deduce contrary empirical propositions from
them. The real test of a theory is its
power in explanation and prediction. I
am sure, for example, that my theory and the power cycle theory of Modelski and
Thompson imply some of the same propositions and some that are contrary, but at
the current level of formalization it is difficult to precisely specify these. Once they are specified, empirical research
may go beyond its current role in examining single propositions and partial
causal structures to become more directly relevant to the theory construction
project. Of course theories do not win
or lose because of "crucial experiments." Thomas Kuhn's (1962) depiction of paradigm
revolutions contends that an accepted "normal science" paradigm only
falls when its failures to account for anomalies accumulate to some tipping
point where a new, more powerful, formulation is proposed and triumphs. But this scenario is only a possible future
for social science. We have not yet
reached sufficient consensus to constitute a normal science paradigm. Rather we live in a world of multi-
paradigm clashes in which empirical research does not have much
affect on the commitment of scientists to one or another body of axioms. This is true partly because social science is
young, and partly because it is more political than are the physical and
biological sciences.
The project of theory
construction and empirical research I am advocating for world-system analysis
cannot be naive about these things, but neither do they vitiate the
effort. Someday social science will come
of age, perhaps in a world in which class struggles are less intense. Theorizing the world-system can help bring
such a world about, and may contribute to the maturity of social science within
that future world.
Spatial and Logical Boundaries
Even within the meta-paradigm in which I am working, historical
materialism, there are several unresolved problems with my formulation. I argue that Wallerstein's totality
assumption, which contends that each world-system has only one mode of
production, creates insoluble problems for understanding contradictions between
modes of production and their transformation. Thus I distinguish between the
spatial boundaries of world-systems and the logical
boundaries of modes of production. But if it is not true that the mode of
production is a feature of the world-system as a whole (as
Wallersteincontends), what is the unit which should be characterized as having
one or another mode of production? Is it
the firm, the household, the nation-state, the core/periphery hierarchy, or
what? I agree with Wallerstein that
exploitation and domination are multilevel phenomena which cannot be adequately
understood by looking only at the "point of production." Many Marxists (e.g. Dobb, 1963; Brenner,
1977) have argued that the logic of a mode of production is determined by the
interaction between a direct producer and his/her immediate overseer. I have no quarrel with the spate of recent
studies which examine control structures within the work place, but I agree
with Wallerstein that we ought not to treat these relations as the kernel (monad)
of the capitalist mode of production.
But if the kernel is neither the world-system as a whole nor the firm,
where is it?
Another related
problem is this. If a world-system can
have more than one mode of production within it, how can we understand the
spatial boundaries between modes? More
specifically: if, as I argue, the capitalist mode of production includes both
core and peripheral capitalism, where are the other modes of production with
which capitalism is articulated? Am I
suggesting that there are locations beyond the periphery, but within the
world-system, which can be understood as the domain of non-capitalist modes?
These problems assume
that modes of production are easily located in space or identified with
particular entities. It would be
conceptually neat to be able to do so, and such a specification would
undoubtedly have important political implications. But the best solution is, I think, another
approach. The notion of a mode of
production is an analytic idea about the logic of social relations, the
systematic way in which the needs of people are met through social organization
and institutions, and the structures of power whichcontrol production and
distribution. As such there is no single
most
appropriate unit of analysis which distinguishes between
different logics. Similarly, spatial
delineations can only correspond roughly to differences in the logic of
production and reproduction. Thus elements
of other modes of production can, in principle, be anywhere within a world-system
which is dominated by the capitalist mode of production.
The logic of
socialism democratic and collective
rationality can exist partially within
the so-called "socialist" states, while at the same time these also
contain elements of capitalism and elements of the tributary modes and while
they remain parts of the larger capitalist world-system. Likewise the logic of socialism may exist to
some extent within workers' organizations in capitalist countries, and indeed,
within some capitalist organizations.
Ernest Mandel (1970) has suggested that street lights (free and publicly
provided) are an instance of socialist logic within the capitalist mode of
production.1 Some aspects of the
tributary mode of production are certainly present in the extant socialist
states,2 and this is also the case within capitalist states to the extent that
some policies are motivated by the fiscal needs of the state. We have also seen that, though the most
successful core states employ a largely profit-oriented approach to
international competition, there are other contending states which choose a
path of conquest which is much closer to the tributary mode. Thus elements of both competitive commodity
production and politicalmilitary domination are present in the capitalist
world-economy.
What makes it a
capitalist world-system is the relative importance of commercialization, market
competition, and capitalist accumulation compared to earlier tributary
world-systems. Also, various elements of
the kin-based (normative) mode of production continue to operate within
households in whichsubsistence production for use and distribution according to
normatively regulated expectations operate.
This kind of income-pooling is not only a characteristic of households,
however. Such normatively regulated
logics can also be seen in larger solidarities such as ethnic groups and
nations in which mutual aid, altruism, and non-market mediated interactions
certainly occur. To point out that these
instances of non-capitalist relations are articulated with and dependent upon
the capitalist mode of production is simply to reiterate that this world-system
is dominated by the logic of capitalist accumulation. These non-capitalist relations are undoubtedly
differentially distributed, but it would be folly to try to specify a single
unit of analysis or a particular spatial location for them.
The above does not
constitute backpedalling on my reformulation of the nature of capitalism. I still maintain that the interstate system
and the core/periphery hierarchy are constitutional parts of the capitalist
mode of production because they are necessary to its reproduction. Remember FosterCarter's (1978) distinction
between articulation and contradiction discussed in chapter 1. Some activities which themselves display the
logic of a non
capitalist mode of production are at the same time firmly
articulated with, dependent upon, and functional for capitalism. Others are in conflict with capitalism and
are potential threats to its continued dominance. My redrawing of the logical boundaries of
capitalism must needs take account of this distinction, although it is a
difficult one to make in many instances.
Constants, Cycles and Trends
In chapter 2 I presented a schema of world-system cycles and
trends which is in the nature of a set of hypothesized features of the
worldsystem as a whole. These features
were produced by eye-balling the history of the modern world-system. They are based on induction, but on the
roughest kind of induction. One obvious
empirical job which needs to be done is to investigate the reality of these
hypothesized cycles and trends. Some of
the work has already been done, and it is reported in other chapters, but
chapter 13 revealed that much more work is needed.
The other problem is
more theoretical. What is the
relationship between the specification of the nature of capitalism in chapter
1, which was mostly adduced from prior theorizing and theoretical disputes, and
the roughly inductive schema hypothesized in chapter 2? Ideally we would be able to specify the links
between the deep structure called
"system constants" in chapter 2
and the patterns which are hypothesized to be observable features of the
world-system as a whole.
Some of this has been
done. For example, chapter 9 discusses
the causation of the rise and fall of hegemonic core powers, a world-system
fluctuation. But most of the linkages
between the system constants and the system cycles and trends are unexamined, a
void which demands further
attention.
This is the main
disjuncture in the book's argument. The
relations among the various structures hypothesized to compose the capitalist
mode of production have been addressed in detail. Part II examines the interdependencies
between the interstate system and capitalist accumulation, and part III
analyzes the links between the interstate system, core capitalism, and the
core/periphery hierarchy.
Chapter 13 proposes a causal model of the relationship among several
world-system variables.
Coreness and Peripherality
Certainly the last word on the analytic definition of the
core/peripheryhierarchy has not been said.
I have already received a critique from Giovanni Arrighi, who contends
that my use of relative capital intensity is theoretically and empirically flawed. It is easy to recommend more research on
coreness and peripherality, but this analytic question is also heavily linked
to theoretical desiderata, and these should be given consideration. How does the proposed definition fit with the
rest of the theory; how easy is it to operationalize; and how heavy a causal
determinant of other outcomes can it be shown to be? These are complicated matters but ones which
ought not to be ignored, or resolved by eclecticism or raw empiricism at least if we are serious about world-system
theory.
On the other hand,
below I will recommend a purely empirical approach to measuring world-system
position. This is not seen as the
complete arbiter over the analytic problem, but it may well have important
implications for theory as well as measurement.
Research Lacunae
I am sure that there
have been important studies which my survey has omitted. So much good comparative work is coming out
that any attempt to be complete is soon obsolete, and I undoubtedly have
overlooked studies which ought to have been included. My apologies to authors and readers.
To my mind the most
gaping empirical hole is the need for a quantitative comparative study of the
world-system position of nation-states and colonial areas in the nineteenth and
twentieth centuries. As explained in
chapter 10, I don't believe it makes sense to spend a lot of effort trying to
locate the empirical boundaries between the core, peripheral and semiperipheral
zones, although studies of the whole core/periphery hierarchy would be useful
in answering questions about trends in inequality and other changes in
structure. In order to do comparative studies of the causes and effects of
world-system position we need measures which can at least rank countries, and
more metric measures would be useful for both cross-national and whole
hierarchy studies. Because so many
countries are in the periphery we especially need data which will allow us to
measure differences among peripheral countries. Once this job is done we will be able to
answer questions about upward and downward mobility, changes in mobility rates
over time,3 and the causes and consequences of world-system position.
Part of the above
study ought to focus on differences among core powers and the hegemonic
sequence, as Thompson's (1986) pathbreaking study of leading sectors of core
production (reviewed in chapter 9) has done.
We need a more exact understanding of the timing of the concentration
and dispersion of both economic and politicalmilitary capacities in the hegemonic
sequence. It would be ideal if the
proposed study could benefit from a theoretically deduced analytic definition
of coreness and peripherality, but realistically that problem is not likely to
be resolved to everyone's satisfaction soon.
So I propose a more eclectic empirical approach which uses the
dimensions suggested by several different theoretical traditions and compares
them. This means
studying military and economic forms of power and their
interaction.
Within these broad
categories there are many subdimensions.
Naval and air force power with a global reach has been studied for
"great powers" by Modelski and Thompson (1988). More conventional forces need to be added and
the comparison of semiperipheral and peripheral militaries needs to be
included. For national economic power we
want to include measures of overall economic productivity such as GNP per
capita, but also different forms of international economic power/dependence
such as investment dependence, the several forms oftrade dependence, aid and
debt dependence, among others.
Dependency here also implies the other side of the coin. Some countries are relatively dominant with
regard to these dimensions. Many of the
measurement problems discussed in chapter 15 will be encountered in this
study. It will take a fairly substantial
amount of resources to overcome these and to produce a series of parallel
indicators which decay as we go back in time and toward the periphery. My guess is that we will be able to do a fair
job back into the middle of the nineteenth century.
In addition, chapters
5 through 13, and also chapter 15 raise many problems which could stand further
comparative research. Rather than my
detailing these here, the interested scholar is encouraged to examine these chapters. But one particularly important problem is the
transformation of modes of production. I
take the liberty of discussing this below despite only a piecemeal understanding
of the processes of the transformation of systemic logic. The best way to understand the transformation
of systemic logic is to study this process as it has actually occurred in
earlier world-systems. This is why it is
important to undertake a comparative study of world-systems and modes of
production, a proposal outlined in the last part of chapter 15.
Political Implications
What are the political implications of the above? First I must qualify every thing that follows
because of the still tentative formulation of my theory and the remaining gaps
in the corpus of systematic world-system research. At this point the following implications are
no more than hints about directions for political practice. I am justified in appending this discussion
because of some of the frightening conclusions about warfare which world-system
theory
implies, and because of the absurdity of material
deprivation in an age when the technological problems for providing basic needs
are obviously solved.
A central question for
any strategy of transformation is the question of agency. Who are the actors who will most vigorously
and effectively resist capitalism and construct socialism? Where is the most favorable terrain, the weak
link where concerted action could bear the most fruit? Samir Amin (1980a) contends that the agents
of socialism are most heavily concentrated in the periphery. It is there that the capitalist world-system
is most oppressive, and thus peripheral workers and peasants, the vast majority
of the world proletariat, have the most to win and the least to lose.
On the other hand Marx
and many contemporary Marxists have argued that socialism will be most
effectively built by the action of core proletarians. Since core areas have already attained a high
level of technological development, the establishment of socialized production
and distribution should be easiest in the core.
And organized core workers have had the longest
experience with industrial capitalism and the most opportunity
to create socialist social relations.
I will contend that
both the "workerist" and the "Third Worldist" positions
have important elements of truth, but there is another alternative which is
suggested by my structural theory of the world-system: the semi-
periphery as the weak link.
Core workers may have
experience and opportunity, but some of them lack motivation because they have
benefited from a non-confrontational relation-
ship with core capital.
The existence of this labor aristocracy divides the working class in the
core and, in combination with a large middle strata, undermines political
challenges to capitalism. Also the
"long experience" inwhich business unionism and social democracy have
been the outcome of a series of struggles between radical workers and the labor
aristocracy has created a residue of trade union practices, party structures,
legal and governmental institutions, and ideological heritages which act as
barriers to new socialist challenges.
Nevertheless it is true that even small victories in the core (e.g.
further expansion of the welfare state, opposition to imperial policies, etc.)
have important effects on peripheral and semi-peripheral areas because of
demonstration effects and the power of core states.
Another reason to
emphasize the importance of even small gains for socialism in the core is
suggested by an analysis of how capitalism came to be a dominant mode of
production. Commercialization, markets,
money, and even the production of commodities by wage labor existed in many of
the precapitalist world-systems. Both
Roman and Chinese world-empires were highly commercialized, but the tributary
mode of production maintained domination over the developmental dynamics of
both of these empires in part because bourgeois forces never obtained
autonomous state power within core areas.
On the other
hand there were "capitalist" states in semiperipheral
areas, autonomous city states such as Dilmun, Tyre, Sidon, Carthage, and Venice
which produced and traded commodities in the interstices of the
world-empires. It was the coming to
state power of bourgeois groups within core states which was the
qualitative leap which distinguished the Europe-centered world-economy from
earlier world-systems.4 This implies
that core areas are indeed crucial for the institutionalization of a mode of
production in a world-system.
The main problem with
"Third Worldism" is not motivation, but opportunity. Socialist movements which take state power in
the periphery are soon beset by powerful external forces which either overthrow
them or force them to abandonmost of their socialist program. Anti-systemic movements in the periphery are
most usually anti-imperialist class alliances which succeed in establishing at
least the trappings of national sovereignty, but not socialism. The low level of the development of the
productive forces also makes it harder to establish socialist forms of
accumulation, although this is not impossible in principle. It is simply harder to share power and wealth
when there are very little of either.
Semiperipheral Socialism
These things are less true of the semiperiphery. Here we have both motivation and
opportunity. Semiperipheral areas,
especially those in which the
territorial state is large, have sufficient resources to be able
to stave off core attempts at overthrow and to provide some protection to
socialist institutions if the political conditions for their emergence should
arise. Semiperipheral regions experience
more militant class-based socialist revolutions and movements because of their
intermediate position in the core/
periphery hierarchy.
While core exploitation of the periphery creates and sustains alliances
among classes in both the core and the periphery, in the semiperiphery an
intermediate world-system position undermines class alliances and provides a
fruitful terrain for strong challenges to capitalism. Semiperipheral revolutions and movements are
not always socialist in character, as we have seen in
periphery. Thus the semiperiphery
is the weak link in the capitalist world-
system. It is the terrain
upon which the strongest efforts and biggest successes to establish socialism
have been made, and this is likely to be true of the future as well. On the other hand, the results of the efforts
so far, while they have undoubtedly been important experiments with the logic
of socialism, have left much to be desired.
The tendency for authoritarian regimes to emerge in the so-called
socialist states betrays the notion of a freely constituted association of
direct producers. And the imperial control of
encourages corruption, consumerism, and political opportunism by
the "new class" of technocrats and bureaucrats who can bolster their
own positions by either competing with the capitalist core or mediating the
importation of sophisticated technology.
But it does not follow
that efforts to build socialism in the semi-
periphery will always be so constrained and thwarted. The revolutions in the
It is important for
all of us who want to build a more humane and peaceful world-system to
understand the lessons of socialist movements in the semiperiphery, and the
potential for future more successful revolutions there. But for me and most of my audience political
action in the semiperiphery is not an option.
As a citizen of a core state I have opportunities and responsibilities
which I would be remiss to give up in an effort to directly engage in political
action elsewhere. Of course I can and
should support such action from where I am, and perhaps with important
effects. But my largest opportunity for
progressive political impact is within my own nation and upon my own
state. Though this fact is itself a
structural constraint of the contemporary world system, it is a structural fact
which I would be naive to ignore.
So what does my theory imply about socialist political practice
in the
hegemonic core state. The
processes of rapid growth and imperial success which have long undermined
socialist politics in the
Revolution, however,
is not the only road to socialism. John
Stephens (1980) has shown that important steps toward democratic socialism can
be achieved by the route of political class struggle within capitalist core
states. In the
estimate their potential contribution to social justice in the
Some critics of the
world-system perspective have argued that emphasis on the structural importance
of global relations leads to political do-nothingism while we wait for
socialism to emerge at the world level.
The world-system perspective does indeed encourage us to examine global
level constraints (and opportunities), and to allocate our political energies
in ways which will be most productive when these structural constraints are
taken into account. We are also urged to
expend resources on transorganizational, transnational and international
socialist relations, but it does not follow that building socialism at the
local or national level is futile.
I have argued
elsewhere that a simple domino theory of the transformation to socialism is
misleading and inadequate. Suppose that
all firms or all nation-states adopt socialist relations internally but
continue to relate to one another through competitive commodity production and
politicalmilitary conflict. Such a
hypothetical world-system would still be dominated by the logic of capitalism,
and that logic would be likely to repenetrate the "socialist" firms
and states. This cautionary tale advises
us to invest political resources in the construction of multilevel
(transorganizational, transnational and international) socialist relations lest
we simply repeat the process of driving capitalism to once again perform an end
run by operating on a yet larger scale.
A Socialist World-system
These considerations lead us to a discussion of socialist
relations at the level of the whole world-system. The emergence of democratic collective
rationality (socialism) at the world-system level is likely to be a slow
process. What might such a world-system
look like and how might it emerge? It is
obvious that such a system will require some kind of a democratically
controlled world federation which can effectively adjudicate disputes among
nation-states and eliminate warfare.
This is a bare minimum. There are
many other problems which badly need to be coordinated at the global level such
as ecological difficulties and a more balanced pattern of economic development.
But a socialist world government should not monolithically centralize every-
thing. The Soviet style
command economy is a poor model even for a nation-
state. Socialist markets
can be used to regulate the production and distri-
bution of many goods and services, and planning and regulation
of many activities can be decentralized.
A socialist world-system ought to be culturally pluralistic, and some
optimal level of multilevel decentralization (as proposed by Galtung, 1980) is
compatible with a constitutionally limited federal world state which regulates
those functions which need to be co
ordinated at the global level.
How might such a world
state come into existence? The process
of the growth of international organizations which has been going on for at
least 150 years will eventually result in a world state if we are not blown up
first. Even international capitalists
have some uses for global regulation, as is attested by the International
Monetary Fund and the World Bank. These
capitalists do not want the massive economic and political upheavals which
accompany collapse of the world monetary system, and so they support efforts
toregulate "ruinous" competition and beggar-thy-neighborism. Some of these same capitalists also fear
nuclear holocaust, and so they may support a strengthened global government
which can effectively adjudicate conflicts among nation-states.
Of course capitalists
know as well as others that effective adjudication means the establishment of a
monopoly of legitimate violence. The
process of state formation has a long
history, and the king's army needs to be bigger than any combination of private
armies which might be brought against him.
This prospect is a frightening spectre to many, but especially to
international capitalists who either consciously or unconsciously realize that
such a level
of the centralization of military power would mean the end of
the interstate
system. Nevertheless I
think we can expect some support for further world state formation from
capitalists and their political allies, and furthermore it is likely that such
an emergent world state would be under the domination of capitalists, at least
for a while.
While the idea of a
world state may be a frightening spectre to some, I am optimistic about it for
several reasons. First a world state is
probably the most direct and stable way to prevent nuclear holocaust, a desideratum
which must be at the top of everyone's list.
Secondly, the creation of a global state which can peacefully adjudicate
disputes among nations will transform the existing interstate system. As I argued in part II, the interstate system
is the political structure which stands behind the manueverability of capital
and its ability to escape organized workers and other social constraints on
profitable accumulation. While a world
state may at first be dominated by capitalists, the very existence of such a
state will provide a single focus for struggles to socially regulate investment
decisionsand to create a more balanced, egalitarian, and ecologically sound
form of production and distribution.
Is this a pipe
dream? Are not other outcomes possible,
such as fascism, continued capitalism, or nuclear annihilation? Of course other outcomes are possible. My structuralism is not inevitabilist
unilinear evolution. A computer
malfunction, completely conjunctural, unrelated to any systemic process of
social change, could kill all life on our planet at any minute. One of the first political tasks is to
dismantle this balance of terror. World
fascism is also a possible outcome of a grave economic and political
crisis. But fascism in
the past has been a desperate effort by capitalists and their political allies
to maintain power in the face of a strong challenge from the Left. Capitalism does not now face a deep crisis or
a strong socialist challenge at the world-system level. Even a collapse of the current international
financial system is not likely to lead to a concerted challenge from the Left
in most countries.
It seems more likely
that capitalism will muddle through, responding to challenges piecemeal,
expanding international and transnational organizations cyclically as it has
done for centuries, until the world political system is relatively
centralized. The rise and decline of
hegemonic core states will either end, or will slow down as the world
bourgeoisie becomes more politically and economically integrated. The gap between the core and the periphery
will begin to diminish, reducing the dampening effect of the core/periphery
hierarchy on class struggles. As efforts
to protect the livelihood of workers, the environment, and human rights become
increasingly oriented toward the emerging world state they will become more and
more organized around internationalism, and the logic of socialism will
eventually gain the upper hand. Socialism
is not the utopian end of history. It is
simply the next progressive step, as was capitalism. Socialism is not inevitable or perfect, and
neither is it immutable. But it is
certainly preferable to the current system of violent conflict, uneven
development, and exploitation.
Glossary of WorldSystem Terms
Colonial empire an empire in which one of the core states
within an interstate system exercises formal political domination over
territories abroad. Distinguished from
"worldempire" (see below) in which a single state apparatus exercises
formal domination over the whole core of a worldsystem.
Commodity chain a treelike sequence of production processes
and exchanges by which a product for final consumption is produced. These linkages of raw materials, labor, the
sustenance of labor, intermediate processing, final processing, transport and
final consumption materially connect most of the people within the contemporary
worldsystem. See chapter 10.
Core a region or "zone" in which most of
the production activity is core production (see below). The core comprises a group of states which
are not necessarily contiguous with one another in space. See chapter 10.
Core/Periphery hierarchy a spatially
differentiated continuum of socially structured inequality based on the
concentration of different kinds of economic production in different
areas. In the contemporary worldsystem
this
corresponds fairly well with the hierarchy of
"developed" and "lessdeveloped" countries. See chapter 10.
Core production the production of core commodities using
relatively (visvis the whole current worldsystem) capitalintensive technology
and relatively killed and highlypaid labor.
See chapters 1 and 10.
Empire formation the transformation of an interstate system
into a worldempire.
External arena a region outside the spatial boundaries of a
worldsystem.
Fundamental goods basic
goods such as food and raw materials that compose the everyday life material
consumption of the masses of people.
Distinguished from "preciosities" (see below and chapter 15).Hegemon the most powerful state in an interstate system. In the capitalist worldeconomy this includes
both economic power (comparative advantage in the production of core
commodities) and military power. See
chapter 9.
Hegemonic core a relatively unequal distribution of power
among core states such that one state has a rather large share of both economic
and military power. Contrasted with
"multicentric core" (below).
See chapters 9 and 13.
Hegemonic sequence a process in which the core of a worldeconomy
oscillates between a hegemonic and a multicentric distribution of economic and
military power among core states. This
corresponds to the rise and fall of hegemons.
See chapters 9 and 13.
Interstate system a system of unequally powerful and competing
states in which none of the states is powerful enough to dominate the whole
system. See "worldempire"
below and chapters 7 and 8.
Kwave the Kondratieff long business cycle in which
production, investment and prices oscillate between growth and stagnation
phases with an approximately 40 to 60 year period.
Logical boundaries major disjunctures between the logics of
modes of
production.
Modes of production the deep structural logics of socioeconomic
systems. Major modes are normative
(kinbased), tributary, capitalist, and socialist.
Multicentric core a relatively equal distribution of economic
and military power among core states.
See chapters 9 and 13.
Peripheral production the production of peripheral commodities
using technology which is relatively low in capital intensity and labor which
is paid low wages and is usually politically coerced compared to labor in core
areas. See chapters 1 and 10.
Peripheralization the processes by which
regions are incorporated into the worldsystem and become located in a
peripheral position in the core/periphery hierarchy.
Periphery those regions in which periphery production
is predominant.
Preciosities luxury goods which have high value due to
scarcity in areas into which they are imported, but much lower value at their
point of origin. Distinguished from
"fundamental goods" (above).
See chapter 7, note 8 and chapter 15, "Spatiotemporal
Mapping."
Semiperiphery those regions which contain either a
relatively balanced mix of core and peripheral production, or in which
production is predominantly at intermediate levels of capitalintensity and
labor remuneration. See chapter 10.
Worldeconomy a type of worldsystem in which the
territorial network of economic exchange is politically structured as an
interstate system.
World economy the total sum of economic relationships
contained in a worldsystem, including intranational, transnational, and
international production and exchange.
Worldempire a
type of worldsystem in which the territorial economic network is largely
contained within a single state apparatus.
Worldsystem a whole social system (not necessarily
global) composed of cultural, normative, economic, political, and military
relations which is bounded by a territorial network of regularized exchange of
material goods. Regarding the problem of
spatial boundaries of worldsystems See chapter 15, "Spatiotemporal
Mapping."
References
Adelman, Irma 1965: "Long cycles fact or fiction?" American Economic Review LV3: 444-63.
Alavi, Hamza 1972: "The state in post-colonial
societies: Pakistan and Bangla-Desh." New Left Review 74, 59-81.
Alker, H. 1982: "Logic, dialectics, politics: some recent
controversies." In H. R. Alker
(ed.), Dialectical Logics for the Political Sciences, Poznan Studies in
the Philosophy of the Sciences and the Humanities, Amsterdam: Kodopi, vol. 7.
Allison, Paul D. 1984: Event History Analysis: regression
for longitudinal event data. Beverly
Hills: Sage.
Althusser, Louis and Etienne Balibar 1970: Reading Capital. Trans. Ben Brewster. London: New Left Books.
Amin, Samir 1974: Accumulation on a World Scale. New York: Monthly Review Press. 2 vols.
1975: Unequal Development. New York:
Monthly Review Press.
1977: Imperialism and Unequal Exchange. New York: Monthly Review Press.
1980a: Class and Nation, Historically and in the Current Crisis. New York:
Monthly Review Press.
1980b: "The class structure of the contemporary imperialist
system." Monthly Review,
31(8), 9-26.
Anderson, Perry 1974a: Passages from Antiquity to Feudalism. London: New Left Books.
1974b: Lineages of the Absolutist State. London:
New Left Books.
1980: Arguments within English Marxism. London: New Left Books.
Arrighi, Giovanni 1978: The Geometry of Imperialism. London: New Left Books.
1982: "A crisis of hegemony." In Samir Amin, Giovanni Arrighi, Andre Gunder
Frank and Immanuel Wallerstein, Dynamics of Global Crisis, New York:
Monthly Review Press, 55108.
Arrighi, Giovanni (ed.) 1985: Semiperipheral Development:
the politics of southern Europe in the twentieth century. Beverly Hills: Sage.
Arrighi, Giovanni and Jessica Drangel 1986: "Stratification
of the world-
economy: an explanation of
the semiperipheral zone." Review,
10(1), 9-74.
Ashley, Richard K. 1984: "The poverty of neorealism." International
Organization,
38(2), 2256.
Bach, Robert L. 1977: "Methods of analysis in the study of
the world-
economy." American
Sociological Review, 42(5), 811-14.
1980: "On the holism of a world-systems
perspective." In Terence K. Hopkins
and Immanuel Wallerstein (eds) Processes of the World-System, Beverly Hills: Sage, 289310.
Bairoch, Paul 1975: The Economic Development of the Third
World Since 1900. Berkeley: University of California Press.
1976: "Europe's gross national product:
1800-1975." Journal of European
Economic History, 5, 273-340.
1986: "Historical roots of economic underdevelopment: myths
and
realities." In
Wolfgang J. Mommsen and Jurgen Osterhammel (eds)
Imperialism After Empire: continuities and discontinuities, London: Allen and Unwin, 191216.
Bairoch, P. and J. M. Limbor 1968: "Changes in the
industrial distribution of the world labour force, by region,
1800-1960." International Labour Review, 98(4), 311-36.
Balibar, Etienne 1970: "The basic concepts of
historical materialism." In Louis
Althusser and Etienne Balibar, Reading Capital. New York: Random House, 199308.
Banaji, Jarius 1980: "Summary of selected parts of
Kautsky's 'The Agrarian
Question.'" In Harold Wolpe
(ed) The Articulation of Modes of
Production,
London: Routledge & Kegan Paul,
4592.
Banks, Arthur S. 1971: Cross-polity Time Series Data. Cambridge: MIT Press.
Baran, Paul and Paul Sweezy 1966: Monopoly Capitalism. New York: Monthly Review Press.
Barbour, Violet 1963: Capitalism in Amsterdam in the 17th
Century. Ann Arbor: University of Michigan Press.
Barker, Colin 1978: "The state as capital." International
Socialism II(1), 16-42.
Barr, Kenneth 1979: "Long waves: A selective annotated
bibliography." Review 2(4),
675718.
1981: "On the capitalist enterprise." Review of Radical Political Economics
12(4), 60-70.
Barrat-Brown, Michael 1974: The Economics of Imperialism. London:
Penguin.
Bellah, Robert N., Richard Madsen, William M. Sullivan, Ann
Swidler and Steven M. Tipton 1986: Habits of the Heart: individualism
and commitment in American Life. New
York: Harper and Row.
Bennett, Douglas C. and Kenneth E. Sharpe 1985: Transnational
Corporations Versus the State: the political economy of the
Mexican auto industry.
Princeton: Princeton University
Press.
Benton, Ted 1984: The Rise and Fall of Structural Marxism. New York: St.
Martin's Press.
Bergesen, Albert 1980: "From utilitarianism to
globology: the shift from the individual
to the world as a whole as the primordial unit of analysis." In Albert
Bergesen (ed.), Studies of the Modern World-System. New York: Academic Press, 112.
1981: "Long economic cycles and the size of industrial
enterprise." In Richard Rubinson
(ed.), Dynamics of World Development, Beverly Hills: Sage, 17991.
1983: "The class structure of the world-system." In W. R. Thompson (ed.), Contending Approaches
to World System Analysis, Beverly Hills: Sage,
4354.
1985a: "Cycles of war in the reproduction of the world
economy." In Paul M. Johnson and
William R. Thompson (eds), Rhythms in
Politics and Economics, New York: Praeger, 31331.
1985b: "How to model the cyclical dynamics of the
world-system." Review 8(4),
501-12.
Bergesen, Albert (ed.) 1980:Studies of the Modern WorldSystem. New York: Academic Press.
Bergesen, Albert and Ronald Schoenberg 1980: "Long waves of
colonial expansion and contraction, 1415-1969." In Albert J. Bergesen (ed.), Studies of
the Modern World-System, New York: Academic Press, 23178.
Bergesen, Albert and Chintamani Sahoo 1985: "Evidence of
the decline of American hegemony in world
production." Review,
8(4), 595-611.
Bergesen, Albert, Roberto M. Fernandez and Chintamani Sahoo
1987: "America and the changing structure of hegemonic
production." In Terry Boswell and
Albert Bergesen (eds) America's Changing Role in the World System, New
York: Praeger, 15776.
Berry, Albert, Francois Bourguignon, and Christian
Morrison 1983: "The level of world inequality: how much can one
say?" Review of Income and
Wealth, 29, 217-41.
Biersteker, Thomas J.
1978:Distortion or Development?: contending
perspectives on the multinational corporation. Cambridge: MIT
Press.
Blanton, Richard and Gary Feinman 1984: "The mesoamerican
world-system." American
Anthropologist, 86(3), 673-82.
Blau, P. M. 1964: Exchange and Power in Social Life. New York: Wiley.
Block, Fred 1977: The Origins of International Economic
Disorder: a study of United States international monetary policy
from World War II to the present.
Berkeley: University of California Press.
1978: "Marxist theories of the state in world systems
analysis." In Barbara H. Kaplan
(ed.), Social Change in the Capitalist World Economy, Beverly Hills: Sage, 2738.
Boli, John 1980: "Global integration and the universal
increase of state dominance,
1870-1970." In A. Bergesen (ed.), Studies
of the Modern World System, New York: Academic Press, 77109.
Bollen, Kenneth 1983: "World-system position, dependency
and democracy: the cross-national
evidence." American Sociological
Review, 48, 468-79.
Bornschier, Volker 1976: Wachstum, Konzentration und
Multinationalisierung von Industrieunternehmen. Frauenfeld und Stuttgart: Verlag Huber.
1980: Multinationale Konzerne, Wirtschaftspolitik und
nationale
Entwicklung im Weltsystem. Frankfurt and New York:
Campus Verlag.
1983: "Dependent reproduction in the world system: a study
on the incidence of 'dependency reversal.'" In
Charles F. Doran, George Modelski and Cal Clark (eds) North/South
Relations: studies of dependency reversal, New York: Praeger, 97116.
1985: "World social structure in the long economic wave." Paper presented at the annual meeting of the
International Studies Association, March 5-9.
1987: "The business of violence and legitimacy." A paper presented at annual meeting of the
International Studies Association, Washington, D.C. April 15. Forthcoming in V. Bornschier, Westliche
Gesellschaft im Wandel. Frankfurt
and New York: Campus Verlag, 1989.
Bornschier, Volker, C. ChaseDunn and R. Rubinson 1978:
"Crossnational evidence of the effects of foreign investment and aid on
economic growth and inequality: A survey of findings and a
reanalysis." American Journal of
Sociology, 84(3), 65183.
Bornschier, Volker and Thanh-huyen Ballmer-Cao 1979:
"Income inequality: A crossnational study of the relationships between MNC
penetration, dimensions of the power structure and income
distribution." American Sociological
Review, 44(3), 487-506.
Bornschier, Volker and Christopher Chase-Dunn 1985:Transnational
Corporations and Underdevelopment. New York:
Praeger.
Bornschier, Volker and Otto Hartlieb 1981:
"Weltmarktabhngigkeit und
Entwicklung: bersicht ber
die Evidenzen und Reanalyse." Bulletin
of the Sociological Institute, 39 (October), University of Zurich.
Bornschier, Volker and Peter Heintz (eds) 1979:Compendium of
Data for World-
Systems Analysis. Sociological Institute, University of Zurich.
Bornschier, Volker and Jean-Pierre Hoby 1981: "Economic
policy and multi
national corporations in development: the measurable impacts in
cross-
national perspective."
Social Problems, 28(4), 363-77.
Borrego, John 1982: "Metanational capitalist accumulation
and the
reintegration of socialist states." In C. Chase-Dunn (ed.), Socialist States
in the World-System. Beverly Hills:
Sage, 11447.Boswell, Terry 1987: "Colonial empires and the capitalist
worldsystem: a quantitative analysis of colonization, 16501960." Paper presented at International Studies
Association meetings, Washington, D.C.
Bourguignon, F., A. Berry and C. Morrison 1983: "The world
distribution of incomes between 1950 and 1977." Economic Journal, 93, 331-50.
Bousquet, Nicole 1980: "From hegemony to competition:
cycles of the core?" In Terence K.
Hopkins and Immanuel Wallerstein (eds). Processes of the WorldSystem,
Beverly Hills: Sage, 4683.
Braudel, Fernand 1972: The Mediterranean and the
Mediterranean World in the Age of Philip II. New York: Harper and Row, 2 vol.
1975: Capitalism and Material Life: 1400-1800. New York: Harper and Row.
1977: Afterthoughts on Material Civilization and Capitalism. Baltimore: Johns Hopkins University.
1984:The Perspective of the World. Vol. 3 of Civilization and Capitalism. Trans. Sian Reynolds. New York: Harper and Row.
Braudel, F. P. and F. Spooner 1967: "Prices in Europe from
1450 to 1750," in E. E. Rich (ed.), The Economy of Expanding Europe in
the Sixteenth and
Seventeenth Centuries,
vol. 4, Cambridge Economic History of Europe, 378-486, Cambridge: Cambridge University
Press.
Braverman, H. 1974: Labor and Monopoly Capital. New York: Monthly Review Press.
Breedlove, William L. and Patrick D. Nolan 1988:
"International
stratification and inequality 19601980." Unpublished working paper, Department of
Sociology, University of South Carolina.
Brenner, Robert L. 1977: "The origins of capitalist
development: a critique of neo-Smithian Marxism." New Left Review 104, 25-92.
Brunt, P. A. 1971: Social Conflicts in the Roman Republic. New York: Norton.Bukharin, Nicolai 1973
[1915]: Imperialism and World Economy.
New York: Monthly Review Press.
Bunker, Stephen 1983: "Center-local struggles for
bureaucratic control in Bugisu,
Uganda." American Ethnologist
10(4), 749-69.
1984: "Modes of
extraction, unequal exchange, and the under-development of an extreme
periphery: the Brazilian Amazon,
1600-1980." American
Journal of Sociology 89(5), 101764 (March).
1985: Underdeveloping the Amazon: extraction, unequal
exchange and the failure of the modern state. Urbana:
University of Illinois Press.
Burawoy, Michael 1979: Manufacturing Consent. Chicago:
University of Chicago Press.
Burke, Peter 1974: Venice and Amsterdam: a study of 17th
century elites. London: Temple Smith.
Cain, P. J. and A. G. Hopkins 1986: "Gentlemanly capitalism
and British expansion overseas I: The old colonial system,
1688-1850." Economic History
Review 2nd Ser., 39(4), 501-25.
Cameron, David R. 1978: "The expansion of the public
economy: A comparative
analysis." American
Political Science Review, 72(4), 1243-61.
Caporaso, James 1981: "Industrialization in the periphery:
the evolving global division of labor."
In W. Ladd Hollist and James Rosenau (eds), World System Structure:
continuity and change. Beverly
Hills: Sage, 14071.
Cardoso, Fernando Henrique 1973:
"Associated-dependent development:
theoretical and practical implications." In Alfred Stepan (ed.), Authoritarian
Brazil, New Haven: Yale University
Press.
1977: "The
consumption of dependency theory in the United States." Latin American Research Review, 12(3),
7-24.
Cardoso, Fernando Henrique and Enzo Faletto 1979 [1971]:
Dependency and Development in Latin America. Berkeley:
University of California.
Carnoy, Martin and Derek Shearer 1980: Economic Democracy:
the challenge of the 1980s.
White Plains, N.Y.: M. E. Sharpe.
Chase-Dunn, Christopher 1975: "The effects of international
economic dependence on development and inequality: a crossnational study."
American Sociological Review 40, 72038 (December).
1976: "Toward a formal comparative study of the
world-system." A working paper and
transcribed talk delivered at the Fernand Braudel Center, December 8. Published as a working paper of Seminar
II: Historical Geography of Social and Economic Structures of the Modern
World-
System, Braudel
Center, SUNY-Binghamton.
1978:
"Core-periphery relations:
the effects of core
competition." In Barbara H.
Kaplan (ed.), Social Change in the Capitalist World Economy, Beverly Hills: Sage, 15476.
1979: "Comparative
research on world-system characteristics."
International Studies Quarterly, 23(4), 601-23.
1980: "The development of core capitalism in the antebellum
United States: tariff politics and class struggle in an
upwardly mobile semiperiphery." In
A. Bergesen (ed.), Studies of the Modern World-System, New York: Academic Press, 189230.
1981: "Interstate system and capitalist world-economy: one logic or two?" International Studies Quarterly 25(1),
19-42.
1982a: "The uses of formal comparative research on
dependency theory and the world-system perspective." In Harry Makler, Albert Martinelli and Neil
Smelser (eds), The New International Economy, London: Sage
Publications, 11740. 1983: "Inequality, structural mobility
and dependency reversal in the capitalist world-economy." In Charles F. Doran, George Modelski, and Cal
Clark (eds), North/South Relations: studies of dependency reversal,
New York: Praeger, 7395.
1984: "Levels of urban primacy in zones of the world-system
since 1800." Paper presented at the
annual meeting of the American Sociological Association, San Antonio, Texas,
August 27.
1985a: "The system of world cities: A.D. 800-1975." In M. Timberlake (ed.) Urbanization in the
World Economy, New York: Academic
Press,
26992.
1985b: "The coming
of urban primacy in Latin America."
Comparative Urban Research 11(12), 14-31.
1986: "Types of
worldsystems." A paper presented at
the annual meeting of the International Studies Association, Anaheim, CA.,
March 26.
1987: "The Korean
trajectory in the world-system." In
Kyong-dong Kim (ed.), Dependency Issues in Korean Development,
Seoul: Seoul National University Press,
270304.
1988: "Comparing worldsystems: toward a theory of
semiperipheral development." Comparative
Civilizations Review (Fall).
ChaseDunn, Christopher (ed.) 1982b: Socialist States in the
World-System. Beverly Hills: Sage.
ChaseDunn, Christopher, Aaron Pallas, and Jeffrey Kentor 1982:
"Old and new research designs for studying the worldsystem: a research note." Comparative Political Studies 15,
3:34156.
Chase-Dunn, Christopher and Richard Rubinson 1977: "Toward
a structural perspective on the world-system." Politics and Society, 7(4), 453-76.
1979: "Cycles, trends and new departures in
world-system development." In John W. Meyer and Michael T. Hannan (eds) National
Development in the World-System, Chicago:
University of Chicago Press, 27696.
Chatterjee, Partha 1975: Arms, Alliances and Stability. New York: John Wiley.
Chaudhuri, K. N. 1985: Trade and Civilisation in The Indian
Ocean. Cambridge: Cambridge University Press.
Chirot, Daniel 1977: Social Change in the Twentieth Century. New York:
Harcourt Brace Jovanovich.
1986: Social Change in the Modern Era. New York:
Harcourt Brace
Jovanovich.
Choucri, Nazli and Robert C. North 1975: Nations in Conflict:
national growth and international violence. San Francisco, CA: Freeman.
Clapham, J. H. 1966: Economic Development of France and
Germany. Cambridge: Cambridge University Press.
Coale, Ansley J. 1974:
"The history of the human population." Scientific American, September, 15-28.
Collier, David (ed.) 1979: The New Authoritarianism in Latin
America.
Princeton: Princeton
University Press.
Collins, Randall 1986: Weberian Sociological Theory. Cambridge:
Cambridge University Press.
Cox, Oliver 1959: The Foundations of Capitalism. New York:
Philosophical Library.
Cox, Robert W. 1983:
"Gramsci, hegemony and international relations: an essay
in method." Millenium: Journal of International Studies,
122), 162-75.
Crosby, Alfred W. 1972: The Columbian Exchange: biological and cultural consequences
of 1492. Westport, CT.: Greenwood Press.
Crouzet, Franois 1982: The Victorian Economy. New York:
Columbia University Press.
Curtin, Phillip 1984: Cross-cultural Trade in World History. Cambridge:
Cambridge University Press.
Davis, Ralph 1973: The Rise of the Atlantic Economies. Ithaca:
Cornell University Press.
Dehio, Ludwig 1962: The Precarious Balance. New York:
Vintage.
de Janvry, Alain 1981: The Agrarian Question and Reformism in
Latin America. Baltimore: Johns Hopkins University Press.
de Janvry, Alain and Frank Kramer 1979: "The limits of unequal
exchange." Review of Radical
Political Economics, 11(4), 3-15.
Deyo, Frederic C. 1981: Dependent Development and Industrial
Order: an Asian case study.
New York: Praeger.
Dixon, Marlene 1982: "Dual power: the rise of the
transnational corporation and the nation-state." Contemporary Marxism 5, 129-46.
Dixon, William J. 1985: "Change and persistence in the
world system: an analysis of global trade concentration, 1955-1975." International Studies Quarterly,
29(2), 171-90.
Dobb, Maurice 1963 [1947]: Studies in the Development of
Capitalism. New York: International Publishers.
Domhoff, G. William 1978: The Powers That Be: processes
of ruling class domination in America. New York:
Vintage.
Doran, Charles F. 1971: The Politics of Assimilation: hegemony
and its
aftermath. Baltimore:
Johns Hopkins University Press.
Doran, Charles F. and Wes Parsons 1980: "War and the cycle of relative
power." American Political
Science Review, 74, 947-65.
Doran, Charles, George Modelski and Cal Clark (eds) 1983: North/South
Relations: studies of dependency
reversal. New York: Praeger.
Dos Santos, Teotonio 1963: "El nuevo caracter de le
dependencia." Santiago: Cuadernos
de Estudios Socio-economicos (10), Centro de Estudios
Socio-economicos, Universidad de
Chile.
Durkheim, Emile 1964 (1893): The Division of Labor in Society. New York:
Free Press.
Duvall, Raymond D. and John R. Freeman 1981: "The state and dependent
capitalism." International
Studies Quarterly, 25(1), 99-118.
Eckstein, Susan 1983:
"Cuba and the capitalist world economy." In
C. Chase-Dunn (ed.) Socialist States in the World System,
Beverly Hills: Sage, 20318.
Edwards, Richard 1979: Contested Terrain: the transformation of the workplace in
the twentieth century. New
York: Basic Books.
Ekholm, Kasja and Jonathan Friedman 1980: "Toward a global
anthropology." In L. Blusse, H. L.
Wesseling, and G. D. Winius (eds) History and Underdevelopment,
Leyden: Center for the History of
European Expansion, Leyden University, 6176.
1982: "'Capital'
imperialism and exploitation in ancient world-systems." Review, 6(1), 87-110.
Emmanuel, A. 1972: Unequal Exchange: a study of the imperialism of trade. New York:
Monthly Review Press.
Engels, Frederick 1935 [1878]: Socialism: utopian and scientific. New York:
International Publishers.
1972 [1884]: The Origin of the Family, Private Property and the
State. New York: International Publishers.
Evans, Peter 1979: Dependent Development: the alliance of multinational, state and local capital in Brazil. Princeton:
Princeton University Press.
1985: "Transnational linkages and the economic role of the
state: an analysis of developing and
industrialized nations in the postWorld War II period." Pp. 192226 in Evans, Rueschemeyer and Skocpol
(eds), 1985.
1986: "State,
capital and the transformation of dependence:
the Brazilian computer case."
World Development, 14(7), 791-808.
Evans, Peter and Michael Timberlake 1980: "Dependence, inequality and growth in
less developed countries." American Sociological Review, 45,
531-52.
Evans, Peter, Dietrich Rueschemeyer, and Theda Skocpol (eds)
1985: Bringing the State Back In.
Cambridge: Cambridge University
Press.
Evans, Peter, Dietrich Rueschemeyer, and Evelyne Huber Stephens
(eds) 1985: States versus markets in the world-system. Beverly Hills: Sage.
Evans, Peter and John Stephens 1988: "Development and the world
economy." In Neil J. Smelser (ed), Handbook
of Sociology. Newbury Park, CA: Sage.
Falk, Richard 1982:
"Contending approaches to world order." Pp. 146-174
Toward a Just World Order, Vol. 1. In Richard Falk, Samuel S. Kim and Saul H.
Mendlovitz (eds), Boulder, CO: Westview
Press.
Feagin, Joe R. 1985:
"The global context of metropolitan growth: Houston and the oil industry." American Journal of Sociology, 90(6),
120429.
Fernand Braudel Center n.d.:
"Spatio-temporal map of the world-economy." Unpublished manuscript, SUNY-Binghamton.
Ferris, Wayne H. 1973: The Power Capability of Nation-States: international conflict and war. Lexington, MA.: D. C. Heath.
Fiala, Robert 1983: "Inequality and the service sector in
less developed countries: a reanalysis."
American Sociological Review 48:42128.Fieldhouse, D. K. 1966: The
Colonial Empires from the Eighteenth Century. New York:
Dell.
Finley, M. I. 1973: The Ancient Economy. Berkeley:
University of California Press.
Firebaugh, Glenn 1985: "Coreperihery patterns in
urbanization." In
Timberlake, (ed), 293304, 1985.
Firebaugh, Glenn and Bradley P. Bullock 1986: "Level of processing of exports: new estimates for 73 less-developed countries
in 1970 and 1980." International
Studies Quarterly 30(3), 333-50.
Fischer, Fritz 1967: Germany's Aims in the First World War. New York:
Norton.
Flannery, Kent V. 1982:
"The golden marshalltown: a
parable for the
archeology of the 1980s."
American Anthropologist, 84(2).
Flora, Peter 1983: State, Economy and Society in Western
Europe, 1815-1975, Vol. 1, The Growth of Mass Democracies and Welfare
States. Chicago:
St. James Press.
Ford, Alec George 1968:
"Overseas lending and international fluctuations: 1879-1914." In A. R. Hall (ed.), The Export of Capital
from Britain 1870-1914, London:
Methuen, 84102.
Forsythe, Dall W. 1977: Taxation and Political Change in the
Young Nation, 1781-1833. New York:
Columbia University Press.
Foster-Carter, Aidan
1978: "The modes of
production controversy." New
Left Review, 107, 47-78.
Fox, Richard G. 1985: Lions of the Punjab: culture in the making. Berkeley:
University of California Press.
1987: "Gandhian
socialism and Hindu nationalism:
cultural domination in the world-system." Journal of Commonwealth and Comparative
Politics, 9, 23347.
Fox-Genovese, Elizabeth and Eugene D. Genovese 1983: Fruits
of Merchant Capital: slavery
and bourgeois property in the rise and expansion of capitalism. New
York: Oxford University Press.
Frank, Andre Gunder 1969:
Latin America: Underdevelopment
or revolution? New York: Monthly
Review Press.
1978: World Accumulation, 1492-1789. New York:
Monthly Review Press.
1979a: Dependent Accumulation and Underdevelopment. New York:
Monthly Review Press.
1979b: Mexican Agriculture 1521-1630: transformation of the mode of
production. Cambridge:
Cambridge University Press.
1980: "Long live transideological enterprise!" In A.
G. Frank, Crisis: In the World
Economy. New York: Holmes and Meier, 178262.
1981: Crisis: In the Third World. New York:
Holmes and Meier.
Frederick, Suzanne Y. 1987: "The instability of free
trade." In George Modelski (ed.) Exploring
Long Cycles. Boulder: Lynne Rienner,
186217.
Friedmann, Harriet 1978: "World market, state and family
farm: social bases of household
production in an era of wage labor."
Comparative Studies in Society and History, XX(4), 545-86.
Frbel, Folker, Jrgen Heinrichs and Otto Kreye 1980: The New
International Division of Labor.
Cambridge: Cambridge University
Press.
Galtung, Johan 1971:
"A structural theory of imperialism." Journal of Peace Research, 8(2),
81-117.
1980: The True Worlds:
a transnational perspective.
New York: The Free Press.
1981: "Western
civilization: anatomy and
pathology." Alternatives, 7,
145-69.
Galtung, Johan, Tore Heiestad, and Erik Rudeng
1980: "On the decline and fall of
empires: The Roman Empire and Western imperialism compared." Review, IV(1), 91-154.
Genovese, Eugene D. 1971: The World the Slaveholders Made. New York:
Vintage.
1974: Roll, Jordan Roll.
New York: Pantheon.
Gereffi, Gary 1983: The Pharmaceutical Industry and
Dependency in the Third World.
Princeton: Princeton University
Press.
Gereffi, Gary and Peter Evans 1981: "Transnational
corporations, dependent development and
state policy in the semiperiphery: a comparison of Brazil and Mexico." Latin American Research Review, 16(3),
31-64.
Gershenkron, Alexander 1962: Economic Backwardness in
Historical Perspective.
Cambridge: Harvard University
Press.
Gibson, Bill 1980: "Unequal exchange: theoretical and
empirical findings." Review of
Radical Political Economics, 12(3), 15-35.
Giddens, Anthony 1979: Central Problems in Social Theory:
action, structure and contradiction in social analysis. Berkeley:
University of California Press.
Gill, Stephen and David Law 1986: The Global Political
Economy. Brighton: Wheatsheat.
Gilpin, Robert 1975: U.S. Power and the Multinational
Corporation. New York: Basic Books.
1981: War and
Change in World Politics.
Cambridge: Cambridge
University Press.
Goldfrank, Walter 1977:
"Who rules the world? class formation at the inter
national level." Quarterly
Journal of Ideology, 1(2), 32-7.
1978: "Fascism and
world-economy." In Barbara H.
Kaplan (ed.), Social Change in the Capitalist World Economy. Beverly Hills: Sage, 75120. 1979a: "Paradigm regained: The rules of Wallerstein's world-system
method." A paper presented at
conference on Methods of Historical Social Analysis, Cambridge, MA., October
18-20.
1981: "Silk and steel: Japan and Italy between the wars." In Richard Tomasson (ed.), Comparative
Social Research, Vol. 4, Greenwich, Conn.:
Jai Press.
1983: "The limits of
an analogy: Hegemonic decline in Great
Britain and the United States." In
Albert Bergesen (ed.), Crises and the World-
System, Beverly Hills:
Sage, 14354.
Goldfrank, Walter (ed.) 1979b: The World-System of Capitalism: past and present. Beverly
Hills: Sage.
Goldstein, Joshua S. 1988: Long Cycles: prosperity and
war in modern age. New Haven: Yale University Press.
Gordon, David M. 1980:
"Stages of accumulation and long economic cycles." In Terence K. Hopkins and Immanuel
Wallerstein (eds), Processes of the World-System, Beverly
Hills: Sage, 945.
Gramsci, Antonio 1971 (1935): Selections from the Prison
Notebooks. New York: International Publishers.
Griffin, Keith B. 1969: Underdevelopment in Spanish America. London:
Allen and Unwin.
Habermas, Jrgen 1970: Toward a Rational Society. Boston:
Beacon.
Hale, Charles A.1968: Mexican Liberalism in the Age of Mora,
1821-1853. New Haven:
Yale University Press.
Hall, Thomas D. 1986:
"Incorporation in the world-system:
toward a
critique." American
Sociological Review, 51(3), 390-402.
1988: Social Change in the Southwest, 13501880. Greenwich, CT: Jai Press.
Hamilton, Nora 1982: The Limits of State Autonomy: post-revolutionary mexico. Princeton:
Princeton University Press.
Hannan, Michael T. 1971:
"Problems of aggregation."
In H. M. Blalock (ed.), Causal Models in the Social Sciences,
Chicago: Aldine.
Hannan, Michael T. and Alice A. Young 1976: "On certain similarities in the
estimation of multi-wave panels and
multi-level cross sections." Technical Report 16, Consortium on
Methodology for Aggregating Data in Educational Research, Vasquez Associates,
Milwaukee.
1977: "Estimation in
panel models: results on pooling
cross sections and
time-series." In David R. Heise
(ed.), Sociological Methodology, San Francisco: Jossey-Bass.
Hannan, Michael T. and Glenn R. Carroll 1981: "Dynamics of
formal political structure: an event history analysis." American Sociological Review, 46,
19-35.
Harris, Marvin 1968: The Rise of Anthropological Theory. New York:
Harper and Row.
Hart, Jeffrey A. 1980:
"The policy of the U.S. toward the NIEO." Paper prepared for the fifth annual Hendricks
Symposium, "U.S. International Economic Policy in an Age of
Scarcities," University of
Nebraska-Lincoln, April 1011.
Hartman, John and Pamela B. Walters 1985: "Dependence, military assistance and
development: a crossnational
study." Politics and Society
14,
43158.
Harvey, David 1982: The Limits to Capital. Chicago:
University of Chicago Press.
Hawley, Amos 1981: Urban Society: an ecological approach, 2nd
edition. New York: Wiley.Hawley, James P. 1983: "Interests, state foreign economic
policy, and the world-system: the case
of the U.S. capital control programs, 1961-1974." In Pat McGowan and Charles W. Kegley (eds), Foreign
Policy and the Modern World-System, Beverly Hills: Sage, 22354.
Hechter, Michael 1975: Internal Colonialism: the Celtic fringe in British national development 1536-1966. Berkeley:
University of California Press.
Heintz, Peter with the collaboration of S. Heintz 1973: The
Future of
Development. Berne:
Huber.
Henderson, W. D. 1965: Britain and Industrial Europe: studies in British influence on the industrial revolution in
Western Europe, 1750-1870.
London: Leicester University
Press.
Henige, David P. 1970: Colonial Governors from the 15th
Century to the Present.
Madison: University of Wisconsin
Press.
Herrera, Amilcar O., Hugo D. Scolnik, Graciela Chichilnisky,
Gilberto C. Gallopin, Jorge E. Hardoy, Diana Mosrovich, Enrique Oteiza, Gilda
L. de Romero Brest, Carlos E. Suarez and Luis Talavera 1976: Catastrophe or
New Society?: a Latin American
world model. Ottawa: International Development Research.
Hibbs, Douglas A., Jr. 1974:
"Problems of statistical estimation and causal inference in
time-series regression models." In
Herbert Costner (ed.), Sociological Methodology 1973-74, San
Francisco: Jossey-Bass.
Hilferding, Rudolf 1981 [1923]: Finance Capital: a study of the latest phase of
capitalist development. London: Routledge and Kegan Paul.
Hilton, Rodney (ed.) 1976: The Transition from Feudalism to
Capitalism. London: New Left Books.
Hindess, Barry and Paul Q. Hirst 1975: Pre-Capitalist Modes
of Production. London: Routledge & Kegan Paul.Hinsley, F. H.
1967: Power and the Pursuit of Peace.
London: Cambridge University
Press.
Hirschman, Albert O. 1980 [1945]: National Power and the
Structure of Foreign Trade.
Berkeley: University of
California Press.
Hobsbawm, E. J. 1959: Primitive Rebels. New York:
Norton
1968: Industry and
Empire. Baltimore: Penguin.
Hopkins, Keith 1978a:
"Economic growth and towns in classical antiquity." In Philip Abrams and E. A. Wrigley (eds), Towns
in Societies, Cambridge: Cambridge University Press.
1978b: Conquerors and Slaves. Cambridge:
Cambridge University Press.
Hopkins, Terence K. 1978:
"World-system analysis:
methodological issues." In
Barbara H. Kaplan (ed.), Social Change in the Capitalist World Economy, Beverly Hills: Sage Publications, 199218.
1979: "The study of
the capitalist world-economy: some introductory
considerations." In
Walter Goldfrank (ed.), The World-System of
Capitalism: Past and Present, Beverly Hills: Sage Publications, 2152.
Hopkins, Terence K. and Immanuel Wallerstein 1967: "The comparative study of national
societies." Social Science
Information, 6(5) 25-8. Reprinted in
Amatai Etzioni and Fredric L. Dubow (eds), Comparative Perspectives: Theories and Methods, Boston: Little, Brown, 1970, 183204.
1977: "Patterns of
development in the modern world-system."
Review 1(2), 11146.
Reprinted in T. K. Hopkins and I. Wallerstein (eds), WorldSystems
Analysis: Theory and Methodology,
Beverly Hills: Sage, 1982,
4182.
1979:
"Cyclical rhythms and secular trends of the capitalist world-
economy." Review
2(4), 483-500. Reprinted in T. K.
Hopkins and I. Wallerstein (eds), WorldSystems Analysis, Beverly
Hills: Sage, 1982,
10420.
1986: "Commodity
chains in the world-economy prior to 1800." Review 10(1), 157-70.
Hopkins, Terence K. and Immanuel Wallerstein (eds) 1982: World-Systems
Analysis: theory and methodology. Beverly Hills: Sage.
Hyden, Goran 1980: Beyond Ujamaa in Tanzania: underdevelopment and an
uncaptured peasantry. Berkeley:
University of California Press.
Hymer, Stephen 1979: The Multinational Corporation: a radical approach. Cambridge:
Cambridge University Press.
Inkeles, Alex 1975:
"The emerging social structure of the world." World
Politics, 27(4), 467-95.
Jacobs, Jane 1984: Cities and the Wealth of Nations. New York:
Random House.
Jenkins, Rhys 1984:
"Divisions over the international division of labor." Capital and Class, 22, 28-57.
Johnson, Dale L. (ed.) 1985: Middle Classes in Dependent
Countries. Beverly Hills, CA.: Sage.
Jones, Gareth Stedman (1971): Outcaste London: a study in the relationship between
classes in Victorial society.
Oxford: Clarendon.
Kahn, Joel S. 1980: Minangkabau Social Formations: Indonesian peasants and the world-economy. New York:
Cambridge University Press.
Kaldor, Mary 1978: The Disintegrating West. New York:
Hill and Wang.
Kay, Geoffrey 1975: Development and Underdevelopment: a Marxist analysis. London:
St. Martin's Press.
Kelly, Raymond C. 1985: The Nuer Conquest: the
structure and development of an expansionist system. Ann Arbor: University of Michigan Press.
Kentor, Jeffrey 1981:
"Structural determinants of peripheral urbanization: the effects of international
dependence." American
Sociological Review, 46, 201-11.
Keohane, Robert and Joseph S. Nye 1977: Power and
Interdependence. Boston: Little, Brown.
Keyfitz, Nathan 1978:
"Understanding world models."
In Karl F. Schuessler (ed.), Sociological Methodology 1978. San Francisco: Jossey-Bass.
Kick, Edward L. 1980""World-system properties and mass
political conflict within nations: theoretical framework." Journal of Political and Military
Sociology, 8, 175-90.
King, Anthony D. 1984: The Bungalow: the production of a global culture. London:
Routledge and Kegan Paul.
Kohl, Philip 1987:
"The use and abuse of world systems theory: the case of
the 'pristine' West Asian state."
In Archeological Advances in Method and Theory, 11, New
York: Academic Press, 135.
Kondratieff, N. D. 1979 (1926):
"The long waves in economic life." Review 2(4), 519-62.
Kowalewski, David 1987:
"Peripheral revolutions in world-system perspective 1821-1985." A paper presented at the annual meetings of
the International Studies Association, Washington, D.C., April.
Krasner, Stephen D. 1976:
"State power and the structure of international trade." World Politics, 28(3), 317-47.
1978: Defending the National Interest: raw materials
investments and U.S. foreign policy.
Princeton, N.J.: Princeton
University Press.
1985: Structural Conflict: the Third World against global liberalism. Berkeley:
University of California Press.
Kravis, Irving B., Alan Heston, Robert Summers 1982: World
Product and Income: international
comparisons of real gross product.
Baltimore: Johns Hopkins
University Press.
Kriedte, Peter, Hans Medick and Jrgen Schlumbohm 1981" Industrialization
Before Industrialization: rural
industry in the genesis of capitalism.
Trans. Beate Schempp.
Cambridge: Cambridge University
Press.
Kugler, Jacek and A. F. K. Organski 1986: "The end of hegemony: says who?" Paper prepared for the annual meeting of the
International Studies Association, Anaheim, CA. March.
Kuhn, Thomas S. 1962: The Structure of Scientific Revolutions. Chicago:
University of Chicago Press.
Kukreja, Sunil and James D. Miley 1988: "The relationship
between world system position and the division of labor: a crossnational
analysis."
Sociological Spectrum
8:4966.
Kuznets, Simon 1971: Economic Growth of Nations: total output and production structure. Cambridge:
Harvard University Press.
LaClau, Ernesto 1977: Politics and Ideology in Marxist Theory. London:
New Left Books.
Landes, David 1968: The Unbound Prometheus. Cambridge:
Cambridge University Press.
Lane, Frederic C. 1966:
"Units of economic growth historically considered." In Venice and History: The Collected Papers of Frederic C. Lane, Baltimore:
Johns Hopkins University Press, 496504.
1973: Venice: a maritime republic. Baltimore: Johns Hopkins University Press.
1979: Profits from Power: readings
in protection rent and violence-
controlling enterprises. Albany:
State University of New York Press.
Lange, Peter 1985:
"Semiperiphery and core in the European context:
reflections on the postwar Italian experience." In G. Arrighi (ed.), Semiperipheral
Development, Beverly Hills: Sage,
179214.
Lpple, Dieter 1985:
"Internationalization of capital and the regional problem." In John Walton (ed.), Capital and Labour
in the Urbanized World, London:
Sage, 4375.
Lattimore, Owen 1940: Inner Asian Frontiers of China. New York:
American Geographical Society.
1962: Studies in Frontier History: collected papers 1928-1958. London:
Oxford University Press.
Lee, Su-Hoon 1986: State-building in the Periphery, 1960-1980. Ph.D.
dissertation, Department of Sociology, Johns Hopkins University.
Lenin, V. I. 1965 [1916]: Imperialism: the highest stage of capitalism. Peking:
Foreign Languages Publishing House.
Lenski, Gerhard and Jean Lenski 1982: Human Societies. New York:
McGraw-
Hill. 4th edition.
Lenski, Gerhard and Patric Nolan 1984: "Trajectories of development: a test of ecological-evolutionary
theory." Social Forces,
63(1), 1-23.
Levy, Jack S. 1983: War in the Modern Great Power System,
1495-1975.
Lexington: University
Press of Kentucky.
1985: "Theories of
general war." World Politics,
27(3), 344-74.
Lipset, Seymour M. 1977:
"Why no socialism in the United States?" In
S. Bialer and S. Sluzar (eds), Sources of Contemporary
Radicalism, Boulder: Westview, 31149.
1981:
"Industrial proleteriat in comparative perspective." In J. Triska and C. Gati (eds), Blue
Collar Workers in Eastern Europe.
London: Allen and Unwin, 128.
Lipson, Charles 1985: Standing Guard: protecting foreign capital in the nineteenth
and twentieth centuries.
Berkeley: University of
California Press.
Lipton, Michael 1977: Why Poor People Stay Poor: urban bias in world
development. Cambridge:
Harvard University Press.
London, Bruce and Bruce Williams 1988: "Multinational corporate penetration,
protest and basic needs provision in non-core nations: a crossnational analysis." Social Forces 63(3), 1632.
Lunday, James E. 1980:
"State formation in the United States, 1837-1860: effects of world market position on political
change and mobilization for war in the Northern states." Ph.D., Department of Sociology, Johns Hopkins
University.
Luxemburg, Rosa 1968 (1913): The Accumulation of Capital. New York:
Monthly Review Press.
McGowan, Pat 1980:
"A political-economy of the world-system perspective on global modeling." Paper prepared for the conference on
Large-Scale Global Modeling, Berlin Science Center, West Berlin, July 14-17.
1985: "Pitfalls and
promises in the quantitative study of the world-
system: a reanalysis of
Bergesen's 'long waves' of colonialism."
Review 8(4), 477500.
McKeown, Timothy J. 1983: "Hegemonic stability theory and
19th century tariff levels in Europe."
International Organization 37, 1:7391.
McNeill, William H. 1963: The Rise of the West. Chicago:
University of Chicago Press. 1964: Europe's Steppe Frontier,
1500-1800. Chicago: University of
Chicago Press.
1982: The Pursuit of Power: technology, armed force and society since
a.d. 1000. Chicago:
University of Chicago Press.
Maddison, Angus 1982: Phases of Capitalist Development. New York:
Oxford University Press.
Malinowski, Bronislaw 1961 [1921]: Argonauts of the Western
Pacific. New York: Dutton.
Mandel, Ernest 1970: An Introduction to Marxist Economic
Theory. New York: Pathfinder Press.
1975: Late Capitalism.
London: New Left Books.
1978: The Second Slump.
London: New Left Books.
1980: Long Waves of Capitalist Development: the Marxist interpretation. London:
Cambridge University Press.
Mann, Michael 1986: The Sources of Social Power: a history of power from the beginning
to a.d. 1760. Cambridge: Cambridge University Press.
Mao Zedong 1967 [1937]:
"On contradiction." In Selected
Works, Vol. 1, Peking: Foreign Languages Publishing House.
Marini, Ruy Mauro 1972:
"Brazilian sub-imperialism." Monthly Review,
23(9), 14-24.
Marshall, T. H. 1965"Class, Citizenship and Social
Development. New York: Anchor Books, Doubleday and Co.
Marx, Karl 1967a [1867]: Capital, Vol. I. New York: International Publishers.
1967b [1894]: Capital, Vol. III. New York:
International Publishers.
1976 [1867]: Capital, Vol. l. Harmondsworth, Middlesex: Penguin.
1973 [185758]: Grundrisse: foundations of the critique
of political economy. New York: Random House.
1978 [1852]: The Eighteenth Brumaire of Louis
Bonaparte. Peking: Foreign Languages
Press.
Mathien, Frances J. and Randall H. McGuire (eds.) 1986: Ripples
in the Chichimec Sea: new
considerations of southwestern-mesoamerican
interactions. Carbondale:
Southern Illinois University Press.
Meillassoux, Claude 1981: Maidens, Meal, and Money: capitalism and the domestic economy. Cambridge:
Cambridge University Press.
Mensch, Gerhard 1978: Stalemate in Technology. Cambridge:
MA., Ballinger.
Merton, Robert K. 1957: Social Theory and Social Structure. Glencoe, Illinois: The Free Press.
Meyer, David R. 1986:
"The world system of cities:
relations between international
financial metropolises and South American cities." Social Forces, 64(3), 553-81.
Meyer, John W. 1987:
"The world polity and the authority of the nation-state." In George M. Thomas, John W. Meyer, Francisco
O. Ramirez and John Boli, Institutional Structure: constituting state, society, and the
individual, Beverly Hills: Sage,
4170.
Meyer, John W. and Michael T. Hannan (eds) 1979: National
Development and the World System: educational,
economic and political change, 1950-1970.
Chicago: University of Chicago
Press.
Meyer, John W., John Boli-Bennett, and Christopher Chase-Dunn
1975:
"Convergence and divergence in development." Annual Review of Sociology, V(1),
223-46.
Meyer, William 1987:
"Testing theories of cultural imperialism: international media and domestic
impact." International
Interactions, 13(4), 35374.
Michalet, Charles-Albert 1976: Le Capitalisme Mondial. Paris:
Presses
Universitaires de France.Milkman, Ruth 1979: "Contradictions of semi-peripheral
development: The South African
case." In Walter L. Goldfrank
(ed.), The World-system of Capitalism: past and present, Beverly Hills: Sage, 26184.
Mills, C. Wright 1959: The
Sociological Imagination. New
York: Grove Press.
Mintz, Sidney W. 1974: Caribbean Transformations. Baltimore:
Johns Hopkins University Press.
1977: "The so-called
world system: local initiative and local
response." Dialectical Anthropology, 2, 253-70.
1978: "Was the plantation slave a proletarian?" Review, II(1), 81-98.
1985: Sweetness and Power: the place of sugar in modern history. New York:
Viking.
Mitchell, Brian R. 1975: European Historical Statistics
1750-1970. New York: Columbia University Press.
1982: International Historical Statistics: Africa and Asia. New York:
New York University Press.
Modelski, George 1964:
"Kautilya: foreign policy
and international system in the ancient Hindu world." American Political Science Review,
58(3), 549-60.
1978: "The long cycle of global politics and the
nation-state."
Comparative Studies in Society and History, 20(2), 214-35.
1982: "Long cycles and the strategy of United States
international economic policy." In
William P. Avery and David P. Rapkin
(eds), America in a Changing World Political Economy, New York: Longman, 97118.
Modelski, George and Thompson, W. R. 1987: "Testing cobweb
models of the long cycle of world leadership." In G. Modelski (ed.), Exploring Long
Cycles, Boulder: Lynne Rienner.
1988: Sea Power in Global Politics, 14941943. Seattle:
University of Washington Press.
Molotch, Harvey 1984:
"Romantic Marxism: Love
is (still) not enough." A
review of Manuel Castells, The City
and the Grassroots. Contemporary
Sociology, 13(2).
Moon, Bruce E. 1983: "The foreign policy of the dependent
capitalist state." International
Studies Quarterly 27, 3:31540.
Moore, Barrington 1966: The Social Origins of Dictatorship
and Democracy. Boston: Beacon Press.
Morganthau, Hans 1952: Politics Among Nations. New York:
Alfred A. Knopf.
Morishima, Michio 1973: Marx's Economics: a dual theory of value and growth. Cambridge:
Cambridge University Press.
Morrison, Donald G. et al. 1972: Black Africa: a comparative handbook. New York:
Free Press.
Mouzelis, Nicos 1986: Politics in the SemiPeriphery: early parliamentarism and late
industrialization in the Balkans and Latin America. New York:
St. Martin's.
Mller, Georg 1988: Comparative World Data: a statistical handbook for social science. Baltimore:
Johns Hopkins University Press.
Murdock, George P. 1957:
"World ethnographic sample."
American
Anthropologist,
59, 664-87.
Murphy, Craig 1984: The Emergence of the NIEO Ideology. Boulder:
Westview Press.
Murray, Martin 1980: The Development of Capitalism in
Colonial Indochina (1870-1940). Berkeley:
University of California Press.
Myrdal, Gunnar 1957: Economic Theory and Underdeveloped
Regions. New York: Harper and Row.Naroll, Raoul 1968: "Some thoughts on comparative method in
cultural
anthropology." In
Hubert M. Blalock, Jr. and Ann B. Blalock (eds), Methodology in Social
Research, New York: McGraw-Hill,
23677.
Nemeth, Roger and David A. Smith 1985: "International trade and world-system
structure: a multiple network
analysis." Review, 8(3).
Neuman, Stephanie G. 1984: "International stratification
and third world military industries."
International Organization 38, 1:16797.
Nolan, Patrick D. 1984:
"Status in the world economy and national structure and
development." In Gerhard Lenski
(ed.), Current Issues and Research in Macrosociology,
International Studies in Sociology and
Social
Anthropology,
vol. 37, Leiden: E. J. Brill, 10920.
1988: "World system status, technoeconomic heritage, and
fertility." Sociological Focus
21, 1:933.
Nolan, Patrick and Gerhard Lenski 1985: "Technoeconomic
heritage, patterns of development, and the advantage of
backwardness." Social Forces,
64(2), 341-58.
Nowak, Leszek 1971: "Problems of explanation in Marx's Capital." Quality and Quantity, 2, 31137.
O'Brien, Patrick 1982:
"European economic development:
the contribution of the periphery."
The Economic History Review, 35(1), 1-18.
O'Conner, James 1973: The Fiscal Crisis of the State. New York:
St. Martin's Press.
O'Donnell, Guillermo 1978:
"Reflections on the patterns of change in the bureaucratic-authoritarian state." Latin American Research Review, 13,
3-38.
1979: "Tensions in the bureaucratic-authoritarian
state and the question of democracy."
In David Collier (ed.), The New Authoritarianism in Latin America,
Princeton: Princeton University Press,
285318.
Ollman, B. 1976: "Alienation: Marx's Conception
of Man in Capitalist Society. 2nd
Edition. New York: Cambridge University Press.
Olson, Mancur 1982: The Rise and Decline of Nations. New Haven:
Yale University Press.
Organization for Economic Cooperation and Development (OECD)
19691974:
Geographical Distribution of Financial Flows to Developing
Countries. Paris:
OECD.
1982: External Debt of Developing Countries. Paris:
OECD.
Organski, A. F. K. 1968: World
Politics. New York: Alfred A. Knopf.
Organski, A. F. K. and Jacek Kugler 1980: The War Ledger. Chicago:
University of Chicago Press.
Paige, Jeffery M. 1975: Agrarian Revolution: social movements and export agriculture in the underdeveloped world. New York:
MacMillan.
Palloix, Christian 1979:
"The self-expansion of capital on a world scale." Review of Radical Political Economics,
9, 1-28.
Palma, Gabriel 1978:
"Dependency: a formal theory
of underdevelopment or a methodology for the analysis of concrete situations of
underdevelopment?" World
Development 6, 7/8, 881924.
Parry, J. H. 1966: The Establishment of European Hegemony
1415-1715. New York: Harper and Row.
Parsons, Talcott 1961:
"Order and community in the international system." In James N. Rosenau (ed.), International
Politics and Foreign Policy. New
York: Free Press, 12029.
1966: Societies:
Evolutionary and Comparative Perspective. Englewood Cliffs: Prentice-Hall.
1971: The System of Modern Societies. Englewood Cliffs, N.J.:
Prentice-Hall.
Petras, James with A. E. Havens, M. H. Morley and P. DeWitt
1981: Class, State and Power in the Third World. Montclair, N.J.: Allanheld, Osmun.
Pfister, Ulrich 1984: "Debt and development: a cross-sectional study of the relationships
between external debt and economic growth in developing countries." Sociological Institute, University of Zrich.
Pfister, Ulrich and Christian Suter 1987: "International financial relations as
part of the world system." International
Studies Quarterly, 31(3), 23972.
Phelps-Brown, E. H. and S. J. Handfield-Jones 1952: "The climacteric of the
1890s." Oxford Economic Papers, New
Series, 4(3).
Phelps-Brown, E. H. and S. V. Hopkins 1955: "Seven centuries of building
wages." Economica, August,
195-206.
Polanyi, Karl 1944: The Great Transformation. Boston:
Beacon.
1957: "The economy
as instituted process." In Karl
Polanyi, Conrade M. Arensberg, and Harry W. Pearson (eds), Trade and Market
in the Early Empires, Chicago:
Regnery 24369.
1977: The Livelihood of Man. Harry W. Pearson (ed.). New York:
Academic Press.
Polanyi, Karl, Conrad M. Arensberg, and Harry W. Pearson (eds)
1957: Trade and Market in the Early Empires. Chicago:
Regnery.
Popper, Karl 1972: The Poverty of Historicism. London:
Routledge and Kegan Paul.
Portes, Alejandro 1981:
"The informal sector:
definition, controversy, and relation to national
development." Review, 7(1),
151-74.
Portes, Alejandro and Robert Bach 1985: Latin Journey: Cuban and Mexican
Immigrants in the United States. Berkeley: University of California Press.
Portes, Alejandro and A. Douglas Kincaid 1985: "The crisis of
authoritarianism: state
and civil society in Argentina, Chile and Uruguay." Research in Political Sociology, 1,
49-77.
Portes, Alejandro and John Walton 1981: Labor, Class, and the
International System. New
York: Academic Press.
Poulantzas, Nicos 1973: Political Power and Social Classes. London:
New Left Books.
1975: Classes in Contemporary Capitalism. London:
New Left Books.
Prebisch, Raul 1949: The Economic Development of Latin
America and its Principal
Problems. CEPAL (E/CN. 12/89/Rev.
1). Reprinted in Economic Bulletin
for Latin America, 7 (1), 1962.
Ragin, Charles C. and David Zaret 1983: "Theory and method in comparative
research: two strategies." Social Forces, 61(3), 731-54.
Ragin, Charles and York Bradshaw 1986: "International economic dependence and human
misery, 1938-1980: a global
perspective." Presented at the
annual meeting of the International Studies Association, Anaheim, CA.
Ramirez, Francisco O. and Richard Rubinson 1979: "Creating members: the political incorporation and expansion of
public education." In John W. Meyer
and Michael T. Hannan (eds), National Development and the World System: Educational, Economic and Political
Change, 1950-1970, Chicago:
University of Chicago Press, 7284.
Ray, James Lee and J. David Singer 1973: "Measuring the concentration of power in
the international system." Sociological
Methods and Research, 1(4), 403-37.
Renfrew, Colin 1984:
"Systems collapse as social transformation." In
C. Renfrew, Approaches to Social Archeology. Cambridge:
Harvard
University Press, 36689.
Rescher, Nicholas 1977: Dialectics: a controversy-oriented Approach to the theory
of knowledge. Albany, N.Y.: SUNY Press.
Riley, James C. 1980: International Government Finance and
the Amsterdam Capital Market 1740-1815.
Cambridge: Cambridge University
Press.
Robertson, Roland and Frank Lechner 1985: "Modernization, globalization and the
problem of culture in world-systems theory." Theory, Culture and Society, 2(3),
103-17.
Robinson, R. and J. Gallagher 1953: "The imperialism of free
trade." Economic History Review,
6(1), 1-14.
Rodney, Walter 1981: A History of the Guyanese Working
People, 1881-1905. Baltimore: The Johns Hopkins University Press.
Rose, Stephen J. 1986: The American Profile Poster. New York:
Pantheon.
Ross, Robert J. S. and Kent Trachte 1983: "Global cities and global classes: the peripheralization of labor in New York
City." Review 6(3), 393-431.
Forthcoming: The New Leviathan: Global Capitalism. Berkeley:
University of California Press.
Rostow, W. W. 1978: The World Economy: history and prospect. Austin:
University of Texas Press.
Rowlands, Michael, Mogens Larsen and Kristian Kristiansen (eds)
1987: Centre and Periphery in the Ancient World. Cambridge:
Cambridge University Press.Rubinson, Richard 1976: "The world-economy and the distribution
of income within states: a
cross-national study." American
Sociological Review, 41, 638-59.
1977a: "Reply to
Bach and Irwin." American
Sociological Review, 42(5), 817-21.
1977b: "Dependence,
government revenue, economic growth, 1955-1970." Studies in Comparative
International Development, 12, 3-28.
1978: "Political
transformation in Germany and the United States." In Barbara H. Kaplan (ed.), Social Change
in the Capitalist World Economy,
Beverly Hills: Sage, 3974.
Rubinson, Richard and Deborah Holtzman 1981: "Comparative dependence and economic
development." International
Journal of Comparative Sociology, 22(1-2), 86-101.
Rueschemeyer, Dietrich and Peter B. Evans 1985: "The state and economic transformation: toward an analysis of the conditions
underlying effective intervention."
In Peter B. Evans et al. (eds), Bringing the State Back In. Cambridge:
Cambridge University Press, 4477.
Rupert, Mark E. and David P. Rapkin 1985: "The erosion of U.S. leadership
capabilities." In Paul M. Johnson
and William R. Thompson (eds), Rhythms
in Politics and Economics, New York: Praeger, 15580.
Sahlins, Marshall 1972: Stone Age Economics. New York:
Aldine.
Sapir, Edward 1949: Selected Writings in Language, Culture
and Personality, David Mandelbaum
(ed.). Berkeley: University of
California Press.
Sassen-Koob, Saskia 1985:
"Capital mobility and labor migration: their expression in core cities." In Michael Timberlake (ed.), Urbanization
in the World-Economy, New York:
Academic Press, 23168.
Schiller, Herbert I. 1969: Mass Communications and
the American Empire. New York: A. M. Kelley.
Schlote, Werner 1952 [1938]: British Overseas Trade from 1700
to the 1930s. Trans. W. O. Henderson
and W. H. Chaloner. Oxford: Basil Blackwell.
Schneider, Jane 1977:
"Was there a pre-capitalist world-system?" Peasant Studies, 6(1), 20-9.
Schumpeter, Joseph 1939: Business Cycles. Vol. 1.
New York: McGraw-Hill.
1955 [1919]: "The
sociology of imperialism." In Imperialism
and Social Classes. New York:
Meridian Books.
Seers, Dudley 1983: The Political Economy of Nationalism. New York:
Oxford University Press.
Sell, Ralph R. and Stephen J. Kunitz 19861987: "The debt crisis and the end of an era
in mortality decline." Studies
in Comparative International Development, 21(4), 330.
Semyonov, Moshe and Noah LewinEpstein 1986: "Economic development,
investment dependence, and the rise of services in less
developed nations." Social
Forces, 64(3), 58298.
Senghaas, Dieter 1985: The European Experience: a
historical critique of development
theory. Dover, New Hampshire: Berg Publishers.
Shils, Edward A. and Morris Janowitz 1948: "Cohesion and
disintegration in the Wehrmacht in World War II." Public Opinion Quarterly, 12, 280-315.
Singer, J. David 1979: The Correlates of War: I:
Research Origins and Rationale. New York:
Free Press.
Singer, J. David, Stuart Bremer, and John Stuckey 1979: "Capability
distribution, uncertainty and major power war,
1820-1965." In J. David Singer
(ed.), The Correlates of War: Research
Origins and Rationale. New York:
Free Press, 26598.Singer, J. David and Melvin Small 1966: "The composition and status ordering of
the international system:
18151940." World Politics,
28, 23682.
Sklar, Holly (ed.) 1980: Trilaterialism: The Trilateral Commission and Elite Planning for World Management. Boston:
South End Press.
Skocpol, Theda 1977:
"Wallerstein's world capitalist system: a theoretical and historical
critique." American Journal of Sociology,
82(5), 1075-90.
1979: States and Social Revolutions. Cambridge:
Cambridge University Press.
1985: "Bringing the
state back in: strategies of analysis
in current research." In Peter Evans, et al. (eds.), Bringing
the State Back In. Cambridge: Cambridge University Press, 343.
Small, Melvin and J. David Singer 1973: "The diplomatic importance of states,
18161970: an extension and refinement of
the indicator." World Politics,
25, 57799.
Smelser, Neil 1959:
"A comparative view of exchange systems." Economic Development and Cultural Change, 7,
173-82.
Smith, David A. and Douglas R. White 1986: "A dynamic analysis of inter
national trade and world-system structure: 1965-1980." A paper presented at the annual meeting of
The International Studies Association, Anaheim, CA. March.
Smith, Joan, Immanuel Wallerstein, and Hans-Dieter Evers 1984: Households
and the World-Economy.
Beverly Hills: Sage.
Smith, Neil 1984: Uneven Development: nature, capital and the production of Space.
Oxford: Basil Blackwell.
Snyder, David and Edward L. Kick 1979: "Structural position in the world-
system and economic growth, 1955-1970: a multiple network
analysis of transnational
interactions." American Journal
of Sociology, 84(5), 1096-126.
So, Alvin Y. 1986: The South China Silk District: local historical trans
formation and world-system theory. Albany: State University of New York Press.
Sokolovsky, Joan 1983:
"China and the struggle to withdraw from the world system."
In C. Chase-Dunn (ed.), Socialist States in the World-system, Beverly Hills: Sage, 15780.
1985: "Logic, space
and time: the boundaries of the
capitalist world-
economy." In Michael
Timberlake (ed.), Urbanization in the World-economy, New York:
Academic Press, 4152.
Stepan, Alfred 1978: The State and Society: Peru in comparative perspective. Princeton:
Princeton University Press.
Stephens, John 1980: The Transition from Capitalism to
Socialism. Atlantic Highlands,
N.J.: Humanities Press.
1987: "Democratic
transition and breakdown in Europe, 18701939:
a test of the Moore thesis."
Working Paper 101, Kellog Institute, University of Notre Dame.
Stevenson, Robert L. and Donald L. Shaw 1984: Foreign News
and the New World Information Order.
Ames: University of Iowa Press.
Stinchcombe, Arthur L. 1968: Constructing Social Theories. New York: Harcourt, Brace and World.
1978: Theoretical Methods in Social History. New York:
Academic Press.
1983: Economic Sociology.
New York: Academic Press.
Strange, Susan 1986: Casino Capitalism. Oxford and New York: Basil Blackwell.Strange,
Susan and Robert Tooze (eds) 1981: The International Politics of Surplus
Capacity: competition for market
shares in the world recession.
London: Geo. Allen and Unwin.
Sunkel, Osvaldo 1973:
"Transnational capitalism and national disintegration in Latin
America." Social and Economic
Studies, 22(1), 132-76.
Suter, Christian 1987:
"Long waves in core-periphery relationships within the international financial system: debt-default cycles of sovereign
borrowers." Review,
10(3).
Suter, Christian and Ulrich Pfister 1986: "Long waves in Third World
indebtedness: the
contribution of country case studies."
Sociological Institute, University of Zurich. Paper prepared for the ESRC Third World
Economic History and Development Group Conference, University of
Birmingham, September.
Sweezy, Paul 1976 [1950]:
"A critique." In Rodney
Hilton (ed.), The
Transition from Feudalism to Capitalism, London: New
Left Books, 3556.
Sylvan, David, Duncan Snidal, Bruce M. Russett, Steven Jackson,
and Raymond Duvall 1983: "The peripheral economies: penetration and economic distortion,
1970-1975." In William R. Thompson
(ed.), Contending Approaches to World System Analysis, Beverly
Hills: Sage, 79114.
Szymanski, Albert 1971: The Dependence of South America on
the United States. Ph.D.
Dissertation, Department of Sociology, Columbia University.
1973: "Military spending and economic
stagnation." American Journal of
Sociology, 79(1).
1975: "Dependence, exploitation and development." Journal of Military and Political
Sociology, l, 15-29.
1981: The Logic of Imperialism. New York:
Praeger.
1983a: Class Structure:
a critical perspective.
New York: Praeger. 1983b:
"The socialist world system."
In C. Chase-Dunn (ed.), Socialist States in the World-system,
Beverly Hills: Sage, 5784.
Taagepera, Rein 1978:
"Size and duration of empires:
systematics of size." Social
Science Research, 7, 10827.
Tardanico, Richard 1978:
"A structural perspective on state power in the capitalist
world-system." A paper prepared for
the annual meeting of the American Sociological Association, San Franciso.
1982: "Revolutionary
state-making in the periphery: Mexico,
1924-1928." Comparative Studies in Society and History, 23(2).
Taylor, Charles L. and M. C. Hudson 1972: World Handbook of
Political and Social Indicators.
New Haven: Yale University Press.
Taylor, John G. 1979: From Modernization to Modes of
Production: a critique of the sociologies of development and underdevelopment. London:
Macmillan.
Teune, Henry 1979:
"Cross-level analysis: a
case of social inference." Quality
and Quantity, 13, 52737.
Thomas, Brinley 1954: Migration and Economic Growth: A Study of Great Britain and the Atlantic
Economy. Cambridge: Cambridge University Press.
Thomas, Clive Y. 1984: The Rise of the Authoritarian State in
Peripheral Societies. New York:
Monthly Review Press.
Thomas, George M., Francisco O. Ramirez, John W. Meyer, and
Jeanne Gobalet
1979: "Maintaining
national boundaries in the world system: the rise of centralist
regimes." In John W. Meyer and
Michael T. Hannan, National Development and the World System,
Chicago: University of Chicago Press,
187206.
Thomas, George M., John W. Meyer, Francisco O. Ramirez, and John
Boli 1987:
Institutional Structure:
Constituting State, Society, and the
Individual. Beverly Hills: Sage.Thompson, E. P. 1978: The Poverty of
Theory and Other Essays. New
York: Monthly Review.
Thompson, William R. 1983a:
"Succession crises in the global political system: a test of the transition model." In Albert Bergesen (ed.), Crises in the
World-System, Beverly Hills: Sage,
93116.
1983b: "Cycles, capabilities and war: an ecumenical view." In W. R. Thompson (ed.), Contending
Approaches to World System Analysis, Beverly Hills: Sage, 14163.
1983c: "Uneven growth, systemic challenges, and global
wars."
International Studies Quarterly, 27(3), 34155.
1986: "Leading sectors, systemic leadership and global
war." Paper presented at the annual
meeting of the American Political Science
Association, Washington, D.C., August 28. Forthcoming in W. R. Thompson, On Global
War: Historical Structural Approaches to
World Politics. Columbia: University of South Carolina Press.
Thompson, William R. (ed.) 1983: Contending Approaches to
World-System Analysis.
Beverly Hills: Sage.
Thompson, William R. and Karen A. Rasler 1988: "War and systemic capability
reconcentration." Journal of
Conflict Resolution, 32.
Thompson, W. R. and L. G. Zuk 1982: "War, inflation and the Kondratieff long
wave." Journal of Conflict
Resolution, 26(4), 621-44.
Tilly, Charles 1984: Big Structures, Large Processes, Huge
Comparisons. New York:
Russell Sage.
1985: "War making
and state making as organized crime."
In Evans,
Rueschemeyer and Skocpol (eds), Bringing the State Back In,
Cambridge: Cambridge University Press,
16991.
Timberlake, Michael (ed.) 1985: Urbanization in the
World-Economy. New York: Academic Press.
Timberlake, Michael and Kirk R. Williams 1984: "Dependence, political
exclusion, and government repression: some cross-national evidence." American
Sociological Review, 49, 141-46.
Timberlake, Michael and James Lunday 1985: "Labor force structure in the zones of
the worldeconomy, 19501970." In
Timberlake (ed.), 32550.
Toynbee, Arnold 1967: "Anarchy by treaty,
1648-1967." In Fred L. Israel
(ed.), Major Peace Treaties of Modern History, New York: McGraw-Hill, xiiixxix.
Trimberger, Ellen Kay 1978: Revolution from Above: military bureaucrats and development in Japan, Turkey, Egypt and
Peru. New Brusnwick, N.J.: Transaction Books.
Trotsky, Leon 1932: The History of the Russian Revolution. Vol. 1.
New York: Simon and Schuster.
Trouillot, MichelRolph 1988: Peasants and Capital: Domenica in the World Economy. Baltimore:
Johns Hopkins University Press.
Tuma, Nancy B. and Michael T. Hannan 1984: Social
Dynamics: Models and Methods. New York:
Academic Press.
Van Duijn, J. J. 1983: The Long Wave in Economic Life. London:
Allen and Unwin.
Vanneman, Reeve 1979:
"Strategies for the Southasian semi-periphery: cashews and crankshafts." Presented at the annual meeting of the Association for Asian Studies, Los Angeles.
Vyrynen, Raimo, 1988: "Economic fluctuations, military
expenditures and warfare in international relations." A paper presented at the 12th Annual
Conference on the Political Economy of the WorldSystem, Emory University,
Atlanta, March 24.Verdery, Katherine 1983: Transylvanian Villagers: three
centuries of
economic, political and ethnic change. Berkeley: University of
California Press.
1971: Sovereignty at Bay.
Vilar, Pierre 1976: A History of Gold and Money. Trans. Judith White. Atlantic Highlands: Humanities Press.
von Braunmhl, C. 1978:
"On the analysis of the bourgeois nation state within the world
market context." In J. Holloway and
S. Piccioto (eds), State and Capital: A Marxist Debate,
Austin:
von Ranke, Leopold 1887 [1824]: Histories of the Latin and
Teuton Nations: 1495-1514. Trans. Philip Ashworth.
von Werlhof, Claudia 1984:
"The proletariat is dead; long live the house
wife?" In Joan Smith
et al., Households and the World-Economy,
Walker, Richard Louis 1953: The Multi-state System of Ancient
Wallerstein, Immanuel 1974: The Modern World-system I: capitalist agriculture and the
origins of the European world-economy in the sixteenth century.
1979a: The Capitalist World-Economy.
1979b: "Kondratieff up or Kondratieff down?" Review, 2(4), 663-74.
1979c: "The
questions for research."
Review, 2(3), 389-400.
1980a: The Modern World-System II: mercantilism and the consolidation of the
European world-economy, 1600-1750.
1980b: "'The
withering away of the states.'" International
Journal of the Sociology of Law, 8, 369-78.
1983a: Historical Capitalism.
1983b: "A comment on O'Brien." Economic History Review, 36(4), 58083.
1984a: "The three instances of hegemony in the history of
the capitalist world-economy." In
Gerhard Lenski (ed.), Current Issues and Research in Macrosociology,
International Studies in Sociology and Social
Anthropology,
vol. 37,
1984b: The Politics of the World-Economy: the states, the movements and the
civilizations.
1986: "The incorporation of the Indian subcontinent into
the capitalist world-economy." Economic
and Political Weekly, XXI(4), 2839.
Walton, John 1981: "The internationalization of capital and
class structures in the advanced
countries: The United States
case." In Alejandro Portes and John
Walton, Labor, Class and the International System,
1985: "The third 'new' international division of
labor." In J. Walton (ed.), Capital and Labor in the
Urbanized World,
Waltz, Kenneth N. 1979: Theory of International Politics.
Ward, Kathryn B. 1984: Women in the World-System: its impact on status and fertility.
Ward, Michael Don and Harold Guetzkow 1979: "Toward integrated global models: from economic engineering to social science
modeling." Journal of Policy
Model, 1(3), 445-64.Warman, Arturo 1980: We Come to Object: the
peasants of morelos and the national state.
Warren, Bill 1980: Imperialism: Pioneer of Capitalism.
Weber, Max 1978 [1922]: Economy and Society. 2 vols. Guenther Roth and Claus Wittich (eds).
1981 [1927]: General Economic History.
Weber, Robert Philip 1981: "Society and economy in the
Western world system." Social
Forces 59, 4:113048.
1987: "Cycles of the third kind." European Journal of Political Research
15:14553.
White, Douglas R. and Michael L. Burton 1987: "Worldsystems
and ethnological theory." Research
proposal funded by the National Science Foundation.
Whorf, Benjamin Lee 1956: Language, Thought and Reality.
Wilkenson, David 1987: "Central civilization." Comparative Civilizations Review, 17,
3159.
Williams, Kristen M. 1985: "Is 'unequal exchange' a
mechanism for
perpetuating inequality in the modern world system?" Studies in Comparative International Development,
20(3), 47-73.
Wilson, C. H. 1957: Profit and Power: A Study of
Winckler, E. A. 1979: "
Wolf, Eric 1982:
Wolfe, Alan 1980: "Capitalism shows its face: giving up on democracy." In Holly Sklar (ed.), Trilateralism: the
trilateral commission and elite planning for world management,
Wolpe, Harold (ed.) 1980: The Articulation of Modes of
Production: essays from economy and society.
Wood, Robert E. 1980: "Foreign aid and the capitalist state
in underdeveloped countries." Politics
and Society, 10(1), 1-34.
1986: From Marshall Plan to Debt Crisis: foreign aid
and development choices in the world economy.
Woodruff, William 1967: Impact of Western Man: a study
of
World Bank (IBRD) 1983: World Tables, 3rd edition. Vol. 1, Economic Data, vol. 2, Social Data.
Wright, Erik O. 1976: "Class boundaries in advanced
capitalist societies." New Left
Review, 98, 341.
1978: Class, Crisis and the State.
1985: Classes.
Wright, Erik O. and Bill Martin 1987: "The transformation
of the American class structure, 1960-1980." American Journal of Sociology, 93(1),
1-29.
Wuthnow, Robert 1980: "The world-economy and the
institutionalization of science in seventeenth-century
System,
Zeitlin, Maurice 1984: The Civil Wars in
Zelditch, Morris, Jr. 1969: "Can you study an army in the
laboratory?"
In
A. Etzioni (ed.), 2nd edition. Complex Organizations: a sociological reader, 52839.
1971: "Intelligible Comparisons." In Ivan Vallier (ed.), Comparative Methods
in Sociology,
Zinn, Howard 1980: A People's History of the
Zolberg, A. R. 1981: "Origins of the modern world system: a
missing link." World Politics,
23(2), 253-81.
Table 6.1 General government consumption as a percentage of GDP
at current market prices: means
_______________________________________________________________________________
Groups of countries 1960 1965 1970 1981
_______________________________________________________________________________
Low income developing economies
8.3a 10.3a 11.1a 10.7
Middle income developing economies 11.2 9.7 12.3 13.5
Industrial market economies 15.1 15.4 16.4 17.3
_______________________________________________________________________________
aExcludes People's Republic of
Source: World Bank, World Tables, 1983 volume
1, p. 502
Table 9.1
Relative distribution of Gross National Product per capita among the major powers.
_______________________________________________________________________________
_______________________________________________________________________________
1870 25.2 19.1 16.8 20.2
9.1
9.6
1880 23.5 19.3 16.8 24.2
7.1
9.1
1890 26.7 19.1 16.5 23.2
6.7
7.8
1900 25.3 19.3 17.1 23.6
6.9
7.6
1913 22.5 18.5 18.0 26.4
7.4
7.1
1925 21.2 18.1 14.7 27.9 10.3 7.8
1938 21.1 14.6 18.7 23.9 11.1 10.5
_______________________________________________________________________________
The proportions of the sum of GNPs for all the countries have
been weighted by each country's
population.
Source: Adapted from
Kugler and Organski (1986:10).
Populations are from
Banks (1971).
Table 9.2
Percentage of national product in agriculture
_____________________________________________________________________________________________________________
c.1790 40 49
c.1810 36 51
c.1820 26 48
c.1830 24 51
c.1850 21 45 47
c.1860 18 45 45 57 38 34
c.1870 15 43 39 57 39 33 48 21
c.1880 11 41 36 57 37 -- 44 16
c.1890 9 37 32 51 32 27 35 17
c.1900 7 7 30 51 28 22 29 17
c.1910 -- 35 25 42 25 23 30 17
c.1920 6 -- -- 48 22 -- 23 15
c.1930 4 -- 18 31 13 17 19 9 39
c.1940 4 22 15 30 12 12 17 9 29 11
c.1950 6 15 10a 32 11 14 20 6 25 13 9
c.1960 4 9
6a 15 8 10 14
5 20 11 7
c.1969 3 6
4a 11 4 6 9
4 19
7
5
_______________________________________________________________________________________________________________
aWest
Sources: Mitchell, 1975:799, table
Table 10.1 Comparison of Population of
______________________________________________________________________________
Colonies
compared to
Year European
European Colonies Europe
(in percent)
______________________________________________________________________________
1700 140 16 11
1750 160 22 14
1800 207 120 58
1830 242 240 100
1860 294 270;680b 92; 252
1900 414 490;
960b 118;
232
1913 481 530;
1030b
110; 214
Population in millions.
aIncludes
bThe first figure excludes
Source: Bairoch (1986:197)
Table 12.1 Transnational corporate penetration indexa: Group
Averages
____________________________________________________________________
Year LICsb MICs NICs OPEC
____________________________________________________________________
1967 5.2 10.1 7.1
6.2
1971 5.7 11.1 8.2
9.1
1975 6.3 13.4 10.7
8.0
1979 13.0 28.3 21.4
19.4
N of
Countries 33 31 10 7
aThis index is a measure of the extent to which a country is
dependent on transnational firms. It uses data on the book value of foreign
direct investment weighted by an
estimate of the stock of domestically owned
capital. See Bornschier and
Chase-Dunn (1985:chapter 4).
bLICs = Low Income Countries, MICs = Middle Income Countries,
NICs = Newly Industrializing Countries,
OPEC = Members of the Organization of Petroleum
Exporting Countries
Source: Bornschier and Heintz (1979) and updates.
Table 12.2: Total debt
outstanding as a proportion of GNP, market prices: group averages
_____________________________________________________________________
Year LICs
MICs NICs OPEC
______________________________________________________________________
1970 .18 .16 .10 .09
1973
.19 .18 .09 .10
1975
.26 .22 .15 .12
1976
.28 .24 .15
.14
1977
.31 .27 .17 .17
1978
.34 .31 .18 .27
1979
.36 .32 .18 .26
1980 .37 .32 .18 .23
Number of countries
29-34 29-32 8-10 5-7
aMeans are calculated for only those countries which had data
available in 1975. Sample size varied from year to year within
the ranges indicated. Data for
Sources: GNP from World Bank (various years) World
Debt Tables. Debt, 1970-74, OECD
(various years), Geographical Distribution of Financial Flows to Developing
Countries: 1975-80 OECD (1982).
Table 12.3 Concentration of resources among countries in the
world-system: 19601980a
___________________________________________________________________________________________________________________
Resources or Social Structural Features Percentages
___________________________________________________________________________________________________________________
1960 1970 1980
Economic Production The proportion of world GNP going to:
the countries highest on GNP per capita with 20% of world
population 79.2 79.8 81.1
the middle countries on GNP per capita with 60% of world
population 19.4 18.9 18.1
the countries lowest on GNP per capita with 20% of world
population 1.4 1.3 .8
the United States 32.1 28.5
26.9
N = 112
The proportion of
world energy consumed by:
the countries highest on per capita energy consumption with 20%
of world population 83.3 80.3
69.9
the middle countries on per capita energy consumption with 20%
of world population 16.1 18.8
29.1
the countries lowest on per capita energy consumption with 20%
of world population .6 .9 1.0
the United States 36.0 32.4 27.8
N = 126
Economic Structure The proportion of the world's agricultural work
force living in:
the countries lowest on % of work force in agriculture with 20%
of world population 3.8 3.3 2.6
the middle countries on % of work force in agriculture with 60%
of world poulation 63.0 58.0 57.0
the countries highest on % of work force in agriculture with 20%
of world poulation 33.2 38.7 40.4
the United States
1.0
.6 .4
N = 129
The proportion of the
world's industrial work force living in:
the countries highest on % of work force in industry with 20% of
world population 42.7 41.1 41.0
the middle countries on % of work force in industry with 60% of
world population 51.1 53.3 51.8
the countries lowest on % of work force in industry with 20% of
world population 6.2 5.6 7.2
the United States 12.9 11.6
9.5
N = 129
Urbanization The proportion of the world city-dwellers
living in:
the countries highest on % urbanization with 20% of world
population 38.4 37.3 34.1
the middle countries on % urbanization with 60% of world
population 56.2 56.6 58.9
the countries lowest on % urbanization with 20% of world
population 5.4 6.1
7.0
the United States
14.3 12.9 11.0
N = 133
___________________________________________________________________________________________________________________
aPercentages are based in sets of countries for which data are
available for all time points. Thus comparisons over time are not confounded by
missing data.
Source: All data are taken from
World Tables, 1983, volume II, except
for GNP which was supplied by the World Bank's Economic Analysis and
Projections Department.
Table 13.1 Colonial governors sent, 1415-1969
_______________________________________________________________________________
Century Expansion Traded or Taken Divided
or Consolidated Totals
_______________________________________________________________________________
15th 6 0 0 6
16th 49 1 0 50
17th 76 11 5 91
18th 47 25 17 89
19th 100 11 2 113
20th 53 13 0 66
_______________________________________________________________________________
Totals 331 61 24 415
_______________________________________________________________________________
Source: Henige (1970).
Table 13.2 Recent trends in export partner concentration
(percentages),
1970-1980
_______________________________________________________________________________
Number of
Type of Country Countries 1970 1980
_______________________________________________________________________________
Industrial 18 24.2 24.1
Centrally planned 6 33.6 29.0
OPEC 6 33.4 37.3
NICs 8 28.4 25.8
Middle Income 25 31.0 28.7
Low Income 20 37.6 30.9
_______________________________________________________________________________
All Countries 83 31.1 28.6
_______________________________________________________________________________
Source: Mller, (1988).
EndNotes
Introduction
1.See the frontispiece
statement on editorial policy in the
2.For a preliminary
overview of the development of stateless, ancient and modern world-systems see
Chase-Dunn (1986).
Part IThe Whole System
1.The
deconstructionists correctly argue that all structural theories and claims to
universalistic science are themselves acts of power. For them it follows that all such theories
should be attacked and destroyed, and
historicist description is the only possible nonelitist form of
explanation. While it is true that
structuralist theories have been used badly
to justify exploitation and
oppression these theories are not themselves a major
cause of exploitation and oppression; nor would evil cease if all ideology were
historicist. A truer and more universal
understanding of the causes of exploitation and oppression can be of use to
those who would like to eliminate them.
2.Wolf's analysis
falters when he arrives at modern capitalism, which he analyzes in orthodox
Marxist terms. He fails to take acccount
ofthe systemic links between capitalist accumulation, the interstate
system, and the core/periphery
hierarchy.
3.Taxed small
commodity producers (yeomen, artisans, etc.) and wage laborers are also to be
found within many societies dominated by the
tributary mode of production.
Chapter 1 The Deep
Structure: Real Capitalism
1.It is questionable
as to whether or not classical European feudalism ought to be considered a mode
of production in its own right. It may
be more fruitfully understood as an instance of the decentralized phase of a
cycle of centralization (empire) and decentralization (feudalism) which had
been spiraling along for millennia during the development and expansion of
tributary modes of production (see Ekholm and Friedman, 1982; Mann,
1986:chapter 5).
2.The problem of dialectically
qualitative transformation of one such system into another is difficult to
model, but here we may get some help from the mathematics of catastrophe theory
(see Renfrew, 1984). The few formal or
mathematical specifications of Marx's model of capitalist accumulation (Nowak,
1971; Morishima, 1973) have not approached the problem of system
transformation.
3.Max Weber's
theoretical approach is less systemic than that of Marx but he does present a
notion of the theoretical essence of capitalism which solves, in a different
way than I am proposing, the problem of the relationship between capitalism and
the state. Weber's notion of capitalism
focuses on "formal rationality" rather than commodity
production. Formal
rationality is the concern for the efficiency of means rather than a direct
focus on ends. Quantitative
calculationand rational decisionmaking are the important thing, while
noncapitalist rationality ("substantive rationality") focuses on the
delineation of ideals. For Weber formal
rationality, especially in the form of rational capital accounting, is the
essential core of capitalism. Markets
and wage labor are only important contextual conditions for the development of
profitoriented industry which employs rational capital accounting. The state is brought in both as an important
contextual variable, but also more directly as a bearer of formal rationality
itself. Rationallegal authority and
bureaucratic administration are additional forms taken by formal
rationality, although
the connection between rationallegal authority and the notion of formal
rationality as concern for efficient means is somewhat strained. Extremely rationalized legal systems are not
necessarily more efficient than common law or judgebased systems. Weber argues that rationallegal law and
courts facilitate rational economic activity by making legal and state action
more predictable than other forms of authority.
It is arguable that it is the actual content of contract, property and
market law which facilitates capitalism rather than the systematization of law.
Weber's focus on
formal rationality biases the evaluation of capitalism by asserting that its
essence is efficiency. Profitmaking is
alleged to be a neutral goal, not a substantive value, which is only an
indicator of efficient economic action.
This has served as an important source of ideological mystification
within sociological theory (especially Parsons) because of the uncritical
linkage between individualized rationality and efficiency. It isoften assumed that individual action is
the only kind which can be rational, while collective orientations are
identified with "traditional" societies or with modern irrationality
as evidenced by affective involvement with collective solidarities (nationalism,
social movements, socalled "collective behavior" mobs, riots, revolutions, etc.). Thus capitalism and individualism are
rational
while socialism and collective orientations focus on vague
ideals such as justice and equality which are necessarily affective and
irrational. This formulation receives
support from economists who prove with elegant mathematical theories that
maximizing the collective welfare is impossible.
Certainly
capitalism has been the motive force for the development of planning and
calculation techniques, just as almost all other technologies, but this does
not prevent socialists from adopting these, especially linear programming, for
the purposes of a more just mode of production.
There is no technical reason why socialist planning cannot maximize both
individual and collective outputs efficiently in terms of resource scarcities
and tradeoffs. The experiments with such
techniques in socialist states has, to date, all but ignored democratic inputs
into such planning although recent reform movements have raised this
issue. Whether democracy will simply
mean more markets and profit incentives (a move toward capitalism) or new forms
of democratic control over collective planning remains to be seen.
4.This definition of
the core/periphery hierarchy is similar to Kautsky's discussion of commodity
circuits between urban and ruralareas within a regional division of labor (see
Lipton, 1977:115-21;
also Banaji,
1980). Alternative conceptualizations of
core and periphery are considered in chapter 10.
5.Jeffrey Paige
(1975:13) correctly understands the peripheral smallholder as "a worker in
an agricultural system controlled by urban financial interests." Paige approvingly quotes Marx's description
of the nineteenthcentury French peasantry as applicable to peripheral
smallholders producing agricultural exports: "The small-holding of the
peasant is now only the pretext that allows the capitalist to draw profits,
interest and rent from the soil, while leaving it to the tiller of the soil
himself to see how he can extract his wages."
6.The problem of
world-system boundaries and modes of production is further considered in my
comparison of stateless, ancient and modern world-systems (Chase-Dunn,
1986). And there is a discussion of competing
definitions of world-system boundaries in chapter 15.
Chapter 2 Constants,
Cycles, and Trends
1.Certain
methodological problems with the analysis of cycles are discussed in chapter
15.
2.The hegemonic
sequence has been shown to correspond to changes in the population size
hierarchy of the world city system (Chase-Dunn, 1985a).
3.These
"precapitalist" forms of production, which are articulated with
capitalist commodity production, have often been analyzed in terms of the way
in which they subsidize the lifetime cost of reproducing the work force. It is certainly true that such forms have
been allowedto exist, or have been actively created by the policies of colonial
powers seeking to sustain a low wage labor force. On the other hand, the struggles of
peripheral workers to maintain some independence of capitalist commodity
production has also been an important cause of the continued existence of these
forms. Excellent studies of the
processes of reproduction and breakdown of these forms of production in
peripheral areas are those by Kahn (1980), Murray (1980), Warman (1980),
Meillassoux (1981), and Rodney (1981).
Chapter 3 Stages of
Capitalism or WorldSystem Cycles?
1.In the preface to
the first German edition of volume 1 of Capital Marx writes "The
country that is more developed industrially only shows, to the less developed,
the image of its own future" (Marx, 1967a:8-9).
2.This was first
argued by Rosa Luxemburg (1968) and has been
elaborated by Mandel
(1975) and Amin (1975). Mandel insists
that primitive accumulation is a separate process from capitalist production
itself, while Amin and Frank (1979a) see "primary accumulation" as a
characteristic of peripheral capitalism.
4 The WorldSystem Since
1945: What Has Changed?
1.David Gordon (1980)
has developed an admirable conceptual structure for a theory of stages,
breaking hypothetical qualitatively different stages into general phases based
on a Marxian theory of capitalist accumulation.
Unfortunately he does not designate the particular stages to which his
scheme applies.
2.Recall the
discussion of the terms "transnational" and
"international"
on p. 81 in chapter 2. 3.Chirot has
revised this assessment in the new edition of his text on the world-system
(Chirot, 1986). He now admits that some
areas of the "
4.Although the larger
semiperipheral countries have also promoted industrial exports, and these have
grown rapidly, they constitute only a small proportion of industrial output in
these countries. Gereffi and Evans
(1981:51) show that exports as a percentage of total sales by transnational
manufacturing firms has indeed risen in both
5.Saskia Sassen-Koob
(1985) has also explored the effects of inter
national immigration
on the structure of
Chapter 5 World Culture,
Normative Integration and Community
1.Normative
integration does not preclude exploitation, oppression, or institutionalized
inequality. Much recent anthropological
work on stateless societies stresses contradictory interests based on gender
and age-based stratification.
Nevertheless most still argue that the magnitude of inequality based on
normative integration not backed up by coercive institutions is limited.
2.Polanyi's
use of the term "redistribution" was somewhat unfortunate, implying
as it does a kind of functional pooling and redistribution of resources. Polanyi apparently adopted the term from the
analysis of semistratified chiefdoms.
Most recent analysts, especially those interested in distinguishing modes
of production, emphasize the coercive nature of states, the use of threats,
conquest, and politically organized exploitation in those societies which are
characterized by the "tributary modes of production."
3.My friend and
colleague Katherine Verdery, an anthropologist, strongly argues against the
presuppositions of a Parsonsian notion of normative systems based on consensus
about values and internalized
norms, even as a bte
noire in the discussion of world culture.
Anthropologists have been moving toward a notion of culture as
negotiations around the interpretation of symbols, and this modifies the
picture of those societies which I have characterized as integrated primarily
by normative consensus. I would agree
that consensus is never complete and unproblematic, and neither do individuals
smoothly carry out the role prescriptions to which they have been
socialized. The process of
consensus-formation is always problematic, but the primacy of normative negotiations
as a producer of social order varies across societies and groups. The degree of consensus and the centrality of
normative integration are variables, and it makes sense to inquire in any
particular case what role they play in determining the dynamics of a system.
4.Subunit solidarities
and the collective values which they exhibit sometimes contradict the
individualistic logic of capitalism andprovide a basis for anti-systemic
movements. But, as noted in previous
chapters, these challenges have so far only driven the expansion of the spatial
scale of capitalist firms and markets.
5.The relationship
between nationalism and the state in peripheral areas only becomes
complimentary following decolonization, and even then there may be nationalist
challenges to particular regimes which are perceived as dominated by a core
power.
6.Efforts to apply
Gramsci's notion of ideological hegemony to the world-system (e.g. Cox, 1983;
Gill and Law, 1986) have had to ignore the fact of the relatively low level of
cultural consensus at the
global level, which
reduces the importance of ideological domination. Albert Bergesen's effort to apply Durkheimian
ideas of moral boundary maintenance to the world-system (Bergesen, ed.,
1980:chapter 1) foundered on this same problem.
Like Parsons and Durkheim, the Gramscian notion of ideological hegemony
assumes that cultural integration is central to the maintenance of social
order.
Chapter 6 States and
Capitalism
1.While some extreme
materialists may view ideology and politics as completely epiphenomenal, I agree
with those who stress their
importance. Politics, like the medical arts, is more an
applied than a pure science. It utilizes
general theory as a guide to action in conjunctural situations in which the particularities
may matter as much as the general conditions.
2.The lists of
countries in each category are given on pp. xixxxi of volume 2, World Bank, World
Tables (1983). A more complete
breakdown of central government finances, beginning for most countries in
1975,is given in the second series of tables in volume 1 of the World Tables. These show current revenues from different
sources as well as expenditures on various categories such as public
enterprise, defense, education, health, social security and welfare, and
housing and community services. A
comparative study of the differing capacities of states in these various types
of activities should be able to put these data to good use.
3.Evans and Stephens
(1988) do not use the vocabulary of core and periphery; rather they prefer to
discuss early and late
industrializers. Nevertheless their synthesis of
democratic Marxists to
explain the emergence of welfare state structures, is an important contribution
to understanding core/
periphery differences
in regime form.
4.The two apparent
exceptions are state socialism and the Islamic revival. The theory of the dictatorship of the
proletariat is, in principle, a further extension of democratization of
political power. It was originally
understood by Marx as a transitional period on the way toward the
"withering away of the state."
Actual authoritarian socialist states, on the other hand, are justified
as necessary means for defending against internal and/or external threats. The idea of socialism constitutes a further
extension of democracy into the realm of the economy. Islamic fundamentalism is a major exception
to legitimation from below, but it may be understood as a reactive movement
against commodification (as is socialism) which utilizes a traditional
hierarchical ideology. Many types of
religious fundamentalism emerge periodically within the development of
capitalism as displaced or threatened groups try to protect themselves from
unwanted changes. Islamic fundamentalism
is this sort of movement, but it receives an extra dose of energy because it
corresponds to peripheral opposition to core domination and also is a reactive
civilizational response to European cultural domination.
5.The Chilean regime
of Pinochet is portrayed by Portes and Kincaid (1985) as the exception to the
rule, being based more on a personal and patriarchal type of authority than on
the more bureaucratic form of authoritarianism found in the other Southern Cone
countries.
7 Geopolitics and
Capitalism: One Logic or Two?
1.Marxist theory has
usually disdained the seemingly artificial distinctions between economics and
politics which have become enshrined in the academic disciplines. Marx asserted a holism of the underlying
processes operating within a mode of production. But the use of a deterministic logic to
justify alleged necessity by Stalinist ideologues added fuel to the revolt
against "economism" which had begun with Luxemburg, Gramsci and
Lukacs. Poulantzas (1973) took this
theoretical tendency to its most extreme point within Marxism. Claudia von Braunmhl (1978) and the other
German "state derivation" theorists have reintegrated the new
theories of the state by utilizing Marx's basic theoretical concepts to account
for the operation of capitalist states.
Von Braunmhl has done the most interesting work at the conceptual level
to integrate the role of the larger interstate system and the operation
of the world market within a framework of basic Marxist concepts. 2.Stephen Krasner notes the similarities
between the structural Marxist "relative autonomy" theory and his own
state-centric "realist" approach to international political economy
(Krasner, 1978:32).
3.The reader is also
reminded of another matter of usage. The
glossary distinguishes between "world-economy," a type of
world-system in which the spatial economic network is overlain by a
multicentric and competitive interstate system, and "world economy"
(unhyphenated) a reference to the
structure of global economic production and exchange. A world-economy is a whole socio-economic
system, while a world economy is only one aspect of such a system, the
territorial network of material production and exchange. It should also be noted that the term
"world economy" includes both international and intranational
economic production and exchange.
4.Thanks to Guenther
Roth for pointing out the connection between Weber's idea and the work of von
Ranke. The quotation from General
Economic History was brought to my attention by Volker Bornschier (1987).
5.See chapter 9 for a
comparison of this view of the interaction of class struggle and economic
growth with Mancur Olson's theory as specified in his The Rise and Decline
of Nations (1982).
6.On the prior origins
of the European interstate system Zolberg (1981) is ambiguous. At one point he argues that the formation of
the European world-economy contributed to the crystallization of the interstate
system, while at other points he refers to the prior existence of states and
the logic of military expansion. Fernand
Braudel
(1984:324) contends that the European balance of power mechanism operated from
the thirteenth century on.
7.Skocpol is inconsistent on the issue of the
autonomy of the interstate system. In
her text quoted above she seems to assert that the interstate system and the
logic of geopolitical domination are substantially autonomous from the capital
accumulation process, but in a footnote (1979: 299) she approvingly quotes Otto
Hintze "the affairs of the state and of capitalism are inextricably
interrelated, ...these are only two sides, or aspects, of one and the same
historical development."
8.The exchange of
"unequals" involves exchanging goods which have little value to their
producers but great value to their consumers.
Ernest Mandel (quoted in Frank, 1979a) refers to the exchange of
"unequals" which occurs before social systems interact substantially
enough such that their divisions of labor come to be sufficiently integrated to
produce what Marx (1967a) referred to as "abstract labor" the equivalence of qualitatively different
types of labor in terms of a single system of value. Of course the formation of abstract labor and
the equivalence of labor values is a slow process which is never complete, even
within an integrated commodity economy.
But in a single interactive commodity economy there is a tendency toward
the exchange of equal values which subjects the division of labor itself to
periodic reorganizations which reflect changes in the average socially necessary
labor time needed to produce a product.
The exchange of unequals tends to become the exchange of labor value
equivalents as merchant capital "acts as a solvent on
precapitalistrelations of production."
Wallerstein's (1979a) distinction between
preciosities vs.
fundamental goods is similar in some ways to the idea of the exchange of
unequals. Preciosities are defined as
luxury goods which have high value in one area (due to scarcity) but much lower
value in their area of origin.
Wallerstein uses this idea to argue that exchange of fundamental goods
(bulk foods and raw materials) bound worldsystem territorial trade networks,
while the exchange of preciosities (such as the longdistance luxury trade
between Europe and
9.The businessmen of
10.Modelski's explanation
of the rise of hegemonic core powers in terms of a "desire to create a
global order...an expression of the will topower, the urge to control and to
dominate, to imprint a pattern on events" (1978: 224) also seems an
unlikely candidate for explaining much of the variance. He does not clearly explain why this desire
emerges in 1500 or how it is differentially distributed across time and space
to explain the rise and fall of hegemonic core powers. Modelski's (1982) more recent work
acknowledges, in principle, the interdependence of political and economic
processes, but his account continues to emphasize a "great power"
theory of history which
ignores the division
of labor between the core and the periphery, as well as the specifically
capitalist nature of central institutions in the modern world-system (see
chapter 8).
8 Warfare and
WorldSystems
1.Most precapitalist
world-empires allowed less "private" autonomy of the ownership of
land from the state than was the case in the Roman world. Thus the law of private property and contract
among
individuals was highly
developed in
2.To say this is not
to argue that market efficiency is just.
Collective rationality and justice requires a social structure which
subjects major investment decisions to a calculus of social needs in which
market efficiency is only one desideratum.
3.Some precapitalist
interstate systems were composed exclusively of citystates while others were
composed of mixes of citystates and empire states. Thus the Greek citystates were part of a
larger interstate system in whic the
4.Indeed Ekholm and
Friedman (1982) have argued that the modern world-
system is not
fundamentally different from earlier core/periphery systems in which political
power was the main determinant of system expansion and contraction. In a very general sense they are right. The 500year expansion of the modern
world-economy may be only its formative period moving toward world state
formation and a more centralized political integration. But the particular dynamics of this
historical system based on capitalist institutions (as outlined above), and our
concern for humanizing the social processes which appear to be beyond our
collective control, make close attention to the particularities of our system
desirable.
Chapter 9 Rise and Decline
of Hegemonic Core States
1.Anderson (1974a)
depicts the relocation of the capital of the empire in Constantinople and the
split between the Eastern and Western Empires as part of the devolution of the
Roman Empire rather than as a process of hegemonic fall and rise which
revitalized the mode of production.
Although
case of a new
upwardly mobile hegemonic power than a contracting survival of the earlier
2.In an insightful and witty discussion of the
origins and methodology of Wallerstein's
world-system perspective, Walter Goldfrank points to what he calls the
"Goldilocks Principle" which is frequently found in Wallerstein's
explanations "not too cold, not too
hot, but just right." Of relevance
here is the point summarized by Goldfrank (1979a:46: "...to the general
correlation of state strength with core production, he (Wallerstein) adds the
nuances that hegemonic core states need less active state machineries than
their rivals, and that advancing
semiperipheral states take a more active economic role than any."
3.Andre Gunder Frank (1979a) points out that a
relatively equal distribution of income, with a large home market for
consumption goods, is a necessary, but certainly not a sufficient, condition
for upward mobility in the world-economy.
4.Frederic Lane (1979:15-19) shows that the
Portuguese entry into the spice trade did not ever employ the lowering of
prices to drive competitors from the market.
The blockade of the movement of spices
through the Red Sea and the
5.While
6.This is not a
mystery because, as inheritors of the Western science of development, we know
most about the English "birthplace of modernization." Thus even students of the worldsystem tend to
generalize from it even though the overall perpective commands us to treat it
no differently than the others.
7.A more detailed
analysis of the upward mobility of the
8.A similar set of
figures with a slightly different comparison base is shown in table 12.3 in chapter 12. This shows the
9.Choucri and North's
(1975) study of the processes which led up to World War I contains an interesting second chapter
which compares the "national capabilities" of the "major
powers" during the period from 1870 to 1915. Graphs depict the populations, iron and steel
production, national income (total and per capita), foreign trade,marine
tonnage, and colonial area for
10.The figures in
table 9.1 should be interpreted as rough indicators only. The main point is that total GNP includes a
factor due to population size, while GNP
per capita is much closer to what we mean by a developed national economy. The use of GNP figures for the nineteenth century is especially hazardous
because of differences in national accounting procedures and the problem of
currency exchange rates. Even postWorld War II national accounting
figures have required major adjustments to take into account differences in
purchasing power of currencies across national boundaries (see
Kravis, Heston, and Summers,
1982). Nevertheless the figures in table
9.1 undoubtedly give a closer estimate of the relative core status of the great powers than the original table
presented by Kugler and Organski, which was based on total GNP unweighted by
population.
11.A similar
explanation is suggested for the high GDP per capita figure for
12.A measure which is
highly correlated with the proportion of national product in agriculture is the
proportion of the labor force employed in agriculture. Maddison (1982:table 2.2) shows that the
Dutch economy was much more industrialized in 1700 than was the economy ofthe
United Kingdom on the basis of the percentage of the work force in agriculture
services and industry. But by 1890 the
____________________________________________________
_____________________________________________________
1700 % %
Agriculture 40 60
Industry 33 15
Services 27 25
______________________
1890
Agriculture 33 16
Industry 31 44
Services 36 40
_____________________________________________________
Source: Excerpted from Maddison (1982:35, table 2.2).
Chapter 10 Core and Periphery
1.Of course this
discussion of nested hierarchies has only considered those structures which are
easily conceived in spatial terms.
The actual nested hierarchy also
contains features which are spatially organized in even more complicated ways,
such as classes and political organizations.
2. Another form of intermediate activity should
also be noted because it has
consequences which are somewhat similar to those of the types of
semiperipherality defined above. Certain political and/or economic
entrepreneurs mediate core/periphery relations.
The New England merchant capitalists involved in the "triangle
trades" in the eighteenth century, and the
3.This notion may also
prove important for understanding the rise of
semiperipheral marcher states in the process of ancient empire-
formation (see
Mann, 1986:chapter 5; ChaseDunn, 1988).
4.See Fernand
Braudel's (1984:299) discussion of national wealth, national income and methods
of estimating per capita national income for the seventeenth and eighteenth
centuries. There are, of course, large
problems of comparability of GNP figures across countries. See note 10 in chapter 9.
5.All researchers
agree that the core/periphery hierarchy is relational in the sense that it
involves relations of power and dependence among entities. The advocates of network analysis argue that
their measures are truly relational and structural, while other less complex
measures are merely attributional, being based on attributes of units rather
than relations among units. My own
position is that there is no hard and fast division between attributes of an
entity and relations between entities.
Sometimes, as with symbolic inter
actionism, the identity
of a unit is a major constituent element of that unit's interaction with other
units. A variable characteristicof a
unit may in fact be an indicator of that unit's relationship with other units,
as with investment dependence of nationstates which is measured by the amount
of foreignowned equity capital within a country. But even attributes which seem to be wholly
"internal" such as GNP per capita may be useful indicators of
relations among units or position in a larger system when they are used in comparative
analysis.
6.Mintz (1974)
produced an early anthropological analysis of local/
worldsystem
links. Other anthropologists who have
done this kind of work are Ekholm and Friedman (1980), Kahn (1980), Warman
(1980), Verdery (1983), Fox (1985), an dTrouillot (1988). Historians and sociologists have also done
similar studies. I already mentioned
Hall (1986, 1988). Others are Murray (1980),
Rodney (1981), and So (1986).
7.Bairoch shows the
following figures for
(Table 10.1 About Here)
8.O'Brien later
discusses the Baltic trade when considering the flows of bullion into Europe,
but he does not distinguish between peripheralized
chapter 6) argues was
an external arena until it became incorporated into a semi-peripheral position
within the European world-economy in the eighteenth century. Wallerstein (1983b) has published a response
to O'Brien's article which, in addition to my criticisms, points out the
deficiences of economic statistics which fail to estimate the inputs of nonwage
labor. 9.Wallerstein argues that plunder
is undertaken in external arenas, while incorporation into the periphery involves
a shift to production of peripheral commodities using coerced labor. When we begin to compare the processes of
peripheralization that operate in different worldsystems we may want to
understand plunder as one important form of peripheral exploitation rather than
simply a prelude to unequal exchange.
Plunder tends to destroy the plundered society, while peripheral
exploitation underdevelops it.
Nevertheless there have been fairly longlasting core/periphery
hierarchies based exclusively on plunder (e.g. Kelly, 1985).
Chapter 11 Reproduction
of the Core/Periperhy Relationship
1.Charles Lipson's
fascinating study of the changing international regime which supports
foreign-held private property and investment in
peripheral countries reveals that the degree to which foreign investment
depends directly on military power has declined, and that the
growing ideology of
sovereign national development has decreased the extent to which core states
can guarantee the property rights of core investors in the periphery (Lipson,
1985).
2.Chase-Dunn (1981)
analyzes the case of the
3.The exception to
this has been the OPEC oil cartel, but the difficulties of keeping this cartel
together and the failure of attempts to create cartels for other raw materials
and agricultural products support the above generalization.
4.The number of
developing countries for which data were available for the described estimates was 33 for GNP per
capita and 26 forelectrical energy consumption per capita. With such small Ns estimates can be heavily
influenced by one or two cases.
Chapter 12 Recent Trends
1.William Dixon's
(1985) study of trends in commodity concentration (the tendency for a country's
exports to be concentrated in one or a few
main commodities) and partner concentration (the tendency for exports to
be sent to one other country) concludes that there has been little or no change in the distribution of these
forms of trade dependence between 1955 and 1975. But
2.Similar upward
trends in debt dependence for the period between 1970 and 1982 are shown for
several types of debt in peripheral countries of Africa, Asia, and the
3.Interpolation is
required between the country closest to 20 percent of the world's population
and the next country listed in order to estimate the proportion of world GNP
controlled by exactly 20 percent of the world's population in countries highest
on GNP per capita. This table follows
the method of an earlier table whichexamined changes in the distribution of
world resources from 1950 to 1970 (Meyer, et al., 1975:table 2). Though it is calculated in a similar way, the
sources used and the cases included are different, and thus the proportions
at any point in time vary to some
extent. These tables are intended to
show change over time. The exact size of
proportion estimates are influenced by the particular countries included. Because data availability differs across
sources the proportions will differ, but changes over time should still
indicate trends in world-system inequality.
In this regard table 12.3 and the earlier version presented by Meyer et
al. (1975) lead to similar conclusions.
4.A study by Breedlove
and Nolan (1988) study reveals a trend toward increasing inequality in the
global distribution of GNP per capita among 120 countries. Breedlove and Nolan compute Gini coefficients
for five year intervals between 1960 and
1980, and while the average GNP per capita increased for the core, the
semiperiphery, and the periphery over this time period, the amount of relative
inequality among all the countries indicated by the Gini coefficient increased
as well.
5.There are a few
surprises, such as the inclusion of
Chpater 13 WorldSystem
Fluctuations
1.Bergesen and
Schoenberg sensibly do not include reorganizations ofexisting colonies as
instances of expansion. Thus they
exclude 22 new Spanish governors sent to head "intendencias" during
the decentralization of the Spanish empire in the late seventeenth and
eighteenth centuries. I would also want
to exclude instances in which European colonial powers acquired colonies from
one another, either by force or through treaty.
As table 13.1 shows, there were 61 cases of this kind in Henige's data,
and they were unevenly distributed over time, peaking in the eighteenth
century. These are probably more an indicator
of contention among core powers than an indicator of territorial expansion of
the world-system.
(Table 13.1 about here)
My examination of the
Henige data also suggests interesting
differences between
core powers in the timing of their colonial expansion. Despite Bergesen and Schoenberg's insistence
that we look only at characteristics of the whole world-system, figure 13.3
shows colonies established by each core power by centuries. Countries with few colonies, such as
(Figure 13.3 about here)
Figure 13.3 shows what
all students of colonialism know the
first wave was the work of
2.Pat McGowan
(1985:488) notes that Europe also had an earlier wave of colonial expansion
"beginning with the Crusades of the twelfth century, that established the
Crusader states in the Holy Land and that culminated in the commercial empires
of Venice and Genoa in the Levant in the fourteenth and fifteenth centuries,
including the colonies of Crete, Cyprus, and many Black Sea ports."
3.McGowan (1985:table
4) does find a positive relationship between the cumulative net number of colonies and
Bergesen and Schoenberg's measure of war, but I do not give this finding much
importance because both the measure of colonialism, which includes both
cumulative colonies and colonial terminations, and the measure of war have
serious problems, as discussed above.
4.Actually the
Southern control of
the Federal state and the presidency was at its
zenith. The election of
5.It should be noted
that visual inspection indicates that this period-
ization of the expansion of capital exports is not closely
related tothe cycles of the number of new colonies established as revealed in figure
13.4 above. The 1860s/1870s boom of
capital exports appears during a trough of colonial expansion, while the pre-
and post-World War I periods of capital export growth occur on the downswing of
a colonial expansion wave which had peaked in about 1890. Thus, the two types of expansion are
apparently neither in phase nor inversely related.
6.The conception of
cycles of world-system development described in chapter 3 would, of course,
contradict this claim in many ways.
First, the industrial revolution which began in England in the
eighteenth century was the most extensive (up to that time) of a number of
previous "revolutions" in which rapid technological development, the
increase in capitalintensity in both manufacturing and agriculture, and the emergence
of new lead "industrial" sectors, had occurred. The earlier industrial revolutions were
undoubtedly of less magnitude as well as harder to measure, but economic
history strongly supports their existence.
This perspective does not, however, provide an explanation for Suter's
finding that inter
national financial
flows do not reveal cyclical characteristics in the first half of the
nineteenth century. Suter himself
observes that international financial centers in earlier centuries moved
geographically with
changes in the hegemonic leader from the
centrality of the
Medici in
Chapter 14 Theory
Construction
1.Among Marxist social
scientists in the United States there are not many who continue to take an explicitly
structuralist approach. A notable
exception is Erik Wright (1985) and the circle of "analytic
Marxists."
2.Karl Popper (1972)
uses the term historicism in a nearly opposite way, and his usage has become
popular. For Popper historicism meant
teleological causation and unilinear evolution, qualities which he identified
with Marxist theory. Structuralist
theory need not be functionalist or teleological, and neither must it embrace a
simple version of deterministic unilinear evolution.
3.Charles Tilly's
(1984: 81) valuable discussion of different
strategies of
comparison adds a second dimension. In
addition to the continuum I have discussed, which he terms individualizing
versus universalizing comparison of instances of a phenomenon, Tilly proposes
that scholarly approaches differ also in terms of their usage of differentiated
forms of a phenomenon. Some analysts
propose only a single form, whereas others conceptualize multiple forms. This second dimension enables Tilly to
cross-classify comparative strategies in a way which yields, in addition to
individualizing and universalizing approaches, encompassing and
variation-finding types. Encompassing
comparisons analyze different levels of analysis simultaneously e.g. the
world-system and national
states. Variation-finding studies examine how
variables are
systematically but
differently interconnected in different cases.
4.I have long expected
that the shift to the world-system level of analysis (as a frame of reference
and as a focal unit of analysis in its own right) would eventually lead to the
application of modernization theory and the Parsonian approach to the
world-system. This has occurred in the
work of George M. Thomas et al. (1987) and Roland Robertson (Robertson and
Lechner, 1985). The only book to
explicitly contrast different theoretical approaches to the worldsystem is
William R. Thompson's (1983) Contending Approaches to World System Analysis.
5.Charles Tilly's
(1984) quite sensible plea for the grounding of
comparative studies in a relatively concrete setting even though
the foci of analysis are "big structures, large processes and huge
comparisons," makes the case that the concept of "society" has
been particularly abused as an abstraction by social theorists. While I am sympathetic with his effort,
Tilly's main difficulty with the term "society" is that it is hard to bound a concrete society in time and
space. Thus "German society"
is a messy idea when we try to say when, where, and which people are included
within it. But I am not convinced that
this proves that the idea of society is a bad idea. My reservations stem from my own somewhat
limited exposure to German society, which despite its complex nature and messy
boundaries, is more than simply a mythical national identification. The concept "civilization" is
similarly problematic, but this is not reason
enough to dispense
with it, at least until a substitute specification
which captures the complexities of the idea has been developed.
6.Karl Marx describes
his own metatheoretical approach in the
Grundrisse:
Thus, if I were to
begin with the population, this would be
a chaotic conception (Vorstellung) of the whole, and I would then, by means of
further determination, move analytically
towards ever more simple concepts
(Begriff), from the
imagined concrete towards ever thinner abstractions until I had arrived at the
simplest determinations. From there the
journey would have to be retraced until I had finally arrived at the population
again, but this time not as a chaotic conception of a whole, but as a rich totality of many
determinations and relations. (Marx, 1973:100)
Chapter 15 Research
Methods
1.This same point is
made with regard to other levels of analysis in Morris Zelditch, Jr.'s essay,
"Can you study an army in the
laboratory?" (Zelditch,
1969).
2.As in any
experimental or non-experimental research design there may be things going on
that we don't know about, that are not included in our model, and these may
account for the relationships we find. A
critic who suggests this, however, is generally required to be more specific,
to outline the source of the hypothesized spuriousness. If the critics' claim is plausible it is
incumbent on the researcher to include the new variable in his model and to
test for its effects on the relationships.
3.The need for an
omnibus crossnational data bank has been the topic of discussion at recent
meetings of the International Studies Association and the American Political Science Association.
Professors Dina Zinnes
and Richard Merritt of the University ofIllinois at Champaign-Urbana have formed a multidisciplinary committee on
"Data Development for International Research" and this group has obtained
support from the National Science Foundation for the improvement and
integration of crossnational data, including that which is relevant for
international political economy.
4.A related point is
made by Doran and Parsons (1980:959) in their
discussion of the measurement of the timing of a "critical
period" in the cycle of relative power among core states.
Part V
1.One implication of
the focus on logical boundaries already discussed in my book on socialist
states (Chase-Dunn, 1982b), is that we cannot be so sanguine about the
definition of socialism as Marx and Engels were. Reacting to the detailed blueprints of the
utopian socialists, they were happy to leave the specification of the contents
of socialism up to the worker's movement.
In light of the history of efforts to establish socialism since then, it
is obvious that things are more problematic.
We want to be able to say more definitely which political initiatives
represent progress toward socialism, and which do not. It is not a simple matter of intention, or
reading the formal program or constitution.
Use of the word socialism does not guarantee its content for example, national socialism. I have only defined socialism very broadly as
democratic collective rationality.
This definition raises many theoretical and empirical questions about
the actual content of socialist programs, institutions, and political
initiatives. This difficult problem is
heavy with social and political philosophy, and beyond the intended bounds of
this book,which has focused on the nature of capitalism. The two problems are not unrelated in
reality, as the question of logical boundaries implies, but I feel that to go
further toward a more elaborate and concrete specification of the content of
the socialist mode of production would take us into a problem which should be
the topic of another work.
2.We need not agree
with Wittfogel's (1957) thesis of "oriental despotism" in order to
see elements of the tributary mode of production in the contemporary socialist
states. To the extent that a technocratic
and bureaucratic class uses state power to extract surplus product from direct
producers we have a version of the tributary mode. Samir Amin (1980a) terms these societies
"statist." None of this is to
deny, however, that they also contain significant elements of the logic of
socialism in the laws which forbid private ownership of large-scale means of
production, welfare, and employment guarantees, and in some cases important
forms of economic democracy.
3.An extremely crude
effort to do this using seat-of-the-pants coding of world-system position was
presented in Chase-Dunn (1983).
4.This occurred first
in the medieval cities of feudal