Walking Towards Justice, Volume 9: Democratization in Rural Life

CONTENTS LIST OF CONTRIBUTORS ix EDITORIAL ADVISORY BOARD xi NATURE, RAW MATERIALS, AND POLITICAL ECONOMY: AN INTRO...

Author: M. Bell | Fred Hendricks

12 downloads 644 Views 3MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

CONTENTS LIST OF CONTRIBUTORS

ix

EDITORIAL ADVISORY BOARD

xi

NATURE, RAW MATERIALS, AND POLITICAL ECONOMY: AN INTRODUCTION Paul S. Ciccantell and David A. Smith

1

PART I: THEORETICAL FOUNDATIONS MATTER, SPACE, TIME, AND TECHNOLOGY: HOW LOCAL PROCESS DRIVES GLOBAL SYSTEMS Stephen G. Bunker and Paul S. Ciccantell

23

ENVIRONMENTAL SOCIOLOGY’S THEORETICAL AND EMPIRICAL PARADOXES Frederick H. Buttel

45

FOR A SOCIOLOGY OF ‘SOCIONATURE’: ONTOLOGY AND THE COMMODITY-BASED APPROACH Paul K. Gellert

65

KEEPING TIME: TEMPORAL HIERARCHIES IN SOCIO-ECOLOGICAL SYSTEMS Charles H. Wood

93

v

vi

CONTENTS

CYCLES OF ACCUMULATION, CRISIS, MATERIALS, AND SPACE: CAN DIFFERENT THEORIES OF CHANGE BE RECONCILED? Denis O’Hearn

113

PART II: COMMODITIES, EXTRACTION AND FRONTIERS STARTING AT THE BEGINNING: EXTRACTIVE ECONOMIES AS THE UNEXAMINED ORIGINS OF GLOBAL COMMODITY CHAINS David A. Smith

141

SUNK COSTS, RESOURCE EXTRACTIVE INDUSTRIES, AND DEVELOPMENT OUTCOMES Bradford L. Barham and Oliver T. Coomes

159

JAPAN’S ECONOMIC ASCENT AND ITS EXTRACTION OF WEALTH FROM ITS RAW MATERIALS PERIPHERIES Paul S. Ciccantell and Stephen G. Bunker

187

A PERPETUAL EXTRACTIVE FRONTIER? THE HISTORY OF OFFSHORE PETROLEUM IN THE GULF OF MEXICO Tyler Priest

209

COMMODITY FRONTIER AS CONTESTED PERIPHERY: THE FUR TRADE IN IROQUOIA, NEW YORK AND CANADA, 1664–1754 Jonathan Leitner

231

Contents

vii

EXTRACTION, GENDER AND NEOLIBERALISM IN THE WESTERN AMAZON Susanna Hecht

253

MATERIAL PROCESS AND INDUSTRIAL ARCHITECTURE: INNOVATION ON THE CUBAN SUGAR FRONTIER, 1818–1857 Dale Tomich

287

PART III: CONNECTING POLITICAL AND ECONOMIC CHANGE WORLD-SYSTEMS IN THE BIOGEOSPHERE: THREE THOUSAND YEARS OF URBANIZATION, EMPIRE FORMATION AND CLIMATE CHANGE Christopher Chase-Dunn, Alexis Alvarez and Daniel Pasciuti

311

COFFEE, REVOLUTION, AND DEMOCRACY IN CENTRAL AMERICA Jeffery M. Paige

333

PEASANTS, PLANTERS, AND THE PREDATORY STATE: EXPORT DIVERSIFICATION IN THE DOMINICAN REPUBLIC, 1970–2000 Andrew Schrank

353

SELLING THE RIVER: GENDERED EXPERIENCES OF RESOURCE EXTRACTION AND DEVELOPMENT IN LESOTHO Yvonne A. Braun

373

LIST OF CONTRIBUTORS Alexis Alvarez

Institute for Research on World-Systems, University of California-Riverside, CA, USA

Bradford L. Barham

Department of Agricultural and Applied Economics, University of WisconsinMadison, WI, USA

Yvonne A. Braun

Department of Sociology, University of Oregon, OR, USA

Stephen G. Bunkery

Department of Sociology, University of Wisconsin-Madison, WI, USA

Frederick H. Buttel y

Department of Sociology, University of Wisconsin-Madison, WI, USA

Christopher Chase-Dunn

Institute for Research on World-Systems, University of California-Riverside, CA, USA

Paul S. Ciccantell

Department of Sociology, Western Michigan University, MI, USA

Oliver T. Coomes

Department of Geography, McGill University, Quebec, Canada

Paul K. Gellert

Department of Sociology, University of Tennessee, TN, USA

Susanna Hecht

School of Public Affairs, University of California, Los Angeles, CA, USA

Jonathan Leitner

Peekskill, NY, USA

Denis O’Hearn

School of Sociology, Queens University, Belfast, N. Ireland and Department of Sociology, Binghamton University, Binghamton, NY, USA ix

x

LIST OF CONTRIBUTORS

Jeffery M. Paige

Department of Sociology, University of Michigan, MI, USA

Daniel Pasciuti

Institute of Research on World-Systems, University of California-Riverside, CA, USA

Tyler Priest

C.T. Bauer College of Business, University of Houston, TX, USA

Andrew Schrank

Department of Sociology, University of New Mexico, NM, USA

Gay Seidman

Department of Sociology, University of Wisconsin-Madison, WI, USA

David A. Smith

Department of Sociology, University of California, CA, USA

Dale Tomich

Department of Sociology, Binghamton University, NY, USA

Charles H. Wood

Center for Latin American Studies, Department of Sociology, University of Florida, USA

EDITORIAL ADVISORY BOARD Norman Long

Wageningen University, Wageningen, Netherlands (International Rural Sociology Association)

Max Pfeffer

Cornell University, USA (Rural Sociological Society)

Hilary Tovey

Trinity College, Dublin, Ireland (European Society for Rural Sociology)

Frank Vanclay

University of Tasmania, Australia (International Rural Sociology Association)

xi

NATURE, RAW MATERIALS, AND POLITICAL ECONOMY: AN INTRODUCTION Paul S. Ciccantell and David A. Smith ABSTRACT In this introductory chapter, we briefly outline the history of the political economy of raw materials, focusing particularly on the relationship between raw materials and economic development. We then introduce the chapters of this volume, and we conclude by discussing future directions for research in this area.

1. INTRODUCTION At the beginning of the twenty-ﬁrst century, nations, economies, and people around the world confront tremendous environmental challenges. Conﬂicts in oil-producing areas threaten production and increase prices at the same time that many argue that global oil supplies have begun a long-term decline. Access to water drives growing numbers of social and military conﬂicts. International efforts to address global warming falter as major polluting nations withdraw from the Kyoto Protocol. Deforestation in the rainforests of Southeast Asia, Africa, and the Amazon threaten to change Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 1–20 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10001-8

1

2

PAUL S. CICCANTELL AND DAVID A. SMITH

global weather patterns and increase poverty in many of the world’s poorest nations. Even the world’s fastest growing economy, China, faces mounting resource shortages and pollution problems that may choke off national and, as a result, global economic growth as well. How can we make sense of the increasing complexity of the interdependence between society and nature and the growing severity of the resulting environmental, economic, and social problems? The chapters in this volume present theoretical and empirical approaches to these problems that can help inform social science analysis of and public policy toward the relationship between society and nature. We have three aims in this introductory chapter. We will ﬁrst briefly outline the historical evolution of the political economy of nature and raw materials, highlighting the critical place that understanding the relationship between society and nature plays in the origins and ongoing theoretical developments in this ﬁeld. We then introduce the overall structure of this volume and the individual chapters, describing the key concerns and themes that guide this research in the diverse industries, nations, and natural conditions that these chapters analyze. Finally, we will discuss some directions for future research that emerge from these works and the critical issues regarding the nature–society relationship that are apparent in the world economy today.

2. THE POLITICAL ECONOMY OF NATURE AND THE CONCEPT OF RENT Political economy analysis beginning with Adam Smith (1776) and David Ricardo (1817) focused on issues of natural wealth and its consequences for national development. Both analyzed the impacts of the huge ﬂows of silver and gold from the Spanish and Portuguese colonies in the Americas on trade, money supply, and investment in Europe. The arguments presented for the gains from free trade and specialization in the areas in which nations held comparative advantages rested on the recognition that nations differed in their natural and other material characteristics (e.g. population, geographic size, access to oceans, etc.). In order to understand the contribution of differential endowments of natural resource wealth and quality in mineral deposits and agricultural productivity, early political economists developed the concept of rent to explain the wealth that accrued to property owners from these differences in

Nature, Raw Materials, and Political Economy

3

natural wealth. Ricardo deﬁned rent as ‘‘that portion of the produce of the earth, which is paid to the landlord for the use of the original and indestructible powers of the soil. It is often, however, confounded with the interest and profit of capital’’ (Ricardo, 1817, p. 91). The relationship between society and nature, as these early political economists recognized, is a critical element of political economic analysis, despite arguments about ‘‘dematerialization’’, the ‘‘new economy’’, and globalization today that largely ignore issues of matter, space, and nature–society relations and how these issues continue to shape political economy (Herman et al., 1989; Bunker, 1996). As the chapters in this volume will show, even some of the most global industries remain profoundly rooted in and shaped in critical ways by the natural production systems on which they depend. Many analysts in different theoretical traditions utilize the concept of rent and the ways in which it is distributed among various social groups as an analytical tool. Marx recognized that ‘‘nature’’ played an essential role in the capitalist mode of production and, indeed, in any human production process, but ‘‘nature’’ is a problematic concept. For Marx, relations between humans and ‘‘nature’’ are mediated by the human labor process: ‘‘labour is, ﬁrst of all, a process between man and nature, a process by which man, through his own actions, mediates, regulates and controls the metabolism between himself and nature’’ (Marx, 1977, p. 283). ‘‘Nature’’ also enters the production process in another fundamental way: ‘‘we may include among the instruments of labouryall the objective conditions necessary for carrying on the labour process. These do not enter directly into the process, but without them it is either impossible for it to take place, or possible only to a partial extent. Once again, the earth itself is a universal instrument of this kind, for it provides the worker with the ground beneath his feet and a ‘ﬁeld of employment’ for his own particular process’’ (Marx, 1977, p. 287). Marx thus deﬁnes ‘‘nature’’ transhistorically as, ﬁrst, an essential underlying condition of production as the location of production, second, as the source of natural products, and, third, as the original source of materials which are converted into raw materials for use by industry through the labor process. Humans and their productive processes are not, however, separate from ‘‘nature’’, as the conventional economics framework maintains. Humans are a part of Marx’ ‘‘nature’’, but a part which has a tremendous capacity to transform this ‘‘nature’’ as a result of conscious labor. Wittfogel (1985) extends Marx’ conception of the relationship between humans and ‘‘nature’’ and the particular form that this relationship takes in the capitalist mode of production. First, Wittfogel argues that ‘‘the

4

PAUL S. CICCANTELL AND DAVID A. SMITH

significance of the natural factor grows together with the social conditions (powers) of the process of production’’ (Wittfogel, 1985, p. 47). The development of industrial production in the capitalist mode of production increased the need for raw materials to the highest level in human history. Second, Wittfogel argues that ‘‘changes in scientific insight and technological practice can thus continually bring new kinds and sites of raw materials to the fore, replacing the old as obsolete and outworn. But this change in the concrete form of application does not invalidate the principle of the fundamental importance of sources of raw material, their ‘location’, and thus the necessity for capitalism to struggle for their control. It is precisely because of this dynamic character inherent in raw material (not by ‘nature’ but by reason of the activity of social labor) that the struggle for sources of raw material (those already known and those which may possibly become significant in the future) in the age of imperialism so feverish and the goals of imperialist politics so boundless’’ (Wittfogel, 1985, p. 52). The depletion of sources of raw materials in the capitalist nations of Western Europe and the United States forced capitalists to look for more fertile sources in other areas, resulting in political and military conﬂicts. Finally, Wittfogel provides an excellent solution to the problem of determinacy usually posed as humans versus ‘‘nature’’ as the key determinant of historical change: ‘‘if the totality of the powers of production determine the character of the mode of production at any given historical moment, it is the social aspects which (being the actively-motivating agents) determine change, whereas the naturally-conditioned agents determine whether and if change is possible and accordingly the direction of this change. Even as man puts nature to his ‘service’, he thereby submits himself to nature (Plekhanov) and follows her’’ (Wittfogel, 1985, p. 55). Within the capitalist mode of production, the ownership of the natural means of production and sources of raw materials and the resulting extraction of rent by the landlord class presents a problem for the labor theory of value: how are these natural products assigned prices? Marx’ analysis of the internal dynamics of the ‘‘pure’’ capitalist mode of production in Volume 1 of Capital ignored this problem, but this problem was very prominent in discussions of agriculture during the eighteenth and nineteenth centuries. In contrast to Volume 1’s model of a two-class mode of production in which the labor theory of value was the key dynamic, in Volume 3 Marx adds a third class, the landlord class, and introduces the concept of rent as this class means of expropriating a portion of the total surplus value from the capitalist class. The members of the landlord class are deﬁned by the ownership of private property in land: ‘‘landed property is based on the monopoly by certain

Nature, Raw Materials, and Political Economy

5

persons over definite portions of the globe, as exclusive spheres of their private will to the exclusion of all others’’ (Marx, 1967, p. 615). The key to analyzing this class ‘‘is to ascertain the economic value, that is, the realisation of this monopoly on the basis of capitalist production’’ (Marx, 1967, p. 616). The landlord class in the capitalist mode of production in England emerged from the destruction of the feudal mode of production. Under capitalist agricultural production, landed property no longer controlled workers tied to the land or even the means of production; this control passed to capitalist tenant farmers. These capitalist tenant farmers introduced rapid technological innovation and produced large agricultural surpluses using wage laborers. The landlords received ground-rent in return for the use of this natural means of production, the land. Marx refutes claims by ‘‘bourgeois economists’’ that this ground-rent is in fact a form of interest and not an important phenomenon in its own right. Marx notes that, to an individual capitalist choosing where to invest a certain stock of capital, buying land and investing in order to receive interest appear as two equivalent alternatives. However, the source of the returns on capital invested in these two alternatives is quite different; investing for interest increases the stock of capital available for capitalist production, while receiving rent from land purchased results from the extraction of surplus value from the capitalist class. The landlord class is thus a distinct class with a distinct source of surplus value and is in conﬂict with the capitalist class over distribution of surplus value (Marx, 1967, pp. 622–624). The rent appropriated by the landlord class takes three forms. The ﬁrst form is Differential Rent I (DRI) which is equivalent Ricardo’s initial conception of rent. DRI occurs when leased natural means of production produces a surplus-profit for a capitalist tenant; this surplus-profit is ‘‘produced as a difference between the products of two equal quantities of capital and labour, and this surplus-profit is transformed into ground-rent when two equal quantities of capital and labour are employed on equal areas of land with unequal results’’ (Marx, 1967, p. 649). The unequal results of the application of equal amounts of capital and labor to equal amounts of land results from two causes: differential fertility between the two pieces of land, whether naturally produced or the result of previous investment in the piece of land; and the locational advantages of one piece of land over the other relative to the location of markets (Marx, 1967, p. 650). This DRI results from the process of formation of market prices for agricultural products. In contrast to capitalist industrial production, in which competition among producers forces the market price down to the social average price of production, the market price of agricultural products is the price of production

6

PAUL S. CICCANTELL AND DAVID A. SMITH

of the highest cost producer. This is true, however, only if the supply of the agricultural product is in rough balance with the demand for the product; excess production forces down prices to the cost of production of the highest cost producer whose production is needed to satisfy demand. Producers whose price of production is higher than this new regulator price are forced to sell at a loss or to stop producing. The market price is thus a social product, not a technical result; the exchange-value of the agricultural product determines its market price (Marx, 1967, pp. 658–661). DRI accruing to the owner of a particular piece of land is thus the difference between the price of production of the highest cost plot of land and the price of production of the lower cost piece of land owned by the individual landlord. The second form of differential rent (DRII) is based on the DRI which accrues to investment to two plots of land of different fertility and location in the same time period. DRII accrues, however, over more than one time period because of differential distribution of capital and access to credit (Marx, 1967, pp. 676–677). The input of capital over time can increase the productivity of a piece of land and the amount of DRII that accrues to the landowner as the result of investments by capitalist tenants. The third form of ground-rent, absolute rent, results directly from the definition of private property in land: ‘‘the mere legal ownership of land does not create any ground-rent for the owner. But it does, indeed, give him the power to withdraw his land from exploitation until economic conditions permit him to utilise it in such a manner as to yield him a surplusyhe can change the quantity of land placed on the marketyhe does not lease his land until he can be paid lease money for it’’ (Marx, 1967, p. 757). Thus, Marx argues, ‘‘landed property itself has created rent’’ (Marx, 1967, p. 755). Absolute rent thus appropriated by landlords appears, Marx argues, as the equivalent of monopoly prices in capitalist industrial production, prices determined strictly by the purchasers’ willingness and ability to pay and not by the price of production or the value of the commodity. This is only one small part of the total absolute rent expropriated by landlords, however. The vast majority of absolute rent results from the lower organic composition of capital in agriculture relative to industry: ‘‘rent, then, forms a portion of the value, or, more specifically, surplus-value, of commodities, and instead of falling into the lap of the capitalists, who have extracted it from their labourers, it falls to the share of the landlords, who extract it from the capitalists’’ (Marx, 1967, p. 771). Nwoke (1984) extends Marx’ concept of rent to the extraction of minerals in peripheral nations. Nwoke deﬁnes mining rent ‘‘as the difference between the market price of a unit of mineral resource sold in the form of ﬁnished

Nature, Raw Materials, and Political Economy

7

products and the total average costs incurred in discovering, producing, transporting, reﬁning, and marketing the unit of that particular mineral resource. Total average costs are current costs, and include amortization and a fair rate of return required by a private ﬁrm in the minerals industries’’ (Nwoke, 1984, p. 42). The market price is deﬁned as the production price of the highest cost producer needed to satisfy social demand. Differential rents occur because of differential natural productivity. Harvey (1982) departs radically from Marx’ conception of the landlord class as a distinct class and the role of this class as mediating the relations between capitalist production and ‘‘nature’’. Harvey begins by citing Marx’ discussion of colonies such as the United States in which landed property did not exist before the entrance of capitalism into the region; private property had a significantly different form in these colonies than it did in England (Harvey, 1982, pp. 345–346). Marx argued that ‘‘when the landlord is himself a capitalist, or the capitalist is himself a landlordyfor him landed property does not constitute an obstacle to the investment of capital. He can treat his land simply as an element of Nature and therefore be guided solely by considerations of expansion of his capital, by capitalist considerations’’ (Marx, 1967, p. 751). Harvey argues that this situation has increasingly become the norm in the capitalist mode of production. Land has become ‘‘a commodity of a rather special sort’’ whose ownership entitles the landlord to receive an annual rent; therefore, ‘‘any stream of revenue (such as an annual rent) can be considered as the interest on some imaginary, ﬁctitious capital. For the buyer, the rent ﬁgures in his accounts as the interest on the money laid out on land purchase, and is in principle no different from similar investments in government debt, stocks and shares of enterprises, consumer debt and so on’’ (Harvey, 1982, p. 347). Land thus becomes a ﬁnancial asset, investment in which is determined by expected future returns on this asset as opposed to expected returns on equivalent assets such as stocks. Landlords, rather than being a distinct class, become merely a part of the fraction of the capitalist class whose capital is invested as interest-bearing capital. Harvey argues that ‘‘the integration of landownership within the circulation of interest-bearing capital may open up the land to the free ﬂow of capital, but it also opens it up to the full play of the contradictions of capitalism’’ (Harvey, 1982, p. 349). Private property in land and rent for Harvey are the means of the social relations of the capitalist mode of production to extend their control over ‘‘nature’’. ‘‘Nature’’ is subordinated to the interests of individual capitalists and to the working out of the contradictions of the laws of motion of

8

PAUL S. CICCANTELL AND DAVID A. SMITH

capital. Harvey does not deny the ‘‘natural’’ element of features such as space as location and land as the site of naturally produced use values, but these elements are clearly secondary in comparison to the working out of the laws of motion of capital on a global scale. Notions of the ‘‘social construction of space’’ and even of the ‘‘social construction of ‘nature’’’ are not far removed from Harvey’s work. Harvey’s development and extension of Marx’ work contain a very large number of powerful insights into the operation of late capitalism and the relations between humans and ‘‘nature’’ as space and location; however, even the minimal constraints imposed on human production by ‘‘nature’’ recognized by Marx seem to disappear in Harvey’s analysis.

3. COMMODITY-BASED ANALYSIS AND LINKAGE THEORY Other political economists have focused on space as a material characteristic of the world economy. Commodity-based analysis of economic development, pioneered by Innis’ (1956) analysis of the economic development of Canada, provides conceptual tools to examine the nature–society relationship. Three key theoretical insights emerge from Innis’ work. First, the particular characteristics of any natural resource structure in important ways the social technology and economics of extracting, processing, and consuming the resource. Second, the naturally produced location of a resource interacts with the socially produced aspects of location, including transport technology, transport costs, and distance from world markets, to partially determine the economic viability of resource extraction in a particular region. Third, the nature and size of the world market for a particular natural resource determine in important ways whether, where, and when a particular natural resource is extracted. Barham, Bunker, and O’Hearn (1994) extend commodity-based analysis by constructing a bottom-up approach for studying raw materials industries that focuses on ﬁve ‘‘basic characteristics of raw materials that shape industry organization and operationydemand (or social uses), relative scarcity, geographic concentration, specific geographic and ecological location, and technology of extraction and reﬁning’’ (Barham et al., 1994, p. 16). Analysts identiﬁed ﬁve key dimensions of this spatiality. First, space provides the location of sites for natural resource extraction and processing, and, second, space as distance and topography plays a crucial role in

Nature, Raw Materials, and Political Economy

9

determining the costs of transporting natural resource-based products from their naturally determined locations to markets (Innis, 1956; Bunker, 1992; Bunker & Ciccantell, 2003). Third, space as the location of natural resource extraction provides the basis for the accrual of rent to the holders of property rights over the location (Marx, 1967; Peet, 1980, 1985). Fourth, the socially random location of natural resources within the social boundaries of particular states provides the basis for natural resource-based linkage development strategies (Bunker, 1992; Auty, 1990). Fifth, and most critically, as Mandel (1975), Harvey (1982), and Bunker (1985) argued, capitalist economic development is fundamentally uneven across space because of the underlying logic of capital accumulation. The uneven development across space of the capitalist world economy is a fundamental structural characteristic of the system as a whole (Chase-Dunn, 1989; Wallerstein, 1974). Innis’ careful commodity-based analysis was appropriated and combined with concepts of growth poles and linkages in the economic development literature. Linkage theory (Hirschman, 1958), the related growth pole model of development (Perroux, 1955), and the staple theory of economic growth presented by Watkins (1963) were combined in theory and practice to argue that natural resource extraction and processing could serve as the engine of development for Third World nations with significant natural resource endowments. In some sense Hirschman’s linkage model is a model for the overcoming of comparative disadvantage in industrial production in noncore nations by using a comparative advantage in natural resource extraction as the basis for industrialization. Hirschman’s version of linkage theory provides a model of national economic development that, the theory argues, will eventually result in a fully articulated national economy with internal capital accumulation and consumption providing the key dynamics of economic development. This model thus proposes a method by which a noncore state can lead its nation out of the key trap seen by dependency theory (Girvan, 1976; Cardoso & Faletto, 1969; Jalee, 1968), the inability of dependent nations to escape the domination of the processes of capital accumulation and consumption in the center nations. Hirschman’s linkage theory provides prescriptions for both external and domestic economic policies. In relation to the world economy, the theory argues that, in order to build efﬁcient and competitive facilities in the noncore nations, these facilities must be of at least some minimum economic size, ‘‘deﬁned in economic terms relative to normal profits and efﬁcient foreign suppliers. In other words, it is the size at which the domestic ﬁrm will be able both to secure normal profits and to compete with existing foreign suppliers, taking into account locational advantages and disadvantages as

10

PAUL S. CICCANTELL AND DAVID A. SMITH

well as, perhaps, some infant industry protection. In this way comparative cost conditions are automatically taken into account’’ (Hirschman, 1958, p. 101). This comparison with the most modern plants in the developed nations in order to minimize economic inefﬁciency (a criterion typically ignored in the actual implementation of linkage-based policies), while not requiring that noncore plants capture all of the potential economies of scale available, means that these plants must not sacriﬁce too large a part of these economies, thus requiring plants to be of relatively large size. This policy has four major implications for the relations between the noncore nation and the world economy: ﬁrst, because of the need to capture at least part of the beneﬁts from economies of scale, unbalanced growth is an inevitable result, since the limited capital available in a particular noncore nation would preclude the construction of plants in many industries at the same time; second, a large share of production from these facilities must be exported to the world market, since the domestic market is likely to be far too small to absorb this production; third, in order to ﬁnance such large investments, the participation of foreign capital is likely to be necessary; and, fourth, foreign technology and expertise will need to be imported in order to construct and operate world-class facilities. In the domestic sector, Hirschman’s linkage model argues that the key to promoting economy development is the creation of opportunities for entrepreneurial activity in one or a few carefully selected leading sectors through state investment and encouragement of private activity. Given the limited capital and human resources available in noncore nations, the theory argues that these limited resources should be concentrated on one or a few such leading sectors in one or a few regions, termed growth poles, in a process of unbalanced growth. The state focuses its investment and development efforts in these regions and sectors with the goal of creating linkages from these leading sectors to other industries within the nation, thus stimulating the creation of and production by other industries and increasing consumption of domestic products (Hirschman, 1958, pp. 62–75). The linkage model as applied to natural resources thus has in the abstract a powerful multidimensional logic. The abstract economic logic of the model is ﬁvefold. First, the natural resource endowments located within the nation’s boundaries represent stocks (such as mineral deposits) and ﬂows (such as rivers with hydroelectric potential) of wealth whose extraction, processing, and consumption can in turn produce even greater wealth. Second, extraction and processing create markets for other goods and services; these new markets make it possible to develop backward linkages into the extraction of other natural resources that are used as raw materials and

Nature, Raw Materials, and Political Economy

11

forward linkages into intermediate and ﬁnished goods production within the nation’s boundaries for domestic consumption and export. Third, the initial natural resource-based industry and its linked industries create entrepreneurial opportunities for the domestic capitalist class and job opportunities for the middle class and working class, all of which contribute to increasing national income. Fourth, shares of the wealth created by natural resource extraction that accrue to the domestic capitalist class and to the government make it possible to create ﬁscal linkages to develop other unrelated industries in the nation through investment in new sectors of the economy. Fifth, the experience and training gained by workers, managers, and entrepreneurs in the initial natural resource-based industry provides improved human capital resources for the development of new sectors. The abstract political logic of the linkage model is twofold. First, the accrual of beneﬁts to the capitalist, middle, and working classes can provide the government with broad-based political support. Second, the extraction and processing of natural resources that are frequently located in remote regions of the nation reinforces national sovereignty both through the consolidation of governmental control over its territory and through the provision of additional resources to the government that permit it to meet the variety of demands placed upon it. The abstract social logic of the linkage model is twofold as well. First, job creation and increased national income result in improved living standards for the population. Second, the extension of national sovereignty incorporates remote populations into national society. The existence of natural resource in a noncore nation is seen by development planners employing linkage theory models as the starting point for the creation of leading sectors. However, linkage theory has several important problems, both as a theoretical construct and, in practice, as the basis for development planning. One crucial problem with linkage theory as a theoretical construct is that, as Bunker (1989) argues, Hirschman (1958) originally based his argument about the logic of the spread of linkages across sectors in a national economy. During the course of the book, however, Hirschman argues without substantial support that the logic also applies across regions in a national economy. The problem, as Bunker argues and Hirschman himself had earlier argued, was that linked industries tended to cluster geographically around the leading sector. Thus, in practice, even if the leading sector created links across sectors, few links across regions would be created because of transport, infrastructural, and other costs associated with geographical dispersion of industries. This confusion of sector and space, Bunker argues, became problematic in practice because

12

PAUL S. CICCANTELL AND DAVID A. SMITH

development planners accepted this supposed equivalence and used the linkage and growth pole model for planning economic development projects intended to resolve interregional imbalances. Second, growth pole theory and application, Bunker (1989) argues, ignores the specific characteristics of the raw materials involved in leading sectors. Innis (1956) pays close attention to the natural characteristics of raw materials, their relative locations, and transportation technology, but growth pole theory and its application typically ignored these characteristics and their impacts on economic development projects. Third, the growth pole model of development ignores the characteristics of the world market and world industry of which an economic development project will be a part. The capitalist ﬁrms based in the core nations play very large roles in the provision of technology and investment for raw materials extraction and processing projects throughout the world and world markets provide the outlet for the production in excess of domestic consumption of the leading sectors of the noncore nations. Moreover, the leading sectors of these noncore nations are likely to be quite different from the leading sectors of the capitalist world economy as a whole, since leading sectors of noncore economies are typically industries in which the rate of profit in the core nations has fallen and costs of production have risen because of increased competition, rising costs of labor and other inputs, and the elimination of technological rents (Vernon, 1971; Frobel, Heinrichs, & Kreye, 1980). Failure to understand the structure, economics, and politics of the world industry can lead to poor development planning in noncore countries. Fourth, linkage theory assumes the existence of an economically and political strong state with a considerable degree of relative autonomy of action from domestic social classes and class fractions and from transnational corporations, core states, and the capitalist world economy as a whole. The various state agencies and domestic classes and class fractions involved in development planning and implementation are also assumed to share some set of goals about the purpose of and appropriate means for these efforts. Moreover, the linkage model assumes the technical rationality of development planning and of state action to implement these plans based on the existence of a group of skilled technocrats; these assumptions are frequently violated in practice (Bunker, 1989). Building on the logic of linkages and growth poles based on raw material wealth and the application of these ideas as development policies in many nations, Richard Auty (1990) examines the impacts of resource-based industrialization in a variety of locations. Auty’s comparative analysis ﬁnds that eight different oil-rich nations that sought to use oil revenues to fund

Nature, Raw Materials, and Political Economy

13

domestic industrialization have all failed in these efforts. This negative result is echoed in Terry Karl’s (1997) analysis of the consequences of petroleumbased industrialization in Venezuela. Some development economists have gone even further in their negative assessment of the impact of raw materials on economic development in recent decades. A number of scholars, including Sachs and Warner (1999, 2001), Gylfason, Herbertsson, and Zoega (1999), argue that natural resource wealth is a ‘‘curse’’ for economic development, claiming that countries with abundant natural resource wealth grow more slowly than resource-poor countries because of their natural resource wealth. This ridiculous claim of causation based on the correlation between natural resource wealth and poverty in former colonies is comforting to the core states and mining ﬁrms that beneﬁt from paying low prices for raw materials, but the absurdity of blaming the victims of imperialism and neocolonialism for misallocating the rent windfall from petroleum and other mineral industries is at best ahistorical and apologetic, rather than analytical.

4. THE CONTRIBUTIONS OF THIS VOLUME TO THE POLITICAL ECONOMY OF NATURE AND RAW MATERIALS The chapters in this volume push the study of the multifaceted nature– society relationship and the socioeconomic consequences of human dependence on nature forward in a variety of areas. The chapters in this volume are divided into three sections. In the ﬁrst section, the ﬁve chapters lay out theoretical models for examining the nature–society relationship. In the ﬁrst chapter, Stephen Bunker and Paul Ciccantell outline the key elements of their new historical materialist model of long-term socioeconomic change. Based on their forthcoming book, Bunker and Ciccantell argue that comparative analysis of the four most rapid and transformative cases of rapid economic ascent over the past ﬁve centuries (Holland, Great Britain, the US, and Japan) reveals a common set of challenges to and strategies for rapid economic ascent. The tremendous materially based economies of space potentially available in the extraction, transport, and processing of the most voluminously used raw materials confront equally important diseconomies of scale as rising industrial economies are forced to meet their growing raw material needs from ever more distant resource frontiers. In these few spectacular cases, coordinated state–sector–ﬁrm strategies overcame

14

PAUL S. CICCANTELL AND DAVID A. SMITH

these challenges and achieved these economies of scale and resulting global competitiveness and trade dominance, while simultaneously creating and exploiting their resource frontiers. In the following chapter, Fred Buttel outlines what he identiﬁes as critical theoretical and empirical paradoxes in the ﬁeld of environmental sociology as it has developed over the past three decades. Buttel argues that environmental sociology is a research program of shared concerns rather than a coherent sociological subdiscipline. By analyzing the paradoxes that characterize research in the ﬁeld, Buttel seeks to lay out a research agenda for environmental sociology. In Chapter 4, Paul Gellert seeks to build a model of nature–society relations that integrates what are typically seen as two distinct spheres of ‘‘nature’’ and ‘‘society.’’ Gellert’s model links the insights of postmodern approaches to social science to materialist analysis of the nature–society relationship. In Chapter 5, Chuck Wood outlines a model of the linkages between the social and natural world by focusing on the relevance of time and what he terms the ‘‘temporal frequencies intrinsic to the spatial and organizational hierarchies that comprise the socio-ecological system.’’ The goal is to introduce careful consideration of time into environmental studies using the concepts of ‘‘temporal grain’’ and ‘‘temporal fallacies’’ to strengthen analysis of environmental issues. In Chapter 6, Denis O’Hearn presents a comparative analysis of competing theories of socioeconomic change, arguing that theories of long-term hegemonic change need to incorporate understandings of local economic change and the path dependencies created by long-term changes. O’Hearn argues that incorporating a focus on materio-spatial process into a multiple variable model of crisis and hegemonic change will produce a more complete model of long-term change in the world economy. In the second section, a series of case studies covering a range of industries, locations, and historical periods present a variety of applications of the political economy of natural resources to critical issues regarding commodities, extraction, and frontiers. In Chapter 7, David Smith links the conceptual model of commodity chains to the research on extractive economies, arguing that the commodity chains literature largely ignores the extractive end of the production system and seeking to create a new integration between these two largely distinct literatures. In Chapter 8, Brad Barham and Oliver Coomes focus on the impact of sunk costs in extractive industries in shaping and constraining the potential developmental impacts of extractive industries. Barham and Coomes argue

Nature, Raw Materials, and Political Economy

15

that sunk costs fundamentally shape ﬁrm strategies, resulting industrial structures, and ultimately the developmental outcomes of regions dependent on extractive industries. The negative consequences emphasized by the ‘‘resource curse’’ literature are largely the result of these ﬁrm strategies, industry structures, and the strategies of states and individuals in extractive regions that are driven by the dynamics created by the existence of sunk costs. In Chapter 9, Paul Ciccantell and Stephen Bunker present a summary of their forthcoming book on Japanese economic development. Their analysis reveals how the very same industries, most notably the steel and shipping industries, simultaneously drove Japanese economic development and the underdevelopment of frontier regions in Australia, Brazil, and Canada that supplied increasing quantities of raw materials at progressively lower cost to the Japanese steel industry. In Japan, state–sector–ﬁrm coordination created a mutually reinforcing ‘‘virtuous circle’’ of economic development based on raw materials’ processing, while in Japan’s extractive peripheries, the steel industry imposed huge costs and risks on exporting states and ﬁrms and extracted billions of dollars in subsidies for Japanese industrialization from these raw materials’ suppliers. In Chapter 10, Ty Priest examines one of the most global and highest technology industries in the world, offshore oil and gas production. Based on his analysis of this industry in the Gulf of Mexico, Priest argues that technological innovations in the industry have opened large new resources in deepwater areas that were previously economically and technologically inaccessible. However, despite efforts by the oil industry to claim that these innovations demonstrate the ability of the industry to free itself from natural constraints and provide an ever-increasing supply of oil, Priest argues that the particular natural characteristics of marine resource development in the Gulf of Mexico that created these technological opportunities, but these conditions are not endlessly available globally. In Chapter 11, Jonathan Leitner shifts the temporal focus back several centuries and the geographic focus to frontier regions that later became part of the U.S. Leitner ﬁnds that the fur trade in Iroquois regions of New York and Canada incorporated these regions as extractive peripheries whose economic and political characteristics were shaped both by the natural and technological characteristics of the industry and by the efforts of various groups to shape this extractive periphery to suit its interests. In Chapter 12, Susanna Hecht examines the emergence of extractive reserves in the western Brazilian Amazon and the role of women in these reserves, focusing particularly on the role of women as processors of Brazil

16

PAUL S. CICCANTELL AND DAVID A. SMITH

nuts. Hecht argues that longstanding misunderstandings of the roles of women in the rainforest have led to poorly designed development policies that place even heavier burdens on women and families. These policies often exacerbate existing inequalities within communities and within families, rather than contributing to development and improved standards of living. In Chapter 13, Dale Tomich examines the sugar industry in Cuba during the ﬁrst half of 1800s. Tomich argues that the spatial and material conditions of sugar production shaped the pattern of capital accumulation and political economic development in Cuba, as sugar planters sought to respond to changing world market conditions under constraints of the material requirements and spatial constraints of growing and processing sugar on an increasingly large scale. Planters created new production spaces in frontier regions and spatially reconstructed their plantations and processing plants to provide increasing quantities of sugar to the world market. In the third section, ‘‘Connecting Political and Economic Change’’, four chapters focus on the relationship between raw materials, economic change, and socioeconomic change. In Chapter 14, Chris Chase-Dunn, Alexis Alvarez, and Daniel Pasciutti take a long-term macrocomparative approach to explaining the impacts of climate change on urbanization and the formation of political empires. Their research reveals synchronous changes in regional interaction networks in East Asia and the West Asian/Mediterranean region. The chapter then presents a plan for analyzing the causes of this empirical phenomenon. In Chapter 15, Jeffery Paige examines the relationship between high levels of dependence on the coffee industry and political patters of revolution and democracy. Paige argues that dependence on this commodity produces an anti-democratic elite that blocks democracy. Democratic transitions in Central America are due to what he terms the ‘‘route to democracy through socialist revolution from below’’ that forces elites to open political systems. Paige’s research thus challenges both of the major models of the transition to democracy, formulating an alternative route to democracy that is rooted in the historical experiences of Central America. In Chapter 16, Andrew Schrank examines efforts to promote nontraditional manufacturing exports within the sociopolitical structure of the Dominican Republic created by the historical reliance on agro-exports. Although this ‘‘Dominican model’’ is now often touted by international ﬁnancial institutions as a model to be emulated by other poor nations, Schrank argues that this model is limited both in terms of its ‘‘success’’ in the Dominican Republic and in its potential applicability in other nations. The preexisting social and ecological characteristics of the Dominican

Nature, Raw Materials, and Political Economy

17

Republic both enabled and constrained the opportunities for the state, Dominican elites, and foreign ﬁrms to utilize free trade zones as centers for economic development and capital accumulation. In Chapter 17, Yvonne Braun analyzes the Lesotho Highlands Water Project, a dam-based development effort that focuses on exporting water from Lesotho to industrial and population centers in South Africa. Braun argues that the rural population of Lesotho, and particularly women in these poor households, are effectively providing large subsidies in the form of lost land and other resources to this project. Rural residents have lost much of their resource base, with little or no compensation, in order to support economic growth and water consumption in South Africa as part of national development strategy in Lesotho.

5. FUTURE DIRECTIONS FOR RESEARCH People in many extractive regions confront the same types of problems and conﬂicts outlined in many chapters in this volume. Indigenous groups and small producers face threatened dispossession by more powerful groups with different uses for their land and resources. In the Brazilian Amazon, oil extraction, gold mining, dam building, logging, and agriculture all threaten to redeﬁne space and resources as sites for extraction and profit for national and transnational ﬁrms, leaving long-term residents without livelihoods. Similar processes are underway in rainforest regions in Asia and Africa (see, e.g. WCD, 2000 on the impacts of large dams). Diamonds, oil, and other resources fuel authoritarian governments and opposition movements in Africa and Asia. Many of these situations merit further research in order to understand the relationship between global processes and local material and social processes. Given the focus in this volume on the critical role of natural resources for society, one question to be asked is, when will the world economy exhaust its supplies of critical raw materials? Some analysts now argue that we have reached Hubbert’s peak in global oil production (Campbell, 1997; Deffeyes, 2001), meaning that increasing prices and declining supply will create tremendous economic and political instability as ﬁrms and nations are forced to seek alternatives to the petroleum-based economy that has dominated the past century (Klare, 2001). Rapidly growing demand for oil and other energy sources in China, India, and the U.S. is motivating an intense competition for gaining access to these resources in the Middle East, Africa, Latin America, Asia, and the former Soviet Union. The Chinese

18

PAUL S. CICCANTELL AND DAVID A. SMITH

government, for example, has created an extensive network of oil supply relations with Iran, Sudan, Venezuela, and other nations that are often in conﬂict with the U.S. The potential socioeconomic impacts of peaking oil production and the potential geopolitical impacts of these national strategies to gain access to oil and, in extractive regions, to use oil resources to promote development are all worthy of further investigation. An important counterweight to this crisis scenario, however, is the tremendous capacity for innovation in the capitalist world economy. Jevons and Flux (1965) once made a convincing case for Great Britain’s exhaustion of its coal supplies in the 1800s, and the Meadows (1972) made similar predictions in the 1970s for a variety of resources, concerns popularized via Julian Simon’s bet with Paul Ehrlich about future prices of commodities. However, capitalist ﬁrms and states have developed new technologies, opened new resource frontiers, and created new ways to transfer the costs of extraction and processing to nations in these resource frontiers that increased supplies and reduced real costs in petroleum, coal, iron ore, agriculture, and many other industries over the past century. The political economies of the nature–society relationship presented in this volume make it clear that there are natural limits to growth, but social structures and strategies make it difﬁcult to precisely specify these limits. Further careful research along the lines of the chapters in this volume is essential to understanding these limits and ﬁnding alternative patterns of nature–society relations. The U.S. invasion of Iraq and other ongoing conﬂicts in the Middle East, Central Asia, and Africa make further analysis of the oil industry and its socioeconomic and geopolitical role imperative. Many ﬁrms, states, and social groups remain critically dependent on the availability of growing volumes of low-cost petroleum and have proven capable of shaping the global political economy in support of their interests and of imposing huge and tragic human, economic, social, and environmental costs on others. The Middle East may once again prove to be the space in which efforts to achieve or maintain economic and political hegemony founder, as it was for Germany during World War II and for Great Britain and France at Suez in 1956. The U.S. neoconservative imperial strategy may end in similar fashion in Iraq and Afghanistan in pursuit of the same goal of securing low-cost oil. Such an end is far from certain, just as the British general Gordon’s disaster at Khartoum did not end British imperialism in the nineteenth century. Further research on the political economy of nature–society relations is needed to understand these strategies and their consequences, as well as the resistance that these strategies generate.

Nature, Raw Materials, and Political Economy

19

REFERENCES Auty, R. (1990). Resource-based industrialization: Sowing the oil in eight developing countries. Oxford: Oxford University Press. Barham, B., Bunker, S., & O’Hearn, D. (Eds) (1994). States, firms, and raw materials: The world economy and ecology of aluminum. Madison: University of Wisconsin Press. Bunker, S. (1989). Staples, links and poles in the construction of regional development theories. Sociological Forum, 4(4), 589–610. Bunker, S. (1992). Natural resource extraction and regional power differentials in a global economy. In: S. Ortis & S. Lees (Eds), Understanding economic process (pp. 61–84). Lanham: University Press of America. Bunker, S. G. (1985). Underdeveloping the Amazon: Extraction, unequal exchange, and the failure of the modern state. Urbana: University of Illinois Press. Bunker, S. G. (1996). Raw materials and the global economy: Oversights and distortions in industrial ecology. Society and Natural Resources, 9, 419–429. Bunker, S. G., & Ciccantell, P. S. (2003). Generative sectors and the new historical materialism: Economic ascent and the cumulatively sequential restructuring of the world economy. Studies in Comparative National Development, 37(4), 3–30. Campbell, C. J. (1997). The coming oil crisis. Brentwood, England: Multi-Science Publishing and Petroconsultants. Cardoso, F. H., & Faletto, E. (1969). Dependencia y Desarollo en America Latina. Mexico City: Siglo Veintiuno. Chase-Dunn, C. (1989). Global formation: Structures of the world-economy. Cambridge: Basil Blackwell. Deffeyes, K. S. (2001). Hubbert’s peak: The impending world oil shortage. Princeton: Princeton University Press. Frobel, F., Heinrichs, J., & Kreye, O. (1980). The new international division of labour: Structural unemployment in industrialised countries and industrialisation in developing countries. Cambridge: Cambridge University Press. Girvan, N. (1976). Corporate imperialism: Conflict and expropriation. New York: Monthly Review Press. Gylfason, T., Herbertsson, T. T., & Zoega, G. (1999). A mixed blessing: Natural resources and economic growth. Macroeconomic dynamics, 3, 204–225. Harvey, D. (1982). The limits to capital. Chicago: University of Chicago Press. Herman, R., et al. (1989). Dematerialization. In: J. Ausubel & H. Sladovich (Eds), Technology and environment (pp. 50–69). Washington: National Academy Press. Hirschman, A. (1958). The strategy of economic development. New Haven: Yale University Press. Innis, H. A. (1956). Problems in Canadian economic history. Toronto: University of Toronto Press. Jalee, P. (1968). The pillage of the third world. New York and London: Monthly Review Press. Jevons, S., & Flux, A. (1965). The coal question: An inquiry concerning the progress of the nation, and the probable exhaustion of our coal-mines (3rd ed.). New York: A.M. Kelley. Karl, T. L. (1997). The paradox of plenty: Oil booms and petro-states. Berkeley: University of California Press. Klare, M. T. (2001). Resource wars: The new landscape of global conflict. New York: Metropolitan Books.

20

PAUL S. CICCANTELL AND DAVID A. SMITH

Mandel, E. (1975). Late capitalism. London: Verso. Marx, K.(ed. F. Engels). (1967). Capital (Vol. 3). New York: International Publishers. Marx, K. (1977). Capital (Vol. 1). New York: Vintage Books. Meadows, D. (1972). The limits to growth: A report for the club of Rome’s project on the predicament of mankind. New York: Universe Books. Nwoke, C. (1984). World mining rent: An extension of Marx’s theories. Review, 8(1), 29–89. Peet, R. (1980–81). Historical forms of the property relation: A reconstruction of Marx’s theory. Antipode, 12–13, 13–29. Peet, R. (1985). The social origins of environmental determinism. Annals of the Association of American Geographers, 75(3), 309–333. Perroux, F. (1955). A note on the concept of ‘growth poles’. In: I. Livingstone (Ed.), Development economics and policy readings (pp. 182–187). London: George Allen & Unwin. Ricardo, D. (1817). Principles of political economy and taxation. Harmondsworth, England: Penguin. Sachs, J. D., & Warner, A. M. (1999). The big push, natural resource booms and growth. Journal of Development Economics, 59, 43–76. Sachs, J. D., & Warner, A. M. (2001). Natural resources and economic development: The curse of natural resources. European Economic Review, 45(4–6), 827–838. Smith, A. (1776). An inquiry into the nature and causes of the wealth of nations (1937 ed.). New York: Modern Library. Vernon, R. (1971). Sovereignty at bay: The multinational spread of U.S. enterprises. New York: Basic Books. Wallerstein, I. (1974). The modern world-system. New York: Academic Press. Watkins, M. (1963). A staple theory of economic growth. Canadian Journal of Economics and Political Science, 29, 141–158. Wittfogel, K. (1985). Geopolitics, geographical materialism and Marxism. Antipode, 17(1), 21–72 (original 1929). World Commission on Dams (WCD). (2000). Dams and development: A new framework for decision-making: The report of the world commission on dams. London: Earthscan Publications.

MATTER, SPACE, TIME, AND TECHNOLOGY: HOW LOCAL PROCESS DRIVES GLOBAL SYSTEMS Stephen G. Bunkery and Paul S. Ciccantell ABSTRACT Incorporating local space, matter, and society into our concepts of the global in analytically compatible ways poses a major challenge for contemporary scholars of both world systems and globalization. Many analysts ignore both materiality and locality of production. They assume the global as their point of departure, and attempt to incorporate the local into it. In this chapter, we aim to reverse that logic. We will take into account and theorize the interaction of natural and social processes. In other words, we will integrate ecologic or materio-spatial logic with sociologic within the economic logic of global markets.

1. INTRODUCTION Understanding the disruptions and inequalities that the extraction of natural resources for globalized raw material markets imposes on local Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 23–44 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10002-X

23

24

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

ecological and social systems requires close attention to the complex interweavings of matter and space with economy and society. This is not an easy task. Most of our life experience of economic process consists primarily of consuming the products of increasingly homogenous, increasingly largescale industry. Most of us live and work in globalized economies, where process and product are increasingly distant from the sources of the raw materials they transform. Few of us have much direct experience or practice in observing or analyzing how multiple naturally produced heterogeneous material forms, extracted and transported from many different places, are combined to produce the standardized commodities we purchase and use. Marx (1967) explained that mainstream theories of capitalist economies fetishize commodities so as to mask the social relations that produce them. More recently, Susan Willis (1991) has pointed out that the growing distance between production and consumption that occurs as ﬁrms with markets in wealthy countries, move production to poorer countries, makes it even more difﬁcult to perceive the social relations of production in the commodities produced. If ideology and distance make it that difﬁcult to discern the social inequities and exploitation incorporated into the commodities we consume directly, then perceiving and thinking about the natural processes of production and the social relations of extraction that provide the raw materials incorporated in the production of commodities must be even more challenging. By the time we experience industrial products as commodities, the raw materials of which they are made would have been transformed and combined in ways that make their natural forms and shapes, as well as their place of origin, extremely difﬁcult to identify. The biology and geology that produce many of these raw materials and that constrain the means of their extraction and transport are alien and exotic to us; the physics and chemistry that transform them into machines and commodities may seem equally obscure. The specific conﬁgurations of matter and space that produce these raw materials and constrain the social processes of their extraction, transport, and processing become even more difﬁcult to observe, imagine, or think through as globalizing capital employs more complex technologies to transform the raw materials it imports from ever more distant sources. Incorporating local space, matter, and society into our concepts of the global in analytically compatible ways, therefore poses a major challenge for contemporary scholars of both world systems and globalization (e.g. ChaseDunn, Kawana, & Brewer, 2000; Tomich, 2000; Robinson, 2001). Most of

Matter, Space, Time, and Technology

25

these authors ignore both the materiality and the locality of production. They assume the global as their point of departure, and attempt to incorporate the local into it. In this chapter, we aim to reverse that logic. We will take into account and theorize the interaction of natural and of social processes. In other words, we will integrate ecologic or materio-spatial logic with sociologic within the economic logic of global markets. Fortunately, there are successful examples of such logic to guide us. Some earlier scholars – perhaps because they wrote before the distance between extraction and production became as great as it is today – offered far more coherent considerations of how society, space, and matter interact in the creation of political economy. We can use their perceptions to sharpen our own, even though their work predates our current concerns with globalization. To accustom ourselves to thinking about the dynamic relations between space, matter, technology, time, trade, and globalization, we will examine selected earlier analyses of how materio-spatial conﬁgurations structured the extraction and transport of raw materials and of how new technologies changed those structures. We will pay close attention to the empirical and to the theoretical construction of analyses of social and ecological production as material process within differentiated spaces. We have chosen authors who analyzed the ways that specific materio-spatial conﬁgurations structured human social, economic, and political activities and organization. We will search in their work for specific theoretical and methodological tools that we can use to build our analysis from local detail into global system. First, though, to assist us in this task, we will try to think through a materiospatial logic that can account for the ways that local production structures and organizes the global economy.

2. MATTER, SPACE, AND THE LOGIC OF PRODUCTION All production is profoundly material and local. Understanding the expansion and intensiﬁcation of the social, material, and spatial relations of capitalism that have created and that sustain the growth of the world economy requires analysis of how material processes of natural and of social production interact with each other and with space. Such analysis necessarily encompasses the ways that technological, social, organizational, and

26

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

political innovations change these interactions over historical time. This means that we will have to examine how multiple heterogeneous, spatially distinct natural systems produce the diversity of material forms that constitute an increasingly homogeneous global economy. We will thus see space as materially differentiated and matter as spatially differentiated by topography, geology, hydrology, and climate, as well as by absolute and geographic distance between places. Many new technologies have used new and different combinations of raw materials, so this examination will also require us to consider historical changes in the ways that technology and social organization respond to and mediate between matter, space, and human purpose. In other words, we will look at how historic changes of technology have changed the relation of society to nature. In the process, we will see how unequally these relations affect different societies and their material economies. As capitalist economies of scale expand, development and implementation of new technologies become increasingly costly. The circuits of capital that initiate, and then profit from, these technologies expand and accelerate as the scale of technology increases. This means that ﬁnance and technology drive each other toward globalization. We therefore need to construct a model of how technology and ﬁnance respond to material and spatial constraints as these affect capitalist competition for trade dominance. We will ﬁrst use this model to inquire why world systems and globalization analysis, and the modern social sciences generally, fail to consider the effects of space as it impinges on the material processes around which social actors organize economy and polity. We will next use the model to show that adequate explanations of globalization must account for how capitalist competition expands and intensiﬁes material processes of production, and how space affects material and economic process.

3. MATTER, SPACE, TECHNOLOGY, AND TRADE: BUILDING A MODEL OF GLOBALIZATION Industrial production has accelerated consumption of raw materials and thus increased the absolute space across which ever larger volumes and more numerous types of matter are transported. Globalization – identiﬁed as a recent or novel phenomenon by Sklair (2000) and others – is in fact the latest and largest of successive and cumulative phases of materio-spatial expansion and intensiﬁcation that have emerged from technologically achieved

Matter, Space, Time, and Technology

27

economies of scale adopted by ﬁrms, sectors, and states in nations competing to dominate world trade. Within nations competing for world-trade dominance, ﬁrms and sectors often collaborate technically, ﬁnancially, and politically with other domestic ﬁrms and sectors and with the state to develop and implement these technologies – even though they compete with each other to capture greater market share and higher profits. This collaboration generates dramatic episodic increases in economies of scale in industrial production, in raw material extraction, and in transport. Economies of scale expand, intensify, and diversify the consumption of materials needed to produce greater quantities of a larger variety of commodities. Constructing the larger, faster, stronger machines and infrastructure for the technologies that make these economies of scale possible in the ﬁrst place also increases and diversiﬁes the consumption of greater volumes of more different kinds of raw materials. The diversiﬁcation and precise speciﬁcation of the types of material that scale-enhancing technologies require has accelerated dramatically since thermally powered engines – progressing from steam pressure to internal combustion of petroleum to nuclear reactions of uranium – have replaced animals, wind, and water as the prime motor force in industry. All of these engines produce and must contain pressure and heat; the more powerful the engine, the greater the heat and pressure. Resistance to intense heat and to heavy pressures requires alloys of specialty metals naturally available in only a few places, so the material requirements of the new technologies drive searches for sources to more distant sites. These material requirements intensify as larger, stronger, faster machines expand technologically achieved economies of scale and speed. Humans must discover new ways of combining and processing the physical and chemical properties of different raw materials in order to achieve the additional strength, ﬂexibility, and resistance to impact and to weight for the larger, stronger, faster machines, vessels, and vehicles that make the expanded economies of scale possible. Expansion, intensiﬁcation, diversiﬁcation, and more precise speciﬁcation combine to make new technologies increasingly dependent on more, larger, uniformly higher-grade deposits of raw materials across broader spaces. This is not a new phenomenon. Trade relies on transport; the complex interplay of matter, space, and technology is as old as competition to dominate trade. Expanding the size of sailing ships to cheapen freight costs in the 17th century required larger, stronger, longer, appropriately shaped curved oak for the keel; longer, stronger straight oak for planking; and larger,

28

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

stronger, lighter pine for masts able to sustain the additional wind pressure on their much broader sails. Large, high-quality pines provided reach and strength without adding the extra weight above decks that would make the ship more prone to capsize. To gain speed and maneuverability without sacriﬁcing strength and cargo space, Dutch boat builders learned to form the hull from a single sheet of oak planks – precisely cut, then joined and caulked around a skeleton frame of much thicker timbers. This technology required a wide variety of precisely speciﬁed shapes and sizes of different types of wood, together with particular kinds of tar for caulking and long, strong ﬁbers of ﬂax for making the sails. All of these materials grew in different parts of the world, so the new technologies that supported Dutch competition in world trade required the complex coordination of differentiated matter over broad space. Similarly, the Japanese engineers who designed and built the supertankers and the enormous dry-bulk carries that enabled Japan to challenge U.S. domination of world trade at the end of the 20th century relied on large, cheap supplies of steel of much higher tensile strength and ﬂexion than could have been mass-produced 50 years earlier. Producing large volumes of uniformly high-grade steel cheaply enough to enable Japanese shipyards to capture over half of the world market required access to huge deposits of iron and coal of precisely speciﬁed purity, grade, carbon and sulfur content, hardness, and moisture content. As with wood, ﬂax, and tar, these material features of iron and coal, as well as the other mineral inputs for high-tensile steel, were naturally produced in different parts of the world. Engines to move these huge ships require equally high grades of matter. Similarly, engines and bodies of supersonic jet-powered aircraft must resist far more heat and stress than DC-3s or B-52s were designed to withstand at the middle of the 20th century. Cobalt and titanium, necessary for supersonic engines, are extremely rare in nature. Capital has had to seek them in remote and difﬁcult places. National states’ dependence on the stronger tax base and the military power that would be impossible without these raw materials makes them ready and anxious to assist ﬁnancially, diplomatically, and militarily in this search. Matter and space are naturally given. Technologies that mediate between them and human actions and goals are socially created, but they can only achieve the human goals for which society invents and ﬁnances them if they conform to the natural – biological, geological, locational, physical, and chemical – features of the raw materials they transform. As human economies expand their political domains, from communities to nations to continental unions, competition between these units to dominate trade – ﬁrst

Matter, Space, Time, and Technology

29

locally, then regionally, nationally, and ﬁnally globally – drives searches for technologies of production and transport that enable increased economies of scale. The technologies that make these increases in scale possible only work with particular material forms of appropriate quality and quantity naturally produced in particular places. Expanding and diversifying raw material supplies requires discovering and exploiting additional, larger, and more precisely uniform sources. This requirement progressively widens raw materials markets in space. Economies of scale in transport make it economically viable to bring large volumes of raw material across great distances. Expanded industrial production, raw material location, and transport technology have thus combined over centuries of capitalist competition to drive the world economy toward globalization. Globalizing technologies only work, though, if their design and operation are adapted to the physical, chemical, biological, and geological features and reactions of the raw materials of which they are constructed and to which they are applied. Many of the most important technical improvements have depended on discovering and extracting from large, uniform deposits of iron, coal, and manganese with specific chemical and physical features. Different coal burns at different temperatures, and minute variations in quantities of sulfur, carbon, moisture, or other impurities can make steel more or less likely to bend, twist, or break under heavier loads, forces, and pressures. Iron formed in natural pellets requires less fuel and is thus far cheaper to smelt, and allows much more regular control of heat, than does large, irregular chunks. Such features of raw materials are locally, and naturally, produced. They form the material bases and the spatial context for the ﬁnancial and political competitions and collaborations that have driven globalization, not just over the past two centuries of competition to improve iron- and steel-based technology, but through the earlier centuries when competition to dominate world trade required scale-enhancing technological innovations in using wood, sail, and wind for production and transport. An economy of scale reduces the cost of production for each unit produced, but increases the total capital sunk in infrastructure and technology. The only way for capital to recover the hugely increased investment in new, expanded machines and infrastructure is to operate them as close to full capacity and as cheaply as possible. This requires access to cheap, stable sources of large volumes of raw materials of uniform high quality. Large, high-grade, uniform deposits of speciﬁed physical and chemical composition are rare in most materials, and broadly scattered across the globe. The

30

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

economically and technically viable sources closest to industrial centers tend to be exploited, and thus depleted or overharvested, ﬁrst. As expanded economies of scale progressively expand and diversify raw material consumption, the savings available from larger, higher-grade, deposits become more important and the number of potential sources of the necessary size and grade available in the world becomes smaller. To summarize, in production as in transport, and with relatively few exceptions as capitalist economies have expanded, both the larger, stronger machines and the commodities they mass-produce require higher grade raw materials with more precisely speciﬁed chemical and physical properties. Volumes produced and sold expand, but so do the volume and value of capital invested in plant and machinery and in maintaining them. As the scale and speed of processing and production increases, adjusting larger machines for variations in purity, grade, and physical composition of raw materials becomes increasingly costly – both in the technical operation itself and in the production lost during the time the machine is down. Capital therefore seeks large, uniformly high-grade deposits of many different materials. The larger and purer the deposits, the more broadly distributed they are in space. Extractive economies thus become more dispersed, while productive economies become more agglomerated. In brief, this means that the natural logic of matter as it is produced in space creates a contradiction with the social logic of capitalist economies of scale. The contradiction is resolved, and then reiterated on a larger scale, with each cycle of technological innovation in expanded plant and machines. These material and technical dynamics combine to assure that expanded consumption of matter requires transport of ever greater volumes from the diverse locations where different material forms are produced, so material intensiﬁcation, and the economies of scale that drive it, require spatial expansion, and increased total volume and cost, of extraction and transport. By simultaneously reducing the unit costs of production and increasing the amount produced, technological and organizational economies of scale increase market share and profits for the innovating ﬁrm and national economy and expand their reproduction of capital.1 The increased absolute space across which the increasing volumes of matter consumed must be transported, however, has an opposing effect – it raises unit costs, reducing profits and market share. Seeking to resolve the contradiction between economies of scale and the cost of space, capitalists and agencies of capitalist states develop and implement cost-reducing, scale-dependent innovations in the technologies, infrastructures, and social organization of

Matter, Space, Time, and Technology

31

transport. Such innovations both cheapen access to raw materials and increase their available volume and variety, thus enabling a new cycle of even greater economies of scale in production. These then reproduce the need to procure more raw materials at even greater distance, thus reiterating and expanding previous cycles. Each such reiteration surpasses in scale and scope the material and spatial problem and the technological solutions of the previous cycle. Each cyclical reiteration of these processes (1) expands the space in which national ﬁrms procure raw materials, and thereby (2) stimulates development of stronger, faster, more capacious technologies – ships, trains, and loading equipment – and larger, more extensive infrastructure – ports, rail-lines, and warehouses – for their handling and transport, which also (3) reduce the cost of exporting products and so expand the space in which each national economy can competitively market its products and, thus, (4) advance the globalization of the world economy while expanding its reproduction of capital. Globalization can therefore best be understood by comparative historical analysis of the discrete, sequentially expanding cycles of contradiction and resolution between economies of scale and diseconomies of space. In each cycle, the technological innovations that create, and then the innovations that resolve this contradiction must be adapted to the local conﬁgurations of the matter they incorporate and transform, to the space from which the matter is extracted, and to the space between the location of extraction and the location of transformation. The interaction between the socially designed technologies used and the naturally produced matter and space on which they are used on shape the opportunities and the constraints that mold local systems of extraction, transport, and production, even as these feed into global systems of trade and ﬁnance.2

4. HOW MODERN SOCIAL SCIENTIFIC ANALYSIS NEGLECTS SPACE AND NATURE In order to explain how capitalist economies intensify material processes across ever broader spaces, we need ﬁrst to correct social science’s failure to theorize matter and space. World-systems analysts (see, e.g., Chase-Dunn, 1989; Wallerstein, 1974) use spatial terms – core, periphery, world, global – as a conceptually central organizing metaphor, but they seldom elaborate the most fundamental mechanisms and processes that expand and intensify the social and material relations of capitalism to the entire globe. This

32

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

oversight is peculiar, given the strong explanations in economic theories, both Marxian and classical, of mechanisms that favor agglomeration or spatial concentration of industrial production. Indeed, central-place theory is founded on a clear understanding that, if economies were not encumbered by the ways that human actors and their labors occupy space, they could operate most efﬁciently on a single point in space. Explaining the anomaly that the accelerating dispersion of integrated production and commerce around the globe directly contradicts the ﬁnancial and political preferences of capital should contribute importantly to our understanding of globalization. We just saw how the expanded reproduction of capital creates the need for greater volumes of a greater variety of raw materials. Varied topographic, hydrological, geological, and atmospheric conditions are needed to produce all of these different types of matter. The world system is not simply a division of labor in space, but a division of production and extraction within materially differentiated space. It is precisely the search for larger, higher-grade volumes of more different types of matter that creates a material imperative to disperse the world economy ever more broadly across the globe, simultaneously and contrary to the socio-economic imperative to agglomerate the world economy in a minimal space to reduce transaction and communication costs. All of this matter is naturally produced in distinct local places. Space is simultaneously a means and a condition of its production and an obstacle to or cost for its transport. The vast array of raw materials consumed in expanding industrial production is therefore dispersed among multiple ecologically different locations that provide the natural means and conditions of their production. Social sciences should be able to theorize the complex ways that society and economy adapt to and exploit the constraints and the opportunities presented by the varied materio-spatial conﬁgurations on which they depend. Social scientists should be able to account for how capitalist states and societies devise laws, institutions, organizational forms, and technologies in response to all three of the ways that space impinges on both natural and social production. Space can provide a means of production, as, for example, the amount of land needed to produce a bushel of corn or a yard of timber. Space may, however, be simply a condition of production, such as the amount of room a worker needs to perform a task with free muscular movements and safety. Finally, space as distance between the source of the material, the locus of transformation, and the points of eventual sale of the commodity becomes an obstacle to production and exchange.

Matter, Space, Time, and Technology

33

The coordination of ﬁrm and state responses to these three forms of space becomes increasingly complex as the world economy becomes more global. Space, both as means and as condition of production, requires increasingly complex organization and engenders increasingly violent struggle as globalization of raw material markets progresses. At the same time, the technologies to reduce the costs of space between the sites of extraction and of transformation require greater investments. Globalizing capital is increasingly constrained to coordinate the technologies it uses in the local spaces to which it controls access with the technologies for reducing the cost of transport across the global space between extractive and productive economies. The scale of investment forces capital to coordinate with increasingly powerful ﬁnancial institutions, and the intensity of conﬂict forces capital to coordinate with increasingly militarized states, both on the local and on the international level. The different dynamics of extraction and of production make industrial economies progressively stronger, and extractive economies progressively weaker – ﬁnancially, politically, diplomatically, and militarily. Industrial economies expand materially while agglomerating spatially.3 Accelerated depletion of the most proximate sources of each type of raw material enhances the need for new sources. The advantages of spatial agglomeration centralize and concentrate industry in space while the dynamics of depletion disperse raw material extraction in space. Expanding industrial economies must therefore procure raw materials across ever greater spaces and distances. The increased costs of transporting raw materials over greater distances stimulate collaborative efforts of states, ﬁrms, and ﬁnancial institutions to develop technological and infrastructural means to (1) cheapen bulk transport and (2) reduce material inputs required for each unit produced. The successive campaigns and strategies of nations striving to dominate world trade – the Portuguese, the Dutch, the British, the United States, and the Japanese – have each superceded the then-established capacities to procure and transport cheaply over great distance the raw materials they used in greatest volume. The cumulative effects of these technological and infrastructural increases have progressively globalized the world economy, while progressively subordinating extractive to productive economies and accelerating the depletion and impoverishment of extracted eco-systems. Technological and social organizational innovations to reduce the inputs needed per unit of production and to reduce the unit costs of transport4 thus reiterate (cf. Haydu, 1998) and surpass (cf. Arrighi, 1994) the sequentially cumulative power and scale of earlier solutions to the contradiction between scale and space. Each solution depends on expanded technical economies of scale, not only in production, but also in extraction and in transport. Each

34

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

scale increase accelerates the depletion of proximate sources of raw material and drives the procurement of new sources at greater distance. Each such episode thus perpetuates and exacerbates the cycle of contradiction between economies of scale and the cost of space. Each such episode also extends larger, more scale-dependent and more capital-intensive extractive and transport technologies across more national boundaries into territories less politically, culturally, and economically incorporated into the world system. Spatial expansion to solve the increasing need for raw materials thus stimulates ﬁrms, states, and ﬁnancial institutions in the ascendant economy to implement strategies to secure supplies and protect investments in a broadening set of territories under diverse systems of political and economic organization and sovereignty. These strategies have varied over time with the geopolitical possibilities of the world system. As the world economy has expanded spatially and intensiﬁed materially, ﬁrms and states in the core nations competing to dominate trade have had to devise new, more effective systems of governance and control to keep their access to raw materials cheap and secure and to control the spaces in which they are produced. Over time, these systems progressed from chartered companies defending their politically granted monopoly rights from fortiﬁed trading posts, through formal colonial conquest and administration, to imperial regimes of free trade enforced unequally on different nations and colonies. After World War II, the United States displaced Europe’s colonial system, promoting instead formally independent regimes of free trade and foreign direct investment sustained by multilateral ﬁnancial agencies like the World Bank and the IMF. Since 1982, the U.S. has pushed the World Bank and the IMF to institute the contemporary system of debt-coerced free trade under structural adjustment programs. In the process, natural-resource rents and prices have declined, while extractive economies have incurred huge debts to pay for the expanded transport infrastructure that global raw materials markets require. The contradiction between scale and space is constant, but each solution to it is rooted in the intersection of (1) geography, demography, and political and ﬁnancial organization, both of the economically ascendant nation and of the nations that supply its raw materials, and (2) the technological innovations that drive the ascendant economy’s growth, with (3) the materiospatial characteristics of the raw materials that these new technologies require. The reiterated solutions to this contradiction can only succeed in each instance if they take topographic, spatial, and material properties of local production into account.

Matter, Space, Time, and Technology

35

The scale of the contradiction increases with each systemic cycle of accumulation (cf. Arrighi, 1994). Each solution therefore tends to be scale dependent (Bunker & Ciccantell, 2001). The reiterated solutions are thus sequentially cumulative, that is, each separate solution is distinct in the characteristics and historic timing of its technological innovations and in the types and sources of the raw materials that these innovations incorporate and transform, but the scale of technology and the distance to raw materials accumulate over the sequence of solutions. The reiterated, cumulatively sequential, but materio-spatially distinct, solutions to the contradiction between scale and space constitute a central mechanism behind the expansion and intensiﬁcation of the world system. In other words, the local – as the naturally and socially heterogeneous collection of specific extractive economies structured by significant material resources – significantly drives the forms of the global as the world economy moves toward homogeneity. As the distance between extraction and production increases, the natural features of space and matter lose concreteness and presence in the awareness of intellectuals who live and work in core societies. This may explain why so many analysts of the world system and of globalization have overlooked the contradiction between scale and space and have thus ignored its dynamic consequences. In this they follow more general trends in the social sciences. Over the past half-century, social scientists have tended to consider space as the passive context, container, or boundary of social organization and activity. Space conceived thus is neither analyzed nor theorized. To the extent that modern geographers attempt to analyze or theorize space, they tend to emphasize the social construction of space. Neil Smith (1984), for example, incorporates Schmidt’s notions of second nature into Harvey’s (1983) discussions of the built environment to declare that the dynamic of capital leads to a social reconstruction of nature, including space. He interprets Marx’s statement, that time annihilates space, within a generalized vision that capital recreates nature and time. Smith’s subsequent critique of renewed attention to the early Marxist writings of Wittfogel (1985[1929]) betrays the extent to which geography’s unhappy battle against the excesses of environmental determinism and geopolitical analysis continues to constrain radical geographers’ attention to the natural characteristics and effects of space on society. Lefevbre’s (1991) and Soja’s (1989) oversocialized views of space manifest a similarly dogmatic rejection of any consideration of what Smith calls absolute, or natural, space. Even though David Harvey (1983) addresses the material characteristics of space when he elaborates Marx’s theories of differential rent, his explanations for the incorporation of new space into economic expansion rest completely on the

36

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

ﬁnancial logic of the overaccumulation and site-specific devaluation of capital, thus ignoring the lessons about material, space, and the cost of distance that he could have learned from Marx (1967) and from von Thunen (1966).

5. HOW EARLIER ANALYSES OF LOCAL MATERIOSPATIAL CONFIGURATIONS CAN ENHANCE CONTEMPORARY ANALYSES OF GLOBALIZATION Earlier social scientific traditions, still inﬂuenced by the more proximate articulation of extraction and production in the early stages of industrialization, took the material effects of space on economy far more seriously. In this section, we review some of the richest analyses of how local conﬁgurations of space and matter molded the ways that particular regions were incorporated into world capitalist economies. Then we will consider how we can expand and update the concepts and methods developed in these earlier analyses to assist us in our present attempts to explain how matter, space, and technology have shaped globalization. Wittfogel (1985[1929]) abstracted from a broad range of Marx’s writing about the extraction and transformation of coal, iron, timber, and wheat a powerful demonstration that, as science and technology advance, industrial capitalism becomes more, not less, dependent on nature. Wittfogel was not alone in his attention to the materio-spatial logic of hydrology and topography, though his exegesis of Marx’s materio-spatial analyses of technological developments remains unsurpassed. In the ﬁrst half of the last century, descriptive economic historians such as Richard Albion (1926) and Harold Innis (1933, 1956) analyzed specific local extractive economies by (1) the material attributes of topography, geology, hydrology, and climate that produced the resource extracted, (2) the topographic and hydrological characteristics of the space between the locus of extraction and the locus of consumption as these characteristics affected the technology and the cost of transport across this distance, and (3) the nature and purpose of the production, transport, and military technologies that create demand for the specific resource extracted. Innis and Albion both used materiospatial analyses in ways that contributed to explanations of how and why efforts of dominant groups in Europe structured the conquest, settlement, exploitation, and defense of the territories they incorporated into their separate national imperialist systems. Though their studies focused on particular binary relations between specific local economies – Albion on

Matter, Space, Time, and Technology

37

materio-spatial dynamics in Britain’s use of colonies and free trade to procure timber and metal for the imperial navy, and Innis on how French, British, and ﬁnally U.S. needs for raw materials inaugurated Canada’s incorporation into the world economy and then continued to mold its economic and political development – these authors elaborated explanatory mechanisms that we may ﬁnd very helpful in our explorations of how local material and spatial features have driven capitalism’s historical progress toward globalization. For Albion (1926), both shipbuilding and metallurgy in Britain were constrained by the location, accessibility, and transport cost of timber.5 The course of navigable rivers and the location of soils and climates conducive to the growth of different tree species determined the location and organization of these economically and politically critical industries. As growing demand for iron and for ships overshot the domestic supply of oak for charcoal and for timbers, these same considerations inﬂuenced the imperial strategies of a British state enormously sensitive to Admiralty needs for ships and cannon. The kinds of trees available in different parts of the world and their location in relation to navigable rivers and ports directly inﬂuenced British imperial interests, administrative strategies, and commercial behavior in each area. Albion (1926) showed that the material features – masts, keel, hull planks, rudder, or caulk and tar – of different kinds of ship required the strengths, ﬂexibilities, shapes, saps, and sizes particular to various species of tree that grew in specific climates, soils, and elevations. Albion thus related technological requirements to materio-spatial details in order to explain the geography of British imperial expansion and trade relations, as well as the organization and relations of labor, transport, and exchange in the zones incorporated into Britain’s raw material periphery. Though he was specifically concerned to explain how shortages of timber supply affected British imperial strategies and policies, the techniques he devised to analyze how material and spatial features of forests in particular places shaped the political and economic behaviors of British imperial control can be extended into analysis of how material and spatial features of other sources of natural resources have affected each of the steps the world economy has taken toward globalization. Innis (1956), like Albion, was concerned with a specific country, though his focus was on Canada, a peripheral source of raw materials, rather than on England, the core of imperial conquest and control. Like Albion, Innis elaborated conceptual and methodological tools for his analysis of one nation that we can use in our broader efforts to elaborate a comparative history of globalization.

38

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

For Innis (1956), the incorporation of Canadian space into the world economy was shaped by the material needs of the British, French, and Spanish as they struggled for economic expansion and military power. Ethnically and nationally varied business and immigrant groups moved into the spaces whose topographies most effectively and easily provided defensible access to and transport of naturally produced material forms demanded by these competing European economies. After they expelled their French competitors, the British shaped Canada’s local extractive, productive, and transport capacities – and its political and ﬁnancial institutions – to satisfy their own needs for foodstuffs, consumer goods, industrial inputs, and military material. The natural processes that transformed matter and energy in distinctive – topographically, hydrologically, geologically, and climatologically structured – spaces into ﬁsh, beaver, different kinds of trees, and different metals and fuels, together with the ways that river systems and land forms interacting with transport technologies created the cost of their extraction and export, all conditioned by the military and economic competition to control these spaces and the material in them – ﬁrst between the European nations and then between England and the U.S. – molded Canada’s social, economic, ﬁnancial, political, and demographic organization in ways that persist today. For Innis (1956), space as a differentiated condition of natural production – combined with space as an obstacle to exchange or export – determined the material composition and the location of different extractive economies. Naturally occurring spatial and material features set the parameters within which socially constructed technologies, markets, and geo-political forces determined the organization of labor, the settlement patterns, and demographic characteristics of different populations, the composition of capital, the infrastructure in which capital was invested, and the organization and structure of the Canadian state. The natural course, ﬂow, and size of navigable rivers presented obstacles to and opportunities for extracting and exporting staples to world markets. Social organization and technological innovation created and adapted various means – vehicles and infrastructure – of transport. These mediated between space as a natural obstacle to access and exchange and rivers as an avenue of cheap access and export. Space, and the topographic and climatological characteristics that determined the type and abundance of material production in that space, set the parameters within which available capital and technology could be effectively and profitably deployed to create transport systems. These systems in turn determined the utility of the resources available and the cost of their extraction and export. They, thus determined not only the structure, cost,

Matter, Space, Time, and Technology

39

and profitability of each separate extractive export economy, but also national, social, and political structure. State’s and society’s obligations to pay for these systems affected ﬁnancial, economic, and political organization of the Canadian nation in enduring ways. Like Albion, Innis (1956) examined interactions between materio-spatial features of both the European core and the North American periphery with the technologies of extraction and transport available for adapting to these features in order to supply European demand for locally produced raw materials. He then explained how these interactions determined patterns of settlement and urbanization, exploitation of natural resources, accumulation and deployment of capital, and the formation of the state in the newly incorporated peripheries. The following section abstracts the kinds of materio-spatial data and the kinds of economic historical logic Innis used to explain how and why Canada’s political economy took the shape it did.

6. INNIS’S MATERIO-SPATIAL EXPLANATIONS OF CANADA’S ECONOMIC HISTORY Britain lacked France’s access to cheap salt, and so was obliged to settle and defend land-spaces to dry cod caught on the Grand Banks while her rival could salt the ﬁsh without landing. This disadvantage in ﬁshing later gave Britain an advantage in moving across the Pre-Cambrian shield surrounding the St. Lawrence. The British ﬁrst used this river for access to the Artic zones. There, cold temperatures and thin human populations created the best environments for the ﬁne thick furs favored for elegant comfort in European winters. Trade routes, transport and warehousing infrastructure, military posts, and settlements developed to support the high-value, luxury commerce of beaver pelts. Locations, sizes, and purposes of these social creations and activities were directly shaped by the dendritic patterns and hydro-logic of the river systems. Forts, docks, warehouses, and eventually urban centers, for example, were built around the afﬂuences of tributary rivers into larger streams. These were the transport nodes for making- and breaking-bulk. Cargos were shifted from or to the canoes that could navigate streams, to or from the rafts and pirogues that could handle tributary rivers, and then similarly to and from the small boats that could manage the main water courses – at least between rapids, and ﬁnally the larger sailing ships that could manage the high seas while providing much lower unit costs for carrying cargo. The infrastructure and population of these settlements

40

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

later facilitated the inﬂux of capital and labor for extraction of the timber required by the expanding military and commercial transport ﬂeet in Britain, and then for the wheat economy that employed loggers off-season to supply cheap food to Britain. The initial advantage that the British acquired in the control of the St. Lawrence River meant that the British traders paid higher prices for beaver than the French could afford. When a French company penetrated Hudson Bay and established that route to Europe as a shorter distance accessible to larger ships than the St. Lawrence provided, the British were able to co-opt the French agents into the Company’s supply networks and thus control this more economical route to markets. The shorter trip and larger ships significantly reduced the cost of raw materials in England. The accumulating materio-spatial advantages contributed to Britain’s eventual expulsion of the French from Canada. The shift to Hudson Bay favored the British economy, but led to significant economic stagnation of the established commercial centers along the St. Lawrence until the national state – under pressure from the wheat farmers initially brought in by the cheap fares, open lands, and seasonal wages made available by the lumber trade – supported the construction of canals and locks to improve, and rail lines to supplement the river’s natural channels. Innis’s (1956) comparative analysis of beaver, ﬁsh, timber, and wheat showed how the physical features of the natural resource and the topography of its natural location interacted with (1) the military and economic strategies and conditions of rival European nations, (2) available technologies of extraction, transport, and production, and (3) the character of the market for each raw material into mold settlement and commercial patterns, rates and locations of capital accumulation, and social, economic, and political organization. The value-to-volume ratio of the raw material exported determined the distribution of costs and profits along the localized nodes of its trade. The physical features of the raw material and of its sources in nature also determined the intensity and organization of labor, its demographic characteristics and location, and the costs and requirements of its reproduction. The different labor requirements, combined with the different value-tovolume ratios, determined the proportions and balance between the volumes of inbound provisions and outbound products for each of these economies. The beaver pelts sent down river and on to Europe, for example, occupied far less cargo space than the goods and provisions sent from Europe for trade with the Indians that trapped the furs. The value of beaver pelts had to be high enough to compensate for the transport costs incurred in return

Matter, Space, Time, and Technology

41

shipping – ﬁrst overseas and then upriver – the far greater volume of provisions trappers and traders would consume while gathering a shipment of beaver. Logs and wheat, both much lower value-to-volume goods reversed this proportion of inbound to outbound cargo. The ratio of imports to exports determined the relative cost of freight or of passage on the inbound and outbound trip. Extra cargo space on a boat from Europe provisioning the beaver trade from Europe was very costly; passage on a timber boat returning relatively empty to Canada was far cheaper. These differential cost ratios directly affected the rate and destination of immigration, and thence settlement patterns and the cost and availability of labor. These in turn determined the cost of imported relative to locally produced goods, and thus the chances of establishing local productive economies. The scale, type, and cost of transport technology responded to the material characteristics of the goods transported and to the size and topography of the rivers transited. City location reﬂected the logic of make- and break-bulk locations determined by the volume of matter and the relative size and navigability of rivers and by the dendritic patterns formed as tributary rivers joined to form larger waterways. Innis’s analysis of migration, demographical patterns, labor relations, urban location and organization, and state formation demonstrate that his theorization of matter and space did not preempt, but rather facilitated and extended, his analysis of social and economic processes. By integrating social and natural mechanisms, he could account for such temporally and spatially consequential phenomena as the changing size and material demands of the world economy, the political and economic dimensions of empire and colonial resistance, or the role of conﬂicts in the core, ﬁrst between Britain, France, and Spain, and later between the United States and Europe. Incorporation of materio-spatial data enriched his analysis of the economic, ﬁnancial, and political trajectory of the Canadian nation and state. We incorporate Innis’s approach to data and to comparative historical logic into our analysis of how the specific materio-spatial features of other locations, as they interact with available technologies and with contemporary competitions for military and trade dominance, determine the form and extent of that area’s participation in the world economy and thus contribute to the shaping of global formations. Some recent authors do provide empirical and theoretical accounts of how new technologies to cheapen bulk transport have changed the economic and political relations between world regions, but even the most precise of them lack Innis’s close-grained analysis of the material and economic effects

42

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

of space on specific commodities in particular places. Mandel’s (1975) discussion of how technological innovations in transport reduced the natural tariff barriers of space shows considerable sensitivity to some of the physical attributes of space as an obstacle to exchange, but misses how the heterogeneity of matter differentiates space. Douglas North (1958), and later O’Rourke and Williamson (1999), in their specific consideration of technical change in transport and its effect on world trade, attend to space as the naturally autonomous locus of socially produced technology and infrastructure, but do not deal with the natural productivity of matter in space. Their works thus do not achieve explanations of the interactions between space as condition of production – and therefore determinant of the kinds of economic activity apt for profitable exploitation – and space as obstacle to exchange – and therefore determinant of the location of materially bound human activity. They, thus cannot achieve the dynamic dialectic that characterized Innis’s or Albion’s work, and that Wittfogel found in Marx’s analysis of naturally produced use values.

7. CONCLUSION: RAW MATERIALS, ECONOMIC ASCENT, AND UNDERDEVELOPMENT IN THE PRODUCTION OF GLOBALIZATION We have shown in this chapter how the long-term process of globalization is fundamentally rooted in the efforts of economically ascendant states and ﬁrms to resolve the contradiction between economies of scale and diseconomies of space. These social, organizational, political, technological, and economic efforts were profoundly challenged by and, in the most spectacular cases of economic ascent, responded to and took into account material and local particularities in increasingly socially remote extractive regions. In a few cases, these strategies were spectacularly successful for ascendant economies in Holland, Great Britain, the U.S., and Japan. For the particular extractive regions that became part of these ascendant economies’ raw materials peripheries, these strategies shaped social, political, and economic life via underdevelopment that harnessed places like Canada, the Amazon, and other extractive peripheries as sites of extraction to supply industrialization and consumption in the ascendant economies. This underdevelopment included massive subsidization and loss of rents from raw materials exports. In other research, we have shown that, as the result of becoming part of Japan’s coal and iron ore peripheries over the

Matter, Space, Time, and Technology

43

past 50 years, Australia, Brazil, and Canada have sacriﬁced tens of billions of dollars in rent and provided billions of dollars in subsidies for the extraction and transport of raw materials that provided a critical ingredient for Japan’s international trade competitiveness and its economic ascent. Our analysis of globalization presented in this chapter allows us not simply to understand the relationship between society and nature, but also to understand the social division of the costs and beneﬁts of the process of globalization between ascendant economies and their raw materials peripheries.

NOTES 1. Capital reproduces itself when the money paid for a commodity produced by labor and raw materials that capital has purchased is invested in more labor and raw material to produce another commodity for sale. Expanded reproduction occurs when the commodity is sold for more than the cost of the labor and raw materials that comprise it and part of the profit plus the original amount of capital is invested in labor and raw materials. Expanded reproduction of capital includes accumulation of capital if some part of the profit is invested in technologies that endure for more than one cycle of production and exchange. 2. Even transport, the movement of matter through space, is at each moment as local as the location of the vehicle and of the capital sunk in the built environment at each point in the space traversed. 3. This remains true even though the internal combustion machine and the diffusion of communication technologies has cheapened the cost of distance enough to allow increased spread of industry into suburban spaces. These are still agglomerated around heavy transport nodes – ports, railheads, and highways. 4. Unit costs of transport can be calculated by ton/mile, that is, weight/distance. 5. Albion notes, for instance, that the cost of animal traction limited the harvest of trees to an area of three miles or less distance from river’s edge.

REFERENCES Albion, R. (1926). Forests and sea power: The timber problem of the Royal Navy, 1652–1862. Cambridge: Harvard University Press. Arrighi, G. (1994). The long twentieth century. London: Verso. Bunker, S. G., & Ciccantell, P. (2001). International inequality in the age of globalization. Japanese economic ascent and the restructuring of the capitalist world-economy. Paper presented at the ISA meetings, Chicago. Chase-Dunn, C. (1989). Global formation: Structures of the world-economy. Cambridge: Basil Blackwell. Chase-Dunn, C., Kawano, Y., & Brewer, B. (2000). Trade globalization since 1795: Waves of integration in the world-system. American Sociological Review, 65, 77–95.

44

STEPHEN G. BUNKER AND PAUL S. CICCANTELL

Harvey, D. (1983). Limits to capital. Chicago: University of Chicago Press. Haydu, J. (1998). Making use of the past. American Journal of Sociology, 104(2), 339–371. Innis, H. A. (1933). Problems of staple production in Canada. Toronto: Ryerson Press. Innis, H. A. (1956). Problems in Canadian economic history. Toronto: University of Toronto Press. Lefebvre, H. (1991). The production of space. Oxford: Blackwell. Mandel, E. (1975). Late capitalism. London: New Left Books. Marx, K. (1967). Capital: A critique of political economy, (Vol. 3), New York: International Publishers. North, D. (1958). Ocean freight rates and economic development, 1750–1913. Journal of Economic History, 18(4). O’Rourke, K. H. O., & Williamson, J. (1999). Globalization and history: The evolution of a nineteenth-century atlantic economy. Cambridge, MA: MIT Press. Robinson, W. I. (2001). Transnational processes, development studies, and changing social hierarchies in the world system: A central American case study. Paper presented at ISA meetings, Chicago Sklair, L. (2000). The transnational capitalist class. Oxford: Blackwell Publishers. Smith, N. (1984). Uneven development. Oxford: Blackwell. Soja, E. (1989). Postmodern geographies. New York: Verso. Tomich, D. (2000). The world after the world system. Paper presented at the Social Science History Association, Pittsburgh, PA. von Thunen, J. (1966). Isolated state. Oxford: Pergamon Press. Wallerstein, I. (1974). The modern world-system. New York: Academic Press. Willis, S. (1991). A primer for daily life. London: Routledge. Wittfogel, K. (1985 [1929]). Geopolitics, geographical materialism, and Marxism. Antipode, 17(1), 21–72.

ENVIRONMENTAL SOCIOLOGY’S THEORETICAL AND EMPIRICAL PARADOXES Frederick H. Buttely ABSTRACT In this chapter, I want to take some stock of the subdiscipline of environmental sociology. I believe that a productive approach to restoring some of the coherence of environmental sociology is to conceive of mainstream environmental sociology as reflecting several paradoxes. The bulk of this chapter will be devoted to a brief explication of environmental sociology’s theoretical and empirical paradoxes. I will begin with three paradoxes that have played a major role in environmental sociology since the 1970s. However, many of the theoretical and empirical paradoxes of the subfield are relatively new ones – and some have not even been thought of as paradoxes. The thrust of the present chapter consists of something of a research agenda for environmental sociology for the next decade or so.

1. INTRODUCTION One of the hallmarks of the work of Stephen Bunker and his close associates is that it is refreshingly interdisciplinary, and that it transcends the Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 45–63 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10003-1

45

46

FREDERICK H. BUTTEL

entrenched subdisciplinary structure of sociology. Stephen Bunker’s work has long involved elements of world-systems theory, the sociology of development, economic sociology, political sociology, agrarian or peasant studies, and several other sociological specialties. Among these specialties that Bunker has drawn on – and contributed to – is the subdiscipline of environmental sociology. In this chapter, I would like to approach the future of this ﬁeld from the same eclectic and synthetic point of view that is represented in Bunker’s published works. Environmental sociology is nearly three decades old, having arisen in the aftermath of the mobilization of the US environmental movement in the late 1960s and early 1970s. Despite the relative youth of environmental sociology, it has changed enormously since its early years (see Buttel & Gijswijt, 2001; Buttel, Dickens, Dunlap, & Gijswijt, 2002; Dunlap, 1997; Foster, 1999 for overviews of the development of the ﬁeld). As I will mention in more detail below, while environmental sociology in the 1970s essentially focused on three topics – environmental values and the environmental movement, the sociology of rural resources and landscapes (especially forests, recreation habitats, mining, and ﬁsheries), and the ‘‘new human ecology’’ (a` la Schnaiberg, 1980; Catton & Dunlap, 1978) – the theoretical and empirical diversity of environmental sociology has increased enormously. The ﬁeld has, if anything, become so fragmented that it may be at risk of losing its coherence as a subdiscipline or research program. In this chapter, I want to take some stock of the subdiscipline of environmental sociology. I believe that a productive approach to restoring some of the coherence of environmental sociology is to conceive of mainstream environmental sociology as reﬂecting several paradoxes. The bulk of this chapter will be devoted to a brief explication of environmental sociology’s theoretical and empirical paradoxes. I will begin with three paradoxes that have played a major role in environmental sociology since the 1970s. However, many of the theoretical and empirical paradoxes of the subﬁeld are relatively new ones – and some have not even been thought of as paradoxes. Before discussing the theoretical and empirical paradoxes of environmental sociology, I will make a few observations about the nature of environmental sociology. Though this is perhaps a minority view, I would argue that environmental sociology is, at root, more of a research program than it is a coherent sociological subdiscipline. This is mainly due to the fact that there can be no a priori definition of environmental sociology’s subject matter, other than at a very abstract level – the study of societal–environmental relationships – that is not really very revealing or useful. Relations

Environmental Sociology’s Theoretical and Empirical Paradoxes

47

between environment and society are in an ultimate sense ubiquitous, ranging across levels of analysis, involving all institutional complexes (state, economy, educational institutions, family, religion), and ranging from the mundane (e.g., the inﬂuence of urban form on automobile fueled consumption) to the profound or fundamental (whether interstate disagreements over global climate protocols are likely to portend massive dislocations two or three generations hence due to global climate change). Environmental sociology’s coherence, to the degree it exists, is a general commitment to materialism and to a realist ontology, but even though this has tended to be the case there have been challenges to and critiques of environmental sociology’s materialist-realism over the past decade or so (see the overview in Buttel et al., 2002). But even if there was agreement within the ﬁeld on ontological posture, there is still enormous room for disagreement on empirical foci and theoretical approach. Environmental sociology is certain to exhibit a remarkable diversity of empirical foci and theoretical approaches because no single theory can pretend to account for, say, the place of nature in human identity in nineteenth century England and the reasons why international environmental regimes are threatened with longterm impotence and irrelevance at the turn of the twenty-ﬁrst century. I might add that the fact that environmental sociology is more a research program than a coherent subdiscipline is not undesirable. This state of affairs will enable environmental sociology to remain a dynamic ﬁeld open to new perspectives and approaches. This said, the thrust of the present chapter consists of something of a research agenda for environmental sociology for the next decade or so. In setting forth this agenda, I respect the fact that not all scholars in the ﬁeld will agree. My reasons for positing this agenda are essentially twofold. First, this agenda for environmental sociology reﬂects a bias toward stressing issues in ways that draw on the ‘‘central tendencies,’’ strengths, and tools of sociology at large. Second, the focal point of this agenda also reﬂects a bias toward focusing on environmental phenomena that are important in a macrosocial and material way to contemporary and future societies. The essence of this research agenda is to understand the politics of environment and resources, the role of the natural world in comparative-national and global development, and the nature and significance of environmental movements, activism, and countermovements. It is my argument that utilizing the most powerful tools of sociology to address the most critical issues relating to the way that groups and societies have related, and will continue to relate to the natural world will make for the most interesting and relevant research agenda for environmental sociology.

48

FREDERICK H. BUTTEL

2. THE ‘‘POLITICAL SOCIOLOGIZATION’’ OF SOCIOLOGY AND ENVIRONMENTAL SOCIOLOGY Pamela Oliver (2001), when she was chair of the American Sociological Association’s Section on Political Sociology, wrote a piece in the Section newsletter in which she wondered out loud whether political sociology was becoming the core subdiscipline in sociology. She noted, for example, that about half of the graduate students in the University of Wisconsin sociology graduate program elected political sociology as their principal area of specialization. Oliver’s speculations about the increasingly central role being played by political sociology have an interesting parallel in the ﬁeld of environmental sociology.1 Two of the points I will stress in this chapter are that, ﬁrst, there is a discernible trend underway toward the political sociologization of environmental sociology; that is to say, there is, in my view, a growing tendency for theory and research in environmental sociology to draw substantially from theories and perspectives in political sociology; and, second, I will suggest that the bulk of the key issues in environmental sociology can be addressed or resolved only if they are explored from a political–sociological perspective, or are framed in political–sociological terms. If I am correct about the growing political sociologization of environmental sociology, this argument should not be construed as a claim that environmental sociology is becoming more coherent. Indeed, it seems apparent that there has been a fragmentation of environmental sociology over the past decade or two (Buttel & Gijswijt, 2001). The growing anchor of environmental sociology in political sociology and the fragmentation of environmental sociology are by no means contradictory trends. There are two major reasons for the fragmentation of environmental sociology. One is that environmental sociologists have expanded the empirical scope of the ﬁeld. The list of environmental-sociology specialty areas is now enormous, consisting of numerous theoretically driven (e.g., ecological modernization, eco-Marxism, critical theories, actor-network analysis, treadmill of production theory, the new ecological paradigm, and post-materialism) and empirically driven areas (e.g., environmental justice, the environmental movement, the environmental state, international environmental regimes, environmental Kuznets analysis, sustainable development practice, sociology of outdoor recreation, and environmental values) of inquiry. A second basis of the fragmentation of environmental sociology is, in fact, its political sociologization, combined with the fact that political sociology has itself become increasingly wide-ranging and, as a result, more fragmented. There are now more than a dozen major theoretical traditions in

Environmental Sociology’s Theoretical and Empirical Paradoxes

49

contemporary political sociology. In addition to pluralism and ‘‘elite’’ or ‘‘power elite’’ theories (those of C. Wright Mills, Theodore Lowi, and Charles Lindblom), there have long been several distinct neo-Marxist theories (instrumentalism, structuralism, and theories associated with James O’Connor and Claus Offe that are based on theorizing the functions of the capitalist state). Neo-Weberian or state-centered theories (e.g., of Theda Skocpol) remain inﬂuential. In recent years there has been a particularly rapid development of cultural and postmodernist theories of power (such as Foucaultian and Habermasian theories, theories of the depoliticization of politics, governmentality theories, and theories of new social movements and transnational or global social movements). ‘‘Repertoires of action’’ perspectives such as those of Tilly, MacAdam, Tarrow, and Zald have led to a remarkable theoretical and empirical cross-fertilization between political sociology and the sociology of social movements. There are now strong traditions of neoinstitutionalism, typiﬁed by Peter Evans’ work on ‘‘embedded autonomy’’ and by civil society theories (e.g., of Robert Putnam). Rational choice theories (e.g., of Michael Hechter and Joel Rogers) have become more prevalent in political sociology, as have a wide variety of globalization theories (e.g., by Wallerstein, Robertson, Sklair, and Wapner).

3. THE PARADOXES OF EARLY AMERICAN ENVIRONMENTAL SOCIOLOGY One of the major points, I wish to make in this chapter is that a good share of the agenda for environmental sociology can be expressed in terms of recognizing and resolving several key paradoxes. It is the case, however, that environmental sociology has long rested on, or has been implicated in, theoretical and empirical paradoxes. Two paradoxes were particularly pivotal from the earliest years of American environmental sociology. One of these paradoxes – the matter of how it can be that there was growing social adherence to environmental symbols and growing concern about environmental issues from the 1960s onwards at the same time that these attitudes and orientations are essentially unrelated to individual or collective environmentally related behaviors – was central to American environmental sociology in the 1970s. Note that for about 15 years this paradox was really not recognized as such. There was well over a decade of research before it came to be recognized that (general) environmental attitudes bear only a small relationship to environmental

50

FREDERICK H. BUTTEL

behaviors. Nonetheless, this paradox was seemingly resolved by research showing that there is some association between environmental attitudes and behaviors when there is a high degree of speciﬁcity in the attitudes involved and in the attitude object. Thus, while there is a considerable literature showing that general environmental attitudes are very poor predictors of environmentally related behaviors (e.g., ‘‘environmental concern’’ is unrelated to membership in an environmental group or to household energy consumption/conservation), the literature has revealed that highly specific attitudes, such as those toward recycling or littering, have modest associations with highly specific attitudes’ objects, such as recycling programs or anti-littering campaigns (Tellegen & Maarten, 1998). But this resolution still begs a number of important questions. For example, what is the actual direction of causality between these specific attitudes and behaviors? Are the attitudes really causal in some sense, or are the attitudes as much justiﬁcations of behavior as they are explanations of behavior? Is this research of any practical relevance? Is it not paralyzing for policy actors to work from a framework that there is no such thing as environmental attitudes and orientations in general, and that the only way to induce pro-environmental behavior is through multiple appeals and programs that affect dozens of highly specific attitudes and attitude objects? For all practical purposes, the environmental social psychology of attitude-behavior linkages declined as a major focal point of environmental sociology research from the early 1980s onwards. This is arguably a welldeserved fate for research of this type, since it had failed to lead to knowledge of either sociological or practical significance. But as I will mention later in the chapter, a case can be made that environmental sociology ought to re-engage this paradox, albeit while jettisoning its 1970s social psychological basis. It is my view that, consistent with the overall perspective of this chapter, this issue can only be explained in a political sociological way. I will return to this issue in a provisional way later in the chapter. In contrast to the ﬁrst paradox, which has essentially disappeared as a major problematic in American environmental sociology, there are two other paradoxes that are alive and well as foci of theory and research. One of these paradoxes is that, while environmental sociology tends to place more emphasis on and faith in organized environmentalism than any other social science discipline as a source of solutions to environmental problems (Buttel, 2003),2 much of the thrust of environmental sociology serves to undermine many of the claims of organized environmentalists. For example, environmentalists tend to presume that natural resource efﬁciency (e.g., higher fuel economy of automobiles, reduced resource requirements of manufacturing) will result in

Environmental Sociology’s Theoretical and Empirical Paradoxes

51

clear environmental beneﬁts. Environmental sociologists, however, have long recognized the substantial validity of what has been known as the ‘‘Jevons Paradox’’ or the ‘‘Jevons effect’’ (Martinez-Alier, 2003) (after Stanley Jevons, the nineteenth century British economic analyst of the coal industry) – that increased efﬁciency in the use of natural resources tends to result in increased consumption of natural resources. The Jevons effect is that natural resource efﬁciency, because it increases productivity and contributes to income growth, will, all things being equal, increase the scope of economic expansion and stimulate more, rather than less, aggregate materials consumption (see Bunker, 1996; Clark & Foster, 2001; Martinez-Alier, 1987). Relatedly, sociologists such as Schnaiberg, Gould, Weinberg, and Pellow (Schnaiberg & Gould, 1994; Weinberg, Pellow, & Schnaiberg, 2000) have been critical of the widespread environmentalist faith in recycling. These scholars have shown that urban recycling, at least in the U.S. context, has generally been of very modest beneﬁt on both social and ecological grounds. A further example of critical sociological reﬂection on environmentalist doctrine concerns the hypothesis that resource scarcity is driving interstate and intrastate social conﬂicts. This hypothesis is most closely associated with scholars such as Homer-Dickson (1999) and with NGO researchers such as Michael Renner (1996). Peluso and Watts (2001), however, argue that it is seldom resource scarcity per se that induces violence. Instead, more often than not violence is initiated or undertaken by states and/or dominant classes, rather than by those deprived of access to resources, as a means to privatize access to and monopolize resources. The 2003 Iraq War is but one recent example of Peluso and Watts’ theoretical and empirical argument. Environmental sociologists have questioned whether environmental movement stress on global warming and global climate change has been an effective movement claim; there is a growing tendency by environmental sociologists to argue that the stress on global climate change, because it cannot be directly experienced, and because the beneﬁts of prospective climate-change mitigation policies will mainly be future generations rather than the people alive today, has served to decrease public allegiance to the environmental movement (Buttel, 2003). Finally, environmental sociologists have also tended to be sharply critical of many of the major postulates of Malthusianism, which for decades has remained very inﬂuential in environmentalist doctrine and discourses. Environmental sociologists, for example, have been critical of the notion that population density or growth is a potent cause of environmental problems, and that individual consumption decision making can be a major lever in reducing environmental destruction (Schnaiberg, 1980).

52

FREDERICK H. BUTTEL

The second long-standing paradox of environmental sociology that continues to generate substantial attention is that while capitalist growth and expansion are socially valued and could well be among the principal means of solving environmental problems, the expansion of capital has been the principal cause or driver of environmental destruction. From the earliest years of environmental sociology there was a general view that, for one or other reasons, economic expansion and capital accumulation tended to lead to major environmental problems (e.g., Molotch, 1975; Schnaiberg, 1975; Catton & Dunlap, 1978). The ‘‘second contradiction’’ (O’Connor, 1994) or ‘‘economic growth versus the environment’’ (Johnson & Hardesty, 1971) conundrum, however, did not become viewed as paradoxical until the 1980s and early 1990s when a set of interrelated notions – ecological modernization, reﬂexive modernization, environment Kuznets curves, dematerialization, and industrial ecology – emerged as hypotheses or arguments about how modernity and capital accumulation might also provide solutions to environmental problems. Thus, many environmental sociologists now preoccupy themselves with questions such as the following: What are the social and ecological possibilities for and limits to ‘‘delinking’’ (of ‘‘growth’’ and environmental destruction)? Can environmental protection – both pollution control and reduced materials throughput – be accomplished within the framework of a capitalist economy and private capital accumulation? Is the ‘‘treadmill of production’’ intrinsic to a capitalist-market economy? Contemporary environmental sociology is perhaps more deeply divided on these questions than on any others that will be discussed in this chapter. Two of the weightiest positions on the paradox of socioeconomically desirable and destructive capital accumulation and expansion have been brought to bear the debate between Allan Schnaiberg, David Pellow, and Adam Weinberg, and Arthur Mol and Gert Spaargaren, concerning treadmill of production and ecological modernization interpretations of the contradictions of the environmental state. The environmental state notion concerns the fact that the state is simultaneously an agent of environmental destruction and an agent of environmental reform, and the arena of the state thus involves an inherent structural contradiction between environmental devastation and protection. In very general terms, the ecological modernization perspective rests on the notion that a number of structural changes of or associated with late modernity (such as advanced industrial technology, state restructuring in response to ‘‘state failure,’’ and the global diffusion of environmental regulatory innovations) serve to position states to overcome the contradiction of being held responsible for achieving rapid rates of resourceintensive economic growth and for achieving environmental protection.

Environmental Sociology’s Theoretical and Empirical Paradoxes

53

By contrast, the treadmill of production position is that the political–economic constraints on the modern capitalist state serve to render impotent the state environmental impulse. Another position, one that is less theoretically elegant but which has suggested itself in macro-statistical studies of nationstates’ ecological footprints, is the neo-Malthusian STIRPAT model (York, Rosa, & Dietz, 2003), which suggests that a combination of population, the level of economic production, and the level of consumption are the major ‘‘drivers’’ of environmental destruction. The notion of the environmental state both reﬂects and provides a potential resolution to the paradox of economic growth and expansion undermining and contributing to environmental protection.

4. THEORETICAL AND EMPIRICAL PARADOXES AND THE FUTURE AGENDA FOR ENVIRONMENTAL SOCIOLOGY The initial paradox that deserves space on the agenda for environmental sociology over the next several years relates to the association of neoliberalism and environmentalism. It is clearly the case that over the past 25–30 years there has been a clear trend toward the rise of neoliberal regimes and policies over virtually the entire globe (Campbell & Pederson, 2001; Huber & Stephens, 2001) in tandem with the increased persuasiveness of environmental discourses and the rise of environmental mobilization and activism (Dryzek, Downes, Hunold, & Schlosberg, 2003, pp. 148–149). Given that these two processes have covaried over time, the question arises as to the relationship between the two. It would seem paradoxical to imagine that neoliberalism has had a causal relationship to the rise of environmental activism (other than environmentalism having been a reaction or response to neoliberalism); environmental movements are typically portrayed as struggling against neoliberalism and neoliberal policies (Faber & McCarthy, 2003). Neoliberalism is often understood as including environmental deregulation and the de-fanging of environmental lobbies as one of its core components, while environmentalism is typically seen as lying at the core of resistance to neoliberalism. But there are some sound reasons for imagining that environmentalism and neoliberalism are not fundamentally antagonistic or contradictory. First, as noted earlier, environmentalism as we know it grew up during the era of neoliberalism. Neoliberalism has established the basic institutional conditions – the ‘‘political opportunity structure,’’ if you will – of environmentalism.

54

FREDERICK H. BUTTEL

Thus, we ﬁnd that mainstream environmentalism seldom places social justice issues front and center, and accordingly stresses either imminent ecological catastrophes or calamities and/or trumpets neoliberal policy instruments such as emissions trading and other market mechanisms. Second, it is perhaps not coincidental that the world’s two major centers of neoliberalism and environmentalism are the U.S. and the U.K. Major environmental groups, particularly those in the U.S., have been at the forefront of the shift of state environmental control practices away from command-and-control policies and toward decentralized, collaborative, market-driven policies. Environmentalism, and the quest for votes among environmentally steadfast publics, has become an increasingly central commitment of left parties; the environmentalization of left parties has doubtless displaced trade unions and workers as the central constituency of social–democratic parties, and in doing so may have reinforced neoliberal ideologies, parties, and regimes. What is the basic essence of the relationship between neoliberalism and environmentalism? In all likelihood, future research will come to an understanding that there has been a recursive historical and complementary relationship between these two deﬁning forces of the late twentieth century and early twenty-ﬁrst century. Thus, for example, one might hypothesize that as Fordism and the welfare state began to run their course in the early 1970s in most of the West, environmentalism progressively replaced working classes and social justice discourses within institutionalized left politics. The growing strength of environmentalism led to conservatives and center-right reformers seeing overzealous and bureaucratically hidebound environmental programs as a major source of inefﬁciency and lack of competitiveness. As environmentalists came to see neoliberal politicians and ideologies as their most immediate adversary, environmentalism then became one of the major voices of opposition to neoliberal regimes, as was the case during the environmental backlash to the Reagan Administration. Further, environmental claimsmaking became the kinder, gentler edge of particular neoliberal regimes such as Germany’s Kohl and Britain’s Blair. Ultimately, further sociological research will likely unravel a contradictory and complementary pattern of coevolution of neoliberalism and environmentalism. The second paradox worthy of a prominent place on environmental sociology’s research agenda concerns globalization and environmental governance. Before proceeding to discuss the paradox of globalization and environmental governance, it is useful to recognize that globalization is itself a contradictory and complex phenomenon. Globalization is a notion that is employed to portray a diversity of processes: global market integration, ‘‘ﬁnancialization,’’ the rise of new governance regimes (e.g., the World Trade

Environmental Sociology’s Theoretical and Empirical Paradoxes

55

Organization, NAFTA, Codex Alimentarius, ISO), the simultaneous weakening and strengthening of national states, cultural homogenization, globalscale social resistance movements, the ‘‘globalization project’’ of national and multinational capitals seeking de-regulation and re-regulation on terms more favorable to capital (McMichael, 2000), and the synchronization of worldsystemic process through trade, ﬁnance, and ‘‘long cycles.’’ Globalization is particularly contradictory from a global governance standpoint. On the one hand, environmental regulations are chief among those that capital seeks to ﬂee. Most environmental activists see the quest for ‘‘free trade’’ as being aimed at, or having the consequence of, reducing the level of environmentalregulatory oversight over production and consumption and reducing the sovereignty of the national environmental regulatory state; some environmental groups are opposed to trade liberalization on principle, while most others strongly favor tying trade liberalization to provisions that ensure the integrity of environmental regulations.3 The raison d’eˆtre of trade liberalization is to promote global economic expansion, which in turn can be expected to result in expanded extraction of natural resources (Martinez-Alier, 2003; Lofdahl, 2002) and more environmental destruction. The paradox of globalization and environmental governance can be addressed, in part, through empirical research on whether, on balance, globalization processes undermine or enhance environmental protection and conservation (see, e.g., Lofdahl, 2002; Vogel, 1995). But even if there emerges some deﬁnitive judgment on this score, the globalization–governance paradox will remain one of the critical issues in environmental politics and environmental strategy. On the one hand, in a globalizing world, it seems apparent that environmental activists need to contest globally footloose capital and the growing global power of capital by organizing at a global level. Thus, it has become an article of faith among mainstream environmental groups that engaging in global-scale activism is integral to their strategy. Thus, over the past 20 or so years there has been a great deal of emphasis by environmental activist groups to catalyze international environmental regimes and to bolster the role of environmental intergovernmental organizations (Buttel, 2003). For many in environmental sociology it has become the conventional wisdom that the progressive elaboration of ‘‘global society’’ leads inexorably to more effective environmental governance (Frank, Hironaka, & Schofer, 2000). As much as the need for international-level mobilization seems apparent, there are several decisive shortcomings of global strategies. The ﬁrst and most dramatic of these is that the track record of international environmental activism over the past decade or so has been mixed at best. Though

56

FREDERICK H. BUTTEL

there have been some successes (e.g., the Cartagena Protocol of the Convention on Biological Diversity and the 1987 Montreal Protocol for mitigation of stratospheric ozone depletion), the failures have been manifold. The Framework Convention on Climate Change’s Kyoto Protocol will in all likelihood come to naught because the U.S. was unwilling to sign, and the Protocol failed to restrain coal consumption and CO2 output in China. Other than the Convention on Biological Diversity, none of the global agreements that came out of the Rio Earth Summit has been very successful. The Johannesburg World Summit on Sustainable Development, the most recent instance of global-scale environmental summitry, is widely considered to have been a failure. In particular, the corporate domination of Johannesburg demonstrated that capital has considerable power superiority over activist groups even in the case of summits such as this one that were established to give voice to agents of environmental protection. Thus, in a nutshell, international environmentalism seems ‘‘necessary,’’ and may be the ‘‘only game in town’’ in responding to the global reach of capital, at the same time that it confronts the impressive power of capital and encounters the social and ideological contradictions of employing a global strategy to attract adherents with predominantly local concerns. These considerations suggest a third contemporary environmentalsociological paradox: the fact that while, on one hand, for most people their most immediate and profound encounters with the natural world occur locally, and are based in ordinary, but intimate and meaningful, experiences, on the other hand, the most potential and inﬂuential political mobilizations with respect to the environment involve spatially and temporally distant environments. That is to say, for example, most Greenpeace ofﬁcials and Greenpeace members are motivated to address environmental problems they do not experience personally. Some of these problems are those that are manifest spatially distant environments – that is, these problems affect areas outside of activists’ own communities of residence. Likewise, groups such as Greenpeace may even stress problems that have not yet occurred, such as global warming; the problems stressed are those that will be experienced by future generations. Or, put somewhat differently, these are problems, which if solved, will confer beneﬁts mainly on future generations. In general, then, environmentalists tend to stress the environmental problems of other people. Environmental movement politics thus tends not to be a politics of grievances, wherein groups that personally experience environmental destruction or ‘‘insults’’ mobilize to redress these situations. Environmentalism tends instead toward representational politics, in which environmental activists call attention to the problems experienced by other groups, or

Environmental Sociology’s Theoretical and Empirical Paradoxes

57

problems that are not directly experienced by anyone at the present time, but which will presumably be experienced eventually by future generations.4 The historical tendency for many environmental sociologists and environmentalists has been to look upon the contemporary environmental challenge as being of self-evident importance. The fact that it is not, even for many among the literate public, can be gauged by reading Mitchell Thomashow’s (2002) Bringing the Biosphere Home: Learning to Perceive Global Environmental Change. Thomashow’s book is premised on the assumption that most citizens in their daily lives do not tend to recognize or see as important the kinds of global environmental problems that are stressed by most environmental movement organizations. Bringing the Biosphere Home sets forth an educational philosophy and program in order to guide members of the literate public to see the connections between their local surroundings and everyday environments on one hand, and broader-scale – essentially global – environmental concerns on the other. Thomashow’s book is in some sense an acknowledgement from environmentalist quarters that recognition of the seriousness of global environmental problems is not self-evident even for perceptive, well-educated persons. Thomashow’s book thus indicates how serious was the miscalculation of so many environmentalists and environmental sociologists during the 1980s and 1990s when they assumed that the global environmental predicament was, in effect, self-evident to all who are aware of the world around them. It clearly was not. A ﬁnal paradox of relevance to environmental sociology’s future research agenda can be stated as follows. In an important way, we can say that environmental sociology is based on a critique of the conventional sociological assumption that the biophysical environment is essentially a neutral backdrop against which social life and action play themselves out. Environmental sociology is thus necessarily materialist and realist in its ontology; human groups, societies, and other social aggregates are seen to shape, and to be constrained, by the biophysical world. At the same time, environmental politics is largely a politics of ideas, values, and ethics. Environmental political sociology thus requires that one understand symbolic politics – or, in other words, that we understand what Weber referred to as ideal interests and ethical orientations to social action (i.e., value-rational [vertational] orientation to action). Interestingly, employing environmental sociology’s materialist ontology, it is easier to understand anti-environmentalism than it is to understand environmentalism. Opposition to environmentalism tends to result from immediate or prospective interests – usually economic ones, but also more

58

FREDERICK H. BUTTEL

general material or ideological interests that ﬂow from a commitment to market liberalism and deregulation – whereas environmentalism tends not to be rooted in immediate interests. Environmentalism, by contrast, tends not to derive from material interests (instrumentally rational or zweckrational conduct), in the sense that environmental activists seldom beneﬁt materially as individuals if they win in struggle against corporate or state adversaries. This is particularly the case with mainstream ‘‘representational environmentalism.’’ Put somewhat differently, after decades of environmental sociological research on environmental movements, it is not clear that we have a convincing and comprehensive explanation of the phenomenon of environmental mobilization. It is not particularly helpful to see environmental activism and mobilization as being value-rational conduct, since this begs the (longstanding) question that has been leveled in critique of Weber’s ideal types of orientations to social action: what are the origins of vertrational or value-rational orientations to social action? Are the origins of vertrational conduct knowable, or do we basically have to accept that it exists without developing a convincing explanation of why this is the case? Can one distinguish clearly between value-rational orientations and traditional and affective orientations to social action? Implicit explanatory recourse to Weber’s category of vertrational orientation to action reﬂects a decidedly nonmaterialist ontology, and is seemingly in contradiction with the fundamental ontology of environmental sociology (and the unthinking idealism of doing so is one of the reasons that Macnaghten and Urry (1998) have dispatched environmental sociology as being hopelessly confused and contradictory). More structural views of the origins of environmental activism, such as Inglehart’s (1997) theory of postmaterialism, may well gain us some purchase on understanding environmental values and retain the integrity of materialist-realism, but it has little to contribute to understanding how a relatively few persons who hold postmaterialist/environmental values decide to orient their energies to environmental activism. Even the soundest and most insightful analyses of contemporary environmentalism, such as the innovative theoretical and empirical work of Dryzek et al. (2003), basically take environmental mobilization to be a given and provide little understanding of why some persons and groups come to make commitments to activism. How is environmental sociology to be able to retain its commitment to a materialist ontology while also generating a satisfactory understanding of what leads individuals and groups to take an activist orientation to the environment? Some possible resolutions of the paradox of ontological

Environmental Sociology’s Theoretical and Empirical Paradoxes

59

materialism and de facto idealism with respect to one of the key environment-society processes may include the following. One pathway to resolution is for environmental sociologists to come to recognize that for most groups most of the time, the biophysical environment is indeed substantially a backdrop of social behavior in which, at least cognitively speaking, behavior occurs largely independently of the biophysical environment. Thus, daily life for Americans living in North Florida does not differ very much from that in Southern Alaska, despite the vast differences in environmental context. People in their daily lives tend to be preoccupied with the rhythms of their daily lives. Greenhouse gases, gas-guzzling SUVs, Arctic oil drilling, and the consumption of various minerals, materials, and fuels are ‘‘noise,’’ as far as shaping social action is concerned. In the context of everyday conduct, ‘‘environmental values’’ are of little significance, having little to do with the nature of consumption or other behaviors. How is it then that some persons and groups become mobilized for environmental action for either brief or lengthy periods? There are likely to be a variety of distinct pathways of mobilization.5 One of these is the occupational-career pathway in which some people, particularly those who are socialized to be skeptical of corporate practices and disinterested in a corporate career, undergo training in environmental science, natural resources, or law in anticipation of an environmental organization career. But as the literatures on environmental justice, the sociology of risk, and the sociology of disasters suggest, the lynchpin of mobilization is that of victimization (see, for example, Couch, 1996; Edelstein, 1988). For example, whether a disaster is endured with little social unrest (as was the case with the 9/11/2001 World Trade Center incident) or results in mobilization or popular uprising (as in the Three Mile Island, Centralia Mine, and Buffalo Creek disasters) depends on whether those who experienced the disaster feel victimized by corporations, public authorities, or other agents. The success that have been registered by the environmental justice movement (see Agyeman, Bullard, & Evans, 2003) are likewise rooted in the emergence of perceptions of selective victimization – that negative environmental consequences are distributed unfairly and disproportionately affect minority groups, the powerless, and the poor. One of the implications of existing research on victimization and mobilization is that the two are most likely to come to the fore in connection with local (particularly community if not neighborhood-based) environmental problems as opposed to transnational or global environmental issues. But as Thomashow (2002) has noted, an exclusively local or person (‘‘NIMBY’’) environmental politics has its limits – inviting criticism (as

60

FREDERICK H. BUTTEL

shortsighted, superﬁcial, and selfish), encouraging corporate ﬂight to more quiescent areas, and failing to connect local concerns to global environmental problems.

5. CONCLUDING REMARKS In this chapter, I have argued that environmental sociology increasingly revolves around a set of theoretical and empirical paradoxes, and that a good share of the research agenda for the subdiscipline will revolve around understanding and resolving these paradoxes. Further, I have suggested that these paradoxes best lend themselves to resolution or understanding through the lens of political–sociological approaches. As mentioned earlier, my image of the future agenda for environmental sociology is to stress the understanding of the macropolitics of environment and resources, the role of the natural world in comparative-national and global development, and the nature and significance of environmental movements, activism, and countermovements. One could make the argument that much of sociology is rooted in paradoxes (e.g., as in the Marxist paradox of the progressive and exploitative character of capitalism, or in the Durkheimian paradox of the individual act of suicide being understandable only in macrosocial context as a social fact rooted in either a surfeit or dearth of social regulation). Thus, it could be said that environmental sociology is no different than the rest of sociology in its being understandable as inquiry anchored in resolving apparent paradoxes. There is no doubt an element of truth to this claim. At the same time, one enduring source of paradox in environmental sociology is that its ultimate validation within sociology, at least for the foreseeable future, lies in what may be depicted broadly as environmental knowledge. The stature or role of environmental sociology in the discipline depends on environmental knowledge. Environmental sociology’s principal justiﬁcation or niche as a sociological specialty is that it focuses on resource or biospheric issues that are of contemporary and long-term societal concern (and perhaps even the issue of survival as a species). Without persuasive social knowledge of environmental challenge and constraint, sociologists and other scholars would not be inclined to see environmental sociology as an important ﬁeld or subdiscipline. But, as noted earlier, the application of the sociological imagination to the study of environment and society often leads to scrutiny of environmental science and knowledge and/or of its uptake within environmentalist quarters. All of the paradoxes stressed in

Environmental Sociology’s Theoretical and Empirical Paradoxes

61

this chapter are paradoxical, at least in part, because they involve actual or potential sociological scrutiny or demystiﬁcation of environmental knowledge. The research agenda of environmental sociology is likely to be shaped indefinitely by the contradictory relationship of environmental sociology to environmental knowledge. And environmental sociologists are likely to continue to differ on whether sociological scrutiny of environmental knowledge undermines either environmental sociology or environmental activism (compare, for example, Dunlap & Catton, 1994; Blu¨hdorn, 2000). These disagreements will not only guide research on environmental-sociological paradoxes, but will continue to generate new ones that will be included within our research agenda.

NOTES 1. Interestingly, Savage (2001, p. 253) has recently lamented the fact that ‘‘the past two decades have seen political sociology in retreat.’’ It should be noted, however, that Savage’s definition of political sociology excludes the study of political movements and research on ‘‘the state’’ – two of the major staples of what I – and most others – would deﬁne as political sociology. 2. Thus, a closely related paradox concerns how it can be that environmental sociology is staunchly materialist – and tends to preach to non-environmental sociologists about the dangers of conceiving of social action as if nature, natural resources, and the biophysical world did not exist – at the same time that its stress on environmental movements, values, and actors represents an idealist position. Macnaghten and Urry (1998) originally posited the paradox, though they represented it as a contradiction in logic rather than a theoretical and empirical puzzle. 3. Sufﬁce it to say, however, that many social scientists believe that the basic essence of global economic integration is that of ‘‘trading up’’ (Vogel, 1995) – the notion that there tends to be a ‘‘California effect’’ (of environmental-regulatory upgrading) among participants in global trade regimes. There is also an increasingly inﬂuential hypothesis, ﬁrst advanced by Karliner (1997) that global economic integration leads to ‘‘grassroots’’ globalization, which in turn involves the intensiﬁcation of local activism that results in the strengthening of environmental regulations. 4. The fact that environmental activism involves representational politics rather than grievance-driven mobilization does not mean that environmentalism is distinctive in this respect. Many contemporary movements are highly professionalized in that they pivot around career activists who strive to seek redress for problems that are experienced by other persons and groups. 5. See, in particular, the useful typology of forms (civic, state, and corporate) and modes (personal, ideal, and instrumental) of environmentalism of Sunderlin (2003, p. 194 ff). Note, though, that this otherwise sophisticated typology leaves largely unaddressed the causes of environmentalist mobilization.

62

FREDERICK H. BUTTEL

ACKNOWLEDGMENTS An earlier version of this chapter was presented at the XV World Congress for Sociology of the International Sociological Association, Brisbane, July 2002. This work was made possible by support from the University of Wisconsin Graduate School.

REFERENCES Agyeman, J., Bullard, R. D., & Evans, B. (Eds) (2003). Just sustainabilities. Cambridge: MIT Press. Blu¨hdorn, I. (2000). Post-ecologist politics: Social theory and the abolition of the ecologist paradigm. London: Routledge. Bunker, S. G. (1996). Raw material and the global economy: Oversights and distortions in industrial ecology. Society and Natural Resources, 9(4), 419–429. Buttel, F. H., & Gijswijt, A. (2001). Emerging trends in environmental sociology. In: J. R. Blau (Ed.), Blackwell companion to sociology (pp. 43–70). Oxford: Blackwell. Buttel, F. H., Dickens, P., Dunlap, R. E., & Gijswijt, A. (2002). Sociological theory and the environment: An overview and introduction. In: R. E. Dunlap, et al. (Eds), Sociological theory and the environment (pp. 3–32). Lanham, MD: Rowman and Littleﬁeld. Buttel, F. H. (2003). Environmental sociology and the explanation of environmental reform. Organization & Environment, 16(3), 306–344. Campbell, J. L., & Pederson, O. K. (2001). The rise of neoliberalism and institutional analysis. Princeton: Princeton University Press. Catton, W. R., Jr., & Dunlap, R. E. (1978). Environmental sociology: A new paradigm. The American Sociologist, 13, 41–49. Clark, B., & Foster, J. B. (2001). William Stanley Jevons and The Coal Question: An introduction to Jevons’s ‘Of the economy of fuel’. Organization & Environment, 14(1), 93–98. Couch, S. R. (1996). Environmental contamination, community transformation, and the centralia mine ﬁre. In: J. K. Mitchell (Ed.), The long road to recovery: Community responses to industrial disaster. Tokyo: United Nations University Press. Dryzek, J. S., Downes, D., Hunold, C., & Schlosberg, D. (2003). Green states and social movements. New York: Oxford University Press. Dunlap, R. E. (1997). The evolution of environmental sociology: A brief history and assessment of the American experience. In: M. Redclift & G. Woodgate (Eds), International handbook of environmental sociology ðChapter1Þ. London: Edward Elgar. Dunlap, R. E., & Catton, W. R., Jr. (1994). Struggling with human exemptionalism: The rise, decline and revitalization of environmental sociology. The American Sociologist, 25, 5–30. Edelstein, M. R. (1988). Contaminated communities. Boulder, CO: Westview Press. Faber, D., & McCarthy, D. (2003). Neo-liberalism, globalization, and the struggle for ecological democracy: Linking sustainability and environmental justice. In: J. Agyeman, R. D. Bullard & B. Evans (Eds), Just sustainabilities (pp. 38–63). Cambridge, MA: MIT Press.

Environmental Sociology’s Theoretical and Empirical Paradoxes

63

Foster, J. B. (1999). The canonization of environmental sociology. Organization & Environment, 12, 461–467. Frank, D. J., Hironaka, A., & Schofer, E. (2000). The nation-state and the natural environment over the twentieth century. American Sociological Review, 65, 96–116. Homer-Dixon, T. (1999). Environment, scarcity, and violence. Princeton, NJ: Princeton University Press. Huber, E., & Stephens, J. D. (2001). Development and crisis of the welfare state. Chicago: University of Chicago Press. Inglehart, R. (1997). Modernization and postmodernization. Princeton: Princeton University Press. Johnson, W. A., & Hardesty, J. (Eds) (1971). Economic growth vs. the environment. Belmont, CA: Wadsworth. Karliner, J. (1997). The corporate planet. San Francisco: Sierra Club Books. Lofdahl, C. L. (2002). Environmental impacts of globalization and trade. Cambridge: MIT Press. Macnaghten, P., & Urry, J. (1998). Contested natures. London: Sage. Martinez-Alier, J. (1987). Ecological economics. Oxford: Basil Blackwell. Martinez-Alier, J. (2003). Mining conﬂicts, environmental justice and valuation. In: J. Agyeman, R. D. Bullard & B. Evans (Eds), Just sustainabilities (pp. 201–228). Cambridge, MA: MIT Press. McMichael, P. (2000). Development and social change. Thousand Oaks, CA: Pine Forge Press. Molotch, H. (1975). The city as a growth machine: Toward a political economy of place. American Journal of Sociology, 82(2), 309–332. O’Connor, J. (1994). Is sustainable capitalism possible? In: M. O’Connor (Ed.), Is capitalism sustainable? (pp. 152–175). New York: Guilford. Oliver, P. (2001). From the chair: Political sociology as the core of sociology? States, power and societies (newsletter of the Section on Political Sociology, American Sociological Association), 8(1), 1–2. Peluso, N., & Watts, M. (2001). Violent environments. In: N. Peluso & M. Watts (Eds), Violent environments. Ithaca, NY: Cornell University Press. Renner, M. (1996). Fighting for survival. New York: Norton. Savage, M. (2001). Political sociology. In: J. R. Blau (Ed.), Blackwell companion to sociology (pp. 253–267). Oxford: Blackwell. Schnaiberg, A. (1975). Social syntheses of the societal-environmental dialectic: The role of distributional impacts. Social Science Quarterly, 56, 5–20. Schnaiberg, A. (1980). The environment. New York: Oxford University Press. Schnaiberg, A., & Gould, K. A. (1994). Environment and society. New York: St. Martin’s Press. Sunderlin, W. (2003). Ideology, social theory, and the environment. Lanham, MD: Rowman and Littleﬁeld. Tellegen, E., & Maarten, W. (1998). Society and its environment: An introduction. Amsterdam: Gordon and Breach Science Publishers. Thomashow, M. (2002). Bringing the biosphere home: Learning to perceive global environmental change. Cambridge: MIT Press. Vogel, D. (1995). Trading up. Cambridge: Harvard University Press. Weinberg, A., Pellow, D., & Schnaiberg, A. (2000). Urban recycling and the search for sustainable community development. Princeton: Princeton University Press. York, R., Rosa, E. A., & Dietz, T. (2003). Footprints on the Earth: The environmental consequences of modernity. American Sociological Review, 68(2), 279–300.

FOR A SOCIOLOGY OF ‘SOCIONATURE’: ONTOLOGY AND THE COMMODITY-BASED APPROACH Paul K. Gellert ABSTRACT The works of Stephen Bunker represent early, sometimes unacknowledged, contributions to a sociological imagination regarding the role of nature and raw material extraction in processes of social change. By engaging debates defined within a world-systems frame of national state power and cycles of capital accumulation, Bunker’s work maintains the hierarchy of core, semi-peripheral, and peripheral states. While bridging the local and the global in unusual ways, his emphasis is predominantly on the external limits posed by nature and global markets on peripheries and the converse advantages offered to cores, thus allowing less room in the analysis for questioning which particular totality(ies) of social structure and relations are open to sociological inquiry. Gleaning the contributions from a socionatural approach and relating them back to Bunker’s commodity-based approach may expand the purview for analyses of the intersection of the natural and the social. In this chapter, I argue that attention to ‘nature’ in its multiple socionatural occurrences contributes to an understanding of the structuring of power in time and place. I rely on Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 65–91 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10004-3

65

66

PAUL K. GELLERT

geographer Eric Swyngedouw’s deployment of the concept of ‘socionature’ and the framework of actor-network theory to explore the benefits as well as challenges of a more relational, nondualistic sociological analysis of society and nature.

1. INTRODUCTION In recent years sociology has moved toward a consensus that takes nature seriously. As opposed to prior penchants to partition off the natural world from the putatively separate social world, analysts are struggling to create more useful ontologies and epistemologies for analyzing what has variously been termed ‘nature-culture,’ ‘hybrid objects,’ ‘quasi-objects’ (Latour, 1993); ‘cyborgs’ (Haraway, 1991), or, most usefully, ‘socionature’ (Swyngedouw, 1999). All of these theoretical interventions are attempts, as Goldman and Schurman (2000) note, to close the ‘great divide’ between society and nature that two decades ago William Catton and Riley Dunlap (1978) dubbed the ‘human exemptionalism paradigm.’1 By recalling earlier sociological contributions and by continuing to learn from the margins of the adjacent ﬁelds of geography, anthropology, and history, I believe that sociology can create ‘a new sociological imagination’ (Goldman & Schurman, 2000, p. 578), which will take socionature and socionatural relations as our objects of inquiry. The works of Stephen Bunker represent early, sometimes unacknowledged, contributions to such an imagination. In his innovative analyses of the extraction of commodities from the periphery in the Brazilian Amazon and the resulting cycles of underdevelopment, Bunker recognized the importance of complex ecological processes and the embeddedness of human societies within such formations, until wrenched by the market forces of global capitalism (Bunker, 1984, 1985).2 From this work, he developed a distinctive commodity-based approach to studying and understanding world-systemic processes that recognized the spatial and topographic determinants of social power (Bunker, 1992, 1994, 1996; Barham, Bunker, & O’Hearn, 1994). ‘‘[H]uman agency and political and economic choice,’’ Bunker (1992, p. 70) stridently observed, ‘‘are more restricted in extractive than in transformative economies.’’ In recent years, Bunker and Ciccantell have expanded this approach to challenge top-down versions of hegemonic transition by offering an alternative explanation based on careful bottom-up attention to the importance of resource extraction and transport in the

For a Sociology of ‘Socionature’

67

formation of ‘generative sectors’ that can more ﬂexibly rearrange space and time through innovation in technology (Bunker & Ciccantell, 1995, 1999, 2003). By engaging debates deﬁned within a world-systems frame of national state power and cycles of capital accumulation, however, Bunker’s work maintains the world-systems hierarchy of core, semi-peripheral and peripheral states developed by Wallerstein (1974). Arguably, these categories are reiﬁed in ways that tend to ignore ‘‘particular location-specific struggles as determinant of change’’ (Feldman, 2001, p. 352). Thus, although interested in the potential capacity of less powerful and (by definition) more peripheral social forms to overcome the constraints they face, Bunker’s emphasis is predominantly on the external limits posed by nature and global markets on peripheries and the converse advantages offered to cores. Additionally, limits and advantages are framed in terms of the potential for capital accumulation and state movement within the global hierarchy. While bridging the local and the global in unusual ways, Bunker nonetheless leaves less room in the analysis for questioning which particular totality(ies) of social structure and relations are open to sociological inquiry (see, e.g., Tomich, 1997). Gleaning such contributions and relating them back to Bunker’s commodity-based approach may expand the purview for analyses of the intersection of the natural and the social. Comparable to Feldman’s (2001) use of a gender lens to exemplify the nondeterminate character of postcolonial thought, I argue that attention to ‘nature’ in its multiple socionatural occurrences contributes to an understanding of the structuring of power in time and place. The ontological and epistemological limitations of modernist frames such as world-systems, including a priori-ism, totalizing schemas, and determinism, are addressed at least in part by emerging perspectives that build on actor-network theory (ANT) (Latour, 1993, 1999, 2000; Law, 1999). In this chapter, therefore, I rely on geographer Eric Swyngedouw’s (1999) deployment of the concept of ‘socionature’ to explore the beneﬁts as well as challenges of a more relational, nondualistic sociological analysis of society and nature (see also Goodman, 2001). The ontological inseparability of nature and society warrants this new vocabulary.3 Like Bunker, Swyngedouw and other writers building on ANT examine the interrelations of nature and society (Castree, 2002; Castree & Braun, 1998), but they differ from his commodity-based analysis in that they attribute more than ‘resistance’ to nature in a social process deﬁned by worldsystemic structures of capital accumulation. In these perspectives, ‘nature’ is not a separate category that is acted upon and then revealed, especially in its

68

PAUL K. GELLERT

commodiﬁed form, by extraction, industry, or capitalism. Rather, proposing that the world is constituted by interconnected actors (or, more neutrally, the French ‘actants’) who form networks and that actants include both humans and nonhumans, united in particular actor-networks, socionature offers a more multifaceted and open-ended understanding of the world. In order to understand why the ontology of socionature is an advance, the ﬁrst two sections examine the advantages of ANT for a sociology of socionature and review two kinds of asymmetrical interrelation. In the third section, I address several challenges to the analysis of socionature, as well as more specifically to actor-network approaches (see Marsden, 2000; Murdoch, 1997; Castree, 2002). To clarify the utility of a balanced ontology, I reexamine Freudenburg et al.’s debate with Pickering about the ‘conjoint constitution’ of nature and society (Freudenburg, Frickel, & Gramling, 1995, 1996; Pickering, 1996). Siding with Pickering reveals the advantages of a ‘strong’ version of conjoint constitution, which recognizes the ontological status of nature as potential agent while rejecting insinuations of anthropomorphism; opens the possibility of a sociology ‘after humanism’ (Breslau, 2000); and asserts the ontological inseparability of ‘local’ and ‘global’ actors. The ﬁrst and the third, especially, are areas that Bunker’s work also emphasizes but where he might be pushed further, especially through a discussion of the challenges of specifying the spatial and temporal bounds of socionature.

2. ADVANTAGES OF ANT FOR THE SOCIOLOGY OF SOCIONATURE There are ﬁve key advantages, emanating from ANT, to altering our endeavor from the study of social facts to socionatural relations, as Noel Castree (2002) has recently argued. Importantly, two major dualisms are overcome. The ﬁrst is the social-natural dichotomy itself, which is rejected as ontologically incomplete in the claim that ‘‘entities are ‘essentially’ either social or natural prior to their interaction with one another’’ (Castree, 2002, p. 118). Instead, as Latour (1993) argues, we might seek an ‘amodern’ ontology and recognition of hybrids or quasi-objects. Based on their emphasis on multiple and ‘relentlessly heterogeneous’ (Murdoch, 1997, p. 745) networks, ANT advocates encourage relational thinking, and as a result, reject causes ‘outside’ socionatural networks. That is why Latour emphasizes the careful description of networks before any attempt is made at explanation.

For a Sociology of ‘Socionature’

69

To be clear, description is not the endpoint but explanations are possible only at a later point in the analysis. The second dualism is the local–global dichotomy that is so ubiquitous, both in studies of globalization and of local community or ‘alternatives.’ It is important to recognize that ‘‘the ‘local’ and the ‘global’, however conceptualized, are mutually constituting’’ (Araghi & McMichael, 2000). One might recall Bunker’s (1985) Underdeveloping the Amazon, which was aimed at transcending the debate on ‘external’ versus ‘internal’ causes of development by examining their articulation. Yet, his narrative structure overwhelmingly stressed the common topographical vocabulary in which ‘local’ resource degradation is attributed to ‘global’ market pressures. Against this, ANT argues for ‘topological’ or relational language taken from network thinking. Building on Foucauldian notions of power, for example, Whatmore and Thorne (1997, p. 291) posit that ‘‘size, or scale of an actor network is a product of network lengthening, not of some special properties peculiar to the ‘global’ or ‘core’ actors.’’ By thus ‘ﬂattening’ our world view, there is recognition that ‘‘all parts of a network (be it long or short) are, at some level, causally important in ensuring that it holds together and endures’’ (Castree, 2002, p. 120). Without relying on Foucault, Bunker’s (2003, p. 219) recent synthetic analysis also moves in the direction of ﬂattening the world by analyzing how ‘‘the localystructures and organizes the worldsystem.’’ While deploying a different vocabulary from ANT’s network lengthening, Bunker nonetheless observes a similar ‘‘progressive spatial dispersion of production’’ (Bunker, 2003, p. 221). The difference is that the direction of change is singular and the ontological status of the worldsystem is unchallenged. This ﬂattening relates to the importance of three other advantages. First, all types of asymmetry are avoided. Neither the anthropomorphism of social constructionism wherein the so-called natural is seen as merely a construct of the social world, nor the anthropocentrism in which ‘‘ultimately, nature can only be understood and valued in human (sic) terms,’’ (Castree, 2002, p. 120), nor the nonhuman eco-centrism of deep ecology is accepted. In their place, a ‘co-constructionist’ tradition is put forth (Murdoch, 1997, 2001; Haraway, 1991; Latour, 1993; Pickering, 1995) that recognizes the lack of purity of either the ‘natural’ or the ‘social’. In Bunker’s words, ‘‘reciprocal dynamics belie notions of the social construction of nature and of space. Humans do not transform or construct space socially; at most they accelerate and intensify their own transit and their transport of matter across it’’ (Bunker, 2003, p. 236). Related to this, second, is a rejection of assumptions that humans are the only significant actors or that action is ‘‘associated with

70

PAUL K. GELLERT

intentionality and linguistic competence (logocentrism)’’ (Castree, 2002, p. 121). One consequence of this position is a preference for the term ‘actant’ over ‘actor’ because agency is understood as a ‘relational effect’ of the actornetwork, as in Callon’s (1986) study of scallops and Swyngedouw’s (1999) study of hydrology in Spain. Lastly, overdetermined, excessively ‘centered’ and structural views of power, from the more traditional to the radical, are rejected for their common assumption that power is ‘held’ by particular actors and nature is merely an ‘effect’ of power. Instead, ANT, like much postcolonial, feminist, and postdevelopment thought, views power more ‘democratically’ as diffuse, as a relational achievement, and therefore, as open to multiple points of resistance (see, e.g., Pile & Keith, 1997). The more open-ended politics of this approach is more optimistic and less despairing than the deterministic stance of world-systems analysis. To be clear, in this open-ended politics, the importance of avoiding asymmetry and structural determinism is not to assert that each actor has ‘equal’ power in a particular historical circumstance but rather that the potential power of actors cannot be ascertained in advance. It is in this emphasis on uncertainty regarding the relative power of (social) actors, especially since their ‘power’ relies on their interaction with putatively separate (natural) actors, that the more symmetrical analysis of ANT opens up multiple possibilities for change. The relative power of actors is not only in relation to the world-system structure but also in relation to other actors or constellations of actors.

3. THE TRANSFORMATION OF SOCIOLOGY’S OBJECT OF INQUIRY To appropriate a phrase from Marx, the so-called primitive separation of nature and society in sociology that bolstered the foundation and early institutionalization of the discipline of sociology has also limited the scope of the discipline in disabling ways. In brief, it has supported the notion that one can ignore ‘nature’ in the study of social relations. Foster (2002, p. 55) observes ‘‘[A]voidance of any concept of nature (not simply subsumed under a radical constructionism) is almost the definition of sociology for many.’’4 In the 1980s, Bunker was the ‘‘lone voice in the proverbial wilderness’’ who sought to integrate environmental and sociological concerns, particularly related to analysis of the world-system (Roberts & Grimes, 1999, p. 61).

For a Sociology of ‘Socionature’

71

Since then other scholars have joined in the effort to (re)integrate society and nature in our analyses (Goldman & Schurman, 2000).

3.1. Asymmetrical Interrelation of Nature and Society The ﬁrst step in bridging the divide is to examine the interaction of society and nature but accept the categories as separate and given. Such approaches remain asymmetrical in that they do not grant equivalent ontological status to human and nonhuman entities (cf. Murdoch, 1997). Here we might distinguish further two divergent streams: social constructionism and environmental determinism, or what Soper (1996) has called ‘nature-sceptical’ and ‘nature-endorsing’ positions. On the one hand, in reductive versions of sociological constructionism, it appears that nature is only a creation in our imaginations or of our conceptualizations.5 Such ‘oversocialized’ conceptions of sociological constructionism have been supported in recent years by the poststructural emphasis on ‘texts’ which consider language as actually constituting and not just representing nature. One of the key thrusts of this view is that there is no objective world waiting to be ‘discovered.’ There are multiple ‘truths’ and multiple ‘natures’ but no real nature. The relativist implication is that it is impossible to adjudicate among the various interpretations (see Cronon, 1994). Defenders of the social constructionist position that landscapes are ‘symbolic environments,’ such as sociologists Greider and Garkovich, argue that ‘‘what is important in any consideration of environmental change is the meaning of the change....’’ (Greider & Garkovich, 1994, p. 21, emphasis added; see also Hannigan, 1995).6 However, we might wonder, with Eden (2001, p. 82), ‘‘if we see nature as (merely) a cultural categorization which is continually, diversely, and mutually renegotiated in its relationship with culture and not as something concrete under threat, then are ‘environmental problems’ merely ﬁctions wrought of constructions and do not require us to try to solve or prevent them?’’ That is, the political implications for socionatural relations are ignored or not delineated. To be sure, there are varieties of social constructivism and the more nuanced recognize the cultural (or cognitive) and natural (or biogeophysical) co-construction of the world (see Smith, 1999, 364 ff; Scarce, 2000, especially Chapter 4). Undeniably, as communicative animals, we rely on language to represent the world, and in this basic sense the world we know is constructed by and through our language. However, as Rolston reminds us,

72

PAUL K. GELLERT

we need to distinguish between and avoid the conﬂation of this epistemological construction and an ontological construction of the world (see Kidner, 2000, p. 349). The ‘epistemic fallacy’ identiﬁed by noted realist Roy Bhaskar is to treat the world as dependent on cognitive choices (Kidner, 2000, p. 343; see also Proctor, 1998; Foster, 2002).7 On the other hand, in works of environmental determinism from the neo-Darwinian population biology of Paul Ehrlich to environmental preservationism, a separate and much more powerful domain of nature is assumed. In such asymmetric eco-centrism it is posited that nature is being destroyed by too many humans and too much human activity, and that, ultimately, humans are subordinate to laws of population and nature. The ‘destruction’ of nature is not only assumed to be objectively deﬁned and measurable but also directly attributable to human activity, usually of the large-scale and industrial sort. Bunker’s work is similar to the ecological Marxism of James O’Connor (1998) in differentiating, for example, between aggregate population effects and social class effects and more generally retaining dialectical aspirations. Yet, the ontological dualism between human society and nature is retained. O’Connor’s ‘‘second contradiction’’ of capitalism depends on a (separate) nature in effect ‘biting back’ on capitalism, which, it is argued, had harmed the natural ‘conditions of production’ in the ﬁrst place. Bunker, too, after richly describing the complex intertwining of nature and society in the Amazon, concludes that comparative historical analysis can (should) separate nature and society in the pursuit of an ‘‘analysis of the relative weight of the various natural and social factors’’ (Bunker, 2003, p. 250).8 In this sense, ‘‘the dialectic between nature and society becomes an external one, that is a recursive relationship between two separate ﬁelds...’’ (Swyngedouw, 1999, p. 446). Following Latour, geographer Erik Swyngedouw argues against such ‘puriﬁcation’ of categories.9

3.2. ‘Socionature’ and the Co-construction of Nature and Society The main advantage of a hybrid understanding of nature-culture, by contrast, is a sociological vision of the co-construction, both materially and discursively, of nature and society. To pursue this, Swyngedouw (1999) usefully combines Smith’s (1984) production of nature with the actornetwork approach of Latour (1993, 1999; see also Law, 1999) to advance the concept of ‘socionature.’10 As he writes, socionature means:

For a Sociology of ‘Socionature’

73

natural or ecological conditions and processes do not operate separately from social processes, and that the actually existing socionatural conditions are always the result of intricate transformations of preexisting conﬁgurations that are themselves inherently natural and social. (Swyngedouw, 1999, p. 445, original emphasis).

What can be achieved by not separating the factors of nature, culture, politics, engineering, ideology, and the like, is a significantly different sociological account of how a new ‘socioenvironmental landscape’ emerges. These accounts are better, I believe, in the sense that they are more dialectical. They do not rely on ﬁxed or primordial starting points in the relation between humans and nature, such as ‘natural’ forests or other ecosystems. Furthermore, they acknowledge the constant change in new conﬁgurations of socionature. Socioenvironmental landscapes ‘‘are simultaneously physical and social...[reﬂecting] historical-geographical struggles and social power geometries, and [interiorizing] the ﬂux and dynamics of sociospatial change’’ (Swyngedouw, 1999, p. 461).11 This hybrid approach to nature and society is informed by ANT, exempliﬁed by Callon’s (1986) oft-cited study of attempts in St. Brieuc Bay, France, to domesticate scallops in the controlled production of previously ‘naturally’ extracted commodities.12 In this case, scientific attempts to increase scallop production in the face of a deﬁned problem of dwindling stocks of scallops relied on the transfer of a technique discovered in Japan: anchoring scallop larvae to collectors immersed in the sea and protected from predators for a time before being ‘sown’ in the ocean bed to be harvested 2–3 years later. In analyzing the emergent actor-network, Callon proposed four ‘moments’ of a general process called ‘‘translation’’ during which ‘‘the identity of actors, the possibility of interaction and the margins of manoeuvre are negotiated and delimited’’ (Callon, 1986, p. 203).13 The most important ‘moment’ is that of enrollment, when the question is whether the scallops will in fact enter into and anchor onto the collectors. In other words, could a new socionature be created? If not, the scallops contest the emergent network, become ‘dissidents’ in Callon’s terms, and the result is ‘betrayal’ of the socionatural network. In point of fact, this is exactly what happened. Although some larvae did anchor to the collectors, after repeated experiments ‘‘the collectors remain[ed] hopelessly empty’’ (Callon, 1986, p. 219), and following the scallops’ betrayal, came the betrayal of scientists whose funding was threatened as a result and the local ﬁshermen who abandoned the long-term plans for replenishment of stocks and sustainable harvests by ﬁshing out the scallops after only 2 years. Bunker (1984), too, discovered the socionatural ‘‘surprises’’ (Boyd, Prudham, & Schurman, 2001) that may attend to extraction from the

74

PAUL K. GELLERT

periphery in his analysis of the cycles of extraction and underdevelopment in the Amazon Basin. He found that a boom in caiman exports from the Amazon basin in the 1950s for human demand in Europe, for example, actually reduced populations of ﬁsh on which the caiman preyed. The ﬁsh had thrived on nutrients from caiman excrement, and their decline, in turn, resulted in decreased protein available for local human populations. In actor-network terms, the ﬁsh were part of a network that was disrupted by the emergence of a new network. One difference is that, although Bunker viewed the complex ecological (including human) interrelations as part of the ‘progressive destruction’ of the Amazon by more powerful social actors, ANT would emphasize the unpredictability and instability of the interactions. It may not be the case that more powerful social actors destroy (separate) nature for those living in it, but that their opportunities to build networks that support their power are challenged by the ‘enrollment’ of (new) actants. For example, as a result of decades of commercial logging in Indonesian Borneo (Kalimantan), forest ﬁres that occur more frequently in El Nin˜o oceanic oscillation years, have become more intense than in the past (Gellert, 1998a). Their increased intensity, especially in reaching into the forest crowns, causes more damage to the large trees that are the interest of commercial logging companies, as well as to young trees in newly established plantations. Ironically, the plantation companies continue to ﬁnd it in their short-term economic interest to use ﬁre for land clearing, despite the risks to the cleared and planted spaces within their networks of interaction. Another example of unpredictability in the forest spaces of Indonesia is the question of what exactly will happen to the vast areas of logged-over forest, now categorized as in various degrees of degradation or deforestation. The state’s capacity to monitor the prescribed Indonesian selective logging and replanting system and create sustainable forest management has been limited not only by its lack of interest in doing so but also by the spatial extensiveness of logging in huge river basins (Gellert, 1998b). Such efforts to manage nature have become even more difﬁcult in recent years following the fall of the authoritarian President Suharto, promulgation of regional autonomy laws, and the expansion of illegal logging (Casson & Obidzinski, 2002). Yet it is an open question, worthy of further research, whether the outcome(s) in the next historical moment will be captured by the idea of a progressive destruction of an extractive periphery or whether it will (also) encompass the creation of a new networks and conﬁgurations of socionature.

For a Sociology of ‘Socionature’

75

4. ADDRESSING CHALLENGES TO A SOCIOLOGY OF SOCIONATURE: BUNKER’S COMMODITY-BASED APPROACH AND CONJOINT CONSTITUTION Despite what I have argued are advantages, there remain serious but not insurmountable challenges to the ontological advantages of ANT and a hybrid notion of socionature. Four challenges stand out: the question of the agency of nature; the problem of intentionality and humanism; linkages between the so-called ‘local’ and ‘global’ in sociological theory building; and the periodization of socionatural change. Bunker’s (1984, 1992) commoditybased approach, wherein sociological analysis begins from the characteristics of raw materials commodities and simultaneously examines the intersection between world-systemic processes and local or regional economies, addresses many of these challenges. Also, more recently, Bunker (2003, p. 250) has shown ‘‘how local materio-spatial relations and processes in the Amazon intersect with, and partially constitute, the world system as it transforms, and is transformed by systemic changes in cycles of accumulation.’’ Bunker does so, moreover, in ways that Pickering found lacking in Freudenburg et al.’s notion of the ‘conjoint constitution’ of nature and society (Freudenburg et al., 1995, 1996; Pickering, 1996). Nonetheless, a few critical questions remain about the determinism of Bunker’s commoditybased approach that a fuller appreciation of socionature might serve to address. Significantly, unlike other theorists, Bunker has taken seriously the social and ecological effects of the production of basic raw material commodities. In anticipatory fashion, he was writing against the grain of literature in the sociology of development that was beginning to be enamored of ‘globalization’ and the ‘space of ﬂows.’ Bunker observed, ‘‘However global exchange systems have become, commodities can emerge only from locally based extractive and productive systems’’ (Bunker, 1984, p. 1019). In this way, he took on the so-called external-internal debate over the causes of underdevelopment by arguing for a spatially integrated approach that simultaneously examined the multifarious ways in which ‘‘different regional social formationsycoevolve in direct interaction with each other and through a world system of exchange’’ (ibid.). In their article on what is now known as Iron Mountain, along the Michigan–Wisconsin border, Freudenburg et al. (1995) advance a notion of ‘conjoint constitution’ that appears the same as socionature. They claim ‘‘it is important to recognize that the social is inherent in what is usually seen as

76

PAUL K. GELLERT

the physical, just as the physical is often integral to what is perceived as the social’’ (Freudenburg et al., 1995, p. 386). To illustrate, they provide a sociological account of ﬁve phases in the representation and use of Iron Mountain: from indigenous hunting ground and living space to source of timber, source of iron, bust of iron mines, and ﬁnally tourist attraction. Further, they use the idea of conjoint constitution to examine how ‘‘both the physical characteristics and the ongoing changes in society and technology played significant roles in shaping the human uses that proved possible and/ or likely’’ (ibid., original emphasis). In discussing mining technology, for example, they argue ‘‘the attempted allocation into ‘social’ and physical categories may have the potential to contribute as much to confusion as to understanding’’ (Freudenburg et al., 1995, p. 370). Their notion of conjoint constitution is ultimately weakened, however, as Pickering (1996) points out in his critique, because they are not symmetrical enough in their analysis. Their unique sociological contribution is to demonstrate how different social meanings, as well as, secondarily, broader political economic importance, is produced from one physical landscape at different periods of time. Although, they intend to generate ‘‘greater awareness of the extent to which (what we take to be) the physical is inﬂuenced by (what we take to be) the social, and vice versa’’ (Freudenburg et al., 1995, p. 366), ultimately Freudenburg et al. place asymmetrical emphasis on the human meanings of Iron Mountain in the different periods. In this weak version of conjoint constitution, in other words, ‘‘the social was not itself at stake’’ (Pickering, 1996, p. 155, original emphasis). A strong version, by contrast, would alter our sociological object of inquiry from social ‘facts’ to socionature and especially the socionatural processes that produce particular temporary crystallizations of socionature.14

4.1. The Agency of Nature The ﬁrst crucial issue is natural agency, which is recognized by Pickering and Bunker but ignored in Freudenburg’s causal story.15 One can readily accept recognition of the social constructedness of the dividing line between ‘nature’ and ‘society,’ but a more symmetrical account would also consider how the particular mixings of natural and social – what I am referring to as socionature – are historical products. More generally, if sociology is to close the nature-society divide and achieve the ontological advances of that move, then it will need to be more amenable to the potential agency of nature.16 Bunker noted a decade ago, ‘‘Our theoretical and our ideological biases are

For a Sociology of ‘Socionature’

77

to emphasize human agency’’ (Bunker, 1992, p. 80), but recent sociological theorizing on agency (see Emirbayer & Mische, 1998; Fuchs, 2001) recognizes the more symmetrical character of agents in the world that ANT also posits. In a strong version of conjoint constitution, it would be recognized that nature and, for Pickering, the technologies such as pipes and compressors of refrigerants for mining, which were created in conjunction with the (socio)natural activities on the mountain, ‘‘actually did things, there on Iron Mountain’’ (Pickering, 1996, p. 157, original emphases). Conversely, the socionatural results (temporary crystallizations) would have been different without the ‘enrolment’ of these particular actors. Put differently, by analyzing the co-construction of the socionatural by natural and human actants, we come to a different sort of sociological account. In summarizing cases related to bauxite and aluminum, Barham et al. (1994, p. 235) noted that ‘‘although the environmental and physical constraints associated with resource extraction weigh most heavily on the extractive regional economy, all the participantsyare constrained to some degree by the characteristics of the resource and its location in natural environments.’’ Still, the separation of resources and natural environments from humans is maintained in this rendering. The issue is not one of asserting that rocks can think or other crude anthropomorphizing of nature.17 To further the discussion, we can qualify by thinking of kinds of agency that are available. Hacking offers the useful distinction between ‘interactive’ and ‘indifferent’ kinds of actors so that quarks – or scallops, rivers and dams, or trees – are indifferent ‘‘in the sense that calling a quark a quark [or a scallop a scallop, a tree a tree, or more generally a natural ‘thing’ a ‘resource’] makes no difference to the quark’’ (Hacking, 1999, cited in Murdoch, 2001, p. 124).18 In his research on water in Spain, Swyngedouw reveals a socionatural process in which hydrological actants are part of the shaping of political economic change. In the course of Spain’s modernization, including the over 800 dams constructed in the twentieth century alone, power is ‘captured’ from the water of rivers and streams and socionature is produced. In this hybridization process, component parts – biochemical physical processes, material practices, cultural practices, social relations, language, discursive constructions, and ideological practices – ‘‘swirl out from the production process itself’’ (Swyngedouw, 1999, p. 447). In his emphasis on process, following Harvey (1996), nothing is ﬁxed, or at least ‘things,’ such as landscapes, are only temporary crystallizations of historical–geographical transformation. In the eighteenth and nineteenth centuries, a great historical

78

PAUL K. GELLERT

struggle occurred over the ‘restoration’ of Spain’s hydrological cycles to their ‘natural’ laws and the ‘‘attempt to ‘naturalize’ political territorial organization’’ along river basin rather than political boundaries. The struggle was (temporarily) resolved in favor of the ‘regenerationists’ under the fascist dictatorships from the 1920s, which created ‘‘a new geography, a new nature, and a new waterscape’’ (Swyngedouw, 1999, pp. 458–460). Similarly, understood in this way, the characteristics of trihydrate aluminum, including its association with wide, ﬂat, and geologically old river basins, which are ironically, or fortuitously, depending on the actant, not the best locations for hydroelectric dams, shaped the extraction, transformation, and exporting processes in the Brazilian Amazon and elsewhere (Bunker, 1994). Because the Brazilian state, nonetheless, was able to force the ‘enrollment’ of the rivers and aluminum into a socionatural network, the results included a large ﬂood plain and a dam perhaps more likely to suffer from sedimentation. Also, with the export of timber from the tropical Indonesian rain forest, we can understand that in order to ‘enroll’ particular tree species in emergent networks requires negotiation among not only the usual (i.e., social) actors but among the so-called natural actors of soils, climate, river ﬂow, forest ecosystem, and seedlings. None of these is ‘purely’ natural, however. The soils have been compacted by tractors, the forests have various size openings in the canopy from previous logging and other human activity, the seedlings are not randomly evolved in nature but propagated in the nursery based on prevailing scientific knowledge and transported by truck to the regeneration sites, and even the climate can be affected not just by global warming or El Nin˜o-Southern Oscillation currents but also by microclimates affected by the burning (or not) of small patches of forest, for example. The trees in the forests of Indonesia and the ﬁres that have destroyed them in various years are attributable to the actions and nonactions of decades if not centuries of human and nonhuman actants in socionatural networks (cf. Gellert, 1998a). The trees are ‘indifferent’ actors. Nonetheless, their actions – whether, for example, they can be made (enrolled) to propagate, outcompete other species of ﬂora, and grow rapidly in the opened spaces of previous logging – can impact on the longevity of the network that to foresters is meant to be ‘sustained yield forestry management.’ The important issue for human actors in the network is that just as in considering the actions of other humans, the actions of the trees cannot be comprehensively known and are subject to unpredictable interactions. Obstacles, opportunities, and surprises come from ‘nature’ (Boyd et al., 2001).19 Or, as Levins and Lewontin (1985, p. 219) observe, ‘‘because nature is complex, any intervention in the rich

For a Sociology of ‘Socionature’

79

network of interacting variables is likely to have many indirect and unexpected consequences, some of which negate the original purpose of the intervention.’’ The point is not that we humans should master nature by overcoming the inadequacies of our current knowledge, but that such mastery is impossible. 4.2. Intentionality and Humanism Unfortunately,20 as the quotation from Barham et al. above indicates, the sociological ‘monopoly’ on human agency is not so easily given up. In their response to Pickering, Freudenburg et al. stick to a position that only the social world has agency and that to go down Pickering’s road, though ‘thrilling,’ is merely an ‘entertaining distraction’ (Freudenburg et al., 1996, p. 170).21 Underlying their observation, however, is an unnecessary emphasis on intentionality and humanism as the basis for sociological analysis. Sociology can be accomplished by equating the agentic qualities of human, nonhuman, and inhuman (quasi-machine) subject-objects (see Murdoch, 1997). Even more boldly, sociology can be accomplished without resort to agency at all. We might join Fuchs in questioning the ‘overrated’ importance of human intentions: If we know someone’s intentions, that knowledge would not get us much closer to explaining or predicting social outcomes, since intentions and outcomes are very loosely coupled, especially in very complex systems and structures. For example, how much of what happens when a major earthquake hits a big city can be explained as the result of intentions? (Fuchs, 2001, p. 27).

First of all, we cannot truly know anyone else’s intentions. Second, even stated intentions are rarely fully achieved because of the unpredictability and complexity of those with whom actors are interacting. By starting with structures, as Fuchs argues, we can understand ‘intention’ as a variable. Then, we can turn our attention to variations in the (second order) attribution of agency. As a result, agency becomes the ‘‘capacity of a system to surprise its observers’’ (Fuchs, 2001, p. 34). Since both nature and humans can surprise, there is nothing ‘wrong’ in attribution of agency to nature (see also Boyd et al., 2001): ‘‘The important sociological difference is not between things and people but between the attribution of interpretivism or determinism’’ (Fuchs, 2001, p. 33).22 The key is therefore to analyze the variation in how much agency to expect (of any actor) in different social structures. To some readers, this would seem to be an anti-humanist and therefore inadmissable approach to sociological knowledge (see Collins & Yearley,

80

PAUL K. GELLERT

1992; Jones, 1996). Thus, Jones argues that Pickering’s anti-humanist ‘posthumanism’ should be rejected precisely for what he sees as its unnecessary assumption of nonhuman agency (Jones, 1996, p. 299). That also seems to be Freudenburg’s stance, when they claim that their focus is ‘‘sociological thinking about ‘the environment’ ’’ (Freudenburg et al., 1995, p. 368, footnote 5). By contrast, however, I am arguing that we should move our sociology beyond the study of ‘social facts’ to the study of socionature. The study of historical–geographical transformation (Harvey, 1996) viewed through the lens of ‘socionature’ is one in which ‘‘modernization through socionatural changes always takes place within already constructed historical socionatural conditions’’ (Swyngedouw, 1999, p. 449). Like intentionality, humanism is not necessary for this kind of sociology. As Breslau (2000) has argued with regard to the sociology of scientific knowledge, it is not ontological humanism that has led to sociology’s achievements in this area. To the contrary, despite the persistent humanism of sociologists on all sides of debate, ‘‘nonhumanist ontology leaves sociology quite intact’’ (Breslau, 2000, p. 290). In fact, humanist ontology leads to inconsistencies or dualisms between the social construction of science and the necessity of the (social constructionist) sociologist of science believing in the reality of his/her existence, that is, an existence separate from other humans and nonhumans. It is the ‘radically relational ontology’ of the actor-network approach that breaks with this notion of agency, recognizing that the ‘‘distinction between humans and others is constructed’’ (Breslau, 2000, p. 300). As a result, ‘‘the differences between human and nonhuman agency are just that, differences between kinds of agents, rather than between humans who act and objects that are acted upon’’ (Breslau, 2000, p. 300). More importantly, particular kinds of agents only come into being through their networks of interaction. The networks that emerged at different times, for example, in relation to (what we now call) Iron Mountain or in relation to tropical timber exports, bring into being different human and nonhuman agents. However, it is important that there are differences, particularly in the production and exercise of unequal power relations among actors. Rather than overemphasizing the idea that, in effect, the social is doing all of the structuring, we should understand the truly conjoint constitution of socionature (not society and nature). As Pickering argues, the trajectory of U.S. development and the fate of Iron Mountain are ‘interactively stabilized’ as part of a ‘triple becoming’ of a natural resource, a technology, and a society (Pickering, 1996, pp. 156–157). The biogeophysical and the technological impinge upon the production of different social groups, their

For a Sociology of ‘Socionature’

81

power and identities.23 Swyngedouw’s (1999) argument about the modernization of Spain via the transformation of socionature also reﬂects this triple becoming. Bunker’s commodity-based approach, which integrates technologies (e.g., of aluminum processing) with topography and hydrological ﬂows, as well as peripheral and core state strategies, stretches the notion of becoming to the world-system, while admittedly maintaining a humanist ontology.24 The resolution of the agency question leads us to the most challenging issues in pursuing a sociology of socionature: spatial and temporal scale in relation to theoretical explanations.

4.3. Linking the ‘Local’ and the ‘Global’ in Theory With so much diversity among networks that are constantly emerging, changing, and ‘dying,’ it is not immediately evident what sorts of generalizations we can make. In fact, in what Castree (2002) calls ‘strong versions’ of ANT, there is no desire to explain. Instead, ‘‘explanation emerges once the description is saturated’’ (Latour, 1991, quoted in Castree, 2002, p. 119). One may wonder, therefore, whether ANT allows us to ‘‘ever do anything more than describe, in a prosaic fashion, the dangerous imbroglios that enmesh us?’’ (Murdoch, 1997, p. 750, original emphasis). I argue, however, that one need not accept this emphasis on ‘saturated’ descriptions; one can still strive for sociological explanation of hybrid, socionatural objects produced through historically (‘temporally’ for Pickering) emergent and more or less persistent (or transient) structures of socionatural relations (see Breslau, 2000). In this light, it is indicative of the potential of and preferred outlines of a sociology of socionature that Pickering challenges Freudenburg et al. to provide more explanation of the ways in which society is ‘at stake’ in their recounting of Iron Mountain. Pickering believes that the story of Iron Mountain implies a reﬁguring of our understanding of U.S. and global history because social transformations were created by – and would have been much different or slower without – the iron extracted. Freudenburg et al.’s reticence to make what they regard as overreaching claims for their data on Iron Mountain rest in an ontological (and methodological) difference. They grant greater power to ‘local’ causal actors and processes than to the historical production of structured actor-networks as a whole. In other words, it is the meaning attributed to one place (Iron Mountain) rather than the length and shape of the network (including actors at Iron Mountain) that is changing. In arguing for ‘causal proximity,’ however, they fall prey to

82

PAUL K. GELLERT

a kind of ‘abstract localism’ that views the local as particular and the global as general (Araghi & McMichael, 2000).25 But the local does not have an ontological existence separate from the global. As Jarosz and Qazi (1999, p. 9) write, ‘‘the local is global. y[T]he global is realized locally through social structures and agency as it unfolds within and is shaped by particular landscapes.’’ Bunker’s work on cycles of hegemonic ascent attempts to do just that (Bunker & Ciccantell, 1995, 1999, 2003; Bunker & O’Hearn, 1993). Crucially, arguing against Arrighi’s (1994) analytical separation of material life from market economy and anti-market state activity, Bunker and Ciccantell (2003) propose a synthesis of the material into a more holistic analysis. The key to economic ascent in the world system, at least for Holland, Great Britain, the United States, and Japan, has been the incorporation of ‘‘cumulatively sequential increases’’ in production (Bunker & Ciccantell, 2003, p. 7). In this analysis, the ‘‘stasis imposed by Wallerstein’s (1974, 1982) dependence on tripartite categories whose relations remain fundamentally unchanged,’’ is called into question (ibid.). This questioning had begun much earlier in Bunker’s analysis of the Amazon, where he argued: Wallerstein’s erroneous insistence on a single world mode of production and on an international division of labor within it does not adequately account for the additional inequalities involved in the transfer of natural values, that is, commodities whose values are primarily extracted from nature rather than produced by labor and capital, from the periphery to the core. y [Wallerstein] ignores the special characteristics of the extractive economies predominant in many peripheral areas. Nor, despite his concern with modes of production as elements of an international division of labor, does Wallerstein indicate how historical sequences within particular modes of production reproduce themselves and the ways in which they are inserted into the world economy over continuous, or sequential, periods of time calculated at the regional, rather than at the global, level. (Bunker, 1984, p. 1055; see also Bunker, 2003).

Arguably, working an analysis from the periphery requires attention to just such indeterminate processes. The capacity of the state in Indonesia to realize sustainable forest management was limited as much by local geography, topography, and complex ecology as by global (i.e., core) demand for timber available in the forests of Kalimantan and Sumatra and insertion of this timber and thus the region into the world economy (Gellert, 1998b). Imagining different and changing spatial boundaries within one analysis is not easy. Yet, at least in the more recent analyses of hegemony that emphasize comparability across the four cycles, how exactly the different

For a Sociology of ‘Socionature’

83

periods of hegemony differ in terms of the shape of the totality(ies) produced (e.g., Tomich, 1997) is not resolved.

4.4. Periodization of Socionatural Change When approaching a sociology of socionature, the question of time and periodization is one of the most difﬁcult analytical issues to consider. Bunker’s rejection of the key thesis of Wallerstein’s world-systems analysis regarding capitalism as the mode of production and a single, systemic division of labor opens a space for thinking about a materialist analysis that might be multi- rather than uni-directional, indeterminate rather than determinate.26 Particularly when working from the periphery, Bunker and his colleagues and students have demonstrated the importance of the characteristics of globally demanded commodities from caimans and rubber to iron and bauxite. Such cycles, as noted above, are thoroughly socionatural. Following the insights of Canadian economic geographer Harold Innis, attention is focused on the importance of characteristics such as the spatial ﬁxity of (especially mineral) raw materials and their extraction and transport through and over topographies that are differentially amenable to human manipulation. One hesitates to add complexity to what is already a complex historical materialist understanding. To be sure, moreover, struggles over resources and the reconﬁguration of space and technology are integral to Bunker’s (and Ciccantell’s) account of resource extraction, processing, transport, and consumption in cycles of capitalist accumulation and hegemonic power in the world-system. Nonetheless, this narrative retains the totalizing schema and determinism of world-systems analysis more generally, in effect following Wallerstein’s (1988) claim that generalization is more difﬁcult than uncovering historical speciﬁcity. In the penchant for generalization, some historical struggles seem to be eclipsed. The periodization of change is deﬁned by hegemonic power. Access to resources – the central problematic of Bunker’s approach – is framed as a problem for the (protean) core nation-states, and three historically important core strategies are identiﬁed: imperialism, technological innovation, and externalizing costs of production onto peripheral and semiperipheral ‘host countries’ (Bunker & Ciccantell, 2003, pp. 12–13). Less attention is paid to the choices among and combinations of these three strategies that are deployed in any historical period. While ‘nature’ ﬁgures prominently in shaping such ‘choices’ by human actors, the overall story is one of

84

PAUL K. GELLERT

overcoming or evading natural obstacles under fortuitous circumstances. As Bunker and Ciccantell (2003, p. 18) summarize the Dutch case, for example, ‘‘Holland’s ascent rested on its cheap access to timber and the ways that it extended that comparative natural advantage through a series of socially created competitive advantages via economies of scale.’’ The a priori delineation of actors engaged in struggle is conﬁned to states, ﬁrms, and raw materials (e.g., Barham et al., 1994)27 and core actors overwhelm the power of peripheral ones in Bunker and Ciccantell’s (2003, p. 4) ‘‘unitary answer’’ to questions of global inequality. Moreover, the questions themselves potentially reify the categories that are being investigated. That is, how were and are ‘national economies’ formed historically? And how and with what boundaries are totalities formed? The tendency to ascribe power and agency to the core powers is partially addressed by continued attention to the roles of peripheral states and ﬁrms and of the conﬂicts between regional fractions of capital in shaping the operation of particular raw materials industries and markets (Gellert, 2003; Leitner, 2003). The social struggles that (might) form counter-hegemonic discourses and practices are, however, almost completely obscured. More work needs to be done to be able to periodize on the basis of what kinds of socionature are produced and what complex networks of humans and nonhumans produce them. Bunker and Ciccantell (2003, p. 26) partially open the path to consider this by noting that there is a ‘‘wider context of a far greater number of partially successful and failed efforts to ascend within the system of global stratiﬁcation.’’ However, even this framing preﬁgures the possible socionatural relations that might exist by deﬁning success and failure in terms of movement in a hierarchy of states, rather than, for example, in the relations in and with nature in shifting and multiple spatiotemporalities (cf. Braun, 2002, p. 26, Harvey, 1996). To imagine a periodization of history in terms of different socionatures would require recognition of the tension between predictability and surprise and between temporary crystallization and durability among various actants forming particular socionatural complexes.28

5. CONCLUSION In rethinking ACT, John Law (1999) observes that one should not underestimate the ‘shock value’ of ACT, especially due to its rejection of essentialist divisions such as that between humans and nonhumans. In this vein, I have argued that the work of Stephen Bunker can be seen to shock

For a Sociology of ‘Socionature’

85

conventional sociologists in different ways. He has perennially reminded us that a truly materialist analysis provides insights into the workings of capitalism and the interstate relations that characterize global capitalism. The agentic qualities of ‘nature’ and the multifarious linkages between actors and processes in so-called local and global locations play important roles in producing what I have labeled ‘socionature’ at different scales and for different durations. At the same time, I have argued in this chapter that Bunker’s commoditybased approach might be pushed further to incorporate the ontological symmetry and nondualist formulations of socionature in at least two ways. First, it might beneﬁt from pushing ‘back’ to the social production of particular commodities as one (temporary) crystallization of socionatural relations. This would require attention to the complex processes of historical production of (socio)nature that Smith (1984) also outlined. Second, it might be pushed ‘forward’ into a less deterministic analysis of the future (and more generally the historicity) of socionatural relations.29 My point here is emphatically not to suppress recognition by sociologists, as well as by policy makers and activists, of the formidable structural constraints Bunker identiﬁes in extractive economies or for states and ﬁrms aspiring to ascend in a world-system. Nor is the point to detract from the causal relationships that Bunker and Ciccantell identify between extraction of key commodities and the reconﬁguration of space and time by ascendant hegemonic states. The point is to open up the universalizing aspirations of the narratives that emerge to recognize the multiple ways, commodiﬁed and otherwise, that humans and nonhumans constitute and struggle over socionatural relations in time and space.

NOTES 1. The divide was opened much earlier by Durkheimian and Parsonian sociology, in reaction to Spencer’s evolutionism. Catton and Dunlap’s attempt to close the divide is preceded, it should be recognized, by the efforts of early material human ecologists and co-evolutionary biologists (e.g., Norgaard, 1994) and, to a much more limited degree, ecologists engaged in long-term ecological research (see, e.g., Collins et al., 2000). 2. Peluso (1996) notes, however, that Bunker does not recognize the social agency of extractive communities who contribute to the construction of particular ecological landscapes. That is, extractive commodities were not completely ‘gifts from nature’ even several hundred years ago. 3. While much of this analysis has occurred within the sociology of science and technology, more recent applications have been in agriculture (FitzSimmons &

86

PAUL K. GELLERT

Goodman, 1998; Goodman 1999, 2001; Whatmore & Thorne, 1997; Juska & Busch, 1994; Busch & Juska, 1997) and the geography of landscape restoration (Eden, Tunstall, & Tapsell, 2000). 4. Many sociologists have accepted such assumptions, for example analyzing the changing organization, operation and governance of a particular industry, without reference to the natural materials that supply it. Or, they analyze conﬂicts between groups in ‘pure’ cultural terms, as in Huntington’s famous ‘clash of civilizations.’ 5. This is the position ridiculed in the so-called ‘science wars’ by the idea that constructivists do not believe in gravity. It is also the one that is criticized strongly by Catton and Dunlap. 6. Greider and Garkovich’s discussion of the social construction of landscapes is rich and suggestive, particularly in its discussion of power as ‘‘the capacity to impose a specific definition of the physical environment, one that reﬂects the symbols and meanings of a particular group of people’’ (p. 17). However, the historical roots of different groups’ symbols is wholly social in their formulation. The accusation that a separate biophysical reality also exists ‘misses the point’ (p. 21), they say, and they are right for their analysis. But, an analysis of hybrid objects or socionature would not separate the two realities so completely. 7. Deep ecologist Arne Naess writes, ‘‘Having been taken at least twice by avalanches, I have never felt them to be social constructions’’ (Naess, 2000, p. 335). 8. Bunker’s recent (2003) article on the Amazon, matter, and space comes very close to the argument I am making here, but without the vocabulary or the ontology. 9. When one considers the material bases of human existence, as well as the literature on the myth of ‘pristine’ wilderness (see Scott, 1998, Chapter 8, footnote 1; see also Cronon, 1995) it becomes impossible to defend the idea of a pure and therefore separate ‘nature’ to be saved from humans – not to mention the contradiction that the ‘saving’ is to be done by certain humans. 10. The term socionature was put forth earlier by Callon: ‘‘If oneywishes to talk about nature and society, it is better to say that translation networks weave a socionature, an in-between that is inhabited by actants whose competence and identities vary along the translations transforming them.’’ (Callon, 1986, p. 58). 11. This simultaneity of causes is also captured in an intriguing study of the restoration of the river Cole in England where ‘‘Neither the ‘restored’ nor the ‘prerestored’ Cole was solely natural or solely cultural but a complex of the two, the product of social and technical change, of centuries of farming, urbanization, drainage, and use’’ (Eden et al., 2000, p. 261). 12. Or, to produce what Martin O’Connor (1994) terms ‘capitalized nature.’ 13. The four are problematization; interessement (stabilizing the identities and roles of other actors or actants); enrollment; and mobilization. 14. One way of conceiving of this difference is that between a Durkheimian positivism and a Marxian dialectical materialism. 15. FFG (Freudenberg, Frickel, & Gramling) claim that they ‘are not arguing that the biophysical characteristics of a resource or a technical system are somehow irrelevant’ (Freudenburg et al., 1995, p. 371) but in their focus on the taken-for-grantedness of the dividing line between natural and social, they do not provide an explanation of how the biophysical and geological are relevant.

For a Sociology of ‘Socionature’

87

16. While Pickering focuses on the ‘material agency’ of particular technologies, I am more interested in the agency of what is commonly understood as pure (nonhuman) nature. 17. Callon (1986, p. 228, footnote 24) observers that the ‘reasons’ for the conduct of scallops do not matter. Pickering too dismisses the ‘ridiculous imputations’ of anthropomorphism (Pickering, 1996, p. 157, footnote 6). While these are serious issues, I also recall that after a seminar on his commodity-based approach in which Bunker admitted coming ‘dangerously close’ to environmental determinism, a political economist of labor coyly remarked ‘Rocks of the world unite!’ 18. Pickering (1993, p. 565) distinguishes the intentionality of human action from that of nonhumans, although other network theorists such as Callon and Law resist that claim (see Murdoch, 1997, pp. 746–888). 19. Consider also Bunker’s story of ecological complexity in the Amazon mentioned above. 20. I say unfortunately from the perspective of advancing a sociology of socionature. 21. This position seems close, in the end, to Greider and Garkovich’s (1994) sophisticated understanding of the social construction of landscapes. However, Freudenberg et al. argue further for a comparative method because they believe it is ‘easier’ to see how the social is at stake if one has multiple research sites with different environments to compare. ANT would hold that one can examine the social through the emergence and longevity (or lack) of one network over time, rather than a comparative sociology that reiﬁes the division between social and natural and simply places more causal efﬁcacy on the side of the natural than on the side of the social. On the other hand, Freudenburg et al. claim that their focus is ‘‘sociological thinking about ‘the environment.’ ’’ (p. 368, footnote 5) That is, they remain true to ‘pure’ sociological goals. 22. Fuchs elaborates through discussion of differing attribution of agency to pets and people with Alzheimer’s disease. 23. The variation on Pickering that I would introduce is to distinguish analytically between a triple becoming of distinct things, where the emphasis is still on the boundaries between them, and a becoming of blurred socionatural complexes as objects of inquiry, where the emphasis might be on the hybrid networks produced. 24. Leitner’s (2003) insightful analysis of intra-capitalist conﬂict over the ﬂows of Wisconsin rivers views the river as a ‘media of struggle’ which recognizes nature’s resistance to the desires of some capitalists but retains an asymmetrical ontology wherein capitalist accumulation and industry are the real actors. 25. In Freudenburg et al.’s more positivistic approach, bits of nature are ‘added up’ and if they do not have ‘significant effect’ on society, then probabilistic conclusions cannot be made. They advocate more comparisons across resource characteristics before jumping to conclusions about the impact of Iron Mountain. 26. Consider Friedmann’s processural thinking in that, ‘‘to create a world market, the colonized ecosystem was subjected to y radical simplifying’’ (Friedmann, 2000, p. 494, emphasis added). On the other hand, note again that the dualism of society destroying or simplifying a separate nature is maintained. 27. Barham et al. use this trio in their title, although I should note that O’Hearn’s chapter pays at least implicit attention to other social actors in the ﬁrms’ acquisition

88

PAUL K. GELLERT

of bauxite-rich Jamaican lands at agricultural land prices. Still, the ‘losers’ in the story are not deemed important actors. 28. See Boyd et al. (2001) and Haraway’s (1991, Vol. 3, p. 199) discussion of the Coyote and the room in feminist objective science for surprises and ironies because ‘we are not in charge of the world.’ 29. After Bunker’s presentation at the conference that formed the basis for this volume, there was considerable discussion of whether classical Marxist concerns with labor struggles in nature, as opposed to struggles over nature, were absent from Bunker’s approach.

REFERENCES Araghi, F., & McMichael, P. (2000). Bringing world-history back in: A critique of the postmodern retreat in Agrarian studies. Draft paper prepared for World Congress of Rural Sociology. Arrighi, G. (1994). The long twentieth century: Money, power, and the origins of our times. London: Verso. Barham, B., Bunker, S. G., & O’Hearn, D. (Eds) (1994). States, firms, and raw materials: The World economy and ecology of aluminum. Madison, WI: The University of Wisconsin Press. Boyd, W., Prudham, W. S., & Schurman, R. A. (2001). Industrial dynamics and the problem of nature. Society and Natural Resources, 14, 555–570. Braun, B. (2002). The intemperate rainforest: Nature, culture, and power on Canada’s west coast. Minneapolis: University of Minnesota Press. Breslau, D. (2000). Sociology after humanism: A lesson from contemporary science studies. Sociological Theory, 18, 289–308. Bunker, S. G. (1984). Modes of extraction, unequal exchange, and the progressive underdevelopment of an extreme periphery: The Brazilian Amazon, 1600–1980. American Journal of Sociology, 89, 1017–1064. Bunker, S. G. (1985). Underdeveloping the Amazon. Chicago: The University of Chicago Press. Bunker, S. G. (1992). Natural resource extraction and power differentials in a global economy. In: S. Oritz & S. Lees (Eds), Understanding economic process (pp. 61–84). Lanham, MD: University Press of America. Bunker, S. G. (1994). The political economy of raw materials extraction and trade. In: Industrial ecology & global change (pp. 437–450). New York: Cambridge Press. Bunker, S. G. (1996). Raw material and the global economy: Oversights and distortions in industrial ecology. Society and Natural Resources, 9, 419–429. Bunker, S. G. (2003). Matter, space, energy, and political economy: The Amazon in the worldsystem. Journal of World-Systems Research, IX, 219–258. http://jwsr.ucr.edu. Bunker, S. G., & Ciccantell, P. S. (1995). Restructuring markets, reorganizing nature: An examination of Japanese strategies for access to raw materials. Journal of World-Systems Research, 1, 1–62. http://csf.colorado.edu/wsystems/jwsr.html. Bunker, S. G., & Ciccantell, P. S. (1999). Economic ascent and the global environment: Worldsystems theory and the new historical materialism. In: W. L. Goldfrank, D. Goodman & A. Szasz (Eds), Ecology and the world-system (pp. 107–122). Westport, CT: Greenwood Press.

For a Sociology of ‘Socionature’

89

Bunker, S. G., & Ciccantell, P. S. (2003). Generative sectors and the new historical materialism: Economic ascent and the cumulatively sequential restructuring of the world economy. Studies in Comparative International Development, 37, 3–30. Bunker, S. G., & O’Hearn, D. (1993). Strategies of economic ascendants for access to raw materials: A comparison of the U.S. and Japan. In: R. A. Palat (Ed.), Pacific Asia and the future of the world-system. Westport, CT: Greenwood Press. Busch, L., & Juska, A. (1997). Beyond political economy: Actor networks and the globalization of agriculture. Review of International Political Economy, 4, 688–708. Callon, M. (1986). Some elements of a sociology of translation: Domestication of the scallops and the ﬁshermen of St. Brieuc Bay. In: J. Law (Ed.), Power, action, and belief: A new sociology of knowledge? (pp. 196–233). Boston, MA: Routledge & Kegan Paul. Casson, A., & Obidzinski, K. (2002). From new order to regional autonomy: Shifting dynamics of ‘‘Illegal’’ logging in Kalimantan, Indonesia. World Development, 30, 2133–2151. Castree, N. (2002). False antitheses? Marxism, nature and actor-networks. Antipode, 34, 111–146. Castree, N., & Braun, B. (1998). The construction of nature and the nature of construction: Analytical and political tools for building survivable futures. In: B. Braun & N. Castree (Eds), Remaking reality: Nature at the millenium (pp. 3–42). New York: Routledge. Catton, W. R., Jr., & Dunlap, R. E. (1978). Environmental sociology: A new paradigm. The American Sociologist, 13, 41–49. Collins, H., & Yearley, S. (1992). Epistemological chicken. In: A. Pickering (Ed.), Science as practice and culture (pp. 301–326). Chicago, IL: University of Chicago Press. Collins, J. P., Kinzig, A., Grimm, N. B., Fagan, W. F., Hope, D., Wu, J., & Borer, E. T. (2000). A new urban ecology. American Scientist, 88, 416–425. Cronon, W. (Ed.) (1994). Uncommon ground: Toward reinventing nature. New York: Norton. Cronon, W. (1995). The trouble with wilderness or, getting back to the wrong nature. Environmental History, 1, 7–55. Eden, S. (2001). Progress reports: Environmental issues: Nature versus the environment? Progress in Human Geography, 25, 79–85. Eden, S., Tunstall, S. M., & Tapsell, S. M. (2000). Translating nature: River restoration as nature-culture. Environment and Planning D: Society and Space, 18, 257–273. Emirbayer, M., & Mische, A. (1998). What is agency? American Journal of Sociology, 103, 962–1023. Feldman, S. (2001). Intersecting and contesting positions: Postcolonialism, feminism, and world-systems theory. Review, XXIV, 343–371. FitzSimmons, M., & Goodman, D. (1998). Incorporating nature: Environmental narratives and the reproduction of food. In: B. Braun & N. Castree (Eds), Remaking reality: Nature at the millenium (pp. 194–220). New York: Routledge. Foster, J. B. (2002). Environmental sociology and the environmental revolution: A 25th anniversary assessment. Organization and Environment, 15, 55–58. Freudenburg, W. R., Frickel, S., & Gramling, R. (1995). Beyond the nature/society divide: Learning to think about a mountain. Sociological Forum, 10, 361–392. Freudenburg, W. R., Frickel, S., & Gramling, R. (1996). Crossing the next divide: A response to Andy Pickering. Sociological Forum, 11, 161–175. Friedmann, H. (2000). What on Earth is the modern world-system? Foodgetting and territory in the modern era and beyond. Journal of World-Systems Research, VI, 480–515 http:// csf.colorado.edu/wsystems/jwsr.html.

90

PAUL K. GELLERT

Fuchs, S. (2001). Beyond agency. Sociological Theory, 19, 24–40. Gellert, P.K. (1998a). A brief history and analysis of Indonesia’s forest ﬁre crisis. Indonesia, (65), 63–65. Gellert, P. K. (1998b). The limits of capacity: The political economy and ecology of Indonesia’s timber industry, 1967–1995. Ph.D. dissertation, Sociology, UW-Madison, Madison, WI. Gellert, P. K. (2003). Renegotiating a timber commodity chain: The politics of the Indonesia– Japan plywood link. Sociological Forum, 18, 53–84. Goldman, M., & Schurman, R. A. (2000). Closing the ‘great divide’: New social theory on society and nature. Annual Review of Sociology, 26, 563–584. Goodman, D. (1999). Agro-food studies in the age of ecology: Nature, corporeality, biopolitics. Sociologia Ruralis, 39, 17–38. Goodman, D. (2001). Ontology matters: The relational materiality of nature and agro-food studies. Sociologia Ruralis, 41, 182–200. Greider, T., & Garkovich, L. (1994). Landscapes: The social construction of nature and the environment. Rural Sociology, 59, 1–24. Hannigan, J. (1995). Environmental sociology: A social constructivist perspective. London: Routledge. Haraway, D. (1991). Simians, cyborgs, and women: The reinvention of nature. New York: Routledge. Harvey, D. (1996). Justice, nature and the geography of difference. Malden, MA: Blackwell Publishers Inc. Jarosz, L., & Qazi, J. (1999). The geography of Washington’s world apple: Global expressions in a local landscape. Journal of Rural Studies, 16, 1–11. Jones, M. P. (1996). Posthuman agency: Between theoretical traditions. Sociological Theory, 14, 290–309. Juska, A., & Busch, L. (1994). The production of knowledge and the production of commodities: The case of rapeseed technoscience. Rural Sociology, 59, 581–597. Kidner, D. W. (2000). Fabricating nature: A critique of the social construction of nature. Environmental Ethics, 22, 339–357. Latour, B. (1993). We have never been modern. Cambridge, MA: Harvard University Press. Latour, B. (1999). On recalling ANT. In: J. Law & J. Hassard (Eds), Actor network theory and after (pp. 15–25). Oxford: Blackwell Publishers. Latour, B. (2000). When things strike back: A possible contribution of ‘science studies’ to the social sciences. British Journal of Sociology, 51, 107–123. Law, J. (1999). After ANT: Complexity, naming and topology. In: J. Law & J. Hassard (Eds), Actor network theory and after (pp. 1–14). Oxford: Blackwell Publishers. Leitner, J. (2003). North American timber economy: Log transport, regional capitalist conﬂict and corporate formation in Wisconsin’s Chippewa Basin, 1850–1900. Review, XXVI(2), 173–219. Levins, R., & Lewontin, R. (1985). The dialectical biologist. Cambridge, MA: Harvard University Press. Marsden, T. (2000). Food matters and the matter of food: Towards a new food governance? Sociologia Ruralis, 40, 20–29. Murdoch, J. (1997). Inhuman/nonhuman/human: Actor-network theory and the prospects for a nondualistic and symmetrical perspective on nature and society. Environment and Planning D: Society and Space, 15, 731–756.

For a Sociology of ‘Socionature’

91

Murdoch, J. (2001). Ecologising sociology: Actor-network theory, co-construction and the problem of human exemptionalism. Sociology, 35, 111–133. Naess, A. (2000). Avalanches as social constructions. Environmental Ethics, 22, 335–336. Norgaard, R. (1994). Development betrayed: The end of progress and a co-evolutionary revisioning of the future. New York: Routledge. O’Connor, J. (1998). Natural causes: Essays on ecological marxism. New York: Guilford Press. O’Connor, M. (1994). On the misadventures of capitalist nature. In: M. O’Connor (Ed.), Is capitalism sustainable: Political economy and the politics of ecology (pp. 125–151). New York: The Guilford Press. Peluso, N. L. (1996). Fruit trees and family trees in an anthropogenic forest: Ethics of access, property zones, and environmental change in Indonesia. Comparative Studies in Society and History, 38, 510–548. Pickering, A. (1993). The mangle of practice: Agency and emergence in the sociology of science. American Journal of Sociology, 58, 559–587. Pickering, A. (1995). The mangle of practice. Chicago, IL: University of Chicago Press. Pickering, A. (1996). Further beyond the society/nature divide: A comment on Freudenburg, Frickel, and Gramling. Sociological Forum, 11, 153–159. Pile, S., & Keith, M. (Eds) (1997). Geographies of resistance. New York: Routledge. Proctor, J. D. (1998). The social construction of nature: Relativist accusations, pragmatist and realist responses. Annals of the Association of American Geographers, 88, 352–376. Roberts, J. T., & Grimes, P. E. (1999). Extending the world-system to the whole system: Toward a political economy of the biosphere. In: W. L. Goldfrank, D. Goodman & A. Szasz (Eds), Ecology and the world-system. Westport, CT: Greenwood Press. Scarce, R. (2000). Fishy business: Salmon, biology, and the social construction of nature. Philadelphia, PA: Temple University Press. Scott, J. (1998). Seeing like a state. New Haven, CT: Yale University Press. Smith, M. (1999). To speak of trees: Social constructivism, environmental values, and the future of deep ecology. Environmental Ethics, 21, 359–376. Smith, N. (1984). Uneven development: Nature, capital and the production of space. New York, NY: Basil Blackwell. Soper, K. (1996). Nature/‘nature’. In: G. Robertson, M. Marsh, L. Tickner, J. Bird, B. Curtis & T. Putnam (Eds), Futurenatural: Nature, science, culture (pp. 22–34). New York: Routledge. Swyngedouw, E. (1999). Modernity and hybridity: Nature, regeneracionismo, and the production of the Spanish waterscape, 1890–1930. Annals of the Association of American Geographers, 89, 443–465. Tomich, D. (1997). World of capital/worlds of labor: A global perspective. In: J. R. Hall (Ed.), Reworking class (pp. 287–311). Ithaca, NY: Cornell University Press. Wallerstein, I. (1974). The modern world system I. New York: Academic Press. Wallerstein, I. (1982). Dutch hegemony in the seventeenth-century world economy. In: M. Aymard (Ed.), Dutch capitalism and world capitalism (pp. 93–146). Cambridge: Cambridge University Press. Wallerstein, I. (1988). Comments on Stern’s critical tests. The American Historical Review, 93, 873–887. Whatmore, S., & Thorne, L. (1997). Nourishing networks: Alternative geographies of food. In: D. Goodman & M. J. Watts (Eds), Globalising food: Agrarian questions and global restructuring (pp. 287–304). New York: Routledge.

KEEPING TIME: TEMPORAL HIERARCHIES IN SOCIO-ECOLOGICAL SYSTEMS Charles H. Wood ABSTRACT The ability to identify the root causes of leading environmental crises, and the capacity to design effective intervention strategies, rely on a proper conceptualization of the linkages that join the social and natural worlds. An indispensable feature of an integrated model concerns the temporal frequencies intrinsic to the spatial and organizational hierarchies that comprise the socio-ecological system. The purpose of this chapter is to assemble ideas from a wide range of disciplines to promote a greater sensitivity to the relevance of time, and to alert us to the conceptual challenges of introducing temporal considerations into interdisciplinary environmental studies.

1. INTRODUCTION The ability to identify the root causes of leading environmental crises, and the capacity to design effective intervention strategies, rely on a proper conceptualization of the linkages that join the social and natural worlds. An Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 93–112 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10005-5

93

94

CHARLES H. WOOD

indispensable feature of an integrated model concerns the temporal frequencies intrinsic to the spatial and organizational hierarchies that comprise the socio-ecological system. The purpose of this chapter is to assemble ideas from a wide range of disciplines to promote a greater sensitivity to the relevance of time, and to alert us to the conceptual challenges of introducing temporal considerations into interdisciplinary environmental studies.

2. SOCIAL TIME, HUMAN BEHAVIOR, AND ENVIRONMENTAL OUTCOMES The notion that every social action embodies within it a subjective temporal structure has long been recognized in the social sciences. Schutz demonstrated that the social construction of meanings and motives is based on subjective anticipations of the future and remembrances of the past (Hall, 1984, pp. 208–209). Accordingly, a shared sense of temporal consciousness is a fundamental aspect of human culture. Shared subjective temporalities imprint upon its members a temporal design that has profound implications for the way individuals understand themselves and the world around them, and for the way that people exploit, manage, or conserve natural resources (Rifkin, 1987, p. 59). The environmental relevance of cultural conceptions of time is graphically illustrated by Oren Lyons’s (1980) study of the Iroquois. Members of this ethnic group share a worldview that compels them to make decisions in a manner that institutionalizes the future into their present choices. The Iroquois view themselves both as servants of the past and stewards of the future. Their ancestors counsel them from the grave, just as their unborn children speak to them from somewhere beyond the horizon. Jeremy Rifkin (1987, p. 78) cites an interview that Lyons carried out with an Iroquois Chief, who explained: ‘‘We are looking aheadyto make sure that every decision we make relates to the welfare and well-being of the seventh generation to come.’’ The seventh generation criterion, according to the Chief, serves to test the soundness of all of their major decisions. Rifkin (1987, p. 79) was quick to note the striking contrast between the expanded time horizon invoked by the Iroquois, and that of political leaders whose ‘‘concept of decision-making responsibility barely extends beyond the four-year period that marks off each new general election.’’ It appears that the Iroquois have long embraced the perspective expressed in the Bruntland Commission’s definition of ‘‘sustainable development’’ as that form of

Keeping Time

95

economic growth that ‘‘meets the needs of the present without compromising the ability of future generations to meet their own needs’’ (WCED, 1987, p. 43; see also Allen, 1980, p. 18). Like the cultures of which they are a part, temporal perspectives are subject to historical change, often in ways that reﬂect fundamental shifts in the way that people interact with nature. E.P. Thompson’s (1967) essay on the demise of feudalism, for example, documented how people’s temporal reckoning, once based on the agricultural calendar, was supplanted by notions of time dictated by industrial capitalism. In the process of industrialization, time became a commodity mechanically recorded by clocks, watches, and newly invented timepieces of various sorts, marking a radical break with earlier forms of temporal orientation. In foraging and agricultural societies, time had always been measured in relation to the tempo of biotic phenomena, to the rising and setting of the sun, and the changing seasons. With the industrial revolution, time became a function of purely mechanical devices adapted to the rhythms of the factory. Divorced from the periodicities of nature, our clocks and schedules now allow us to impose science and technology on the tempos of the biological and physical world. The result is a temporal orientation that is accelerated, predictable and expedient, predicated on an artiﬁcially controlled environment (Rifkin, 1987, pp. 9, 101, 229). Comparable in significance to the agricultural revolution, the transition to industrial and global time marks a fundamental break with the past, reorganizing the cognitive and material bases of human interactions with the world of nature. Microeconomists provide another angle on the environmental signiﬁcance of temporal subjectivities by invoking the concept of a ‘‘discount rate’’ to measure individual perceptions of the future. In microeconomic theory, it is the high discount rate in developing countries that determines, among other things, a landholder’s propensity to mine the land as rapidly as possible rather than invest for sustainability (see Wachter, 1992). Even when the discount rate is modest, values erode severely over periods of biological significance. The value of conserving resources for ecologically meaningful lengths of time is set by markets, and not by biology (Lee, 1993, p. 192). The likelihood of engaging in long-term investment, and the propensity to defer the immediate exploitation of natural capital in the interest of conservation, are enhanced when people can be sure that, at some point in the future, they can reap the beneﬁts of present investment and conservation. Conﬁdence in the ability to claim ownership over future outputs is promoted via the ofﬁcial (but sometimes informal) establishment of clear and enforced property rights. In rural areas this usually takes the form of land

96

CHARLES H. WOOD

titles, the provision of which is assumed to create a predictable future, and thereby improve the stewardship of land and natural resources. These observations suggest that microeconomic explanations based on discount rates, the property rights paradigm, as well as numerous other theses proposed to explain the human causes of environmental change, including the ‘‘tragedy of the commons’’ hypothesis, are but special cases of the broader issue of the behavioral consequences of subjective temporalities. Discount rates and property rights are themselves contingent on political realities that inﬂuence people’s relationship to land and natural resources. Studies of the migrants who trekked into the Brazilian Amazon in the 1970s and 1980s, for example, found that many of the ﬁrst wave of pioneers who cleared the forest cover to make way for agricultural plots and pastures did so with the explicit expectation of selling their ‘‘improvements’’ to the land as quickly as possible, then moving on to repeat the cycle further down the road (Schmink & Wood, 1992). The itinerant behavior typical of newcomers to the Amazon was an eminently rational response to the intense land conﬂicts that took place on the frontier. Pioneers may have longed for permanence, yet the early migrants also knew fully well that once wealthier and more powerful farmers and ranchers began to arrive on the frontier it was just a matter of time before they would become dispossessed. The strategy was to sell the value of their labor investments at the ﬁrst opportunity, then move on to other forested areas, where the cycle was repeated. The result was a process of rapid and extensive deforestation unrelated to the agricultural potential of the land. When the transience that marked the early days on the frontier gave way to the possibility of longer time horizons, the stage was set for collective action on the part of individuals and communities to resist dispossession. Numerous factors accounted for the rise of organized resistance (Schmink & Wood, 1992), yet the propensity for collective action was partly contingent on group-held conceptions of time, the inﬂuence of which has been observed in other contexts. In his study of utopian communities, for example, John Hall (1978) found that communities most likely to survive were those whose members shared an apocalyptic conception of history in which true believers construed their temporal existence as lying beyond the present in a timeless eternity. Bryan Roberts (1995) similarly drew on Robert Merton’s (1984) notion of ‘‘socially expected durations’’ to explain the behavior of immigrants to the United States. Whether immigrant groups thought of themselves as transient or permanent went a long way to explain group differences in terms of a number of factors such as the tendency to become naturalized citizens, the likelihood of buying a home, and the probability of investing in a business venture.

Keeping Time

97

The idea that socially expected durations of events may profoundly inﬂuence the behavior of individuals and social groups points to possible connections between subjective temporalities and a number of leading issues within the social sciences. As the previous examples suggest, group held ideas of the transitory or permanent character of their present situation can affect the likelihood of social movements (see Foweraker, 1995), the development of ‘‘social capital’’ as a development resource (Putnam, 1993; Portes, 1998), the exploitation of the natural environment (Wachter, 1992), and the feasibility of community-based conservation efforts (Agrawal, 1997).

3. TEMPORAL GRAINS AND TEMPORAL FALLACIES If subjective temporal horizons have been shown to inﬂuence human behavior, theoretical conceptions of time have a strong bearing on the way in which different scholarly disciplines conceptualize the real world, thereby affecting their sensitivity to environmental processes. At its most elemental level, conceptions of physical time permit such notions as ‘‘earlier than,’’ ‘‘simultaneous with,’’ and ‘‘later than,’’ relative to some ﬁxed zero point. Simple chronology, in turn, is the basis of more complex treatments of time when analysts lump temporal spans into distinct periods that share some deﬁning characteristic (Denbigh, 1989, pp. 143–144). Time is thus carved into slices of varying thickness, the relative sizes of which are entirely determined by a priori assumptions about the phenomenon of interest. Given the wide divergence of interests across disciplines, it is not surprising to ﬁnd corresponding divergences in the vocabularies of time spoken by practitioners of different specialties. Meteorologists observe oscillations in temperature and precipitation, which take place at virtually any scale, yet the distinctions they draw between ‘‘microclimate,’’ ‘‘weather,’’ and ‘‘climate’’ connote phenomena that are witnessed over increasingly larger areas and longer time spans (Urban et al., 1987, p. 120). Historians speak of eras, epochs, and stages. Economists concerned with prices and production talk of ﬂuctuations that are of varying length or, as in Kondratiev’s classic study, to cycles that last half a century. In common parlance, people refer to the short, medium, and long term, conveniently leaving aside the precise span of time invoked by each expression. Whatever the basis of periodization, all such attempts invoke objective time as a reference against which events are plotted. It is evident that the lumps of time that depict a given periodization scheme make sense only with respect to prior conceptual assumptions. Every discipline thus has a

98

CHARLES H. WOOD

penchant for a particular temporal horizon that is expressed, if only implicitly, in periodizations that are, by definition, sensitive only to a limited range of frequencies. In this sense, the time lumps constructed by a given conceptual scheme can be likened to the concept of ‘‘grain.’’ As used by geographers and landscape ecologists, grain refers to ﬁnest level of spatial resolution possible within a given data set (Turner, Gardner, & O’Neill, 2001, p. 29). By analogy, perhaps there is some utility in thinking of ‘‘temporal grain’’ as the smallest span of time that is possible within a given conceptual framework. The significance of temporal grain becomes evident in studies that ﬁnd that extending the time frame of observation can lead to radically different conclusions about the functioning of a system. Because different processes become apparent at different temporal scales, interpretations of change may shift dramatically as observational time expands. The problem is analogous to research by geographers on the ‘‘Modiﬁable Aerial Unit Problem’’ that demonstrated that the conclusions reached at one spatial scale change significantly when the same analysis is performed over different spatial partitions (Openshaw, 1984). The notion that conclusions based on data with a large temporal grain may change significantly when the same analysis is performed on data with a smaller temporal grain suggests the notion of a ‘‘temporal fallacy,’’ a concept akin to the ‘‘ecological fallacy’’ ﬁrst noted by Warren Robinson in 1950. Robinson’s (1950) landmark study demonstrated that the high correlation between the percentage of blacks and the percentage of illiterates across U.S. counties (aggregate level) did not mean that individual blacks were illiterate (individual level). The correlation at the aggregate level only meant that low-income people (black and white) tended to live counties populated by a large number blacks (literate and illiterate). Since the publication of Robinson’s seminal article, analysts have been keenly aware of the ecological fallacy of imputing aggregate spatial relationships to individual-level associations. The analogous temporal fallacy of imputing relationships based on large temporal grids to relationships that take place at more rapid frequencies is only beginning to be appreciated.

4. TIME, UNCERTAINTY, AND THE BURDEN OF PROOF Time assumes special status when it ﬁgures into the very conceptualization of prediction and explanation. For neoclassical economists the idea is

Keeping Time

99

expressed in terms of the power of the competitive market to clear away obstructions ‘‘in the long run.’’ Even if no one knows just how long that might be, time invoked in this fashion goes a long way toward insulating the paradigm against the kind of empirical ‘‘anomalies’’ that Thomas Kuhn identiﬁed as the cause of scientific revolutions. So entrenched is the notion of the long run that it provoked Keynes to make his often-quoted remark that ‘‘in the long run we are all dead.’’ His comment achieves its irony by counterpoising two different theoretical times – that of the marketplace and that of individual life spans. If some economists ﬁnd comfort in the passage of open-ended time as the ﬁnal arbiter, less patient analysts, and those skeptical of the model itself, are inclined to argue that if and when the long run does arrive, land will be degraded and countless animal species will have become extinct, along with Keynes and the rest of us. The theoretical utility of open-ended temporal horizons is not limited to neoclassical economists. Analysts in the Marxist-Althusserian tradition, in their attempt to salvage the materialist framework in the face of contradictory observations, are fond of asserting that, despite short-term anomalies, ‘‘the economic is determinant in the last instance.’’ Responding to this kind of reasoning, Ian Roxborough (1979, p. 51) noted that it raises as many problems as it appears to solve: ‘‘Like Minerva’s owl, the coming of the last instance remains a remarkably elusive event of dubious epistemological status and utility.’’ Faced with evidence that contradicts theorized expectations, what remains standing is the conceptual framework, rescued by the invocation of the last instance, a mythical future when all shall be revealed. Abstract discussions notwithstanding, real money has been won and lost on wagers that, in the end, were decided by the passage of time. In 1980, economist Julian Simon and biologist Paul Ehrlich decided to bet on their competing predictions regarding the future price of natural resources (Carnell, 2000). For years Ehrlich had been predicting massive shortages in various resources as a consequence of population growth. Simon, on the other hand, claimed that natural resources were inﬁnite, basing his reasoning on neoclassical economic principle that man-made capital is a near-perfect substitute for natural resources. Simon challenged Ehrlich to a bet based on the market price of metals. Ehrlich could pick a quantity of any ﬁve metals worth $1,000 in 1980. If the 1990 price of the metals, after adjusting for inﬂation, were more than $1,000 (i.e., the metals became more scarce), Ehrlich would win. If the value of metals after inﬂation was less than $1,000, Simon would win. Ehrlich bet on copper, chrome, nickel, tin, and tungsten. Ten years later, after the price of the ﬁve metals declined by an average of 38.2%, Ehrlich plainly lost the best to Simon.1 Despite the outcome of the wager, the

100

CHARLES H. WOOD

conclusion that Simon’s victory supported the premise that resources are inﬁnite was rejected by many, including Herman Daly (1991). For Daly (1991, p. 264) the results of the bet only showed that ‘‘speculators are interested only in the short run, or that there is a sucker born every minute!’’ An expanded time horizon, in Daly’s view, would prove Ehrlich right. Simon’s wager with Ehrlich, like Simon’s (1981) grandstand offer in his book The Ultimate Resource (in which he bet anyone any amount up to $10,000 that the real price of any natural resource will not rise), calls to mind an intractable dilemma in the ﬁeld of ecological management – the relationship between time and burden of proof. For example, those who claim that global warming is anthropogenic have difﬁculty satisfying skeptics who contend that the supporting evidence is weak and controversial, and that a warming trend is part of a long-term natural cycle. As with so many other environmental issues, the problem is that hard evidence may only be available in the future, when disaster has struck and it is too late for remedial action.2

5. HIERARCHIES OF SOCIAL TIME Although time would appear to be a central concern to social scientists, and particularly to historians, explicit attempts to theorize the significance of temporality appear to be infrequent. Notable exceptions include Charles Tilly (1984) and Fernand Braudel (1980). In the preface to his book On History, Braudel (1980) postulates three levels of historical time. The ﬁrst, which brackets the activities of individual people, is attuned to the ‘‘short and nervous’’ vibrations of daily life. But micro-history for Braudel is only ‘‘a surface disturbance, the waves stirred up by the powerful movement of the tides’’ that run at a deeper and slower pace. The tidal movements of history – those that concern the rise and fall of civilizations and the gentler rhythms of changing social structures – comprise what Braudel referred to as the longue dure´e. At a third level, time passes even more slowly, ‘‘a history which almost stands still, a history of man in his intimate relationship to the earth which bears and feeds himyas though it were somehow beyond time’s reach and ravage’’ (Braudel, 1980, p. 12). The three levels, of course, are only a manner of speaking since, as Braudel (1980, p. 74) himself notes, there are ‘‘ten, a hundred levels to be examined, ten, a hundred different time spans.’’ What Braudel leaves unresolved is the methodological challenge of linking one temporal level to another. If Braudel presents a three-tiered

Keeping Time

101

scaffolding of temporalities, and enjoins us to arrange and interpret events along each plane, his work does not present ‘‘a tight analytical model that guides the movement from one analysis to the next’’ (Tilly, 1984, p. 86). ‘‘We end our journey delighted with all we have seen, grateful to our guide’s wisdom and perspicacity, inspired to revisit some of the hidden corners he has revealed, but,’’ as Tilly (1984, p. 73) wrote, ‘‘no more than dimly aware of the master plan.’’ A more successful attempt to link temporal hierarchies is presented in Tamara Hareven’s (1982) book, Family and Industrial Time, in which the author uses life-course analysis to understand how the relatively slow cadence of capitalism, manifest in the rise and fall of profits, employment, and technological change, conditions the faster rhythms of individual and family life. For the most part, however, the social sciences tend to focus either on fast variables or on slow variables, but give short shrift to their interaction, a complaint that can just as easily be made of the natural sciences (Pritchard & Sanderson, 2002, p. 148). Cross-level relationships become increasingly salient in light of the growing attention given to such concepts as ‘‘globalization’’ and its presumed effect at the local level. While it is facile to assert the importance of tracking the effects of global and regional processes on local environmental outcomes, actually making the journey from one level of analysis to the next presents a host of challenges that too often remain implicit or unrecognized. As many analysts have noted, the task of conceptualizing a multi-leveled framework, and of developing general methodological guidelines for documenting the relationships between phenomena within and across the levels of a complex system, is one of the most vexing issues in the social sciences (e.g., Turner et al., 1996, pp. 44–45).

6. HIERARCHIES OF ECOLOGICAL TIME Explicit attention to time, space, and cross-scale analysis comes mainly from geographers but especially from ecologists who have proposed a hierarchical framework for the analysis of ecosystems (Allen & Starr, 1982). Hierarchy theory draws attention to the significance of time and frequency for identifying different levels of analysis within an ecosystem. Pine needles, for example, cycle with a generation time of 1 year, foliage cycles with a generation time of 10 years, and trees cycle with a generation time of 100 years and more (Holling, Gunderson, & Peterson, 2002, p. 71). Moreover, a fundamental principle in hierarchy theory is the notion that larger, slower levels

102

CHARLES H. WOOD

constrain behavior at lower, faster levels within the system. A forest stand, for example, moderates the climate within the stand to a narrow range of temperature variation that is experienced by individual trees (Holling et al., 2002, p. 73). Explicit attention to temporal frequencies has clariﬁed the hierarchical character of ecosystem processes. The properties observed at a higher level in the hierarchy operate so slowly that they can be considered constant, at least for analyses cast at lower levels of the system. By the same token, the rapid frequency of processes that occur at the lowest levels of the system represent little more than background noise relative to higher tiers of interest (see Fig. 1). In this way, the framework recognizes levels of interacting forces, so that ‘‘variables that appear to be drivers at one scale may seem constant at another’’ (Turner et al., 1996, p. 29). Hierarchy theory thus dissects a complex system into a series of hierarchically arranged entities and processes. A system is deﬁned as hierarchical if it is composed of more or less stable subunits that are uniﬁed by superordinate relations (Ahl & Allen, 1996, p. 32). The central idea is that it is possible to understand the behavior of the entire system by understanding

1 2

Frequency of behavior

3

0

1

2

3

4

5

6

7

8

9

10

Fig. 1. Hierarchies of Time. The Top Line (1) is a Variable that Changes so Slowly that it can be Treated as a Constant by Analyses at Level (3). The Lowest Level Represents a Fast Variable that Changes so Quickly that it would be Perceived as Background ‘‘Noise’’ by Analyses at Level (1) (adapted from Turner et al., 2001, p. 35).

Keeping Time

103

the constraints that operate above and below each of the levels that comprises the system. For any given level (call it level 0), there is a level immediately above (+1) and below it ( 1) (see Fig. 2). Hierarchy theory assumes that levels lower than 1 produce changes that are either too small or too fast to be of much significance in the measurement of processes at level 0 (reference level). Similarly, it is assumed that levels above +1 are too large and too slow to affect measurement and understanding at level 0. The levels immediately above and below the referent level produce a ‘‘constraint envelope’’ in which the phenomenon of interest is embedded (Gibson, Ostrom, & Ahn, 2000). Because predictability in complex systems is achievable only if several levels are taken into account, the chief focus is on the processes that take place within each level, as well as the cross-level dynamics that make up the system. Decades of research on regional ecosystems show that it is the interaction of slow and fast processes that establish key features of ecosystem structure (Holling et al., 2002, p. 68). Geophysical controls dominate at scales larger than 10 km. At smaller scales, biotic and abiotic processes control the structure, volume, and pattern of vegetation. These are also the scale ranges where human land use transformations mainly occur. Hence, the arenas within which plant and animal-controlling interactions unfold are the same arenas where human activities interact with the landscape. Human behavior

Larger

Slower

Level + 2

Level + 1 Level

0

Level − 1 Level − 2

Smaller

Fig. 2.

Faster

Levels of Analysis in Ecological Hierarchy Theory.

104

CHARLES H. WOOD

and terrestrial ecosystems are thus inexorably linked at critical levels within social and natural hierarchies (Holling et al., 2002, p. 15). The challenge, then, is to invite socio-economic and bio-physical levels into the same framework as a means to conceptualize the intersection of the human and ecological processes that produce the environmental outcomes we witness in the real world. Wood and Porro (2002) provide an example by depicting the spatial, temporal, and organizational hierarchies that lead to deforestation and environmental change in the Brazilian Amazon.

7. SOCIO-ECOLOGICAL HIERARCHIES AND DEFORESTATION Whereas early treatments of deforestation stressed a single causal factor, such as the effect of population growth, today it is understood that an array of variables account for the scope, pace and pattern of land use, and land cover change. History, politics, and demography are thoroughly implicated (Schmink & Wood, 1992), as are exchange rates, currency inﬂation, legal institutions, road construction, colonization schemes, tax laws, ﬁnancial markets, commodity prices, and tenure security (Capistrano, 1994; Fearnside, 2001; Hall, 2000; Homma et al., 1992; Lambin, 1994; Smith et al., 1995), to name only the more salient variables noted in the literature. It is further recognized that the bio-physical context – deﬁned by such variables as soil quality, water availability, temperature range, and the presence of pests and pathogens – mediates the way that socio-economic drivers play themselves out in a particular location (Nepstad et al., 1999; Liverman, 1990). The daunting complexity implied by these empirical generalizations underscores the need to develop a conceptual framework capable of rendering a coherent picture of the interconnections between the social and the natural worlds. Inspired by the principles of hierarchy theory, Wood and Porro (2002) proposed a three-tiered approach, as shown in Fig. 3. The framework draws a broad distinction between two systems of variables – the ‘‘socioeconomic drivers,’’ shown in the upper portion of the diagram, and the ‘‘bio-physical drivers,’’ shown in the lower portion. To conceptualize the hierarchy of driving forces within each domain, the framework further distinguishes between the ‘‘Proximate,’’ ‘‘Intermediate,’’ and ‘‘Distant’’ scales.3 The box labeled ‘‘Land Cover Outcomes’’ is positioned to the right of the ﬁgure in order to convey the idea that deforestation, as well as other forms of land cover change, are the direct result of land use decisions made by landholders, including households and commercial ﬁrms.

Keeping Time

105

Distant Socio-Economic Drivers (Global scale)

• Int’l regimes • Exchange rates • Commodity prices

• Trade policies • External debt World markets

Intermediate Socio-Economic Drivers (National/Regional scale) • Fiscal incentives • National and regional markets

• Population growth • Colonization policies • Tax laws

Proximate Socio-Economic Drivers (Local scale) • Commodity prices • Input costs • Labor costs

Land Cover Outcomes

• Transportation • Tenure security • Access to credit

Kinship networks

Land Use

Household/Firm Resource allocation

Community networks Proximate Bio-physical Drivers (Local scale/farm site) • Soil fertility • Precipitation • Geomorphology

• Pests • Pathogens • Micro-climate

• Undisturbed forest • Harvested of nontimber products • Selectively logged • Cleared for - annual crops - perennial crops - pasture - mining • Regrowth - managed fallow - abandoned plots

Intermediate Bio-physical Drivers (Landscape) • Topography • Altitude • Precipitation

• Water table • Drainage • Meso climate

Distant Bio-physical Drivers (Global scale) • Global climate • Atmospheric chemistry

Fig. 3.

Socio-economic and Bio-physical Levels in the Analysis of Deforestation in the Amazon.

106

CHARLES H. WOOD

Households (based mainly on family labor) and ﬁrms (that hire permanent labor) are positioned at the center of the framework, wedged, as it were, between the socio-economic and the bio-physical contexts. This depiction is intended to capture the idea that land managers allocate the resources at their disposal by engaging in a complex decision process that takes into account (however imperfectly) the opportunities and constraints, and the incentives and disincentives presented to them by the proximate socio-economic and bio-physical drivers. A powerful consequence of the perspective is that it allows the analyst to focus on an event at a particular level, while recognizing that there are other levels of the system that are relevant to the observed event (Urban et al., 1987, p. 123). The three-tier framework underscores the idea that processes that operate at one level of the system are embedded in contexts that operate at higher levels. An observed change in land cover – for example, the act of a small farmer felling a tree to make way for an agricultural plot – is therefore understood to be the net result of multiple forces, both social and biophysical, many of which operate indirectly and within arenas that are distant from the site of the observation itself, ranging from the local, to the regional/national, to the international/global levels. Compared to the socio-economic system, the hierarchical nature of the bio-physical domain, depicted in the lower portion of the ﬁgure, is more easily grasped. Local-level variations in the biological and atmospheric environments are understood to be embedded features of larger-scale processes that take place more slowly at the landscape, and, ultimately, at the global level. Hence, as one conceptually travels ‘‘outward’’ from the proximate, to the intermediate and distant drivers, the analytical trajectory progressively encompasses successively higher levels of ecological organization, larger areas of geographic coverage, and slower temporal frequencies. The return arrows in Fig. 3 depict the feedback mechanisms between the land use outcomes and proximate, intermediate, and distant biophysical drivers. Temporal considerations again enter the picture in as much as the pace of the feedback effects vary according to the level involved. Land cover outcomes are likely to have strong and rapid feedback effects on the proximate biophysical drivers. However, both the strength and the speed of the feedback effects grow weaker and operate more slowly as the analytical focus moves from the proximate to the intermediate and distant drivers. The varying thickness of the lines in Fig. 3 is intended to reﬂect variations in the strength and pace of the feedback mechanisms. Not all feedback effects modify ecological structure. Ecosystems are able to withstand the effects of disturbances that occur within a certain range of

Keeping Time

107

intensity. The degree of adaptive ﬂexibility in the face of external shocks is captured by Hollings’s (1973) notion of ‘‘ecosystem resilience,’’ deﬁned as the magnitude of the disturbance that can be absorbed before the system changes its structure by changing the variables and processes that control system behavior. Disturbances that are routine in the evolution of the ecosystem’s species are endogenized to become a normal feature of successional processes. Forest ecosystems can thus withstand an unusually warm day by evoking response mechanisms that operate with rapid cycles. Plants cool their leaves by transpiring more water; animals modify behavior and physiology to maintain body temperature, and so on. However, an exceptionally prolonged warming trend that exceeds the tolerance of the species pool produces a wholesale change in ecosystem structure (Westley, Carpenter, Brock, Holling, & Gunderson, 2002, pp. 106–107). It is evident from this example that a conclusion about the resilience of an ecosystem is contingent on the temporal grain of the analysis. Put another way, whether a variable is considered endogenous or exogenous depends on the time horizon the analyst uses. The socio-economic system, by contrast, is the product of the human capacity to manipulate symbols, and the ability of people to collectively invent and reinvent a meaningful order around them. Human consciousness and reﬂexivity thus permit a high degree of self-organization, and a unique capacity for rapid change. If the pace of organic evolution is determined by the rates of mutation and selection of the ﬁttest, changes in the social order occur much faster. New ideas travel quickly, and are passed on to other individuals, and across generations. Social change is essentially a ‘‘Lamarckian process’’ that has no parallel in organic evolution (Westley et al., 2002, p. 111). The human capacity for innovation and rapid change has profound implications for the natural environment. The appearance of a new opportunity to exploit nature can motivate people to rearrange their productive efforts and modify institutional and ideological arrangements with astonishing speed. Such innovations often have disastrous environmental outcomes, at least when the disturbances provoked by human action exceed the limits of ecosystem resilience. Alternatively, human ingenuity can also mitigate the negative effects of their own actions through social learning, adaptive management, and technological innovation. In further contrast to ecological hierarchies, the social system does not always conform to the ‘‘small-fast/large-slow’’ principle that applies to ecosystems. Changes often originate at higher levels, the effects of which reverberate throughout the lower levels of the system. The collapse of a country’s exchange rate, or a decline in the international price of an export

108

CHARLES H. WOOD

commodity – both at global level – can occur virtually overnight, with dramatic downward effects on regional and local economies. Similarly, national-level policies, directives, and priorities are subject to rapid change when one political party loses power to another. Similarly, the notion of a ‘‘constraint envelope’’ posited in ecological hierarchy theory may capture the idea that higher levels of a complex biophysical system constrain behavior at lower levels, as in the earlier-cited example of foliage, trees, and forests. However, the emphasis on the constraining features of higher levels within a natural system is certain to be unwarranted in complex social systems in which the cross-level relationships are looser and less deterministic. Within the social system, higher and lower order phenomena both limit as well as enable the courses of action available to social actors. Hence, in the case of social systems, we can modify the notion of a constraint envelope and consider, instead, a ‘‘decision envelope,’’ thereby accounting for the creative behavior of human actors. The decision envelope is deﬁned by the opportunities and constraints, the costs and beneﬁts, and the incentives and disincentives that confront social actors positioned at a given level within the system (level 0), and that are taken into account in resource use decisions. Each level in the system (proximate, intermediate, distant) is comprised of institutions and actors that operate according to their particular rules and temporal horizons. The hierarchies of time, schematically depicted in Fig. 1, are therefore worth keeping in mind in analyses of complex systems. A municipal politician, for example, is likely to invoke a shorter time horizon compared to those at the state and federal level. Similarly, a small farmer may base his resource allocation decisions on the rapid ﬂuctuations of the local labor market, while taking more or less as given the market for labor at the national level. The general point is that a sensitivity to the tempo of processes at different levels of a complex system, and an attempt to show how the temporalities at one level inﬂuence outcomes at another level, have the potential to advance an understanding of the system as a whole.

8. CONCLUSION The connections between natural and social systems entail a symphony of cycles, tempos, and periodicities. The temporal underpinnings of the market-based industrial age – with its standardized work routines, expanding technology, and accelerated production – is increasingly at odds with the

Keeping Time

109

rhythms of the biological and physical world. It is not surprising, therefore, that the fundamental cause of many environmental tragedies can be traced to the differences in pace between natural and social processes. It follows that effective resource management, and successful environmental action, crucially depend on a deep knowledge of the temporal complexity of socioecological systems. The discussion in this chapter is primarily an attempt to call attention to various aspects of the temporal relationships between social and natural systems. To this end, I noted the cultural definitions of time that inﬂuence the way human beings exploit the natural environment. I proposed the concepts of temporal grain and temporal fallacy as a means to highlight conceptual issues and methodological pitfalls that often go unrecognized in social-environmental research. And I applied insights from ecological hierarchy theory to develop a multi-leveled conceptual framework that illustrates the spatial and temporal complexities pertinent to understanding the causes of deforestation in the Amazon. Taken together, the various observations noted in this chapter promote the conclusion that a greater sensitivity to time in its diverse forms is a precondition for understanding critical aspects of the socio-ecological system, and to the development of policies that affect the environment across time and space.

NOTES 1. Ehrlich claimed that the price of metals fell not due to declining scarcity, but because a recession in the ﬁrst half of the decade of the 1990s slowed demand for industrial metals worldwide. Other factors may have included the disruption in the cartel that, until the 1980s, was able to keep prices inﬂated (Carnell, 2001). 2. Forays into novel legal territory have tried to link future risk to present behavior by endorsing the ‘‘precautionary principle’’ (see O’Riordan & Cameron, 1994; Costanza & Perrings, 1990). 3. The three-tiered framework shown in Fig. 3 is primarily heuristic. Hence, the focus on three levels is intended to be suggestive rather than deﬁnitive. Instead of invoking the realist notion that the three-tiered model depicts what actually exists in the material world, the approach is to be treated as an abstraction that can be adapted to different phenomena, some of which may call for additional tiers in the hierarchy, others fewer. By the same token, the lists of variables shown in the boxes that comprise Fig. 3 are intended to be illustrative of the kinds of factors to be taken into account and are not presented as an exhaustive list of relevant variables. That being said, the three-tiered framework nonetheless reﬂects the familiar micro, meso, and macro scheme, and is sufﬁciently detailed to encompass the main arguments that have been advanced to explain the process of land use and deforestation in the Amazon. More generally, the deforestation example suggests the utility of drawing

110

CHARLES H. WOOD

on hierarchy theory as a means to conceptualize the levels of the socio-economic and bio-physical systems that interact to produce environmental change.

ACKNOWLEDGEMENTS Helpful comments were generously provided by Kent Redford, Marianne Schmink, Rick Stepp, and Laura Goulart.

REFERENCES Agrawal, A. (1997). Community in conservation: Beyond enchantment and disenchantment. CDF Discussion Paper. Gainesville, Florida: Conservation and Development Forum (CDF). Ahl, V., & Allen, T. F. H. (1996). Hierarchy theory: A vision, vocabulary and epistemology. New York: Columbia University Press. Allen, R. (1980). How to save the world. Totowa, NJ: Barnes and Noble. Allen, T. F. H., & Starr, T. B. (1982). Hierarchy: Perspectives for ecological complexity. Chicago: University of Chicago Press. Braudel, F. (1980). On history. Chicago: University of Chicago Press. Capistrano, A. D. (1994). Tropical forest depletion and the changing macro-economy, 1967–85. In: K. Brown & D. W. Pearce (Eds), The causes of tropical deforestation: The economic and statistical analysis of factors giving rise to the loss of the tropical forests (pp. 68–85). Vancouver: University of British Columbia Press. Carnell, B. (2000). Julian Simon’s bet with Paul Ehrlich. http://www.overpopulation.com/faq/ People/Julian_simon.html Costanza, R., & Perrings, C. (1990). A ﬂexible assurance bonding system for improved environmental management. Ecological Economics, 2, 57–75. Daly, H. (1991). Steady-state economics. Washington, DC: Island Press. Denbigh, K. G. (1989). Physical time and mental time. In: J. T. Fraser (Ed.), Time and mind: Interdisciplinary issues. Madison, Connecticut: International Universities Press, Inc. Fearnside, P. M. (2001). Land-tenure issues as factors in environmental destruction in Brazilian Amazonia: The case of southern Para. World Development, 29(8), 1361–1372. Foweraker, J. (1995). Theorizing social movements. Boulder, Colorado: Pluto Press. Gibson, C. C., Ostrom, E., & Ahn, T.-K. (2000). The concept of scale and the human dimensions of global change: A survey. Ecological Economics, 32, 217–239. Hall, A. (Ed.) (2000). Amazonia at the crossroads: The challenge of sustainable development. London: Institute of Latin American Studies, University of London. Hall, J. R. (1978). The ways out: Utopian communal groups in an age of Babylon. London: Verso. Hall, J. R. (1984). Temporality, social action, and the problem of quantiﬁcation in historical analysis. Historical Methods, 17(4), 206–218. Hareven, T. K. (1982). Family time and industrial time. London: Cambridge University Press. Holling, C. S. (1973). Resilience and stability of ecological systems. Annual Review of Ecology and systematics, 4, 1–24.

Keeping Time

111

Holling, C. S., Gunderson, L. H., & Ludwig, D. (2002). In quest of a theory of adaptive change. In: L. H. Gunderson & C. S. Holling (Eds), Panarchy: Understanding transformations in human and natural systems (pp. 3–22). London: Island Press. Holling, C. S., Gunderson, L. H., & Peterson, G. D. (2002). Sustainability and panarchies. In: L. H. Gunderson & C. S. Holling (Eds), Panarchy: Understanding transformations in human and natural systems (pp. 63–102). London: Island Press. Homma, A. K. Y., Walker, R. T., Scatena, F., de Conto, A. J., de Amorim Carvalho, R., de Rocha, A. C. P., Palheta, C. A., Dos Santos, A. I. M. (1992). A dinaˆmica dos desmatamentos e das queimadas na Amazoˆnia: uma ana´lise microeconoˆmica. Unpublished manuscript. Lambin, E. F. (1994). Modeling deforestation processes: A review. Research Report No. 1, European Commission Joint Research Center/European Space Agency, Brussels. Lee, K. N. (1993). Compass and gyroscope: Integrating science and politics for the environment. Washington, DC: Island Press. Liverman, D. M. (1990). Drought impacts in Mexico: Climate, agriculture, technology, and land tenure in Sonora and Puebla. Annals of the Association of American Geographers, 80, 49–72. Lyons, O. (1980). An Iroquois perspective. In: C. Yecsey & R. W. Venables (Eds), American Indian environments: Ecological issues in native American history. New York: Syracuse University Press. Merton, R. K. (1984). Socially expected durations: A case study of concept formation in sociology. In: W. W. Powell & R. Robbins (Eds), Conflict and consensus (pp. 262–286). New York: The Free Press. Nepstad, D. C., Moreira, A. G., & Alencar, A. A. (1999). Flames in the rain forest: Origins, impacts and alternatives to Amazonian fires. Brası´ lia, Brazil: Pilot Program to Conserve the Brazilian Rain Forest, Quick Printer. O’Riordan, T., & Cameron, J. (Eds) (1994). Interpreting the precautionary principle. New York: Island Press. Openshaw, S. (1984). Concepts and techniques in modern geography number 38: The modifiable areal unit problem. Norwich: GeoBooks. Portes, A. (1998). Social capital: Its origins and applications in modern sociology. Annual Review of Sociology, 24, 1–24. Pritchard, L., Jr., & Sanderson, S. E. (2002). The dynamics of political discourse in seeking sustainability. In: L. H. Gunderson & C. S. Holling (Eds), Panarchy: Understanding transformations in human and natural systems (pp. 147–169). London: Island Press. Putnam, R. (1993). Making democracy work: Civic traditions in modern Italy. Princeton, NJ: Princeton University Press. Rifkin, J. (1987). Time wars. New York: Simon and Schuster. Roberts, B. (1995). Socially expected durations and the economic adjustment of immigrants. In: A. Portes (Ed.), The economic sociology of immigration (pp. 42–86). New York: Russell Sage Foundation. Robinson, W. (1950). Ecological correlations and the behavior of individuals. American Sociological Review, 15, 351–357. Roxborough, I. (1979). Theories of underdevelopment. London: MacMillan Press. Schmink, M., & Wood, C. H. (1992). Contested frontiers in Amazonia. New York: Columbia University Press. Simon, J. (1981). The ultimate resource. NJ: Princeton University Press.

112

CHARLES H. WOOD

Smith, N. J. H., Serra˜o, E. A. S., Alvim, P. T., & Falesi, I. C. (1995). Amazonia: Resiliency and dynamism of the land and its people. New York: United Nations University Press. Thompson, E. P. (1967). Time, work-discipline and industrial capitalism. Past and Present, 38, 56–176. Tilly, C. (1984). Big structures, large processes, huge comparisons. New York: Russell Sage Foundation. Turner, M. G., Gardner, R. H., & O’Neill, R. V. (2001). Landscape ecology in theory and practice. New York: Springer. Turner, B. L., II, Skole, D., Sanderson, S., Fischer, G., Fresco, L., & Leemans, R. (1996). Land use and land-cover change: Science/research plan. Geneva: The International GeosphereBiosphere Programme (IGBP) and the Human Dimensions of Global Environmental Change Programme (HDP) of the International Social Science Council (ISSC). Urban, D. L., O’Neill, R. V., & Shugart, H. H., Jr. (1987). Landscape ecology: A hierarchical perspective can help scientists understand spatial patterns. BioScience, 37(2), 119–127. Wachter, D. (1992). Farmland degradation in developing countries: The role of property rights and an assessment of land titling as a policy intervention. Madison: University of WisconsinMadison, Land Tenure Center Paper 145. WCED (The World Commission on Environment and Development). (1987). Our common future. New York: Oxford University Press. Westley, F., Carpenter, S. R., Brock, W. A., Holling, C. S., & Gunderson, L. H. (2002). Why systems of people and nature are not just social and ecological systems. In: L. H. Gunderson & C. S. Holling (Eds), Panarchy: Understanding transformations in human and natural systems (pp. 103–119). London: Island Press. Wood, C. H., & Porro, R. (2002). Deforestation and land use in the Amazon. Gainesville, FL: University Presses of Florida.

CYCLES OF ACCUMULATION, CRISIS, MATERIALS, AND SPACE: CAN DIFFERENT THEORIES OF CHANGE BE RECONCILED? Denis O’Hearn ABSTRACT In this chapter I outline different theories of ascent to and decline from hegemony. In particular, I discuss Arrighi’s finance-based theory of cycles of accumulation and Stephen Bunker’s materio-spatial theory of hegemonic ascent and decline. Then, I ask whether the different explanations of ascent and decline can be reconciled and/or how we can begin to adjudicate between them in terms of their relative explanatory powers throughout the history of the world-system or at different times in history. Finally, I discuss the implications of theories of hegemonic change for our understanding of local economic change, and particularly the relationship between path dependencies and points of exit from them.

1. INTRODUCTION: HEGEMONY AND ACCUMULATION In this chapter I outline different theories of ascent to and decline from hegemony. In particular, I discuss Arrighi’s ﬁnance-based theory of cycles of Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 113–137 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10006-7

113

114

DENIS O’HEARN

accumulation and Stephen Bunker’s materio-spatial theory of hegemonic ascent and decline.1 Then, I ask whether the different explanations of ascent and decline can be reconciled and/or how we can begin to adjudicate between them in terms of their relative explanatory powers throughout the history of the world-system or at different times in history. Finally, I discuss the implications of theories of hegemonic change for our understanding of local economic change, and particularly the relationship between path dependencies and points of exit from them. Hegemony has been deﬁned in various ways but it usually implies overwhelming productive, commercial, and ﬁnancial domination of the worldeconomy. Such economic power implies political and military domination as well. And it also implies a level of comparative advantage that enables the hegemon to produce core commodities more cheaply than its competitors who are yet to make a higher profit nonetheless (Wallerstein, 1984; Chase-Dunn, 1989, pp. 169–171). Hegemons appear to come and go in cycles, and their going is caused by a combination of at least one central contradiction in the regime of accumulation upon which they ascended along with the inability to overcome that contradiction as well as at least one hegemonic contender. The loss of hegemony, which Arrighi (1994) and others have associated with cycles of accumulation, is thus a result of periodic capitalist crises. It is surprising that a literature that relies so heavily on explanations of contradiction/crisis and resolution/ascent adverts so little to the existing literature on capitalist crisis. For decades, Marxian political economy has debated the sources of crises and countertendencies to crises, from both theoretical (including mathematical) and statistical perspectives (on different Marxian perspectives of crisis see, Weisskopf, 1979; Mandel, 1990; Glyn, 1990; Moseley & Wolff, 1992). There are many shortcomings in this literature, ranging from an over-reliance on abstract theory (often to support what appears to be a preconceived afﬁnity to one perspective or another) to the use of national level (particularly the U.S. and British) data to explain crises that are, essentially, global in scope. Yet a number of authors have made valiant attempts to use historical data to adjudicate between one theory of crisis and another. Although marked by elements of path dependency, the outcome of a crisis and the coming of a new expansionary cycle led by a new hegemon is best viewed as an exercise in iterative problem-solving rather than a historical law (see, Haydu, 1998; O’Hearn, 2001). The successful ascendant is the one that most effectively solves the central contradiction(s) of the crisis, in conditions of extreme uncertainty that often include not just competition but

Cycles of Accumulation, Crisis, Materials, and Space

115

war. In doing so, it must mobilize not just its own ﬁrms and population but also the populations and resources of multiple localities throughout the world-system. There are several possible outcomes to crises and hegemonic contention. The most straightforward possibility is that the successful hegemonic contender identiﬁes the key or central cause of crisis and introduces solutions (they may be good solutions or simply sufﬁcient and better than all the rest from a hegemonic point of view). Alternatively, it may introduce certain ‘‘solutions’’ to the general crisis, perhaps even to solve a different problem than that which is the ‘‘central’’ cause of crisis that precipitated the previous hegemon’s decline, which simultaneously solve several problems. Perhaps the new hegemon is just so powerful in the face of the weakness of contenders (the U.S. following the Second World War) that it is the only real choice in town. Even so, it will have to solve problems of resource access, labor access, market access, and technical innovation. It is important to remember that the iterative problem that must be solved is not just two-way between hegemonic contenders from core regions. The new hegemon must also recreate conditions of domination over peripheral and semiperipheral regions (and, indeed, over other key core nations), whether it be to gain access to raw materials, to open up new markets, to obtain cheaper labor, or (probably) some combination of all of the above. Thus, the importance of hegemony – both the nature of its origin in crisis and its solution in terms of a new competitive advantage – depends heavily on the reason why one wants to understand it. For activists in a given locality, certain aspects of crisis and of hegemonic power may be salient even though they may not be considered to be its ‘‘central’’ deﬁning feature. Both hegemonic contenders and localities are involved in phased process of iterative problem-solving. The hegemonic contender attempts to maintain or ascend to a position of power within the world-economy, while its major ﬁrms attempt to increase their profits by producing and selling more commodities/services more cheaply. And different localities and groups within them are either attempting to achieve upward mobility (e.g., through industrializing), to attain class advantages, or (defensively) to minimize the difﬁculties of a new hegemonic regime for its residents.

2. ARRIGHI’S CYCLES OF ACCUMULATION Arrighi (1994) gives a compelling account of how successive regimes attained and lost hegemony identifying key characteristics that enabled their ascent and led to their eventual descent. He identiﬁes four systemic cycles of

116

DENIS O’HEARN

accumulation centered on Genoa (15th–early 17th centuries), Holland (late 16th–18th centuries), Britain (last half of 18th–early 20th) and the United States (late 19th–present). Each cycle included identiﬁable phases of material expansion (establishment of new trade routes and incorporation of new areas of commercial exploitation) and financial expansion (consolidation of capital’s dominance over the enlarged world-economy). A clearly identiﬁable class beneﬁted from both expansions within each cycle. And each cycle included new organizational forms along with transformed revivals of previously superseded forms. These patterns of continuous and discontinuous changes, expansions and restructurings, were led by alliances of state and business leaders who were ‘‘uniquely well placed to turn to their own advantage the unintended consequences of the actions of other agencies’’ (p. 9). Each new ascent to hegemony follows the specific nature of decline of the previous hegemon. British ascent to hegemony was enabled by contradictions of the previous Dutch regime. With the aid of a strong state coercive apparatus, Dutch capital successfully centralized the storage and exchange of the most strategic supplies of European and world commerce in Amsterdam, which became the central money and capital market of the European world-economy. The Dutch merchant class led and governed the European capitalist engine, largely due to their control of critical trades such as Baltic grain and naval stores. Large-scale joint stock companies like the Dutch East India Company (est. 1602) exercised exclusive trading and sovereignty rights over huge overseas commercial spaces from 1610–1620 to 1730–1740. But Holland’s success led other Europeans to imitate its trade, warmaking, and state-making techniques, eventually drawing competition and diminishing returns. War making increased state demands throughout Europe for money and credit, allowing the Dutch to use their competitive advantages in ﬁnance to capture rising returns. But the imperial success of the Dutch East India Company strengthened its managerial bureaucrats, who diverted the company surplus into bureaucratic expansion and managerial corruption at the expense of stockholders. As rates of return fell in Dutch enterprise, English stocks and shares became more attractive and surplus Dutch capital ﬂowed to England, ﬁnancing its own ascent as an economic and military rival to Holland. Britain followed the Dutch example by wresting control of world trade from competing European powers like the French. It bought to sell and took in to send out, but increasingly sent out goods that were reformed through manufacture. The key to British success was its use of state (especially sea) power to control and shift the core axes of trade away from Amsterdam.

Cycles of Accumulation, Crisis, Materials, and Space

117

The English Channel became the place where American and Asian commodities met Baltic supplies. Movement toward rule of the high seas began with the consolidation of royal power after the war of roses, accelerated with the construction of a superior English ﬂeet under Henry VIII and Elizabeth I, and came into its own with the English defeat of the armada in 1588. The English beat Holland at its own mercantilist game. With the aid of marauding naval ﬂeets and joint-stock companies (the Levant, East India, Royal African, and Hudson Bay companies), the English merchant class expanded their foreign investments through privateering and adventure. Unlike Holland, England collected a far-ﬂung empire, which was both agroindustrial and commercio-ﬁnancial. Its role as clearinghouse for the worldeconomy outlasted its role as workshop for the world, but industrialization helped it act as entrepot on a far larger scale than Holland. According to Arrighi, the basis of British hegemony was its control of commerce and ﬁnance. Its recycling of plunder beginning with Elizabeth allowed England to maintain ‘‘sound money’’ and long-term monetary stability. This enabled it to become creditor and ﬁnancier of the world-economy to a much greater extent than Holland or Genoa before. Moreover, the project was achieved by and on behalf of the merchant-adventurer class, establishing the commercio-military alliance at the center of the British regime. The rise of plunder-ﬁnanced expansion coincided with the revival of European trade in wool textiles, which provided a leading sector to impel ﬁnance, commerce, and industry together. Thus began the crucial alliance between English banking capital, merchant capital, and the state. As this integrated economy began to expand more rapidly at the end of the 16th century, industry was key ‘‘not per se but as an instrument of capital accumulation’’ (Arrighi, 1994, p. 194). In this process of material expansion, English industry moved beyond massproduced goods into high value luxury and armaments production: silk, glass, and cannon (Hill, 1967, pp. 63; 71–75). Arrighi claims that the Elizabethan strategy to redirect industrial expansion from cloth to luxuries showed an understanding that industrial expansion translates into expansion of national wealth and power only when associated with breakthrough into high valueadded activities. English expansion, however, was still limited by its subordination to Dutch commercial supremacy. Dutch superiority in cloth dyeing and ﬁnishing shifted the centers of highest textile profitability from England to Holland. English attempts in the early 1600s to capture cloth-ﬁnishing processes failed when the Dutch simply cut off British cloth from the allimportant Baltic trade. This is a clear example of the link between commerce and industry. As Israel (1989, p. 410) argues, ‘‘Dutch superiority in dyeing,

118

DENIS O’HEARN

bleaching, grinding, and reﬁning was hard to challenge when it was the Dutch who had the stockpiles of dyestuffs, chemicals, drugs, and rare raw materials on which all these processes depended’’. England, then, had to remove the entrepot from Amsterdam if it was to capture competitive industrial advantage. The English merchant class and its state allies ﬁnally challenged Dutch commercial supremacy by capturing the Atlantic triangular trade. This brought unprecedented material expansion based on tobacco, sugar, cotton, gold and, especially, slaves (and on the labor required to produce them). Where Dutch commercial supremacy was based solely on a capitalist logic of power (MTM’), which invested money in a given unit of territory to acquire additional money, English commercial supremacy added a territorialist logic of power (TMT’)2, where money was used as a means of expanding territorial control. By limiting its territorial expansion to essential trading ports, Holland avoided problems of administering large territories and populations. But it also became dependent on the entrepreneurship and labor of foreign countries that were outside of its direct control. The English, on the other hand, built the Atlantic economy on direct colonial control of production. England used the Navigation Acts (1651–1660) and other key policies to build superiority in sea commerce and gain effective control over the resources of new territories, which enabled its Atlantic commercial empire to outcompete the Dutch Baltic-centered economy. Arrighi argues that England superseded the Dutch by internalizing its production costs through industrialization and colonization – as Holland had superseded the Genoese in the previous cycle by internalizing ‘‘protection costs’’ within its armed chartered trading companies (Steensgaard, 1974). England brought production within the empire and subjected it to the economizing that was enabled by innovations in the organization of commerce and industry including, crucially, economies in the transactions costs of distribution and transport. By the mid-18th century these English advantages pushed the Dutch into ﬁnance and investing in English production and British empire, even ﬁnancing the Seven Years War (1756–1763) during which England established superiority over France. English industrial/commercial ascent was a path-dependent process that spanned the previous Genoese and Dutch cycles of accumulation. Industrial ascent began with the expansion of (woolen) textiles after Edward III destroyed the Flemish textile industry. Under Elizabeth, it diversiﬁed into metal and luxury industries after England failed to capture high-profit stages of the textiles commodity chain. England became a hegemonic contender when it integrated its industrial powers with overseas commercial and

Cycles of Accumulation, Crisis, Materials, and Space

119

territorial expansions, culminating in the industrial revolution. England was more competitive in each successive phase because of the industrial, commercial, and military techniques it built up in previous phases. In Arrighi’s (1994) words, ‘‘each moment of industrial expansion in England was integral to an ongoing ﬁnancial expansion, restructuring, and reorganization of the capitalist world-economy, in which England was incorporated from the very start’’ (p. 209). England, already the greatest industrial power in Europe, was unbeatable once it controlled world trade. Despite its strength, British hegemony had its own contradictions. Its success invited rivalry. The profitability of its machinofacture, especially after cotton, depended on ever-increasing supplies of raw materials and everincreasing foreign markets. Its attempts to protect domestic technologies, even as it promoted global free trade in other spheres, limited the markets for British capital goods while encouraging European and U.S. capital to produce their own technologies. Already the German and U.S. competitors were developing integrated corporate structures that would challenge British competitive advantages. But the main threat to British hegemony was interstate rivalry, ﬁrst over the territories of the non-European world and then in world wars centered in Europe itself. Such rivalries raised protection costs over and above their beneﬁts for England and its European rivals. Britain’s wartime demands for armaments, machinery, and materials became so great that its main supplier, the United States, turned from being a net debtor to build up substantial claims on British incomes and assets. The U.S. productivity outstripped the British and European economies and foreign assets ﬂowed into U.S. ownership. By the end of the Second World War, the U.S. enjoyed a ‘‘virtual monopoly of world liquidity’’ (Arrighi, 1994, p. 275). Its favorable geographic position between the European and Asian trading blocs and the strength of its vertically integrated corporations enabled it to exercise hegemonic control of the world-system. At the world-systemic level, it did this through the establishment of a new regulatory regime for world trade and money (Bretton-Woods, the IMF, the World Bank). In localities and regions, it used the Marshall Plan and other regional or bilateral agreements to claim free entry not only into European and Asian markets but also into the former colonies (Bunker & O’Hearn, 1993). This was material expansion within a new cycle of accumulation, centered on the U.S. foreign direct investments in manufacturing, which later was branded the new international division of labor (Frobel, Heinrichs, & Kreye, 1980). A major reason for the relative U.S. strength was that it internalized transaction costs through an ‘‘organizational revolution’’ of corporate

120

DENIS O’HEARN

managerial hierarchies, which began with railroads in 1850s and totally transformed the U.S. enterprises by the start of 1900s (Chandler, 1977). Corporations that instituted these structures imposed organizational barriers to new entrants in leading branches of the U.S. economy. Once they exploited opportunities for domestic vertical integration, the U.S. corporations, with the aid of the expansionist U.S. state, moved into foreign countries to increase their powers of accumulation. Vertically integrated corporations could economize on costs of moving intermediate inputs through commodity chains (Coase, 1937; Williamson, 1970; Chandler, 1977). Transnational corporations (TNCs) manipulated internal transfer prices to reduce local taxes and resource royalties. Direct sourcing of raw materials and semifabricates enabled economies of speed by which companies with greater throughput could amass surpluses even without significant advantages in average profit rates. This was especially true of sectors such as aluminum where resource extraction and mass production could be associated with mass distribution in a single enterprise (Barham, Bunker, & O’Hearn, 1994). The U.S. also achieved hegemonic advantage by re-externalizing protection costs. Rather than controlling areas through direct or indirect colonialism, the U.S. spread the costs of regulation across the core states through their participation in interstate bodies (GATT, the IMF, the UN, NATO). Costs of administering colonies and former colonies where the U.S. opened up new investments were borne by the colonial power and the local state. Rearmament to police the new order, internally and against the Soviet threat, provided an additional economic boost for the U.S. and for consumers of its industrial products. These factors made the period from the Korean war to the 1970s oil crises ‘‘the most sustained and profitable period of economic growth in the history of world capitalism’’ (McCormick, 1989, p. 99).

3. BUNKER’S MATERIO-SPATIAL MODEL OF ASCENT So far, I have summarized Arrighi’s model of hegemonic ascent/decline in associated cycles of accumulation. The key aspect of Arrighi’s model is his placement of high ﬁnance at the center of hegemonic ascent and decline, as the driving force of cycles of accumulation. Having said that, however, Arrighi also appears to have an eclectic model of crisis. Each hegemon faces

Cycles of Accumulation, Crisis, Materials, and Space

121

its specific limiting problem that drives down profit rates, and the replacing hegemon is the one that is best placed to attract investments (i.e., it is where ‘‘high ﬁnance’’ moves its money). This is perhaps ironic insofar as Arrighi appears to agree with Wallerstein’s (1984) point that the one function the old hegemon retains is its position as ﬁnancial hub of the world-system. Rather, the captains of high ﬁnance are patriotic to profits rather than their homeland, and they desert the sinking ship for the ascendant power. For its part, the successful ascendant achieves dominance by combining new forms of organization with older forms from previous cycles. Arrighi’s explanation is quite attractive in a historical sense: success follows a creative combination of the new and the old, in ways that create a qualitatively more efﬁcient mode of controlling the world-economy or those parts of it that are under the new hegemon’s control. Stephen Bunker (often writing with Paul Ciccantell) largely agrees with Arrighi about the importance and shape of historical succession of hegemons (although he begins with Portugal rather than Genoa). In the absence of sufﬁcient empirical data for rigorous econometric estimation, Bunker’s (and Ciccantell’s) method, like Arrighi’s, is comparative historical. They search for and compare ‘‘explanatory regularities and mechanisms’’ across several cases of hegemonic decline and ascent. Like Arrighi, they consider that ‘‘expanded production at lowered unit cost constitutes the most basic mechanism of national development’’ and, therefore, that is where one would look to ﬁnd both the crisis and solutions for the problem of ascent and descent in hegemony, which is at the center of a cycle or regime of growth/development. Hegemons lose position when they can no longer control the expanding and accelerating circuits of capital. They are replaced by the ascendant that most successfully transforms the world-system to its own advantage to overcome the contradictions that have impeded the control of accumulation by the previous hegemon. Bunker and Ciccantell search for regular mechanisms in this process of decline and ascent by triangulating: (a) abstract categories of, and implicit laws of motion in, world-system analysis; (b) recent proposals about how mechanisms and regularities conﬁgure into unique events rather than universal outcomes; and (c) ‘‘materially speciﬁed processes of industrial transformation’’ in each case of ascent (Bunker & Ciccantell, 2005). Crucially, they ‘‘root their search’’ for explanatory mechanisms and regularities in the material process of production. This, it could be said, determines the outcome of their search for mechanisms, because they deﬁne material mechanisms (indeed, materials) as central to the process of accumulation. Then, if they ﬁnd them and can show they are important in each historical case of ascent,

122

DENIS O’HEARN

they declare them to be the central constraint on the existing regime and the central problem for which the best solution is rewarded with ascent to a new hegemony. Thus, they take issue with Arrighi for putting high ﬁnance at the center of his model of hegemonic ascent and its embedded cycles of accumulation. Like David Harvey before him, says Bunker, Arrighi emphasizes capitalist crises based on overaccumulation leading to a falling rate of profit, where the ‘‘spatial ﬁx’’ for crisis is for big investors (high ﬁnance) to move their money outward to regions where higher rates of profit can still be achieved. This, argues Bunker, misses the real logic of outward spatial movement. By emphasizing ﬁnance, Arrighi ignores the prior and crucial mechanism of material production. Especially, he fails to problematize materio-spatial constraints on accumulation. Expanded reproduction requires more and more materials and, to get them, producing regions must go further and dig deeper for attractive resources. This makes it more difﬁcult and expensive not just to extract resources but also to move them efﬁciently from the point of extraction to the point of production. Moreover, as production expands and technology advances, the requirement heightens for materials of a certain quality, thus intensifying the materio-spatial problem even further. Thus, the productive region/nation that succeeds in economic ascent and, ultimately, in dominating world trade (Bunker’s definition of hegemony) will be the region that most succeeds in capturing ‘‘economies of scale and space’’ in the extraction and bulk transport of materials. More specifically, it will create these economies in the extraction and bulk transport of key raw materials for the shipping industry, which Bunker identiﬁes as the (repeated) generative sector of the ascendant economy. Together, the capture of supply of raw materials for the shipping sector that moves them creates a ‘‘virtuous cycle’’ of accumulation that not only grows in itself but which is also linked to other sectors throughout the economy. Because the capital needs for infrastructure are so high, the whole system is subsidized by the state, which develops a supportive partnership with key industrial ﬁrms, without which they would be unable to sustain the rapid levels of expansion. Although Bunker spends less time than Arrighi in modeling the crisis/ contradictions of a maturing hegemon, his conception of hegemonic decline is based on a combination of two things. First, it faces problems achieving the economies of scale and space necessary for extracting and moving bulk raw materials for larger ships and associated industries from larger distances (its technologies of shipping and its state-supported institutions of raw materials access reach their limits). Second, and simultaneously, another

Cycles of Accumulation, Crisis, Materials, and Space

123

economy has achieved economic ascent based on better organization and technologies of materials supply, and thus outcompetes the old hegemon. The transition takes different specific historical forms. Britain ‘‘displaces’’ Holland by building bigger/better warships. The U.S. displaces Britain because it owns many of its raw materials sources, on which the latter has become dependent. Japan displaces the U.S. because it has developed a more efﬁcient contractual means of achieving supply of more materials at a cheaper cost. And in all cases, the new technologies of shipping are simply ‘‘bigger’’ in the sense that they can move more materials for subsequently bigger shipping and linked industries. In one ﬁnal, ironic twist, Bunker also suggests that the success of each hegemonic ascendant begins with its use of the old hegemon’s technologies and with it being placed by the old hegemon in a key role that enabled it to build its generative sectors. Thus, each hegemon is the creator of its own succession. The Dutch do heavy bulk hauling for Portugal; the British recruit Dutch shipbuilding craftsmen and steal Dutch ships; the U.S. uses British technology and capital in steel and rail; the Japanese use U.S. shipbuilding and steel technology and U.S. diplomatic support to gain raw materials access (these processes are most distinctly laid out in Ciccantell and Bunker, 2003). Besides taking issue with other models of accumulation that are not directly based on materials, Bunker also disputes recent theorists of globalization who assume that the latest phase of accumulation has been based on dematerialization (see Colombo, 1988 but similar conclusions may be inferred by, e.g., Castells, 1997). This is an illusion, he claims, and the importance of new strategies for much larger-scale extraction and bulk transport of materials shows that the most recent phase of capitalist accumulation is not dematerialized but rather the most materially intensive and spatially most expansive yet. Although Bunker does not provide evidence, he appears to be arguing not only that each successive cycle uses more materials in absolute terms but also in relative terms: i.e., each successive cycle is more materials intensive. Historically, each new phase of hegemonic decline (crisis) and ascent (new regime) has a different specific technological and social organizational solution. Yet in each case the crisis is the same (materio-spatial constraint) and the general solution is the same (ﬁnding economies of scale and scope in the extraction and transport of bulk materials for shipping). Even the generative sector is the same (shipping). Where Britain reaches a territorial solution by bringing material-producing regions into empire (formal and informal), while simultaneously improving its technologies of extraction and transport,

124

DENIS O’HEARN

the U.S. accesses materials through the reorganization of vertically integrated ﬁrms with yet better technologies of extraction/transport and a more densely networked materials/industrial complex. Finally, Japan (and perhaps China) ascends by attaining access through ‘‘new forms of investment’’ and even bigger and better technologies of extraction and transport (Bunker & Ciccantell, 2005). Notice that two things are happening simultaneously. On the one hand, ascending nations achieve a qualitative reorganization of how they mobilize trade and investment to access reserves of materials – especially the dominant bulk material of a given cycle, whether wood or iron or steel. For example, U.S. ﬁrms in association with the U.S. state achieved access to critical materials through displacement of colonial control by foreign direct investments. Japanese ﬁrms achieved access through ‘‘new forms of investment’’ where the state guaranteed multiple sources of access mainly through joint ventures between Japanese conglomerates and local mining companies (Bunker & O’Hearn, 1993). On the other hand, in each successive cycle, technology gets quantitatively bigger and better at handling bulk transport of greater amounts of materials from further distances. ‘‘Economies of scale thus created their own contradiction in the rising cost of distance across space, the reiterated solutions to this contradiction could only be achieved through even greater economies of scale, which then drove further scale increases in consumption of raw materials, reiterating the contradiction anew.’’ Space expands from river basin and coastal; to lake–river–canal systems and continental; to oceanic and global. Scale expands both at the site of extraction (bigger deposits and mines) and in the capacity/efﬁciency of transport. Again, Bunker shows us through the example of the Amazon how the Japanese used bigger ships and bigger ports (at each end) to achieve economies of scale in the raw materials for steel and aluminum production that were unique in the world at its time. In addition, it monopolized them so that contending powers such as the United States could not hook on to these economies. The two movements, qualitative and quantitative, are connected because Bunker’s is not a crude materialist account. Behind each hegemonic innovation is a series of new social, economic, and political relationships that reorganize trading and productive structures. Hegemons need bigger ships and ports, but they also need more effective ways of organizing people and ﬁrms to manage extraction, transport, and trade. Bunker’s cycle of hegemonic change begins with a materio-social constraint on essential raw material(s). Although his examples differ, it appears that essential here refers to materials that are necessary to solve the problem

Cycles of Accumulation, Crisis, Materials, and Space

125

of raw materials access and transport and not necessarily for the leading industrial sector of a hegemon. That is, problems arise in accessing and transporting the raw materials that are necessary for the leading transport technology (usually ships) and infrastructure. Rising costs of accessing materials make capital from the existing hegemon less competitive (at least) and probably induce a crisis in the form of a falling rate of profit. These costs include the direct costs of construction of vessels and of infrastructure for transport, bulk-break handling, and storage. And they include the indirect costs of protecting the materials against losses through attack or loss of property rights, etc. Both these become more problematic through time because capital must go further for supply of materials of a sufﬁcient quality and quantity and because the legal/social structures under which such materials are extracted and transported are outside of the sovereign control of the dominant national economy. In response to the crisis (which is simultaneously opportunity for hegemonic contenders), state/capital alliances from a small group of countries that aspire to ‘‘trade dominance’’ attempt to mobilize social and cultural organizations so that they can achieve innovations that enable them to solve the materio-spatial constraint. This they do by capturing economies of scale and space through innovations in technologies of extraction and transport, markets, ﬁnancial instruments, managerial organization, and labor skills. The contending power that captures economies of scale and space ascends to become the new hegemon until new socio-material constraints arise at the now higher level of materio-spatial intensity at which it is dominating the world-system. In place of high ﬁnance, then, Bunker puts raw materials at the center of his own model of hegemonic ascent and decline. Each cycle of hegemonic ascent, decline and then ascent by a new hegemon is characterized by (1) a dynamic whereby hegemonic contenders must overcome materio-spatial constraints and (2) in doing so they make the world-economy more material-intensive and the spatial problems of access to raw materials more extreme. Thus, in the solution to the materio-spatial problem the hegemon sows the seeds of its own destruction and replacement. Bunker uses a careful historical and sectoral analysis of a succession of hegemons – from Portugal to Amsterdam to Britain to the U.S. to Japan, but all with respect to the Amazon – to demonstrate (1) that solving the materio-spatial constraint is central to hegemonic ascent and (2) that each new hegemon intensiﬁed the problem anew by bringing on more materials from more far-ﬂung places. This is both the central strength and weakness of Bunker’s analysis. His focus on the conditions of extracting and transporting materials in an

126

DENIS O’HEARN

expanding and intensifying series of cycles of material production enable him to introduce a (perhaps the) key variable that has been missing from most analyses of cycles of accumulation. It is also mostly missing from Marxian crisis theory, despite some years of attempts to merge Marxian and ecological analyses (e.g., Foster, 2000). By historically identifying the process of accessing and moving materials, Bunker shows that the ‘‘materiospatial’’ problem is a key problem that each successive hegemon must solve and not just an adjunct, for example, to the labor theory of value. Yet Bunker is also constrained by his a priori selection of the materiospatial problem as his focus. Since he places his focus on it and ﬁnds that it is present in every repeated cycle of hegemonic decline and ascent, and because the materials problem is ‘‘prior to’’ other causes of crises in commodity chains, he therefore assumes that it is the central problem that must be solved in each case. Had he chosen to concentrate on another variable (as does Arrighi, for example, with high ﬁnance) might he not have found that variable to be central? In other words, by opening his eyes to the materiospatial problem, Bunker risks losing sight of the importance of other problems or other possible sequences of crisis and solution.3 This is ironic because Bunker and Ciccantell make such strong claims in their methodological work about the importance of uncovering regularity and variance, of considering the likelihood of multiple causation.4 Yet, unlike Arrighi, both their causes of crisis and their solutions remain invariant (see Table 1). Hegemons fall because they fail to resolve the problem of extracting bulk materials and moving them cheaply across longer distances. New hegemons rise because they do resolve the materio-spatial problem. And so it goes, from one hegemon and one cycle of accumulation to the next. Where Arrighi looks at historical situations and then derives crisis – thereby running the risk of identifying a superﬁcial cause by not looking deep enough – Bunker and Ciccantell begin with the materio-spatial problem as their hypothesized source of crisis and ascent and then identify it historically and show that it is important. In doing so, they risk mis-specifying their model from the beginning, by demoting or even excluding contending explanations of crisis and ascent, or, even if the crisis is one of materials access, excluding solutions that may not be optimal from the materio-spatial standpoint but which are optimal from the broader perspective of restoring the general profit rate. In other words, even if the materio-spatial problem is the central cause of crisis, there may be other secular forces that drive down profit rates, and the successful ascendant may be the one that solves several problems simultaneously instead of a single problem best. In this regard, it is

Cycles of Accumulation, Crisis, Materials, and Space

Table 1.

127

Arrighi’s and Bunker’s Models of Hegemonic Decline and Ascent. Basis of Hegemony

Arrighi’s Schema of Hegemonic Rise and Decline Holland (from Genoa) Internalizing protection costs (chartered company) Britain Internalizing production costs (colonial empire) U.S. Internalizing transactions costs; Re-externalizing protection costs (TNC, international monetary system) Bunker’s Schema of Hegemonic Rise and Decline Holland (from Portugal) Control of river routes for raw materials for shipbuilding; state support for Dutch shipbuilding Britain Control of sources of timber and then coal/iron/steel for bigger ships/steamships U.S. Combined mining/shipping/ rail network with nearby state-supported steel industry Japan Control of huge iron/coal deposits through longterm contracts moved by larger ships owned by major industrial groups

Contradiction

Bureaucratic expansion, managerial corruption Interimperialist rivalry/rising protection costs Overaccumulation, market saturation

Displaced by British warships

Growing reliance on U.S. shipbuilding and shipping industries Expanded access to raw materials through wholly owned FDI

useful to remember that the ‘‘material conditions of production’’, which Bunker and Ciccantell emphasize so much, comprises considerably more than ‘‘materials’’ or the ‘‘materio-spatial problem’’. Among other things, it comprises labor, and the possibilities for a crisis that is rooted in the conditions of production run far beyond the costs of materials. Let us look at a recent example to illustrate this point. The recent Japanese/East Asian crisis was often superﬁcially described as a ﬁnancial crisis. The popular media and the business press focused on Japanese and South Korean banks, arguing that they followed highly questionable practices, particularly by giving out too many loans without sufﬁcient

128

DENIS O’HEARN

guarantee of repayment. Bunker could rightly take these pundits to task for their failure to understand that the crisis was rooted in the material conditions of production. Yet this does not mean that it was a crisis of high costs of extraction and movement of bulk materials. Instead, it was a crisis of overcapacity in production, preceding but encouraged by the cheap cost of money. The crisis originated when profit rates fell because Japanese and other East Asian goods remained unsold and, as a result, loans were not repaid and banks went into crisis. Indeed, although this was not really a crisis of ‘‘high ﬁnance’’, it nonetheless had closer similarities to Arrighi’s description of overaccumulation than to Bunker’s analysis of the materiospatial problem. The point is not that cyclical or secular capitalist crises are always due to overproduction. Rather, crises that appear to be one thing may be something else, and in a specific historical conjuncture there are multiple possible sources of crisis. I return, then, to the question that is presented in the title of this chapter: can different theories of change be reconciled?

4. RETURN TO CRISIS THEORY At the heart of each of the contending explanations of hegemonic decline and ascent (in that order) is a theory of crisis and countertendency. It is important to separate these two critical stages of the change of cycles. Just because one factor (e.g., high materials costs or overproduction) may precipitate a crisis, does that necessarily mean that the nation that emerges from the crisis as the new hegemon will have the optimal solution for that specific problem? A falling rate of profit is a general crisis, the solution to which may involve changes in several of the components that comprise the rate of profit, whether it be cheaper supply of materials, cheaper labor (either directly or by substituting unproductive labor) or better market access. In addition, capitals are concerned with their own rates of profit (and indirectly with those of others that impact on them) rather than the general rate of profit. States are primarily interested in the rates of profit of their national capitals rather than those of other regions (although, again, the health of subsidiary regions may impact such things as market access, as in the case of U.S. reconstruction of post-war Western Europe). Not only can a ﬁrm, a group of ﬁrms, or sectors in a whole economy increase their profit rates by targeting a combination of components, there are also different routes to solving a crisis in each component. The problem of materials supply can be solved by cheaper/better access and more efﬁcient

Cycles of Accumulation, Crisis, Materials, and Space

129

transport and/or by dematerialization. Labor costs can be cheapened by a spatial ﬁx (by moving capital or people) or by a technological ﬁx (replacing unproductive labor with technology). Market access has both spatial ﬁxes (entry into new regional markets) and technological ﬁxes (new or improved or cheaper products). What may be necessary, therefore, is a multivariate approach to crisis and solution that recognizes multiple routes to and from hegemony. Mandel (1990, p. 32) puts it well with respect to crises, It is correct that in the last analysis, capitalist crises of overproduction result from a downslide of the average rate of profit. But this does not represent a variant of the ‘‘monocausal’’ explanation of crises. It means that, under capitalism, the ﬂuctuations of the average rate of profit rate in a sense the seismograph of what happens in the system as a whole. So that formula just refers back to the sum-total of partially independent variables, whose interplay causes the ﬂuctuations of the average rate of profit.

In a sense, both Arrighi and Bunker can be right, even in a specific instance of hegemonic decline and ascent, but only if they demote their own version of crisis and solution to the status of partial explanation rather than determinant. There are variant or multiple causes of crisis, or historically specific causes for a given crisis in a given cycle of accumulation. And the solution to a given crisis (i.e., the path that leads to hegemonic ascent) may be different from or incorporate other solutions than the optimal solution to the central cause of the crisis. This involves separating crisis and solution. One of the interesting aspects of the debate about sources and solutions to crises is that they often center on the importance (or not) of overaccumulation. Arrighi cites overaccumulation as the central form of crisis, as regimes lose control of expanded reproduction and old institutions that were once superior are transcended, through crisis, by the new ones. But he has a specific meaning and outcome to overaccumulation. It is a situation where ‘‘more capital is seeking investment in the purchase and sale of commodities than the structure of the trading system can accommodate without provoking a drastic reduction in the overall profitability and security of trade’’ (Arrighi, 1994, p. 94). Thus, a crisis ensues, where ‘‘networks of high ﬁnance’’ recognize that they can no longer attain high profit rates in the existing center and they begin to seek alternative regions that will provide better rates of return. Bunker refutes this by insisting that the materials crisis is prior to this accumulation crisis and therefore not the central cause of either hegemonic decline or ascent. In a sense, this is the nub of the disagreement that Bunker introduces with Arrighi. But is it really such a profound one? Actually, the concept of overaccumulation is a confused one that appears to have various meanings or implications in the literature. Glyn (1990,

130

DENIS O’HEARN

p. 282) associates overaccumulation directly with the form of crisis that he and others earlier called the ‘‘profits squeeze’’. In this form, overaccumulation impacts on the supply of labor by eroding the reserve army of labor and thereby causing tight labor markets, rising wages, and thus falling profits and recession (see also Glyn & Sutcliffe, 1972; Weisskopf, 1979; Rowthorn, 1980; Itoh, 1980). As I have already shown, Arrighi (1994) associates overaccumulation with a ﬁnancial crisis, where high ﬁnance faces eroding profit rates and reacts by moving its investment elsewhere, generally to the rising hegemon. Brenner (1998) argues that the ‘‘overaccumulation’’ crisis since the 1960s is one of overproduction and realization crisis. Increased foreign competition from West Germany and Japan created excess capacity and excess supply of manufactured goods in the U.S. and in the world-economy. As a result, downward pressure on prices caused a reduced rate of profit in manufacturing. And, if we add Bunker’s analysis to the list, overaccumulation is a materio-spatial problem. The intensiﬁcation and rapid expansion of accumulation beyond the current material capacities of the world-system makes materials access a ‘‘problem’’, threatens profit rates and hegemony and, eventually, contributes to (or, in Bunker’s analysis, determines) the new hegemon. So what is this overaccumulation crisis? Is it about a shortage of labor, is it a ﬁnancial crisis, a crisis of production and overcapacity, a realization crisis, or a materials crisis? Again, Mandel (1990) has a useful angle that is based in Marx’s crisis theory. ythe question to know whether the crisis ‘‘centres’’ on the sphere of production or the sphere of circulation is largely meaningless. The crisis is a disturbance (interruption) of the process of enlarged reproduction; and according to Marx, the process of reproduction is precisely a (contradictory) unity of production and circulation. For capitalists, both individually (as separate ﬁrms) and as the sum total of ﬁrms it is irrelevant whether more surplus value has actually been produced in the process of production, if that circulation cannot be totally realised in the process of circulation (1990, pp. 31–32).

And, taking this further, it matters neither whether the crisis is a ‘‘ﬁnancial crisis’’ because ﬁnance capital in the face of a profits crisis will either restrict their investments or will move them (regionally or sectorally) to where they will earn a higher profit. In this sense, both Arrighi and Bunker are partly correct. Arrighi has a more realistic view of the crisis and recovery in the sense that, although he analyzes the results in terms of the behavior of high ﬁnance, he also recognizes that the crisis itself is caused by rising costs (from variant sources) due to institutional characteristics of the old hegemon while the new hegemon ‘‘solves’’ the problem of rising costs by creating new, more favorable institutions. ‘‘Finance’’ (or investment) simply follows best

Cycles of Accumulation, Crisis, Materials, and Space

131

practice in production (including provision of materials). Bunker, on the other hand, is correct to insist that the cause of crisis in the old hegemonic regime and the factor(s) that enables hegemonic ascent are rooted in material production. Accumulation outruns the existing technique and social organization of the old hegemonic regime in its ability to maintain the factors that are necessary for the central goal of increasing production at reduced costs. On the other hand, if Arrighi is wrong to overemphasize ﬁnance (as Bunker asserts), then Bunker fails to recognize the variant possible causes of crisis and successful hegemonic ascent. Indeed, one could make a persuasive argument that failure to recognize one aspect of successful hegemonic ascent can invalidate success in another variable. Failure by a nation to achieve sufﬁcient market access for the products of its capitals, for instance, can invalidate spectacular success in solving the materio-spatial problem. To show this, we need look no further than Japan, which is Bunker’s last example of ‘‘successful’’ hegemonic ascent. According to him, Japan achieved remarkable economic ascent because it did so well in solving the materio-spatial problem and achieving economies of scale and space. The Japanese state used ‘‘new forms of investment’’ to draw multiple states to invest in huge mining projects in the Amazon and elsewhere, thus creating excess capacity of certain key materials on world markets and driving down prices. At the same time, it invested in new transport technologies (ships and ports) that enabled it to take advantage of this cheap supply of bulk materials by moving it around the world in substantially greater quantities than had been possible heretofore. Together, this cheapened Japan’s costs of producing key products for the world market while enabling it to obtain sufﬁcient materials cheaply enough so that Japanese producers could stay a step ahead of the intensive material demands that were created by rapid accumulation (Bunker & O’Hearn, 1993). But instead of attaining hegemony (yet), Japan instead appears to have engaged in another form of crisis management by overexpanding credit and enabling business to invest more capital than really accumulated surplus value would have enabled them to invest, causing further overcapacity in an already overaccumulating global economy. Japanese ﬁrms were unable to pay back the debt out of the profits of the expanded production that was made possible by the credit expansion, resulting in a crash and a deep crisis.5 It was overaccumulation crisis, but it was simultaneously a production crisis, a circulation crisis, and a ﬁnancial crisis, all (arguably) encouraged by the cheap supply of raw materials to Japanese industry and the sunk costs involved in the infrastructure of materials supply. Or, an alternative way of formulating

132

DENIS O’HEARN

the previous sentence is that the failure to solve the overaccumulation crisis in terms of production/marketing/ﬁnance overwhelmed all of the Japanese state’s spectacular successes in solving the materio-spatial problem. Brenner (1998) has an interesting extension of this analysis of crisis to the whole world-system (or, as he calls it, to the general process of globalization). According to him, overaccumulation is exacerbated because, in this phase of crisis perhaps unlike previous ones, ﬁrms refused to close in sufﬁcient numbers in response to falling profits because of their huge sunk costs. Instead, large U.S. manufacturing ﬁrms in particular responded to the crisis not by exiting but by intensive development of newer technologies with even lower costs. In doing so, they continually bought themselves some time each time they introduced a new technology or product, but their solution also intensiﬁed the problem of overcapacity. The whole process was further exacerbated by the rise of the East Asian NICs in the 1980s, which increased the capacity further. Ironically, Bunker’s concentration on Japan, as his most recent case of ascent and center of the highest form of solution to the materio-spatial problem yet, may disprove his materio-spatial theory in its strong version. Japan managed the materio-spatial problem so well yet it did not become hegemonic. The closest it came was as a partner in a possibly emergent but never realized bigemonie where Japan was the leading world economic power in coalition with the U.S. as the leading political and military power. This example appears to demonstrate that it is necessary to solve several problems of accumulation simultaneously, which Japan never did, nor could it do along with the U.S. because of their ﬁrms’ continuing drive to compete and create new capacity. Perhaps it is too early to make such a judgment. Most recently, Bunker and Ciccantell (2003) argue that China is solving the materio-spatial problem, perhaps at a scale even beyond Japan, and that this could be the basis of their future ascent to hegemony. Perhaps history, as usual, will be the judge of whether these two cases support or refute the strong version of their theory of materials, space, accumulation, and hegemonic change.

5. HEGEMONY AND LOCAL CHANGE Although it is not one of the main foci of Bunker’s research, his analysis of the regularities of hegemonic cycles opens up some very interesting concepts for comparative research on local economic change. One of the central conceptions of world-systems analysis and of other critical approaches to

Cycles of Accumulation, Crisis, Materials, and Space

133

economic change is that the structures of the world-economy restrict local development in important ways. For many years, there was a move toward the study of concrete historical cases so that we could begin to chart the different developmental effects of the world-system on local change. Slowly, we moved from cruder notions of underdevelopment to dependent development to richer notions of historical speciﬁcity. The hardest thing to do, however, was to regeneralize, to move back upwards from the local to the regional to the global, so that we could make some statements about mechanisms and regularities across historical cases. The most interesting recent work in this regard is the reﬁnement of the study of world-systemic change from the encompassing comparison to the incorporating comparison (McMichael, 1990) and in the broader economic sociology literature from path dependency to reiterative problem solving (Haydu, 1998). In these newer approaches, systemic change is not simply given by the structure of the system or by ﬁxed laws of structural change. Rather, change takes place through mechanisms and laws that are worked out in episodes of iterative problem-solving. The precise nature of an outcome of, say, crisis and recovery, hegemonic shift, and local position within the newly organized world-system is contingent, not only on relative positions of regional actors but on the strategies they choose at crucial junctures. In particular, localities are not just subjugated into a path-dependent position of underdevelopment. Rather, they have limited options, some of which are worse and some of which are less worse and, just possibly, some of which are actually better. These switching points are probably more likely to occur during periods of hegemonic shift when world-system positions are changing and some localities are being reintegrated into new roles that are deﬁned by the new regional or hegemonic powers. In these situations, it may even be possible for certain regions to ‘‘hook on’’ to core accumulation processes and to achieve upward mobility within the world-system. Not only has Arrighi used such conjunctural approaches to analyze the regularities of systemic cycles of accumulation, others have used them to study the relationship between global and local change. Among these new studies, O’Hearn (2001) attempted to tie systemic cycles of accumulation in the Atlantic Economy to local economic change, iterative problem-solving and the possibilities of exploiting ‘‘switching points’’ in Ireland’s historical development. Bunker (forthcoming) does the same thing for the Amazon, as he shows the limiting effects of successive regimes of raw material extraction on the local economy, ecology, and society. But Bunker also does something else that may point the way toward a reconceptualization of general patterns of systemic and local changes. Not

134

DENIS O’HEARN

only does he show the underdeveloping impact of extraction on certain localities that are integrated as extractive economies, he also shows that hegemonic ascent is tied to a certain kind of ‘‘virtuous’’ but still subservient relationship between the rising economy and the existing hegemon. Holland attained hegemony through the route of transporting bulk materials for Portugal. Britain attained hegemony by using Dutch technologies eventually by displacing Holland. The U.S. used British capital and rail and mining technologies to develop a Midwestern material–industrial complex that eventually outstripped British hegemony. And Japan turned U.S. plans to provide and mold its material supply, as a subservient economy, to its own advantage. All of these examples taken together suggest that there are ‘‘generative path dependencies’’, where the immediate gains made by the hegemon in subsuming certain regions has certain characteristics that eventually sow the seeds of its own destruction and replacement. This is very different from our usual conceptions of domination and underdevelopment, even in the ‘‘dependent development’’ or ‘‘semi-peripheral’’ developmental varieties. While it may have limited policy implications – many are called but few are certainly chosen for hegemony – it is a key form of historical economic change. Associated with this boundary case may be other, less spectacular forms of generative path dependencies. Senghaas (1985), in his analysis of the economic rise of small West European states in the late 19th century, shows how their connections to the regional core powers lead to the possibility of ‘‘piggy backing’’ on core development so that these small states become associated core economies in their own right. Here, as in the boundary cases above, there are important ‘‘switching points’’ where the locality must break away from core domination before it can re-enter in a new, more mature capitalist form. The small Scandinavian states developed where Ireland and Romania did not because they recognized the importance of strategically withdrawing from their traditional roles as suppliers of agricultural staples to the main core states and developed their local markets under state protection, developing niche technologies until they could re-enter the European market on a more equal basis. But let us not get carried away by such possibilities. As Bunker and others have shown, the common patterns of local change are more restrictive forms of path dependencies where the ‘‘switching’’ possibilities are less obvious or even absent. Over time, as the extractive regions become more and more subsumed – essentially underdeveloped – by their associations with core extractors, they may move further and further away from a point where they can do much at the next potential ‘‘switching point’’ to change the nature of

Cycles of Accumulation, Crisis, Materials, and Space

135

their dependent relations and the degradation of their local political ecologies. At a general level, however, it should be possible to model the possible local paths through hegemonic change according to past position in the world-economy, the resource and labor endowments of a region and the interests that new hegemonic contenders will have in a given region (i.e., as markets, as suppliers of materials, labor, or technologies).

6. CONCLUSION Stephen Bunker’s analysis of the importance of the materio-spatial problem is clearly one of the most important reﬁnements of the existing models of world-systemic change and, particularly of hegemonic cycles and the cycles of accumulation that are associated to them. Whether or not the weight that Bunker places on the materio-spatial problem – as the singular most important cause of both crisis and ascent – is sustainable, it is without doubt that future models of hegemonic change will have to place materio-spatial problems at their very center. They are at least one of several critical determinant variables, the relative importance of which will depend on the specific historical conjuncture with which one is concerned. After Bunker, we can no longer view the cycle of accumulation as labor plus capital (as technology) plus high ﬁnance, with a bit of a marketing problem thrown in. Even from the standpoint of crises of overaccumulation, future models must include materials and the materio-spatial problem at the center, although not the only variable, of our theories of crisis and hegemony. In doing so, not only will we have more realistic models of hegemonic decline and ascent, we will also have more realistic understanding of the limitations and possibilities of localities within processes of hegemonic change.

NOTES 1. Stephen Bunker’s work on hegemonic cycles, usually with Paul Ciccantell, is unpublished at the time of this writing. I have nonetheless, with Bunker’s assent, taken the liberty of critiquing it anyway. The most important work is in Bunker and Ciccantell (2005). 2. These circuits of expansion are adapted from Marx’s original circuit of capital, where money capital (M) is invested in commodity capital (C) to obtain expanded money capital (M’). In Arrighi’s adaptation, money capital (M) can be invested in territory (T) to obtain expanded money capital (M’) or money (M) can be used as an intermediate link in a process of acquiring additional territory (T’).

136

DENIS O’HEARN

3. Perhaps, this is also why Bunker’s conceptualization of hegemonic crisis is less clear than Arrighi’s, and is almost replaced by succession, i.e., the new hegemon simply does it better (‘‘it’’ being the materio-spatial problem) rather than the old hegemon facing a critical barrier based on an internal contradiction of its own system. 4. On variation and regularity, Bunker and Ciccantell favorably cite Tilly (1995), who argues that analysis of regularities conﬁgures into unique events rather than universal or invariant outcomes; as well as McMichael (1990) on the use of ‘‘incorporating comparisons’’, where outcomes of socio-economic processes are built through a process of iterative problem-solving rather than being given by the invariant structural imperatives of a system. On multiple causation, Bunker and Ciccantell insist that the general mechanism of expanded production at lowered cost ‘‘comprises multiple mechanisms – social and material – that can be variably speciﬁed between cases, such as technological innovation, infrastructural growth, ﬁnancial markets, control, motivation and training of labor, and distribution networks’’ (Bunker & Ciccantell, forthcoming). 5. Mandel cites Marx (Capital volume 2) as the source of an early version of this form of crisis extension, where the business cycle becomes entwined with a credit cycle that acquires relative autonomy from what is happening in production. Unpaid debt is produced, which cannot be repaid because, although the system is producing sufﬁcient surplus value capitalists are unable to realize it in the form of profits due to overaccumulation (Mandel, 1990, p. 33).

REFERENCES Arrighi, G. (1994). The long twentieth century. London: Verso. Barham, B., Bunker, S., & O’Hearn, D. (Eds) (1994). States, firms and raw materials: The world economy and ecology of aluminum. Madison: University of Wisconsin Press. Brenner, R. (1998). The economics of global turbulence. New Left Review, 229, 1–264. Bunker, S., & Ciccantell, P. (2005). Globalization and the race for resources. Baltimore: Johns Hopkins University Press. Bunker, S. & Ciccantell, P. (unpublished). Theory and method in analyzing comparative national development and global systemic change. Manuscript. Bunker, S., & O’Hearn, D. (1993). Strategies of economic ascendants for access to raw materials: A comparison of the United States and Japan. In: R. A. Palat (Ed.), Pacific-Asia and the future of the world-system. Westport, CT: Greenwood Press. Castells, M. (1997). The power of identity. Oxford: Blackwell. Chandler, A. (1977). The visible hand: The managerial revolution in American business. Cambridge, MA: Belknap Press. Chase-Dunn, C. (1989). Global formation: Structures of the world economy. London: Basil Blackwell. Ciccantell, P., & Bunker, S. (2003). The economic ascent of China and the potential for restructuring the capitalist world-economy. Paper delivered to annual meeting of American Sociological Association, Atlanta, Georgia. Coase, R. (1937). The nature of the ﬁrm. Economica, 4(15), 386–405.

Cycles of Accumulation, Crisis, Materials, and Space

137

Colombo, U. (1988). The technology revolution and the restructuring of the global economy. In: J. H. Muroyama & H. G. Stever (Eds), Globalization of technology: International perspectives. Washington, DC: National Academy Press. Foster, J. (2000). Marx’s ecology: Materialism and nature. New York: Monthly Review. Frobel, F., Heinrichs, J., & Kreye, O. (1980). The new international division of labor. Cambridge: Cambridge University Press. Glyn, A. (1990). Marxist economics. In: J. Eatwell, M. Milgate & P. Newman (Eds), The new palgrave: Marxian economics. New York: W.W. Norton. Glyn, A., & Sutcliffe, B. (1972). Capitalism in crisis. New York: Pantheon. Haydu, J. (1998). Making use of the past: Time periods as cases to compare and as sequences of problem solving. American Journal of Sociology, 104(2), 339–371. Hill, C. (1967). Reformation to industrial revolution: A social and economic history of Britain (pp. 1530–1780). London: Weidenfeld & Nicolson. Israel, J. (1989). Dutch primacy in world trade (pp. 1585–1740). Oxford: Clarendon Press. Itoh, M. (1980). Value and crisis: Essays on Marxian economics in Japan. London: Pluto. Mandel, E. (1990). Marxian economics. In: J. Eatwell, M. Milgate & P. Newman (Eds), The new Palgrave: Marxian economics. New York: W.W. Norten. McCormick, T. (1989). America’s half-century: United States foreign policy in the cold war. Baltimore: Johns Hopkins. McMichael, P. (1990). Incorporating comparisons within a world-historical perspective. American Sociological Review, 55, 385–397. Moseley, F., & Wolff, E. (Eds) (1992). International perspectives on profitability and accumulation. Aldershot: Edward Elgar. O’Hearn, D. (2001). The Atlantic economy: Britain, the US and Ireland. Manchester: Manchester University Press. Rowthorn, R. (1980). Capitalism, conflict and inflation. London: Lawrence and Wishart. Senghaas, D. (1985). The European experience. Leamington Spa: Berg. Steensgaard, N. (1974). The Asian trade revolution of the seventeenth century: The East Indian companies and the decline of the caravan trade. Chicago: University of Chicago Press. Tilly, C. (1995). To explain political processes. American Journal of Sociology, 6, 1594–1610. Wallerstein, I. (1984). The politics of the capitalist world-economy. Cambridge: Cambridge University Press. Weisskopf, T. (1979). Marxian crisis theory and the rate of profit in the postwar US economy. Cambridge Journal of Economics, 69, 2. Williamson, O. (1970). Corporate control and business behavior. Engelwood Cliffs, NJ: PrenticeHall.

STARTING AT THE BEGINNING: EXTRACTIVE ECONOMIES AS THE UNEXAMINED ORIGINS OF GLOBAL COMMODITY CHAINS David A. Smith ABSTRACT This chapter relates Bunker’s innovative view of nature, raw materials and political economy to the ‘‘global commodity chains’’ (GCCs) approach. The chapter contends that his view of the importance of primary material extraction, shipping, and energy used in various ‘‘transformations’’ of these crucial products can be fruitfully married to a GCCs perspective.

1. INTRODUCTION Stephen Bunker is a singular sociologist whose ideas about comparative sociology, the essence of social change and development, and the way nature is deeply implicated into the unevenness and inequalities of global political economy will have a lasting intellectual impact on our understanding of these things. This chapter will attempt to relate Bunker’s innovative view of nature, raw materials and political economy to the ‘‘global commodity Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 141–157 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10007-9

141

142

DAVID A. SMITH

chains’’ (GCCs) approach. My contention is that his view of the importance of primary material extraction, shipping, and energy used in various ‘‘transformations’’ of these crucial products can be fruitfully married to a GCCs perspective. My essay begins in an unorthodox way, with a little biographical vignette, which helps explain my ties to Professor Bunker and his exciting ideas. Stephen Bunker is a very peripatetic person! My ﬁrst conversation with him was at an annual meeting of the American Sociological Association. He grabbed me (ﬁguratively, if not literally) shortly after a panel where I had presented a paper on global commodity trade networks. My recollection is that the conversation was rather one-sided, with Stephen expounding his approach to unequal exchange based on ‘‘modes of extraction,’’ telling stories from his studies of the Amazon, and explaining details about mineral mining and smelting. Although my interest in this sort of thing was a much more generic effort to ‘‘model’’ overarching world-system structure using network analysis of trade patterns, I was intrigued and fascinated by both Stephen and his ideas! Our initial discussion led to further correspondence. It led to an invitation to be a discussant for a conference Professor Bunker organized in Madison on the international bauxite industry in 1990. I knew virtually nothing about aluminum or bauxite. But the focus on a specific commodity, the details of how it was extracted, processed and transported, and the organizational infrastructure necessary to do all of this seemed right on target, and a necessary complement to the sort of more nomothetic ‘‘modeling’’ of trade in the world-system I was immersed in at the time. My own research was focused on network analysis of commodity trade data from the United Nations Statistical Ofﬁce; the goal was overarching images of world-system structure based on global trade patterns. But Stephen Bunker encouraged me to take a more focused look at specific commodity trade data for some raw material transformations, including the bauxite – alumina – aluminum sequence, with the idea that we could look at those particular networks and get a better image of the way they ﬂowed across global space. It was a different approach to the problem that redirected attention from the ‘‘central tendencies’’ of all trade networks to the mapping of particularly important raw material transformations. In the mid-1990s, I spent a memorable weekend out at the Bunker farm with Stephen that included many hours talking about extractive industry, economies of scale in transportation, and Japanese strategies for cornering raw material markets. During this same period, I was also involved with a project, with my colleague Rich Appelbaum at the University of California – Santa Barbara

Starting at the Beginning

143

on ‘‘commodity chains and industrial districts’’ in East Asia funded by the University of California Pacific Rim initiative. This research was designed to apply the then relatively new GCCs model of economic development and inequality to a world region that seemed to be on the cusp of great dynamism. Gary Gerefﬁ was the main proponent of the commodity chains perspective (and had served on my doctoral committee during the mid-1980s). He was trying to formulate a new way of thinking about ‘‘development’’ in light of various changes that we refer to as ‘‘global restructuring,’’ the ‘‘New International Division of Labor,’’ etc. – and Gerefﬁ was particularly interested in how to conceptualize the ‘‘economic miracles’’ in the ‘‘newly industrialized countries’’ of East Asia. The role of ‘‘industrial upgrading,’’ and attempts to promote growth via export manufacturing and the increase of ‘‘global sourcing’’ in a wide array of products that are consumed in the U.S. and other core countries all pointed to the utility of a reconceptualization of ‘‘development.’’ Drawing on work by Hopkins and Wallerstein, Gerefﬁ argued that the idea of GCCs was particularly useful, deﬁned as ‘‘a network of labor and production processes whose end result is a ﬁnished commodity’’ (Hopkins & Wallerstein, 1986, p. 159, quoted in Gerefﬁ, Korzeniewicz, & Korzeniewicz, 1994, p. 2). This imagery allows us to address ‘‘space’’ and local development issues by exploring the impact that the GCCs have in the places where they ‘‘touch down.’’ These analyses stress that value is realized at differential rates on the various nodes of the chain, with more profits/surplus realization, generally, made at the later steps, particularly in sales and marketing of consumer goods. Despite the great promise of applying network analysis to raw material transformations, Stephen Bunker and I never found the time or funding to do the collaborative project we envisioned. But it was very clear to me that he and Gary Gerefﬁ were talking about different ends of the same ‘‘chain.’’ However, there is a surprising lack of dialogue between the proponents of GCCs (who direct most of their attention toward consumer manufactures and marketing) and the Bunkeresque view (with a heavy extractive emphasis). So, even though everyone would probably agree with Bunker’s argument in his 1984 article in the American Journal of Sociology that ‘‘commodities can emerge only from locally based extractive and productive systems,’’ (Bunker, 1984, p. 1017), there is virtually no attention to this ‘‘end’’ of the commodity chain in most of the GCC literature. Instead, the predominant focus in much of the research that followed Gerefﬁ’s lead focuses either on the phases of manufacturing of relatively simple commodities like athletic shoes and apparel or more complex ‘‘assembled’’

144

DAVID A. SMITH

products like automobiles, computers and other electronic goods. Schematic representations of GCCs often included a ‘‘box’’ for original raw materials (like wool from Australian sheep or various metals for auto components), but studies very rarely focused on these nodes of the chain. The provision of raw materials in most of these cases is taken as a precondition, so it is rarely even empirically investigated, much less ‘‘theorized’’ as an important link. Indeed, as serious research elaborating on the GCC model developed, the push was not only toward studying the later steps of the transformation of ﬁnished – and primarily consumer – goods, but toward incorporating marketing as part of the chain. This is a key insight, since logos and ‘‘branding’’ are perhaps the most important way that profits are realized in many consumer markets (see Klein, 2000). However, it pulls us even farther away from the ‘‘beginning’’of the commodity chain. And Bunker’s work is a clarion call to a lesson that many of us were initially taught as children: we can often learn more about things if we do ‘‘start at the beginning’’! This chapter is a preliminary attempt to use this insight to begin to build a more comprehensive GCCs framework. It will briefly sketch the GCC model as promulgated by Gerefﬁ et al. Then I will highlight how this approach can be enriched in a Bunkeresque way by ‘‘starting at the beginning’’ with the extraction of various raw materials and energy that are essential foundations for the rest of the production chain. This discussion will draw on both Bunker’s older work and my recent reading of the forthcoming Globalization and the Race for Resources book by Stephen Bunker and Paul Ciccantell (2005). While this may make my chapter less empirically rich and a bit more ‘‘theoretical’’ than some others in the volume, my goal is not large abstractions but a call for an integrated approach to looking at GCCs, which has been missing so far.

2. GLOBAL ECONOMIC RESTRUCTURING AND COMMODITY CHAINS Recent research highlighting the importance of a ‘‘new international division of labor’’ (NIDL), the industrial restructuring of global capitalism, and the reconﬁguration of global ‘‘commodity chains,’’ gives the sociology of development a new angle into understanding spatial dynamics and offers an opportunity to reconceptualize the role of science, technology, and information in world development and global inequality. However, the current literature with its emphasis on ‘‘industrial upgrading’’ and changing regimes

Starting at the Beginning

145

of export manufacturing, while shedding new light on some key aspects of recent world-economy dynamics, provides a less than satisfying account of how geography, space, transport, and technology matter. All these factors are central to Bunker’s emphasis on the key role of extractive processes in world-system processes. This is not the place to recite the various twists and turns of two decades of controversies and debates about the nature of global economic restructuring. In 1980 a path-breaking book argued for the emergence of a NIDL based on a globalized production process which emerged as the result of the rise of world capital markets, regional trading blocs, and declining communications and transportation costs (Frobel, Heinrichs, & Kreye, 1980). Consistent with the broad strokes of this argument were descriptions of ‘‘the global assemblyline’’ (Feuntes & Barbara, 1984) which emphasize ‘‘deindustrialization’’ of high-wage core nations through capital ﬂight to low-wage semiperipheral or peripheral societies (Ross & Trachte, 1990). One variant of this approach labeled this process ‘‘global Fordism,’’ arguing that the growing contradictions of the regime of capital accumulation in core nations (premised on relatively well-paid workers/consumers) leads to the relocation of low-skilled manufacturing to cheap wage areas in the Third World (Lipietz, 1982, 1987). For NIDL theorists, technology is important primarily because innovations in both communications and production techniques facilitate the ability of core capitalists to coordinate and control geographically dispersed production. The overriding focus is on how transnational capital takes advantage of differential labor costs, once they have the technical means to do so. While the NIDL thesis helps to describe the recent rapid rise of manufacturing in the Third World, it is too one-dimensional, over-emphasizing the determinant inﬂuence of low wage labor in decision-making about industrial location. Particularly in the most profitable ‘‘cutting edge’’ industries, standardized mass production has given way to small batch runs that emphasize rapid design changes, which either reﬂect the latest technological innovation or shift in fashion. So even in the global garment industry, Frobel, Heinrichs, and Kreye’s (1980) prototypical industry exemplifying the NIDL, locational decisions (particularly for production of ‘‘high fashion’’ segments of the market) are much less tied to labor costs than they supposed (Elson, 1989; Dicken, 1998, Chapter 9; Appelbaum, Smith, & Christerson, 1993). Furthermore, it is obvious that some non-core areas are increasingly engaging in manufacturing activities that are relatively ‘‘high technology’’ (Henderson, 1989). An alternative formulation of global economic restructuring claims that the changes taking place represent the crossing of a new ‘‘industrial divide’’

146

DAVID A. SMITH

(Piore & Sabel, 1984) leading to the rise of ‘‘post-Fordist production’’ (Storper & Walker, 1989) and ‘‘ﬂexible production complexes’’ (Scott, 1988). These authors stress the role of knowledge, technology, and the importance of skilled ‘‘symbolic analysts’’ (Reich, 1991), along with the advantages of ‘‘ﬂexibility’’ in modern competitive industry. Ironically the precise meaning of ‘‘ﬂexibility’’ in this literature is, itself, rather pliable. Malecki (1991, p. 202) claims, ‘‘increasingly, flexible, as opposed to rigid and stable, links allow ﬁrms to adjust their activities to uncertainty and to rapidly shifting conditions brought about by competition and technology, as well as government regulation’’ (italics in the original). Dicken (1998, p. 166) further speciﬁes: ‘‘the key to production ﬂexibility lies in the use of information technologies in machines and operations. These permit more sophisticated control over the production process.’’ While the most obvious examples of industries where this type of ‘‘ﬂexibility’’ becomes important are the knowledge-intensive ‘‘high technology’’ industries like semiconductors (Scott, 1987; Henderson, 1989) or automobiles (Schoenberger, 1988), it is also critically important in ‘‘traditional’’ industries where rapid changes in styles or models require close coordination between design, production, and marketing (for instance, for ‘‘high fashion’’ clothing – see Dicken, 1992, Chapter 9). To succeed in this new ‘‘post-Fordist production,’’ ﬁrms need (1) access to information and innovation and (2) linkages to other ﬁrms (as suppliers, buyers, or production partners), R&D organizations, and marketing outlets. The ‘‘competitive advantage of nations’’ will be held by those who control key technologies and production networks’’ (Porter, 1990). However, these theorists ignore the importance of access to the raw materials that go into the productive processes, something that Bunker and Ciccantell stress in their new book (Bunker & Ciccantell, 2005). In an effort to understand the changing realities of the late twentieth century, both the NIDL and the ‘‘ﬂexible production’’ literature challenge some basic assumptions of previous development theory. They problematize the simplistic assumption that ‘‘industrialization’’ will lead to genuine economic development, urging us to closely examine the nature of manufacturing activities and the specific linkages connecting manufacturing enterprises to global markets and local, state and transnational capital. While some peripheral countries remain primarily ‘‘export platforms’’ for simple low-technology, labor intensive goods made by low-wage unskilled workers, ‘‘industrial upgrading’’ in many of the NICs has led to a shift from commodities like textiles, apparel and footwear to ‘‘higher value-added items that employ sophisticated technology and require a more extensively developed, tightly integrated local industrial base’’ (Gerefﬁ, 1992, p. 92)

Starting at the Beginning

147

such as computers, semiconductors, numerically controlled machine tools, VCRs, and televisions. This understanding of an increasingly integrated global economy where countries come to occupy distinct ‘‘export niches’’ and where ‘‘industrial upgrading’’ is a key strategy leads Gerefﬁ to argue that global ‘‘commodity chains’’ should be the central objects of analysis (Gerefﬁ, 1990; Gerefﬁ & Korzeniewicz, 1990). This idea, which in many ways parallels the ‘‘value chain’’ of economist Michael Porter (1990) or the ‘‘production chain’’ of geographer Peter Dicken (1998), draws on Hopkins and Wallerstein’s (1986) definition of a commodity chain as ‘‘a network of labor and production processes whose end result is a ﬁnished commodity’’ (1986, p. 159). Gerefﬁ and Korzeniewicz (1990) conceptualize these chains as consisting of a number of ‘‘nodes’’ that comprise the pivotal points in the production process: extraction and supply of raw materials, the stage(s) of industrial transformation, export of goods, and marketing. Each node is itself a network connected to other nodes concerned with related activities; local, regional, and world economies are seen as ever more intricate web-like structures of these chains. Ultimately, global inequality and Third World development/underdevelopment are deﬁned by the positions societies (or their ﬁrms, localities or ‘‘industrial districts’’) occupy in these multiplex networks of world-wide economic production and exchange. All commodities undergo a sequence of transformation from raw materials to ﬁnished products. Heavy industrial products (like metals) may end up as girders in buildings, the bow of a ship, the foundation of a highway, or the pilings of a deep-water port. Some strategic raw materials may be processed into military hardware or weapons systems. For consumer products (the overwhelming focus of the existing GCCs literature), commodities end up as packaged and marketed goods. In all of these cases the processes of transformation and transportation lead to geographic linkages and connections that in a real material sense create a spatially bounded structure for the world-economy. In the global economy of the late twentieth century, these ‘‘export networks and export niches are becoming key units of analysis in the contemporary global manufacturing system’’ (Gerefﬁ, 1992, p. 90). Differential profit and surplus are generated at various nodes along these commodity chains. But these patterns are not uniform; while there may be a tendency for the highest profits and the most surplus extraction to occur at the later points along the commodity chain, this is not always the case. Instead, this pattern can vary across particular commodities (or even across nodes producing the same commodities in different ways). There also may

148

DAVID A. SMITH

be a tendency for the production nodes with the highest process technologies, turning out the most sophisticated and innovative products, to be the places where the most profits/surplus are captured. But, while this may have been the assumption of some early conceptual approaches to ‘‘the commodity question’’ and unequal exchange (Emmanuel, 1972) and some efforts to measure ‘‘levels of processing’’ in international trade (Jaffee, 1985; Firebaugh & Bullock, 1987), Gerefﬁ and other recent proponents of global commodity chain approaches claim that things are not quite this simple. GCCs are more than just sequences of production. There is a need to understand how both marketing of products (particularly to highly profitable core consumer markets) and coordination and control of the integrated global production/marketing networks are crucial. Indeed, it is now clear that for leading edge consumer goods, the marketing and ‘‘command and control’’ functions are where the big profits are made, while actual manufacturing is much less lucrative. On the other hand, it is also important to understand that there are a number of ‘‘logistical’’ processes that must be coordinated and controlled for those heavy industrial goods that, in a very concrete sense, make up the foundation of the contemporary global economy. Here ‘‘marketing’’ may not be crucial in the same way, but ‘‘command and control’’ (tempered by things like the physical properties of the materials, the cost of transporting them, business strategies to ‘‘corner the market’’ and ensure supply of strategic commodities, etc.) may still be paramount. This GCC formulation is very useful for students of international development and global inequality. It shifts our attention away from ‘‘national economies’’ and the old images of ‘‘take off’ and ‘‘modernization’’ as the key to growth, arguing that the real ‘‘action’’ occurs at the production nodes along transformation chains that today often traverse national borders and exhibit genuine global reach. This can help us debunk some common misconceptions about the current prospects for development in various parts of the world. For instance, in the late 1990s the Los Angeles Times ran a series of stories on industrialization in China (and, indeed, this tends to be a recurring theme for this newspaper in subsequent years). Appearing on several consecutive days, these pieces highlighted the growth of such ‘‘hightech’’ industries as computer assembly and micro-chip fabrication in that country. The underlying theme was that China was rapidly ‘‘industrializing’’ and developing ‘‘advanced industry’’ so that it could follow the path of economic growth already blazed by core countries. However, the journalists who wrote these stories seemed completely unaware of how the factories and ﬁrms in China that they proﬁled remain situated in subordinate roles along

Starting at the Beginning

149

GCCs. In nearly all cases the cutting edge technology embedded in the products or manufacturing processes originates and is controlled from places beyond China’s borders, as are the marketing networks and knowhow to sell Chinese-made products in wealthy core markets. There was also precious little attention to the massive infrastructure of transoceanic shipping that was necessary to bring all these goods to the shores of North America. There is no doubt that China has become a manufacturing powerhouse, driven by export-oriented industrialization (with a large, rapidly growing, share of world trade in a variety of consumer goods). But the GCC perspective helps us maintain a healthy skepticism of the claims that, based on these developments, China is certain to become the next great economic power. While the GCC perspective is very useful for interpreting the contemporary world-economy, it also is limited in its focus. The global economy is one of those entities that is particularly hard to grasp in a holistic way, and all of us tend to be like those blind men and that elephant, describing the part with which we are most familiar. While the conceptual descriptions of GCCs, invariably acknowledge that raw material inputs are crucial at the beginning of the chain, the unquestionable focus of most of the research that this approach inspires is on manufacturing. Indeed, it tends to periscope in further on the production of ﬁnished products and consumer goods, with much less emphasis on capital intensive heavy industries.

3. BRINGING IN BUNKER: EXTRACTION AND TRANSPORT OF RAW MATERIALS Stephen Bunker’s approach to ‘‘the commodity problem’’ is very different. His interest is in explaining underdevelopment in extractive export economies, with much of his focus over the years on the vast, resource-rich Amazon basin. In his writing, Bunker tends to discuss ‘‘modes of extraction’’ on the ground, how these are compatible or incompatible with the procurement of other raw materials from the same environment, and the need for state and transport infrastructure to support extractive activities. He never really speaks in terms of ‘‘commodity chains’’ (though he comes close at times, referring for example to ‘‘global chains of extraction and production’’ in Bunker & Ciccantell, 2005). But the key components of any ‘‘mode of extraction’’ are the ‘‘tangible commodities whose material differences typify the more abstract mechanisms by which the material properties

150

DAVID A. SMITH

of a commodity determine the way it is incorporated into global markets’’ (Ibid, p. 58). Bunker points out that those of us ‘‘who live in advanced industrial economies are not sufﬁciently accustomed to thinking about the ways that the nature of raw materials structure the relations of land, labor and capital as they are absorbed as commodities’’ (Ibid, p. 61), nor do we usually think much about how these extractive economies are anchored in space at points where raw materials occur in nature, which often impose major barriers to transportation of these commodities. Bunker also argues that ‘‘societies organized around extraction for export remain subordinate, ﬁrst to the natural processes of material production in local ecosystems, and second, to the vacillation of demand and price in the industrial economies’ (Ibid, p. 62). Interestingly, this analysis of extraction resonates with some insights from the GCC perspective. First, the emphasis on a ‘‘close grained analysis of specific commodities in particular places’’ (and the attendant dilemma of reconciling a tension between idiographic and nomothetic logics of analysis) parallels Gerefﬁ’s call to study transformation sequences that are ‘‘situationally specific, socially constructed, and locally integrated, underscoring the social embeddedness of economic organization’’ (Gerefﬁ et al., 1994, p. 2). Bunker’s insistence that we need to thoroughly understand the properties of various minerals (including their chemistry and the details of how they are reﬁned) is rather reminiscent of discussions of the pliable nature of cloth, the need for garment buyers to touch fabrics and hold clothing in their hands, and the intricacies of the process of sewing and ﬁnishing garments from analyses of the global apparel commodity chain. However, Bunker would probably argue that the natural, physical, properties of primary products impose more limits on the ‘‘social construction’’ of transformative processes than Gerefﬁ might consider. Second, there are some striking similarities between extractive economies and low-wage manufacturing sites. Bunker’s descriptions of the relative powerlessness of the peripheral raw material production areas, the way that control and technology are held far away in the core, and the manner in which extractive enclaves might be ‘‘played off’’ against each other by core interests, are quite analogous to the subcontracted low-wage manufacturing enterprise in far-ﬂung apparel, toy, shoe, or other consumer good production network described in the GCC literature. Bunker and Ciccantell assert, ‘‘control of these techniques and of the social organizational, political, and ﬁnancial systems to implement them therefore tends to remain in the regions where these materials are transformed and consumed. So do the major share of the profits that accrue from these processes’’ (Bunker & Ciccantell, 2005).

Starting at the Beginning

151

While this statement refers to raw material processing in extractive peripheries, it might apply equally well to many types of routine low-wage assembly industries scattered about the underdeveloped world. Beyond these parallels and compatibilities, however, there are aspects of Bunker’s formulation of extractive economies that can improve and augment the GCC framework. I want to outline a number of these. (1) Identifying the loci of growth. One of the key challenges of the GCC framework is to understand how and why particular types of activities occur at certain nodes in particular locations. The question of shifting ‘‘industrial location’’ is one of the animating issues behind efforts to understand contemporary global economic restructuring. For most of the ‘‘downstream’’ nodes later in the transformation sequence a number of social, economic and political factors inﬂuence decision-making about where to produce what. Bunker’s work makes it rather clear that the natural endowment of a surrounding geographic region, the conﬁguration of rivers, and physical topography are major factors in determining the success or failure of nodes that depend on extractive advantages. This is obviously true for mining communities and rather immediately determinant of the growth of towns and cities linked to ﬁshing or farming. It is much less clear why certain types of manufacturing are found at particularly locations (and, even if we accept a NIDL inﬂuenced argument about the importance of ‘‘cheap labor’’ for late twentieth century export industries, that this type of locational logic is much more complex). In the most recent manuscript, Bunker and Ciccantell demonstrate that a major reason for Amsterdam’s prosperity, the rise of its world leading ship building industry, and, ultimately, the Netherlands’ advantage that led to a period of world-system hegemony was the city’s easy access to high-quality lumber easily ﬂoated to the city on a dense network of nearby rivers. Of course, the extractive peripheries (like Amazonia) are not centers for dynamic growth. But one of the key points that the authors make is that securing access to a relatively cheap, dependable, supply of essential raw materials is just as crucial for emerging hegemonic economic power as cutting edge technology or military might (and, in fact, resource access often requires and/or reinforces these other two factors). Using commodity analysis to explain hegemonic rise and fall is far beyond the ken of most GCC analysis. (2) Explaining technological change and ‘‘time/space compression.’’ Most of the approaches to global restructuring, including the GCC framework, take the technological changes that make for a ‘‘smaller world’’ as a given. Impressive advances in telecommunications, information processing, and transportation occurred in the mid-twentieth century that allow for an

152

DAVID A. SMITH

epochal ‘‘global shift’’ in economic activities. By focusing on the key role of extractive economies and looking at the very beginning of the commodity chain, where raw materials are extracted from the natural environment, transformed and puriﬁed, and transported to the centers of ‘‘the industrial world,’’ Bunker provides insight into how and why technological innovation takes place. First, in the actual process of extraction, the raw materials tend to get more and more difﬁcult to harvest as time goes on; for example, surface deposits of minerals are used up and people have to dig deeper, the most pure ores are depleted and users must shift to more amalgamated sources, etc. This requires the application of bigger, more powerful equipment, new techniques, etc. Bunker argues that many of the crucial technological innovations in the past several hundred years, many of which had a broad application across a wide range of industries (including all manner of manufacturing), were initially stimulated by the necessity to develop better ways to extract raw materials from nature. Second, primary goods by their very nature tend to be bulky and often occur in nature far from the industries and populations that will consume them. Therefore, transportation becomes another major problem that must be solved through technological change and the provision of infrastructure. Indeed, the long distance and eventually global transportation systems that evolve to handle the movement of bulky raw materials become the sinews of the increasingly ‘‘globalized’’ world, and the circuits through which the industrial products ﬂow around the world, too. The lack of attention to improved transportation systems and to the role of various technological and organizational innovations in shaping contemporary globalization strike me as a particularly gaping hole in most GCC analyzes. There is no doubt that the availability of various worldwide networks of both communication and transportation provide the essential ‘‘backbone’’ for the current ‘‘global assemblyline.’’ For instance, the rise of export-oriented manufacturing in East Asia, the motor of some of the purported ‘‘economic miracles’’ in that region, requires the availability of fast, reliable and relatively cost efﬁcient shipping across the Pacific. Indeed, the key role of this containerized sea trade became apparent recently with the dock strikes on the U.S. west coast; the enormous ﬂow of consumer goods between Asian manufacturing sites and North American retailers and consumers was temporarily shut off, and even a short break in this ﬂow threaten various enterprises with dire economic straits. But this long distance shipping – and the other technologies that facilitate consumer-driven GCCs – is not the object of much sociological research. Instead, studies tend to focus

Starting at the Beginning

153

on the organization of manufacturing or marketing, while taking the technical infrastructure on which they depend on for granted. Bunker’s arguments about the genesis of the necessary technology and transportation systems for various types of ‘‘globalization’’ in worldwide sourcing of raw materials are insightful. It should provide a starting point for more research on phenomena like long distance shipping and the other technologies that facilitate consumer-driven GCCs. (3) The role of states (which still matter). One of the major debates about the nature of the current global condition involves the role of states and their purported ‘‘decline.’’ A number of trends seem to portend ‘‘the retreat of the state’’ (Strange, 1996): the rise of ‘‘global neoliberalism’’ as the dominant ideology of contemporary ‘‘globalization,’’ the move toward lower trade barriers and more open economies, the rising power of transnational corporate players of various stripes, a generalized enhanced reliance on market mechanisms and on the private sector – all supposedly in the service of upgrading national competitiveness, etc. (Milner & Keohane, 1996, p. 24; Cox, 1996, p. 22; Gummett, 1996, p. 15). Various conferences and edited volumes debate whether the state still matters (for instance, see Smith, Solinger, & Topik, 1999) and high-proﬁle debates about new theoretical approaches to globalization assume that the role of states is fundamentally different in today’s world-economy and that we need to escape passe ‘‘nation state centric thinking’’ (Robinson, 1998, 2001). The industry studies that predominate in the GCCs literature rarely ‘‘weigh in’’ on these larger debates. However, implicitly they seem to ‘‘go with the ﬂow’’: states and government policies rarely appear as major considerations in these analyses. Bunker’s analysis of the materio-spatial processes integral to raw material extraction at the beginning of the commodity chain directly challenges the proponents of the ‘‘destatization thesis.’’ While it certainly is true that the roles and power of some nation states ‘‘have changed over time with the technological, economic, ﬁnancial, and political changes that have moved the world-system through successive cycles of systemic accumulation that led to globalizationy(this) does not mean that we can dismiss the state as an important actor’’ (Bunker & Ciccantell, 2005). Indeed, Bunker and Ciccantell’s detailed account of the regional impacts of the giant Carajas mine in the Amazon make it abundantly clear that government ofﬁcials at the national, provincial and local level are all important actors shaping the trajectory of development, even if, in the long-run, the local forces seem more likely to ‘‘lose’’ in the face of national and international pressure for ecological destructive and economically unsustainable growth. It is obvious that the sort of massive investment in equipment and infrastructure that a

154

DAVID A. SMITH

giant mining operation demands will require state involvement at a variety of levels. This underlines a key point that sometimes gets forgotten in discussions of ‘‘destatization’’: capitalists need states, or, at minimum, some entity that will take over the functions that states have long performed. These political organizations set up the ‘‘rules’’ of the capital accumulation game, facilitate various types of monopoly, and minimize egregious sorts of corporate risk. An interesting aspect of the story of successive extractive regimes in Amazonia, evocatively told in the Bunker and Ciccantell book, is how state regulations not only protect the interests of key property owners, but also how these regulations need to renegotiated via contentious – and sometimes violent! – politics as the crucial commodity to be evacuated shifts over time. This is another instance where carefully peering into the raw material ‘‘box’’ at the beginning of the GCC helps us to see the key role of states. The dramatic imposition of giant ‘‘machines in the jungle’’ and a long rail line from the mine to the coast, traversing a number of jurisdictional and ecological zones, makes it impossible to ignore the highly political dimension of what is happening at this origin point of the productive chain. But, of course, it would also be just as true that ‘‘close analysis of how states respond to or are manipulated by different agents acting in a variety of power domains provides critical insights into the interaction of local and global processes’’ (ibid) at the various nodes further ‘‘downstream’’ on the GCC, as well. This is another contribution that a Bunkeresque analysis can offer. One modest caveat: my argument in this chapter is that ‘‘starting at the beginning’’ via Bunker’s analysis of extractive economies can greatly enhance our understanding of GCCs. Looking at this ‘‘upstream’’ source, which is generally ignored by most sociologists in this subﬁeld, helps us better understand the origins of globalization, a major impetus for technological change, the importance of states and politics, etc. But, once again, the elephantine dimensions of the world-system make it difﬁcult for scholars working at one end to fully understand what is going on at the other. This may have something to do with a series of claims that Bunker and Ciccantell make about ‘‘industrial economies.’’ They argue – very persuasively – that extractive activities need to locate near naturally occurring raw materials, so these places are widely dispersed, heterogenous, and diverse. But they make another argument that, at least in the contemporary period, is more dubious: ‘‘industrial economies expand materially while agglomerating spatiallyy The advantages of spatial agglomeration centralize and concentrate industry in space while the dynamics of depletion disperse raw material extraction in space.’’ In a note to that passage, the authors correctly argue that there is a tendency in a society like ours to ‘‘agglomerate around heavy

Starting at the Beginning

155

transportation nodes – ports, railheads, and highways.’’ Later on in their book they even more baldly claim that the history they examined reveals the evolution of ‘‘a singular, increasingly homogenous industrial, ﬁnancial and political core and multiple, increasingly heterogenous and far-ﬂung peripheries.’’ Frankly, I am uneasy about this idea of a unitary and uniform ‘‘industrial core’’ and have serious doubts about the ‘‘agglomerative’’ tendencies of core urban areas and/or manufacturing establishments. Indeed, one of the main messages of the economic restructuring literature is that various types of manufacturing are also becoming ‘‘far-ﬂung’’ (both around the globe and in ‘‘greenﬁeld’’ locations within the core countries). Furthermore, urban sociologists know that there a many centrifugal forces (including many related to today’s transportation technologies) that create sprawling, as opposed to agglomerating, built environments. These dubious claims about the nature of ‘‘industrial economies’’ do not seem crucial to their arguments, so I am not sure that they are worth making.

4. CONCLUSION This chapter is a ﬁrst cut at arriving at a better conceptualization of GCCs that incorporates the insights that Stephen Bunker offers on the nature and dynamics of extractive economies. It explores the feasibility of constructing a synthetic approach to GCCs that combines the insights of people working on ‘‘the beginning of the chain’’ (like Bunker and some of the other scholars in this volume) with analyses like those of Gary Gerefﬁ on export manufacturing. Both of these strands of work are useful and can help us deal with various dilemmas of contemporary ‘‘globalization.’’ Clearly work on extractive economies has a great deal of insight to offer. It answers a number of questions about technology, transportation, and the role of politics and the state that are hardly broached by many GCCs scholars. It is time for the two perspectives and ‘‘literatures’’ to enter into a serious, long-overdue dialogue. Doing so could enhance our conceptual grasp of the dynamics of global development and inequality.

REFERENCES Appelbaum, R., Smith, D., & Christerson, B. (1993). Commodity chains in the pacific rim: Garment trade and manufacturing. In: G. Gerefﬁ & M. Korzeniewicz (Eds), Commodity chains and global capitalism (pp. 187–204). Westport, Conn: Greenwood.

156

DAVID A. SMITH

Bunker, S. (1984). Modes of extraction, unequal exchange, and the progressive underdevelopment of an extreme periphery: The Brazilian Amazon, 1600–1980. American Journal of Sociology, 89(5), 1017–1064. Bunker, S., & Ciccantell, P. (2005). Globalization and the race for resources. Baltimore, MD: Johns Hopkins University Press. Cox, R. W. (1996). A perspective on globalization. In: J. Mittelman (Ed.), Globalization: Critical reflection (pp. 21–30). Boulder, CO: Lynne Rienner Publishers. Dicken, P. (1998). Global shift: The internationalization of economic activity (3rd ed.). New York: The Guilford Press. Elson, D. (1989). Bound by one thread: The restructuring of UK clothing and textile multinationals. In: A. MacEwan & W. K. Tabb (Eds), Instability and change in the world economy (pp. 187–204). New York: Monthly Review Press. Emmanuel, A. (1972). Unequal exchange: A study of the imperialism of trade. New York: Monthly Review Press. Feuntes, A., & Barbara, E. (1984). Women in the global factory. Boston, MA: South End Press. Firebaugh, G., & Bullock, B. (1987). Level of processing of exports: Estimates for developing nations. International Studies Quarterly, 30, 333–350. Frobel, F., Heinrichs, J., & Kreye, O. (1980). The new international division of labour: Structural unemployment in industrialised countries and industrialisation in developing countries. Cambridge: Cambridge University Press. Gerefﬁ, G. (1990). Paths of industrialization: An overview. Manufacturing miracles: Paths of industrialization in Latin America and East Asia. Princeton, NJ: Princeton University Press. Gerefﬁ, G. (1992). New realities of industrial development in East Asia and Latin America: Global, regional, and national trends. In: R. Appelbaum & J. Henderson (Eds), States and development in the Asian Pacific Rim (pp. 85–112). Beverly Hills: Sage. Gerefﬁ, G., & Korzeniewicz, M. (1990). Commodity Chains and Footwear Exports in the Semiperiphery. In: W. Martin (Ed.), Semiperipheral states in the world-economy (pp. 45–68). New York: Greenwood Press. Gerefﬁ, G., Korzeniewicz, M., & Korzeniewicz, R. (1994). Introduction: Global commodity chains. In: G. Gerefﬁ & M. Korzeniewicz (Eds), Commodity chains and global capitalism (pp. 1–14). Westport, Connecticut: Greenwood. Gummett, P. (Ed.) (1996). Globalization and public policy. Cheltenham, UK: Edward Elgar. Henderson, J. (1989). The globalisation of high technology production: Society, space, and semiconductors in the restructuring of the modern world. New York: Routledge. Hopkins, T., & Wallerstein, I. (1986). Commodity chains in the world-economy prior to 1800. Review, 10(1), 157–170. Jaffee, D. (1985). Export dependence and economic growth: A reformulation and a respeciﬁcation. Social Forces, 64, 102–118. Klein, N. (2000). No logo: Taking aim at the brand bullies. New York: Picador. Lipietz, A. (1982). Towards global fordism. New Left Review, 132, 33–47. Lipietz, A. (1987). Mirages and miracles: The crisis of global fordism. London: Verso. Malecki, E. (1991). Technology and economic development: The dynamics of local, regional, and national change. New York: Longman Scientific and Technical. Milner, H. V., & Keohane, R. O. (1996). Internationalization and domestic politics: An introduction. In: R. Keohane & H. Milner (Eds), Internationalization and domestic politics (pp. 3–24). Cambridge, UK: Cambridge University Press.

Starting at the Beginning

157

Piore, M., & Sabel, C. (1984). The second industrial divide: Possibilities for prosperity. New York: Basic Books. Porter, M. (1990). The competitive advantages of nations. New York: Free Press. Reich, R. (1991). The work of nations: Preparing ourselves for the twenty-first century capitalism. New York: Alfred A. Knopf. Robinson, W. (1998). Beyond nation-state paradigms: Globalization, sociology, and the challenge of transnational studies. Sociological Forum, 13(4), 561–594. Robinson, W. (2001). Social theory and globalization: The rise of a transnational state. Theory and Society, 30(2), 157–200. Ross, R., & Trachte, K. (1990). Global capitalism: The new leviathan. Albany: State University of New York Press. Schoenberger, E. (1988). From fordism to ﬂexible accumulation technology, competitive strategies and industrial location. Environment and Planning D: Society and Space, 6, 245–262. Scott, A. (1987). The semiconductor industry in South East Asia: Organization, location and the international division of labor. Regional Studies, 21, 143–160. Scott, A. (1988). Flexible production systems and regional development. International Journal of Urban and Regional Research, 12, 171–186. Smith, D., Solinger, D., & Topik, S. (Eds) (1999). States and sovereignty in the global economy. London, UK: Routledge. Storper, M., & Walker, R. (1989). The capitalist imperative: Territory, technology and industrial growth. New York: Basil Blackwell. Strange, S. (1996). The retreat of the state: The diffusion of power in the world economy. New York, NY: Cambridge University Press.

SUNK COSTS, RESOURCE EXTRACTIVE INDUSTRIES, AND DEVELOPMENT OUTCOMES Bradford L. Barham and Oliver T. Coomes ABSTRACT Sunk costs are a key feature of extractive industries that profoundly shape regional economic development outcomes. In this chapter, we argue that sunk costs do so by influencing both the investment behavior of firms and the organization, as well as the performance, of extractive industries in ways that often deviate substantially from traditional neoclassical models of competitive markets with resource mobility. Sunk costs are defined, and the features that give rise to such costs are identified, followed by an analysis of the impacts of sunk costs on firms, regions, and economies. Sunk costs are shown to underlie two important phenomena associated with the economic experience of resource extraction – ‘‘Dutch Disease’’ and the ‘‘resource curse’’. The chapter concludes with a discussion of the need for development policy to incorporate often overlooked sunk cost considerations into efforts to promote economic development in extractive economies.

Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 159–186 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10008-0

159

160

BRADFORD L. BARHAM AND OLIVER T. COOMES

1. INTRODUCTION Few sights are as impressive as the massive port works, open-pit mines, and 500-mile railways developed to tap the natural resources of frontier regions, or so bittersweet as the relic landscapes left behind in the wake of resource booms. Abandoned mines, idle processing facilities, vacant warehouses, empty ports, disused railroads, boarded-up buildings, and underemployed residents in once vibrant regions speak not only to the capricious nature of resource economies but also to the salience of ‘‘rigidities’’ in investments in extractive industries. Stripped bare of all that could be profitably moved, the remaining specter of idle physical and human resources in once booming areas provides a poignant reminder that investment patterns profoundly affect industrial and regional development paths. A basic, but often overlooked, feature of extractive investments is that they are predominately sunk, giving rise to the phenomena of ghost towns and idle capital and human resources; once they are made, their salvage value for alternate uses in situ or in other locales is less than the costs of transfer. This ‘‘sunkenness’’ of extractive investments fundamentally shapes the investment decisions and actions of ﬁrms and governments, inﬂuences the subsequent industrial organization of the sector and its connections to the regional economy, and thus holds major implications for economic development, especially in economies with high levels of dependence on natural resources. This chapter, inspired by Stephen Bunker’s work on natural resources, develops an integrated view of sunk costs in extractive industries, one that stretches from the micro level of the spatial, biophysical, and basic market features of the resource to the intersection of extractive industries with national development outcomes and the global political economy. It also aims to provide a ﬁrmer microeconomic foundation for the recent ﬂurry of writing on the ‘‘resource curse’’ (Sachs & Warner, 1995, 2001; Gylafson, 2001; Neumayer, 2004; Atkinson & Hamilton, 2003), and the Dutch Disease (Corden & Neary, 1982), which explore the hypothesis that resource abundance contributes to low growth rates and thus to divergence in development performance between resource-rich and resource-poor economies. Inspired by the seminal work of Harold Innis on the role of primary commodity industries in Canadian economic development, the theme of investment rigidities, or ‘‘sunk costs,’’1 in extractive industries is prevalent in this volume. For most scholars since Innis, sunk costs create ‘‘path dependence’’ for regional development: future prospects for growth and structural change are contingent upon the characteristics of industries that propelled the regional economy in the past. Where few options are available

Sunk Costs, Resource Extractive Industries

161

for making use of the related ﬁxed investments once the boom begins to wane, prospects for economic development are likely to be limited; as such, the fortunes of the region remained tied rather strictly to those of the original boom activity. Given the often irreversible nature of investments in extractive industries, the resulting path dependencies are especially important to understanding development outcomes in resource-rich regions. The potential of extractive activities to produce highly problematic development outcomes reaches beyond those captured by the notion of Innisian path dependency. Sunk costs fundamentally shape both the investment behavior of ﬁrms and the organization and performance of industries in ways that can deviate substantially from neoclassical models of competitive markets with resource mobility. In particular, a high degree of sunk costs can generate investment levels that are lower, or paradoxically even higher, than optimal levels predicted by conventional models. If the extractive industry is important to the region, the inefﬁciencies and inequities that result from these investment patterns can limit the industry’s contribution to regional economic development, both during its heyday and thereafter. In addition, sunk costs in a leading extractive sector can act to distort investment patterns of both private and public agents in the surrounding regional economy, again in ways that limit the leading sector’s contribution to development. This tendency for regional economic distortions to arise in the presence of extractive industries can be especially important if the leading sector undergoes a major boom era of high returns, which when combined with sunk costs in the extractive and perhaps surrounding sectors can set the economy on an unsustainable course. In that respect, this chapter provides much needed microfoundations for the ‘‘resource curse’’ and ‘‘Dutch Disease’’ literature by identifying the deeper logic of distorted investment patterns that tend to be associated with extractive industries. We begin our presentation with an overview of the nature and causes of sunk costs. In Section 2, we explore the wide range of factors that can make investments sunk, and conclude that sunk costs are far more pervasive than is commonly recognized. In Section 3, we build on Innis (1930, 1933, 1956) and our work elsewhere (see Barham, Chavas, & Coomes, 1998; and Barham & Coomes, 1994a, b) to explain further why sunk costs are of special significance in extractive industries. We do this by exploring the potential degree of sunk costs associated with investments of capital, labor, and information in different extractive industries and in the various stages of the extractive process. We then examine in Sections 4 and 5 the effects of sunk costs on the efﬁciency and equity of industry and economy-wide performance. We identify four major ways that sunk costs lead to problematic

162

BRADFORD L. BARHAM AND OLIVER T. COOMES

development outcomes. In the conclusion, we discuss the need for an improved understanding among scholars and policy-makers of the broad implications of sunk costs in natural resource industries, and the potential value – in this era of market liberalization, free trade, and liberalized foreign direct investment – of well-managed regulation for promoting economic development in resource-rich regions.

2. THE NATURE AND CAUSES OF SUNK COSTS 2.1. What are Sunk Costs? All economic activities involve some degree of investment, which the investor anticipates will generate a ﬂow of favorable returns. In considering investment in extractive industries, observers tend to focus most attentively on the more obvious setup costs; e.g., outlays necessary to secure rights to the resource and to construct (or purchase) the physical infrastructure for the extraction, processing, storage, and transportation of raw materials and their byproducts. Other forms of investment, however, are also important. Substantial expenditures are made to acquire information on the resource’s location and characteristics, to recruit, screen, hire, and train workers, to comply with government regulations and registration requirements for initiating extractive activity, and to form contracts and business relations with input suppliers and buyers. Generally, as long as an expense incurred by a ﬁrm is undertaken with the intention that it will bring value over time (i.e., because it generates a stream of future returns), then that expense can be thought of as an ‘‘investment.’’ Sunk costs refer to those investments that, once undertaken, cannot be fully recovered through their transfer or sale. The ‘‘irreversibility’’ of investments is frequently viewed as resulting from their ﬁrm- or industryspecific nature (Dixit, 1980; Spence, 1977). However, sunk costs can arise even when investments are transferable to other economic activities or sectors; such is the case where transfer costs make a certain portion of the value of the original investment essentially unrecoverable. To estimate sunk costs, therefore, an analyst would ﬁnd the difference between the value of the original investment and its salvage value (i.e., the value if sold or transferred to another use). Where the salvage value of an investment is high, the sunk costs by definition will be low. Thus, for any given investment a wide range of reversibility can exist: from ‘‘full reversibility’’, where sunk costs are zero because the salvage value equals the original investment, to ‘‘full

Sunk Costs, Resource Extractive Industries

163

irreversibility’’, where sunk costs equal the full value of the original expenditure because there is no salvage value to the investment. The extent of sunk costs in a given investment also can vary over time and place. Through time, the salvage value will ﬂuctuate with changes in the activity’s expected net revenue stream (independent of the original investment costs) or with the value of alternative uses of the investment. Thus, a price shock, an input cost shock, or changes in the viability of alternative uses of the investment can all affect the salvage value. The extent of sunk costs also vary with place, in that the actual location of the investment can affect the prospects for its sale or transfer and hence the salvage value. A more precise measure of sunk costs would also incorporate, at any given time, a value to reﬂect the physical-economic depreciation of the investment, which would be deducted from the original investment cost (Dixit & Pyndick, 1994).2 A fully reversible investment under this modiﬁcation would have (at any given time) a salvage value equal to its original cost less the physical-economic depreciation that has occurred during its lifetime. So far, our attempt to deﬁne sunk costs has been in terms of the extent of sunk costs associated with a particular investment. However, any given investment activity is likely to involve a set of distinctive yet interrelated investment expenditures (e.g., locating the valuable resource, acquiring rights, building the physical infrastructure for its extraction, hiring the workers, and so on). Thus, the extent of sunk costs associated with a set of investments may not simply be the sum of the sunk costs of each distinctive form of investment, treated separately. The salvage value of different forms of investment could well be affected by the sale or transfer of other interlinked investments, so that the salvage value of the set of investments must be analyzed jointly. The contingent and conjunctural nature of sunk costs means that generally no single measure or estimate can be identiﬁed at the moment of original investment decision that will apply over the entire life of the investment. Instead, the sunk cost measure must be continually updated with information about the evolving salvage value of the investment in order to have a proper evaluation of the worth of the assets in question.

2.2. What are the Causes of Sunk Costs? 2.2.1. Physical Characteristics of Investment Perhaps, the most commonly understood cause of an investment being sunk is that the physical characteristics make the installation costly to set up, remove, and relocate or industry-specific in its potential use. For example,

164

BRADFORD L. BARHAM AND OLIVER T. COOMES

the task of extracting, separating, and moving large volumes of raw materials will require sturdy, secure facilities with high installation and deinstallation costs. In addition, these investments, especially in the mining and energy sectors, must often be quite large to achieve a minimum efﬁcient scale. Large investments in secure removal, separation, and transport equipment and infrastructure characterizes many extractive activities.

2.2.2. Investment Specificity and Remoteness Sunk costs are also caused by the speciﬁcity of investment to a ﬁrm or particular activity, even if the investment does not involve physical capital. Of course, expenses that are not inherently ﬁrm- or industry-specific also can be sunk. Investments can have alternate uses to other industries or ﬁrms, yet their transferability may be limited by the physical costs of moving them from their original location, the transaction costs of arranging for their sale or transfer, or by the economic conditions in other sectors of the economy which might be most likely to use them. Any factor that increases the transfer costs of an investment to an alternative use can raise sunk costs by reducing its potential salvage value. Of particular importance here is the location of investment, or more specifically its remoteness. Remoteness contributes to raising sunk costs in two fundamental ways. First, the actual cost of the original investment is likely to be higher, the more remote is the locale; indeed the notion of ‘‘remote’’ reﬂects such access costs. Second, remoteness will also tend to reduce the salvage value of investments – in several ways described later below – , thereby heightening the associated sunk costs. In particular, to move such investments from remote areas, or transfer them to other uses in situ, will tend to be expensive, thereby lowering the salvage value. Although perhaps somewhat counterintuitive, remoteness can increase sunk costs even for such seemingly mobile investments as human resources. People are mobile and many will recoup what they can and seek out new economic opportunities when local opportunities fade. However, in a remote location undergoing a major industry decline, the salvage value for local ﬁxed investments by its residents (e.g., housing, woodlots, piers, not to mention social ties and community) will be low. As a result, many residents may choose to stay, despite the prospect of low returns on their labor and business capital, because the consumption value of their local material and social investments remains greater than the value they could salvage upon moving. In this way, the remote locale potentially imbues an otherwise nearly fully reversible investment of human resources with a significant degree of sunk costs as well.3

Sunk Costs, Resource Extractive Industries

165

2.2.3. Transaction Costs Informational (or transaction) costs also create sunk costs. Specifically, the transfer of an investment entails certain costs which, in turn, reduce the potential salvage value of the investment. Such transaction costs arise from a variety of sources (see Williamson, 1985), including the process of negotiating the terms of the transfer, informational asymmetries between the buyer and seller concerning the real value of the transferred asset, adjustments the buyer will have to make in terms of hiring or retraining labor and management or reorganizing the operation’s human resources, and developing specific relations with input suppliers and buyers that may not be perfectly transferable. All such transaction costs reduce the potential salvage value of the investment and thereby increase sunk costs. Of the various sources of transaction costs, the relationship between informational asymmetries and sunk costs is perhaps least obvious. A resale problem arises for many types of equipment, Akerloff (1970) argues, because of the asymmetry in information between what the seller knows and what the buyer does not know about the condition and effectiveness of the item under consideration. Because of the difﬁculty (or costs) of evaluating the quality of second-hand equipment, buyers will tend to offer a price that corresponds to what they perceive to be the average quality of goods in the market. This, in turn, makes sellers reluctant to sell equipment of aboveaverage quality, because they are unwilling to take the loss in salvage value associated with selling equipment of above-average quality for average prices. Thus, the asymmetry of information between seller and buyer tends to lower the average quality, and hence the price of resale markets for many types of equipment and services. When otherwise transferable inputs are subject to the ‘‘Market for Lemons Effect,’’ and are therefore more costly to transfer, investors face sunk costs in these expenses.4 Other types of informational asymmetries also create sunk costs. Uncertainty or inexperience on the buyer’s side with the technological processes involved in equipment use, or the underlying market conditions, can open gaps in beliefs between sellers and buyers about the value of physical or organizational capital. Where buyers have more uncertainty about the degree of down-side risk in the market or more doubts about the efﬁcacy of the technological processes, then sunk costs for the original investor will also tend to be higher than they would be in the absence of information asymmetries. Again, sunk costs associated with informational asymmetries may be more likely to occur in remote locales, where because markets are thinner, information about key features of distant markets may be less available.

166

BRADFORD L. BARHAM AND OLIVER T. COOMES

2.2.4. The Investment Package and Same Boat Effects Two ﬁnal features make investments more sunk than they would otherwise appear – what Barham, Chavas, and Klemme (1994) refer to as the ‘‘Investment Package Effect’’ and the Same Boat Effect. The Investment Package Effect arises where investments are made that are essential to the continuing operation of a facility, and hence in preserving the value of other sunk investments. The ‘‘Same Boat Effect’’ occurs when simultaneous efforts by ﬁrms (or individuals) to sell off similar investments drives down salvage values, thereby increasing the level of sunk costs. Such conditions are most likely to arise when down-side risk is realized (e.g., a sharp price drop) and widely felt across an industry or across an economy, prompting ﬁrms to sell off their investments and secure their salvage value. To the extent that the shock occurs in a specific industry, the magnitude and scope of impacts will depend on the size of that industry relative to the economy as a whole and the potential utility of sunk investments to other industries. Still, the many sellers of similar investments, essentially all in the same boat of seeking to reduce their down-side losses, can drive down resale prices and hence salvage values, as purchasers of the transferable equipment ﬁnd themselves in a buyers market ﬁlled with many prospective suppliers. A similar effect may be experienced with non-industry specific investment goods when related sectors are also suffering reduced demand for goods and hence create a buyers’ market. All such effects are likely to be heightened in remote regions. When adverse conditions are conﬁned to a locale or region, then prospective relocation of physical and human capital will force net resale prices downward. In more remote areas, transportation costs to relocate capital will reduce resale value, but so will the fewer options to offset the Investment Package Effect, and the greater potential for the Market for Lemons Effect of a product in a distant market. Moreover, the ‘‘Same Boat Effect’’ will be magniﬁed in isolated regions as the pool of potential buyers is smaller than they would be in a well-integrated area. Thus, once more, the remoteness of the original investment becomes important in reducing the potential salvage value of investments, even if such investments are not inherently ‘‘sunk’’ by their actual physical nature.

3. EXTRACTIVE INDUSTRIES AND SUNK COSTS The task of assessing the potential for sunk costs in a specific extractive activity is deceptively difﬁcult. Sunk costs vary substantially not only across

Sunk Costs, Resource Extractive Industries

167

industries, but also within industries over space and time. In some cases, the presence of sunk costs is quite apparent and difﬁcult to dispute, as in a mineral processing facility in a remote locale. In others, though, the variety of causes of sunk costs and their potential variability over space and time make the actual identiﬁcation of sunk costs a major challenge. Moreover, within particular extractive industries, sunk costs will be highly activityspecific and contingent, both geographically and historically. Such complexity suggests that the task of developing a comprehensive, systematic analysis of the occurrence and causes of sunk costs across extractive industries lies well beyond the scope of this chapter. Instead, we focus in this section on exploring the causes and pervasiveness of sunk costs in extractive industries.

3.1. The Stages of Extractive Activity Sunk costs arise in each of four basic stages in extractive activity: discovery and appropriation of the raw material, extraction, processing, and transportation. Initially, ﬁrms invest capital, labor, and managerial effort in ﬁnding the natural resource of interest; discovery costs can be substantial, as can investments needed to secure ownership rights to the raw material. Labor investments can include not just the wage bill in this stage but also the costs of recruiting, training, and monitoring labor. Second, investments are made to extract the raw material from nature. Such investments include not only equipment and labor for extraction but also for information and technology management, and for long-term cleanup costs that are not compensated by future use or regulatory authorities. Finally, investments are made to process the raw material and transport it to markets. Clearly, processing and transportation can occur in either order, depending on whether some or all of the processing is handled more economically near the site of extraction or the market.5 Again substantial costs are incurred in these last two stages for physical capital, labor, information, and clean up.

3.2. Investments in Extractive Industries: Specific, Remote and Lumpy Extractive industries are also prone to sunk costs because of the particular nature of investments. In each stage of extraction, investments tend to be rather activity-specific due, in part, to the unique physical features of raw materials, their specific geographical location, and the large-scale technologies involved in their discovery, extraction, processing, and

168

BRADFORD L. BARHAM AND OLIVER T. COOMES

transportation. Most exploration for rich concentrations of raw materials involves the use of equipment and personnel oriented toward a certain resource; such speciﬁcity in orientation may reduce the prospects of discovering prime reserves of other materials. Depending on the resource in question, a considerable portion of the total costs can be incurred at this early stage, and such costs are quite likely to be sunk, especially when the investment yields no significant discovery. In the extraction stage, the infrastructure used to remove the resource from nature is engineered to extract the material from the locale as efﬁciently as possible. Extractive equipment often is designed to conform precisely to the material properties of the particular raw material and its natural setting; consequently, both the capital and technology used are highly specific to the industry and sometimes even to the site of extraction. The costs of dismantling major installations usually make it cheaper to buy and install new infrastructure and equipment than to relocate extant installations. As such, both the installation and de-installation costs involved in the extraction stage reduce the salvage value of these investments and hence increase the potential level of sunk costs. Reﬁning and initial processing also often require technologies designed to conform to the physical and chemical characteristics of the extracted material. Thus, the processing equipment is also likely to be highly specific to the material at hand, which further increases the potential for sunk costs in the activity. In the case of certain raw materials, such as bauxite, the speciﬁcity of design in the processing equipment is so closely linked to the chemical features of the raw material that the processing equipment is designed to handle the resource from a single site or source (Stuckey, 1983). In these cases, the extreme speciﬁcity involved in the processing investment makes likely a high degree of vertical integration between the raw material site and the processing ﬁrm (in the form of a long-term contract or direct purchase). Either way, such a commitment can mean a higher level of sunk investment for the economic agent involved. Transportation and storage facilities may also be designed around the particular physical characteristics of a resource in a way that promotes more sunk costs. Different resources have particular ‘‘volume to value ratios’’ and thus require quite distinctive transport and storage systems. Thus, the transport and storage infrastructure developed to serve one type of extractive activity in an area may not be suitable for what turn out to be reasonable alternatives (Priest, 1995, 2003). In this manner, investments in a stage of extractive activity which may not seem to entail much in the way of sunk costs can actually be highly prone to sunk costs.

Sunk Costs, Resource Extractive Industries

169

Because raw materials are ﬁxed spatially by their occurrence in nature, only by chance are they likely to occur near important markets or other industries that may serve as potential alternative users of the inputs employed for raw material extraction and transport. As much of Stephen Bunker’s work suggests, increased isolation of many raw materials is also a socially constructed phenomena to the extent that the more proximate reserves have been exhausted, and so extraction moves to more remote locales (see Bunker, 1994a, b). This remoteness tends to make investments in extractive activity more sunk than otherwise would be the case for the reasons explored in the previous section. Remoteness may especially deepen the sunkness of investments in transport and storage facilities which inherently might have alternative uses, but no such options are available when the infrastructure serves a remote site. A third reason why sunk investments are pervasive in extractive activity is that investment requirements, particularly in minerals and non-mineral energy resource developments, often tend to be ‘‘lumpy’’ – that is, investments are only economical at a relatively large scale requiring tightly coordinated and integrated stages (Bunker, 1994b; Priest, this volume). This is so, in part, because of the scale economies that arise in mining or extracting and reﬁning of raw materials. Transport infrastructure requirements also tend to promote larger scale operations, because there is a substantial ﬁxed cost imposed by these installations which can be minimized on a per unit basis by increasing the volumes of materials transported. The lumpiness of extractive investments also occurs in another sense, in that investment in the different stages of the operation have to be tightly coordinated to avoid having investments in one stage generating no revenue as they wait for other investments to be completed. As Bunker (1994b) demonstrates, this need for timeliness in extractive industry investments can make them even larger in their initial costs, which for the reasons given above (and below) are also likely to be sunk.

4. SUNK INVESTMENTS, INDUSTRY ORGANIZATION, AND DEVELOPMENT OUTCOMES Sunk costs can inﬂuence development outcomes in extractive industries in several ways that are distinct from the predictions based on neoclassical models of competitive markets with resource mobility. Two effects of sunk costs occur at the microeconomic level of industry organization and

170

BRADFORD L. BARHAM AND OLIVER T. COOMES

performance. First, sunk costs create the potential for ﬁrms to engage in strategic behavior aimed at inﬂuencing the actions of other participants or potential entrants and hence the degree of competition within the industry. Second, sunk costs, when combined with uncertainty about future net revenue streams, can lead to socially inefﬁcient investment outcomes, i.e., either too little or too much investment, depending on the interactions of markets and public policies in these industries. Sunk costs thus constrain competitive and ﬂexible resource allocation at the industry level, which in turn can undermine both efﬁciency and equity outcomes in developing countries where many extractive industries play major roles in regional and national economies.

4.1. Sunk Costs and Strategic Behavior in Extractive Industries In terms of strategic behavior, sunk costs essentially represent a form of commitment to participate in an industry that would not be present if the costs could be readily reversed. The higher is the portion of sunk costs to total costs, the less the ﬁrm has to lose by choosing to stay in business once the sunk costs are incurred. Sunk cost commitments thus hold strategic value to the incumbent ﬁrm by creating barriers to entry. Erstwhile competitors, who have yet to sink their costs, may be deterred from entering the industry because of the prospect of making losses once their output and sunk cost capacity commitment are added to those of the incumbent. If an incumbent ﬁrm faces zero costs in the future, entrants face a ﬁrm with nothing to lose. If, in fact, entry is deterred because of existing sunk cost investments, incumbent ﬁrms gain some degree of market power, which can then be used to raise prices and profits above competitive levels. Research by Krugman (1985), Auty (1993, 1994), Barham (1994), Woo, Glassburner, and Nasution (1994), Mikesell (1997), Barham et al. (1998), Davis (1997), and Aggarwal and Narayan (2004) show why sunk costs and strategic behavior are likely to be of special importance to the industrial organization and performance of extractive activities in developing countries. One fundamental mechanism highlighted by Krugman and Barham, Chavas and Coomes is how sunk costs help ﬁrms, which by luck or design can secure a ﬁrst-mover advantage, to invest strategically to raise rivals costs and thus to limit or deter future entrants. Extractive industries, especially minerals and energy resources, are strong candidates for this type of preemptive sunk cost investment strategy. This preemptive strategy is most likely to be viable for natural resources with

Sunk Costs, Resource Extractive Industries

171

certain physical features: specifically, where the variation in the quality of the appropriated resources is sufﬁcient to create major cost differences across sites, where the prime reserves are concentrated in a few sites, and where the resource can be readily and relatively cheaply appropriated as private property. These features are shared, for example, by mineral ores; grades vary spatially, are often highly concentrated, and are more readily appropriated than other inputs, such as information or skilled labor for other economic activities. In such cases, the incumbent ﬁrm(s), by securing a small number of strategic purchases or concessions, can deter potential rivals by pushing them toward inferior reserves of key inputs. Whether incumbents actually accomplish this in a given industry depends on the cost and feasibility of securing a ﬁrst-mover advantage. In the early stages of such a strategy, establishing a ﬁrst-mover advantage is likely to require some imbalance in the bargaining position of the original holders of the prime reserve sites (say, a national government) and the ﬁrm seeking to establish its strategic position in an industry, with the imbalance favoring the ﬁrm. Without this type of bargaining advantage, the original holder would be in a position to play off rival ﬁrms against one another, as each sought to establish control over the prime sites. The resulting increase in the cost of capturing the prime resource sites would quickly obviate the viability of a preemptive strategy aimed at capturing control over reserves of the scarce input. First-mover advantages can originate by accident, luck, innovation, asymmetric information, or entry barriers (for example, patents and state licenses). Historically, prime reserves of scarce natural inputs often have been located in isolated regions of developing countries; a paucity of domestic information about their extent and market value combined with relatively weak states and other local institutions may have increased the prospects for this sort of asymmetry to arise, and thereby favor strategic capture of key sites by well-informed and well-capitalized foreign (multinational) ﬁrms (O’Hearn, 1994). Moreover, the fact that many innovations and the resulting patents for the reﬁnement and processing of raw materials were made by these ﬁrms in more developed countries also may have conferred on them an initial bargaining advantage. An initial advantage captured in one or two prime locales, as Krugman (1985) points out, tends to feed on itself, providing the ﬁrm with an even more advantageous market position. The incumbent ﬁrm becomes willing to pay a higher price than a later entrant for a prime source, because it can capitalize into the purchase price some portion of the value of the market power protected by the additional capacity investment in scarce reserves. If,

172

BRADFORD L. BARHAM AND OLIVER T. COOMES

as a result of this or other advantages, rivals drop out of the bidding, initial holders of the prime reserves (such as national governments) may actually ﬁnd themselves on the long side of the market, bargaining with the dominant ﬁrm to have their reserve purchased ahead of other holders of prime reserves. The development implications of sunk costs and strategic capacity efforts in extractive industries therefore depend on several factors. Certain features of extractive industries historically have favored domination of prime reserves by multinational corporations based in more developed countries. Multinational ﬁrms have been able to exploit, to their advantage, initial asymmetries in information about markets, technologies, and resources; to establish initial monopoly positions in the major consuming economies by technological advantages, mergers, or other factors present in those stages of the market; and to outmaneuver and outbargain weak states in remote regions, at times with the help of their home country or international institutions. Just as sunk costs and capacity investments offer strategic opportunities for foreign ﬁrms, so might they also do so for host countries. The ﬂip side of the strategic side of concentrated reserves would, for example, seem to be the opportunity afforded developing countries to acquire market power. Indeed, this prospect certainly has been a motivating force behind a large number of strategic initiatives by resource-rich countries during recent decades. Several factors seriously limit this potential, not withstanding the broad historical trend of recent decades toward rising national ownership and control of mineral and energy reserves.6 One well-recognized limitation is the problems that developing country industry managers face in trying to stay competitive in discovery, extraction, and transport technologies, product development, and marketing and distribution. Perhaps a more fundamental constraint on host countries realizing strategic advantage is that of location or jurisdiction. Even in those cases where prime reserves are concentrated in a few sites, they are rarely contained within one nation or region of a nation. Compared with a multinational company that can strategically integrate within a single organization the use of reserves across many regions and countries, whatever private, public, or quasi-public domestic institution might control the prime reserve in a given resource-rich region will have much less capability to act strategically in the industry. Coordination with other similar bodies in other countries or jurisdictions will run directly into conﬂicting objectives, not the least of which will be the basic contradiction that arises when attempting to agree on which sites will be assigned a reserve or underutilized status in an excess capacity strategy.

Sunk Costs, Resource Extractive Industries

173

Historical evidence of this fundamental limitation in coordination among prime exporting countries in resource industries abounds, as revealed in the difﬁculties associated with recent supply management efforts in bananas, bauxite, coffee, cocoa, copper, and tin. Only those cases – such as oil and diamonds – where a single, low-cost producer (Saudi Arabia) or a deeply inventoried producer (De Beers of South Africa) has notable capacity to discipline the rest of the industry, have we seen ‘‘coordination’’ proving to be more durable. Otherwise, the capability of developing countries to build strategic advantage over scarce reserves seems inherently hamstrung by coordination problems. Sunk cost and strategic capacity investments continue to be dominated by multinational ﬁrms. Therefore, the efﬁciency and equity implications associated with sunk costs and strategic capacity in extractive industries tend to be highly negative for the host region. On the efﬁciency side, if a viable excess capacity strategy is achieved via initial ﬁrst-mover advantages enjoyed by a multinational corporation, then output can be restricted well below competitive levels. The extent of under or even non-utilization of a given locale, the so-called ‘‘captive mine,’’ will depend significantly on how prime its reserves are relative to the rest of the company’s holdings, the extraction, reﬁning and processing capacity originally installed by the ﬁrm, and relatedly the production or export levels which the ﬁrm perceives to be necessary to maintain its property rights to the resource. Given that the strategy calls for excess holdings somewhere, clearly some, if not all, host regions could have output levels reduced below levels than would prevail in a more competitively organized industry. Although the remoteness of the extractive industry may limit the spillover effects of reduced output levels on the national economy, where the extractive activity’s role is significant, restrictions in output would have important local and regional impacts. On the equity side, the terms of the concession (i.e., the original purchase price of the reserve and accompanying taxes or trade duties) will determine the share of the beneﬁts captured by the ﬁrm and host country of controlling prime and strategic reserves. If the multinational is (or was) able to exploit historical conditions that enabled the ﬁrm to secure control over these resources at low cost, the equity impacts on the host country can be quite adverse. Both the strategic rents arising from market power and the resource rents due to the prime nature of the resource are at stake (Barham, 1994). Clearly, without regulation both the equity and efﬁciency impacts of sunk costs in extractive industries can be strongly negative.

174

BRADFORD L. BARHAM AND OLIVER T. COOMES

4.2. Sunk Costs, Uncertainty, and Investment in Extractive Industries Somewhat paradoxically, sunk cost commitments also create the potential for major losses and thereby make strategic investments quite risky. If a ﬁrm or its industry experiences an adverse price or cost shock, the ﬁrm may ﬁnd itself forced to abandon a major capital investment; alternatively, the ﬁrm may be in position to cover variable costs but not recover the costs of the original sunk investment. Thus, by restricting the incumbent’s exit option, sunk costs can evoke either market power or major down-side losses, depending on industry and market conditions. How then do uncertainty and sunk costs inﬂuence the industrial organization and performance of extractive activities?7 Microtheoretic research in economics on sunk costs (Lambson, 1991; Dixit & Pyndick, 1994; Chavas, 1994; Lambson & Jensen, 1995, 1998) has examined how uncertainty (and especially downside risk) discourages investment and distorts entry and exit decisions, where investments are less than fully reversible. Such effects are probably more common than the strategic capacity issues raised in the previous section, because the interaction of sunk costs and uncertainty is likely to affect a much broader class of economic activities. Whereas strategic capacity models can require significant ﬁrst-mover advantages, limited prime reserves, and/or significant cost differences across prime reserves, the prospect of meaningful down-side risks (such as price, output, or cost shocks) is all that is needed in an industry with significant sunk costs to create under- or possibly over-investment by limiting resource mobility. The mere presence of sunk costs and down-side uncertainty about future industry outcomes is sufﬁcient to ensure that ﬁrms will not allocate or reallocate resources ﬂexibly or smoothly according to price signals. The economic logic of the interaction of sunk costs and uncertainty is relatively simple (Chavas, 1994; Dixit & Pyndick, 1994). Taken together, sunk costs and uncertainty create two potentially important investment considerations to incorporate into profitability analyses; ﬁrst is the option value in exiting, which decreases with the extent of sunk costs, because greater sunk costs limit the ability of investors in bad times to avoid losses by selling off investments for a good salvage value; and second, is the option value to enter later, which arises because by delaying investment in a sunk cost activity, the potential investor is able to acquire additional information about future industry conditions before sinking capital. Interestingly, both options are generally overlooked in standard long-term profitability analyses, yet they are potentially inﬂuential.

Sunk Costs, Resource Extractive Industries

175

For participants who have already invested, the more sunk the investment, the more difﬁcult it is to sell off and avoid down-side losses, if industry profitability erodes. On the other hand, this exit constraint provides the incentive for investors or entrants to choose a ﬂexible, ‘‘wait-and-see’’ strategy. The value in waiting arises to the extent that industry conditions permit the investor to enter if long-term profitability conditions improve, or otherwise to avoid the bad times, if industry prospects deteriorate or continue to have large down-side risks. By taking a wait-and-see approach, the investor can earn an overall higher return, cashing in on the good prospects while avoiding the bad.8 The result of these two effects is that private investment in industries with sunk costs and uncertainty is discouraged, yielding socially sub-optimal levels of investment and entry. More efﬁcient levels could be achieved if the uncertainty facing individual investors is reduced via social insurance or other risk-reducing mechanisms; however, these private or public institutions are not established without costs, and thus will not necessarily emerge. Both the investment and entry–exit depressing effects of sunk costs under uncertainty may be highly salient for extractive industries in developing countries (Hum & Wright, 1994; Garland, Sandefur, & Rogers, 1990). Without measures aimed at reducing the interactive effects of sunk costs and uncertainty, private investment patterns in extractive industries will tend decidedly toward underinvestment. Private and public institutions serve to mediate this result when they reduce uncertainty or subsidize sunk cost investments, but there is no assurance these conditions will arise in a given situation without public intervention. If institutions are created to reduce uncertainty or subsidize sunk cost investments, there is of course the distinct possibility that the subsidy involved could induce the contrary outcome: too much investment in the sector (Dixit, 1991). Over-investment appears to have occurred recently on an international level in many extractive industries. In their book, Bunker and Ciccantell (2005) show how Japan seems to have succeeded in achieving an advantageous outcome for their consumers of raw materials by encouraging several resource-rich regions (Brazil, Australia, Canada, and South Africa) to invest in superports sized to ﬁt Japanese ships. For encouragement, Japanese ﬁrms supported by their government typically offered initial seed-money investments for the infrastructure with medium- or long-term guarantees to purchase large volumes of raw materials from the countries. Cumulatively, these inducements resulted in major capacity commitments by governments in these regions, and the unique ﬁt between the efﬁcient scale of operation of their superports and large Japanese ships has helped to assure Japan a steady, low-cost supply of prime raw materials.

176

BRADFORD L. BARHAM AND OLIVER T. COOMES

The susceptibility of resource-rich governments to this arrangement arises, in part, from the hesitance of private investors to move into an industry with sunk costs and uncertainty. Thus, in many developing countries, the state moves to invest heavily in these extractive industries, either directly through state agencies or indirectly through the provision of funds and special support to quasi-state enterprises. In so doing, their actions can combine to produce the reverse problem of over-investment, especially when these government-sponsored efforts do not account for the actions of participants in other regions and countries, so that extractive industry expansion occurs in several places simultaneously. In such a case, the sunk cost nature of these projects leads paradoxically to industry-wide over-investment and depressed prices, because the sunk costs involved act to slow subsequent adjustments to depressed prices. Either way, under the free market or by government intervention, inefﬁcient levels of investment in extractive activities are likely to arise unless careful attention is paid to the interactive effects of sunk costs and uncertainty. Choosing the most efﬁcient forms of intervention – from government sponsored investments for select price stabilization techniques to promotion of private-sector based, risk-mitigating institutions – would depend on a much deeper understanding of these interactions than is available in the current literature. Given the scale of many extractive industry investments in minerals and non-mineral energy resources, advances in this area of research could be of significant value to governments around the world, especially those banking on extractive industries for development.

5. SUNK COSTS AND THE STRUCTURAL EVOLUTION OF THE ECONOMY Sunk costs can also inﬂuence development outcomes at a macroeconomic level by constraining the future options available to the overall economy of a locale, region, or country. This inﬂuence may be felt in terms of the ‘‘path dependencies’’ established by the sunk cost industry, which guide the economy toward certain futures and away from others that would otherwise be more attractive. In the extreme, a sunk cost industry can become important enough in the overall economy that the patterns of investment in the booming sector and other non-tradeables can alter radically the entire economic structure. The resulting economic structure is less ﬂexible and less responsive to price shifts than conventional economic theory would suggest,

Sunk Costs, Resource Extractive Industries

177

which creates the potential for major economic decline when the boom sector declines or fails. In this sense, it may be the sunk cost nature of investments in the booming sector that underpins the kinds of deleterious outcomes predicted by the ‘‘resource curse’’ hypothesis and ‘‘Dutch Disease.’’

5.1. Sunk Costs and Path Dependency in Extractive Industries Much of Stephen Bunker’s research on resource industries (Bunker, 1994b; Bunker & Ciccantell, 2005) demonstrates how extractive industries tend to generate ephemeral development episodes, in which a region’s economic fortunes rise and fall with the rhythm of the leading resource sector. To some readers perhaps, these outcomes may seem to be a direct and inevitable product of the remoteness of extractive activities, which by accident of geography and natural surroundings are often too far from the resources, technologies, and markets needed to stimulate alternate economic activities. However, ‘‘remoteness’’ per se does not make regional economic development outcomes depend on the fortunes of extractive activities. Rather, remoteness reﬂects both the cost of the access and the lack of alternate options, which inﬂuence the level of sunk costs in local investments, and thereby development outcomes. Moreover, sunk costs limit future options even in regions where remoteness does not limit alternative options. Indeed, the negative impact on economic development, or the path-dependence created by sunk costs in extractive industries, is quite commonplace in the less extreme cases where regional economic alternatives are locationally possible but circumscribed by the speciﬁcity of the original investments. In these cases, the key issue lies in the degree to which existing investments in the sector, and other ancillary activities, can be reallocated toward new activities when opportunities change; the more sunk the existing investments, the less readily they will be transferred, and the more dependent the local economy will be on the original investments. The key features which lend themselves toward path-dependent development outcomes in extractive industries are: 1. Sunk costs tend to be high, both because of the inherent nature of the activity and other factors explored above; 2. Where extraction is the economic core sector of a region, ancillary infrastructural and institutional investments, including the investments of workers in their homes and ﬁrms in supporting sectors, tend to be

178

BRADFORD L. BARHAM AND OLIVER T. COOMES

strongly oriented toward and often specific to the core sector’s needs; and, 3. Regional policies of economic diversiﬁcation are thus often difﬁcult to prosecute because of the inherent limitations of sunk investments, the political-economic resistance that fundamental change might elicit from those private agents for whom the salvage value of their investments hinges (or seems to hinge) on the core sector, and the collective action challenge of pursuing a self-insurance scheme aimed at diversiﬁcation in what are often frontier-oriented communities. The factors which account for the ﬁrst two features should be relatively self-evident from our earlier discussion. The last feature, concerning the challenge of economic diversiﬁcation, is discussed below.

5.2. Sunk Costs and Dutch Disease Effects in Extractive Economies Resource booms often generate intersectoral investment and resource allocation patterns that beget rather fragile paths of economic development, a phenomena widely referred to as the ‘‘Dutch Disease’’ (see Corden & Neary, 1982). High economic returns in a booming resource sector (such as natural gas in the Netherlands during the 1970s, hence the expression) tend to attract capital and labor into that sector as well as into the production of nontradeable goods and services. If investments can ﬂow smoothly back into the production of other tradeable goods after the boom, then the Dutch Disease effects of intersectoral resource allocation on longer-run development possibilities would be largely inconsequential, albeit useful in explaining temporary structural shifts in the regional economy. If, however, investments in these favored sectors exhibit a high degree of sunk costs, then the resulting structure of the economy becomes quite rigid and vulnerable to future shifts in the price of the boom resource relative to other tradeable goods. The Dutch Disease phenomena thus helps account for why regional economies tend to become so hypertrophied and dependent on extractive industries. Boom sectors tend to arise, and in turn propel significant capital and labor ﬂows across sectors in the economy, where the potential is greatest for large gaps to arise between the price and the cost of the good. Extractive industries are prime candidates for large price–cost gaps for two reasons: ﬁrst, they have the potential for certain high-quality reserves to be exploited at significantly lower cost than at other sites; and second, extractive industries often face a high inelasticity of demand in consumer markets which can

Sunk Costs, Resource Extractive Industries

179

generate large price rises. Furthermore, the sunk costs of expanding production in new areas, when prices change abruptly, is often high enough to delay rapid entry from areas that are not currently in production. This retarding effect on supply affords extant exporters longer periods of large price–cost gaps and high returns. When high returns are realized by industry participants, fresh capital and labor are pulled strongly into the booming sector, thereby deepening the local reliance on that sector. In addition, the effect of the boom on other economic sectors is highly uneven. Local spending generated by increased incomes during a boom stimulates the production of more non-tradeable goods and services, where local producers can raise their prices to ensure the necessary returns to the factors employed there. Labor and capital are thus drawn both directly (via re-investment) and indirectly (via income increases) into the non-tradeables sector and away from the production of (non-boom) tradeable goods and services. The other tradeables sectors (e.g., agriculture or manufacturing) cannot pay the higher returns now demanded by labor and capital because of the price discipline they face from imports (tradeables), and are forced to reduce production in an attempt to match the more generous returns now available elsewhere in the booming economy. In terms of maintaining a ﬂow of regional exports that can support the rising consumption of imported tradeables and vibrant production activity in the non-tradeable sector, the regional economy becomes even more reliant on the booming sector. If the boom lasts long enough, the local economy can become almost entirely dependent on the prospects of the booming sector, with virtually all of its tradeable exports coming from that sector. If the investments which directly and indirectly gave rise to this unbalanced economic structure are not easily reversible, because of the high degree of sunk costs in investments in the booming extractive sector and the non-tradeables sector, and/or because there are substantial learning costs to re-entering production of other (nonboom) tradeable goods that were abandoned previously, then the Dutch Disease effects caused by the boom can result in an economic structure that is exceptionally vulnerable to a price decline in the boom sector. The Amazon Rubber Boom provides a quintessential historical case of the Dutch Disease phenomena with sunk costs (Barham & Coomes, 1996; Barham et al., 1998). But, by no means does the historical case of the Amazon Rubber Boom stand alone as an example of the Dutch Disease phenomena. In the 1970s and 1980s, numerous resource-rich countries, including Cameroon, Costa Rica, Colombia, Indonesia, Mexico, Nigeria, and Venezuela experienced Dutch Disease episodes of varying degrees of severity, as coffee and oil prices boomed. The consistent difﬁculties experienced by regional and

180

BRADFORD L. BARHAM AND OLIVER T. COOMES

national governments in coping with resource booms and overdependence on a booming sector suggest perhaps that the significance of sunk investments in extractive industries and economic development may be underestimated, a result, in part, of the lack of emphasis given to sunk costs in conventional microeconomic and macroeconomic theory. If investments are not recognized as sunk (in the fuller sense implied above), then the resulting economic structure could be viewed as more supple and resilient than is the case. Also, if the implications of sunk costs are not fully understood, then sunk investments will not be seen by policy-makers as being particularly problematic for longer-term economic development. Again, it is in this sense that sunk costs are potentially a key microfoundation of what has recently been labeled as the ‘‘resource curse.’’ Even with a heightened awareness of sunk costs, however, powerful political-economic forces come into play during resource booms that may make it difﬁcult for regional and national economic policy-makers to avoid succumbing to the Dutch Disease. The realization of high economic returns by those groups linked specifically to the booming and non-tradeables sector enable their rapid ascendance and the advancement of their interests in political spheres. As such, even prescient government ofﬁcials, political leaders, or civil organizations would experience difﬁculties in securing broad-based support to follow a ‘‘go slow’’ or ‘‘save for a rainy day’’ approach to economic development. Rather, state ofﬁcials, particularly, are likely to be under considerable pressure to eliminate infrastructural or institutional obstacles that may be slowing down the expansion of the booming sectors, driven by the promise of new and easy wealth. Furthermore, if the state is capturing a share of the boom’s rents through taxation, trade duties, or other measures, these burgeoning ﬁscal resources would attract considerable pressure from various other groups in society to have outstanding social and economic needs addressed, to say nothing of the opportunities created by these rents for improving the status and living standards of government ofﬁcials and of workers. In such a milieu, state expenditures will probably do more to encourage further sectoral shifts in the economy toward the booming sector and the non-tradeable sector, deepening in the process the future vulnerability of the region to price shocks in their leading sector. The ability of governments to pursue diversiﬁcation policies during extractive resource booms is also constrained by the remoteness and ‘‘frontier’’ character of resource-rich locales. In remote areas, fewer economic alternatives will appear to be viable, even in an economy with radically different relative prices than one undergoing a boom. Moreover, the relatively low returns in almost any investment oriented toward longer-term diversiﬁcation

Sunk Costs, Resource Extractive Industries

181

into the production of tradeable goods will also make it harder to justify the expenditure when much more money can be made in the near term in the booming sectors. Concerned individuals or groups who may advocate economic diversiﬁcation, pointing to the risks associated with sunk cost investments in the boom sector, are likely to earn reputations as ‘‘naysayers’’ and would be hard pressed to garner significant support for their initiatives. The other problem with forging broad support for economic diversiﬁcation lies in the limited stock that the population may put in collective social action. Where resource-rich regions are frontier regions, rugged individualism and fortune-seeking behavior among residents, most of whom have migrated recently to the region, are more likely than in areas with long-standing and durable community institutions and social ties. Forging collective responses in ‘‘boom town’’ environments to pursue diversiﬁcation or savings for some distant ‘‘rainy day’’ will be difﬁcult, especially given the economic and political forces that would tend to rise up against such efforts. The lure of immediate private fortunes, or at least higher returns, will hardly be outweighed by the vague promise of long-term ‘‘social well-being.’’ Moreover, the pursuit of such a longer-term goal would require a level of individual investment in social outcomes that is unlikely to be seen until the boom is over. Then, with the value of everyone’s sunk capital investments riding on alternatives or subsidies from external sources, collective action may become much more accepted as an appropriate response to a common predicament. Sunk costs in extractive industries can thus evoke notably negative development outcomes at both industry and economy-wide levels. Such outcomes stand in stark contrast to the common expectations that follow from ‘‘free market thinking,’’ which rests on the presumption of the optimality of markets, where competition and factor mobility yield ﬁrst-best welfare outcomes. In extractive industries, sunk costs are pervasive and act to undercut competition within the sector and factor mobility in the economy as a whole. For this reason, policies surrounding the development of extractive industries must fully incorporate the notion of sunk costs if outcomes are to coincide with basic efﬁciency and equity goals.

6. POLICY AND RESEARCH IMPLICATIONS FOR EXTRACTIVE INDUSTRIES WITH SUNK COSTS The broad policy implication of our conclusion – that sunk costs matter in extractive industries and thus the process of resource-based development – is

182

BRADFORD L. BARHAM AND OLIVER T. COOMES

essentially straightforward. Regulatory mechanisms, institutional innovation, and other measures aimed at overcoming the negative development outcomes associated with sunk costs in extractive industries are needed to complement any such measures that may be delivered by private markets. For example, some combination of private and public institutional innovation may be needed to reduce the uncertainty that private investors confront in extractive industries with sunk costs if underinvestment is to be avoided. Similarly, at the economy-wide level, state policies may be needed to encourage both private and public agents to include the option value of ﬂexible economic investments to avoid over-reliance on a currently lucrative activity with attendant high levels of sunk costs. The real challenge for policy comes in the specific practice – that is, in designing appropriate policies that will not over-adjust or create even worse distortions. Our analysis suggests that the economic performance problems created by sunk costs cannot all be remedied by the same prescription. This type of challenge is common to a wide range of economic issues, where markets do not work efﬁciently because of market-level distortions (see Carter & Barham, 1996 on policy challenges of agro-export development models). In the case of sunk costs, even the task of identifying the presence or extent of sunk costs for a given activity is a challenging undertaking, one that cannot be simply completed by examining the physical properties of the invested capital, or by referring to a ‘‘once and for all’’ chart of sunk costs associated with different activities or stages of extraction. Sunk costs are both contingent and conjunctural; they can vary substantially over time, space, and even within a specific activity, depending on all of the various causes outlined in Section 3. To take the next analytical step, which involves assessing how the extent of sunk costs in a given extractive activity may inﬂuence industry and economy-wide outcomes, also represents a significant undertaking. Both tasks need to be completed before specific policy prescriptions can be identiﬁed. The task of pursuing the effects of sunk costs on development outcomes promises to be challenging for at least four reasons. First, sunk cost measures are inherently dynamic (the reader will recall that estimating the salvage value of investments requires an assessment of possible future net revenue streams in current or alternative uses), and are therefore challenging to generate because they must rely on forecasts of contingent price and cost outcomes. Second, the effects of sunk costs on ﬁrm behavior and hence on industry organization and performance also are not static outcomes; consequently, the conceptualization and measurement of these outcomes also will be difﬁcult. Third, because private and public agents may act

Sunk Costs, Resource Extractive Industries

183

strategically to take advantage of the commitments offered by sunk costs or attempt to reduce the uncertainty that may hamper sunk investments, there is room for considerable agency and adjustments over time, which can in turn inﬂuence contingent price and cost outcomes in ways that subsequently affect the sunk cost measures. Fourth, attempts to relate features of an extractive industry and the attendant investment patterns that result in path dependence to the overall economic structure will represent a complex undertaking in many cases, especially in those cases where no single sector is ‘‘booming’’ and thus, in some sense, ‘‘overdetermining’’ outcomes in the regional economy. Given the magnitude of the challenges posed for ex ante analyses, we believe that more in-depth ex post empirical studies are much needed to advance our understanding of the nature and role of sunk costs. Because of the prevalence of sunk costs in extractive industries, and the continuing importance of extractive industries to many resource-rich developing areas, research and policy directed toward improving development outcomes is a crucial and urgent task. No hyperbole about the ‘‘magic of markets’’ can ﬁnesse the efﬁciency and equity problems that sunk costs create in extractive activities. The sooner the importance of sunk costs is recognized, and the fact that they are geographically, historically, and socially constituted, the sooner economic analyses and policy proposals will approach a ‘‘real-world’’ economics of resource-based development. By getting on with the difﬁcult task of understanding sunk costs, we advance the day when policy-makers can design and implement appropriate economic development policies to deal with their distortionary effects.

NOTES 1. There are a variety of strands to the sunk cost literature, and several terms are used including asset ﬁxity, irreversible investment, rigid investments, and ﬁxed capital, with sunk costs and irreversible investment being the two most common at present. Asset ﬁxity is a terms used by some agricultural economists, as discussed in Chavas (1994). 2. Depreciation here is taken to reﬂect the physical life of the investment, taking into consideration normal wear and tear and periodic maintenance, and technological obsolescence that occurs as newer, more efﬁcient vintages are introduced. 3. In particular, among lower and middle income families, most of their wealth investment is likely to be in their homes and the land surrounding that home. Even though the market value of that home may fall dramatically following the collapse of a one industry town, the consumption (and psychic) value provided by maintaining the home and the remaining local connections to family, friends, and community may help to explain why families will accept much lower returns to labor and capital

184

BRADFORD L. BARHAM AND OLIVER T. COOMES

than they might be able to earn elsewhere. The sunk nature of the investment in housing and community is a fundamental component of this decision. 4. The moral hazard problem of not being able to observe the buyer’s use often prevents the seller from issuing a warranty or some other signal which guarantees quality. This ‘‘Market for Lemons effect’’ is often used to explain why the resale value of almost new equipment, such as a recently purchased automobile or truck, are often well below the purchase cost. 5. Although the site of processing has been a major issue for developing countries and academic analyses of the role of primary commodities in economic development, the ordering of these two stages is not a key issue in understanding the basic pervasiveness of sunk costs in extractive industries. That said, the desire to avoid the sunkeness of a processing investment next to a mine, and the associated strategic and stochastic risks, may actually play a major role in the decisions of ﬁrms about where to locate the processing stage of the operation. 6. Barham et al. (1994) explore the limitations on resource nationalism in raw material industries, with primary emphasis on the international aluminum industry. The issue is explored in more general terms in the book’s epilogue. 7. The potential interaction between strategic behavior and uncertainty is not explored here, though they are potentially ﬂip sides of the same coin and together could be quite inﬂuential in shaping industry organization and performance. The theoretical complications that may arise by combining these two microeconomic effects are quite complex (Bresnahan & Reiss, 1993 offer an early attack on this issue. More recently, this issue has been given attention in the industrial organization literature associated with deregulated electricity generation). These analyses require more analytical and numerical exploration than can be provided in this presentation. 8. Barham et al. (1994) offers a simple example that demonstrates this attribute of industries with sunk costs and uncertainty where participants make decisions in an atomized, competitive environment. Chavas (1994) and Dixit and Pyndick (1994) also demonstrate this outcome more formally.

ACKNOWLEDGEMENT The authors would like to thank Stephen Bunker for useful suggestions and Paul Ciccantell for excellent editorial assistance.

REFERENCES Aggarwal, R. M., & Narayan, T. A. (2004). Does inequality lead to greater efﬁciency in the use of the local commons? The role of strategic investments in capacity. Journal of Environmental and Economic Management, 47(1), 163–182. Akerloff, G. A. (1970). The market for lemons: Qualitative uncertainty and the market mechanism. Quarterly Journal of Economics, 84, 488–500.

Sunk Costs, Resource Extractive Industries

185

Atkinson, G., & Hamilton, K. (2003). Savings, growth, and the resource curse hypothesis. World Development, 31(11), 1793–1807. Auty, R. M. (1993). Sustaining development in mineral economies: The resource curse thesis. London: Routledge Press. Auty, R. M. (1994). Industrial policy reform in six large newly industrializing countries: The resource curse thesis. World Development, 22(1), 11–26. Barham, B. L. (1994). Strategic capacity investments and the Alcoa–Alcan monopoly, 1888–1945. In: B. Barham, S. Bunker & D. O’Hearn (Eds), States, firms, and raw materials: The world economy and ecology of aluminum. Madison: University of Wisconsin Press. Barham, B. L., Bunker, S. G., & O’Hearn, D. (1994). States, firms, and raw materials: The world economy and ecology of aluminum. Madison: University of Wisconsin Press. Barham, B. L., Chavas, J. -P., & Coomes, O. T. (1998). Sunk costs and the natural resource extraction sector: Analytical models and historical examples of hysteresis and strategic behavior in the Americas. Land Economics, 74(4), 429–448. Barham, B. L., Chavas, J. -P., & Klemme, R. (1994). Low capital dairy strategies: Lessons from a new economic approach to measuring profitability. University of Wisconsin, Department of Agricultural Economics, Staff Working Paper no. 381, October. Barham, B. L., & Coomes, O. T. (1994a). Wild rubber: Industrial organization and the microeconomics of extraction during the Amazon rubber boom (1860–1920). Journal of Latin American Studies, 26(1), 37–72. Barham, B. L., & Coomes, O. T. (1994b). Reinterpreting the Amazon rubber boom: Investment, the state, and Dutch disease. Latin American Research Review, 29(2), 73–110. Barham, B. L., & Coomes, O. T. (1996). Prosperity’s promise: The Amazon rubber boom and distorted economic development. Boulder, CO: Westview Press. Bresnahan, T. F., & Reiss, R. C. (1993). Measuring the importance of sunk costs. Annales D’ Economie et de statisque, 31, 181–217. Bunker, S. G. (1994a). Flimsy joint ventures in fragile environments. In: B. Barham, S. Bunker & D. O’Hearn (Eds), States, firms, and raw materials: The world ecology and economy of aluminium. Madison: University of Wisconsin Press. Bunker, S. G. (1994b). The political economy of raw material extraction and trade. In: R. Socolow (Ed.), Industrial ecology and global change. Cambridge: Cambridge University Press. Bunker, S. G., & Ciccantell, P. (2005). Globalization and the race for resources. Baltimore, MD: Johns Hopkins University Press. Carter, M. R., & Barham, B. L. (1996). Level playing ﬁelds and laissez faire: Postliberal development strategy in inegalitarian agrarian economies. World Development, 24(7), 1133–1149. Chavas, J.-P. (1994). Production and investment decisions under sunk cost and temporal uncertainty. American Journal of Agricultural Economics, 76, 114–127. Corden, W. M., & Neary, J. P. (1982). Booming sector and deindustrialization in a small open economy. The Economic Journal, 92, 825–848. Davis, G. A. (1997). The minerals sector, sectoral analysis, and economic development. Resource Policy, 23(4), 191–199. Dixit, A. (1980). The role of investment in entry deterrence. Economic Journal, 90, 95–106. Dixit, A. (1991). Irreversible investment with price ceilings. Journal of Political Economy, 99(31), 541–557.

186

BRADFORD L. BARHAM AND OLIVER T. COOMES

Dixit, A. K., & Pindyck, R. S. (1994). Investment under uncertainty. Princeton: Princeton University Press. Garland, H., Sandefur, C. A., & Rogers, A. C. (1990). De-Escalation of commitment in oil exploration: When sunk costs and negative feedback coincide. Journal of Applied Psychology, 75(6), 721–727. Gylafson, T. (2001). Natural resources, education, and economic development. European Economic Review, 45, 847–859. Hum, A. S., & Wright, R. E. (1994). Geology or economics? Testing models of irreversible investment using North Sea oil data. The Economic Journal, 104, 363–371. Innis, H. A. (1930). The fur trade in Canada: An introduction to Canadian economic history. New Haven: Yale University Press. Innis, H. A. (1933). Problems of staple production in Canada. Toronto: University of Toronto Press. Innis, H. A. (1956). In: M. Q. Innis (Ed.), Essays in Canadian economic history. Toronto: University of Toronto Press. Krugman, P. (1985). Multinational enterprise: The old and new in history and theory. Mimeo, Massachusetts Institute of Technology. Lambson, V. E. (1991). Industry evolution with sunk costs and uncertain market conditions. International Journal of Industrial Organization, 9, 171–196. Lambson, V. E., & Jensen, F. E. (1995). Sunk costs and the variability of ﬁrm value over time. The Review of Economics and Statistics, 77, 535–544. Lambson, V. E., & Jensen, F. E. (1998). Sunk costs and ﬁrm value variability. American Economic Review, 88(1), 307–313. Mikesell, R. F. (1997). Explaining the resource curse, with special reference to mineral exporting countries. Resource Policy, 23(4), 191–199. Neumayer, E. (2004). Does the ‘resource curse’ hold for growth in genuine income as well? World Development, 32(10), 1627–1640. O’Hearn, D. (1994). Producing imperialism anew: The United States, The United Kingdom, and Jamaican bauxite. In: B. Barham, S. Bunker & D. O’Hearn (Eds), States, firms and raw materials: The world economy and ecology of aluminum. Madison: University of Wisconsin Press. Priest, R. T. (1995). An open vein: Manganese ore and the central do Brasil railway, 1894–1920. Business and Economic History, 24(1), 164–175. Priest, R. T. (2003). Global gambits: Big steel and the U.S. quest for manganese. Westport, CT: Praeger Publishers. Sachs, J. D., & Warner, A. M. (1995). Natural resource abundance and economic growth. NBER Working Paper, no. 5398. Sachs, J. D., & Warner, A. M. (2001). Natural resources and economic development: The curse of natural resources. European Economic Review, 45(4–6), 827–838. Spence, A. M. (1977). Entry capacity investment and oligopolistic pricing. Bell Journal of Economics, 8, 534–544. Stuckey, J. A. (1983). Vertical integration and joint ventures in the aluminum industry. Cambridge: Harvard University Press. Williamson, O. (1985). The economic institutions of capitalism: Firms, markets, relational contracting. New York: Free Press. Woo, W. T., Glassburner, B., & Nasution, A. (1994). Macroeconomic policies, crises, and longterm growth in Indonesia. 1965–1990. Washington, DC: World Bank.

JAPAN’S ECONOMIC ASCENT AND ITS EXTRACTION OF WEALTH FROM ITS RAW MATERIALS PERIPHERIES Paul S. Ciccantell and Stephen G. Bunkery ABSTRACT How did Japan rise to challenge the U.S. economic supremacy? We argue that the foundation of Japan’s rise from a defeated nation in 1945 to an economic powerhouse is the raw materials that Japanese firms have turned into cars, ships, consumer electronics, and of other industrial products. A small island nation that lacked adequate domestic supplies of virtually all the raw materials essential to industrial production became a world leader in the production of steel and of products which required millions of tons per year of raw materials. Japanese firms and the Japanese state turned an apparent material and economic disadvantage, the need to import large volumes of raw materials, into a competitive advantage over the U.S., Europe, and the rest of the world economy by driving down the cost of importing raw materials over long distances. We argue that the strategies of Japanese firms and the Japanese state to resolve the problems of procuring bulk cheaply and reliably from multiple distant locales drove the technical and organizational innovations that underlay Japan’s rapid industrial development and restructured the world economy Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 187–207 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10009-2

187

188

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

in support of Japan’s development. Contrary to claims that globalization supercedes the national state, we find that the actions of the Japanese state, in coordination with firms and industry sectors, were crucial in developing and applying these strategies. The linchpin of these strategies were the MIDAs (Maritime Industrial Development Areas) built on land reclaimed by the Japanese state. This economic success in Japan was also critically dependent on the extraction of billions of dollars of wealth from its raw materials peripheries, most notably Australia, Brazil, and Canada.

1. INTRODUCTION How did Japan rise to challenge the U.S. economic supremacy? Management techniques, just-in-time delivery, Deeming’s ideas on quality, the role of keiretsu industrial groups as risk-sharing cooperative mechanisms, state– ﬁrm cooperation, and the Japanese work ethics are a few of the answers that have been proposed. Arrighi (1994) constructs a powerful model of the stages from maturity to decadence of different hegemons, the contradictions of hegemonic overaccumulation being partially resolved by investment in other rising economies; in this perspective Japan’s economic ascent is closely tied to hegemonic decline in and investment in Japan from the U.S. All of these analyses provide insights into Japan’s dramatic rise from a defeated nation in 1945 to an economic powerhouse, but we believe that the foundation of Japan’s rise is much more basic: low-cost raw materials that Japanese ﬁrms turned into cars, ships, consumer electronics, and a plethora of other industrial products made the Japanese economy a global leader. A small island nation that lacked adequate domestic supplies of virtually all the raw materials essential to industrial production became a world leader in the production of steel, copper, and aluminum and of products that required millions of tons per year of these materials. Japanese ﬁrms and the Japanese state turned an apparent material and economic disadvantage, the need to import large volumes of raw materials, into a competitive advantage over the U.S., Europe, and the rest of the world economy. We argue that the strategies of Japanese ﬁrms and the Japanese state to resolve the problems of procuring bulk cheaply and reliably from multiple distant locales drove the technical and organizational innovations that underlay Japan’s rapid industrial development. As the result of Japanese raw materials access strategies, coal prices fell in real (inﬂation-adjusted) terms from US$86.65 in 1959 (in 1992 dollars) to

Japan’s Economic Ascent

189

US$43.63 in 1998 for coal imported into Japan, and iron ore prices fell in real terms CIF Japan from US$49.80 per ton in 1964 in 1998 dollars to US$24.44 per ton in real terms in 2000. Japanese ﬁrms and the Japanese state collaborated to achieve huge economies of scale and tight integration in transportation and the processing of raw materials. What started in the 1950s as a strategy to resolve the fundamental obstacles to economic ascent in the face of U.S. hegemony drove both these changes and led to absolute trade dominance by the 1970s (Ciccantell & Bunker, 2002). Contrary to claims that some have made in the debate over whether globalization supercedes the national state (for alternative perspectives on this debate, see Amin, 1996; Arrighi, 1998; Biersteker, 1998; Boxill, 1994; Ciccantell, 2000; Garrett, 1998; Harvey, 1995; Kiely, 1998; Shaw, 1997; Sklair, 1998; Yaghmaian, 1998), the actions of the Japanese state, in coordination with ﬁrms and industry sectors, were crucial in developing and applying these strategies. The linchpin of these strategies of scale and integration were the MIDAs (Maritime Industrial Development Areas) built on land reclaimed by the Japanese state. The MIDAs incorporated port facilities to import raw materials and export industrial products and the steel mills and their major customers, including shipyards and automobile plants. The MIDAs’ deepwater ports allowed the use of larger and larger ships (often built in the adjacent shipyards owned by the same industrial group) to import raw materials at low costs to brand-new steel mills that were vastly larger than mills in the U.S. or Europe. These Japanese mills, built with state subsidies, incorporated state-of-the-art technology that produced the world’s largest, lowest cost, most integrated, and most competitive steel industry. The technologies that increased economies of scale in processing and transport were rigidly interdependent, with larger steel mills dependent on larger ships to supply low-cost raw materials over longer distances and on shipbuilding and other large industries for consuming steel, and technologies for building larger ships and higher quality industrial products depended on the increasing availability of progressively lower cost steel. These scale-increasing technologies in processing and transport simultaneously drove a fundamental restructuring of the built environment in Japan and in the areas that became raw materials peripheries supporting Japanese economic ascent. The other critical element of Japan’s economic ascent was the creation of its raw materials peripheries, most importantly in Australia, Brazil, and Canada, to supply Japan’s rapidly growing industries. The iron ore and coal mines in these three nations extracted subsidies of tens of billions of dollars from these raw materials peripheries, lowering production costs, and

190

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

increasing profits for Japanese ﬁrms in many industries and thus supporting Japan’s rise to global trade dominance in automobiles, ships, electronics, steel, and other industries. The following section discusses existing explanations of Japan’s economic ascent. The next section outlines our theoretical model, which we term the new historical materialism. Section 3 outlines the key issues in the development of the Japanese steel industry, the central generative sector that drove Japanese economic ascent from the 1950s through the 1980s, and its relocation to the MIDAs. Section 4 focuses on the MIDAs and transportation, examining the port areas where Japan’s state-of-the-art industries were built and that linked the internal and external dimensions of Japan’s economic ascent. Section 5 analyzes the creation of Japan’s raw materials peripheries and their subsidization of Japan’s economic ascent. The conclusion highlights the impacts of Japan’s restructuring of the world economy over the last 50 years.

2. NEW HISTORICAL MATERIALISM AND GENERATIVE SECTORS In any rising economy, strategies for economic ascent must respond to and take advantage of contemporary technological, geopolitical, environmental, and market conditions in the rest of the world and of the nation’s position and location within that particular global economy. They must also coordinate the physical characteristics and location in space and in topography of the various raw material resources actually or potentially available with the physical characteristics and location in space and topography of the national territory (Ciccantell & Bunker, 2002). Solutions to the raw materials problem require the coordination of multiple physical and social processes across geopolitical and physical space with domestic relations between ﬁrms, sectors, the state, labor, and new technologies. Rising economies resolve these problems at the same time as or even before they increase industrial competitiveness. These solutions stimulate complex processes of learning and of institutional change that fundamentally mold the organization of the national economy at the same time that they change international markets and the rules binding participants in them (Ciccantell & Bunker, 2002). The challenges and the opportunities presented by the basic raw materials industries and by the transport systems on which they depend foster what

Japan’s Economic Ascent

191

we call ‘‘generative sectors’’, sectors that, beyond creating the backward and forward linkages that underlie the concept of a leading sector, also stimulate a broad range of technical skills and learning along with formal institutions designed and funded to promote them, vast and diversiﬁed instrumental knowledge held by interdependent specialists about the rest of the world, ﬁnancial institutions adapted to the requirements of large sunk costs in a variety of social and political contexts, specific formal and informal relations between ﬁrms, sectors, and states, and the form of legal distinctions between public and private and between different levels of public jurisdiction (Ciccantell & Bunker, 2002). Generative sectors will be more numerous, more easily observed, and more efﬁcacious in those national economies that are growing so rapidly that they must achieve massive increases in throughput and transformation of raw materials. The concept is relational, however, within a world-systems perspective, and thus implies that generative sectors in rising economy will have significant consequences for economies that export raw materials or trade in other kinds of goods (Ciccantell & Bunker, 2002). The semantic form of the notion ‘‘generative sector’’ connotes that other sectors, or changes in those sectors, are generated, as Chandler (1965, 1977) demonstrated for the relationship between the railroads in the U.S. and the telegraph companies that, faced with the need to coordinate and regulate the ﬂows of multiple messages from and to multiple locations, adopted the ‘‘template’’ that the railroads had developed. Our model of generative sectors, therefore, includes consideration of feedback, or auto-catalytic loops (Prigogine, 1984), between generative sectors and the sectors that they foment or stimulate to change. If raw materials access strategies require new scales of transport, and thus new technologies, new design, and new infrastructure, successful access to cheaper iron ore and coal will both require and generate technical solutions to transport costs, such that the design and construction of larger ships, better railways, larger ports, more capacious warehouses, and other infrastructure will create demand for more steel, and potentially for higher quality steel, in volumes, or at levels of throughput, that both make possible technological innovation and allow full use of capacity required to reduce unit costs of steel. Reduced unit costs of steel make shipbuilding cheaper and more competitive and further reduce unit costs of iron and coal transport. These feedback loops cheapened, deepened, and accelerated the synergies between iron ore, coal, computerized continuous casting and hot rolling, vastly larger harbors, the construction and deployment of large dry bulk carriers, and the relocation of heavy industries and

192

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

their downstream consumer industries around deep-water ports in Japan a century later (Ciccantell & Bunker, 2002). In this sense, steel may be treated as a generative sector, but transport is not simply a generated sector. It is rather in a feedback or auto-catalytic loop that intensiﬁes the growth and innovation in the generative sector, the steel industry. The progression toward increasingly tight coupling at greater technical economies of scale across greater spaces and multiple sovereignties with greater levels of production that characterizes the unfolding of the world-system means that only economies where effective generative sectors emerge can rise to economic competitiveness or even dominance. As we will demonstrate in the following sections, the Japanese state and Japanese ﬁrms created these generative sectors in the MIDAs and restructured the world economy to support Japan’s economic ascent.

3. STEEL AND MIDAS IN JAPAN After Japan’s defeat in World War II, the U.S. initially sought to prevent Japanese remilitarization and the reconstruction of its key industrial suppliers, steel and shipbuilding. The geopolitics of the Cold War forced the U.S. to ‘‘Reverse Course’’ and support economically and diplomatically Japanese reindustrialization. This joint U.S.–Japanese effort, however, confronted a myriad of obstacles, most notably the exhaustion of domestic raw materials, capital shortage, long ocean voyages from potential supply sources, and bitter resentment among potential Asian raw materials suppliers, particularly Australia, to trade and investment relations with Japan as a result of Japan’s actions in World War II (Panda, 1982; McDougall, 1993; Nafziger, 1995; So & Chiu, 1995; Marshall, 1995; U.S. State Department, 1949; Pauley, 1945; Borden, 1984; Bisson, 1949; Ball, 1949; Bisson, 1954; Hadley, 1970). Japanese ﬁrms and the Japanese state, supported by the U.S. and World Bank ﬁnancial assistance, created a new model of domestic development based in the steel, shipbuilding, and shipping industries. In the external sector, without which these industries could not develop, U.S. ﬁnancial and diplomatic assistance helped to create a new model of raw materials supply via long-term contracts and minority joint venture partnerships in Australia that Japanese steel ﬁrms and the Japanese state, led by the Ministry of International Trade and Industry (MITI) reﬁned and expanded into a highly coordinated global model of raw materials supply and capital accumulation

Japan’s Economic Ascent

193

that drove Japanese economic ascent (see Ciccantell & Bunker, 2002 for an extended discussion of this external dimension). MITI in the early 1950s actively opposed the targeting of the steel industry by the government because of concerns over the industry’s ability to be internationally competitive. The economic boom that began in 1955 led to a reevaluation of the potential of the steel industry as a leading economic sector. MITI became involved as a coordinating agent for the steel sector in a number of areas, including control over capacity expansion in an effort to keep steel plants operating at full capacity without severe price competition (Yonekura, 1994, pp. 212–237; O’Brien, 1992; Johnson, 1982). This new model of state–ﬁrm–sector coordination only developed out of a protracted series of conﬂicts between these groups. MITI confronted the vested interests of the old steel companies and their still highly inﬂuential leaders, but the industry was highly dependent on MITI and the Export–Import (ExIm) Bank of Japan for access to raw materials, negotiations with the U.S. government, and capital. MITI was able to parlay this leverage into regulatory powers over the entire industry, which it used to promote new technologies of unprecedented scale and efﬁciency. The Japanese government focused its resources on promoting economic development through heavy industrialization in steel and shipbuilding during the 1950s and 1960s. In addition to ﬁnancing and export promotion, the Japanese government also set out to establish huge industrial parks. The ﬁrst was on land reclaimed from Tokyo Bay. Kawasaki Steel, a new company, was given 3 m(illion) sq(are) m(eters) of land on which it built the most modern integrated steel facility in the world. Located close to a new, modern harbour, a continuous production line was established covering all stages of production from raw materials to ﬁnished products on the same site and using the most modern technology in the world. With labour still relatively cheap, Kawasaki steel became the cheapest in the world. Here the results of dividing up the old zaibatsu came into play. Neither Yawata Steel nor Fuji Steel was prepared to allow a newcomer, Kawasaki Steel, to steal a march on them. Both launched similar developments, creating a large and modern Japanese steel industry in the world-beating class (Reading, 1992, pp. 70–71).

The Maritime Industrial Development Area (MIDA) program begun in the 1950s coordinated ﬁrm and state investment in new greenﬁeld ports and steel plants utilizing the latest technological advances developed in Japan and imported from other nations to reduce costs and increase Japanese economic competitiveness in steel, shipbuilding, and all other sectors that used steel and the steel-based transport infrastructure. In many Japanese ports, facilities for importing coal and iron ore and the steel plants that used

194

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

these raw materials became major components of the industrial area (see, for example, Osaka & Bureau, 1987, 1999; Yuzo, 1990). The MIDA ports were intended to serve as growth poles that would generate a variety of linked industries in the port and the region, and this economic development model was used to justify a combination of national, prefectural, and city government subsidies for the MIDAs. The MIDA program built on the model developed during the Meiji Era to use port development as a means of economic development and of gaining the support of local warlords for the national government. By 1982, there were twelve iron and/or coal ports in Japanese MIDAs capable of unloading 100,000+ dwt ships in Japan, and all of the major Japanese steel ﬁrms have their own dedicated large-scale ports (CSR, 1982, pp. 37–38; Penfold, 1984, p. 29). By 1990, a total of 19 Japanese ports could unload vessels of 100,000 dwt or larger, with 10 able to unload 200,000 + dwt ships and the largest, Oita, capable of unloading ships up to 310,000 dwt (Lloyd’s, 1990, pp. 314–360). The steel mills built in the MIDAs led the world in technological innovations and scale increases, driving down costs and providing increasingly high-quality steel at the lowest cost in the world. At the ﬁrst stage of processing iron ore and metallurgical coal, the blast furnace that produced pig iron, Japanese steel ﬁrms beginning in the 1950s undertook a long series of efforts to increase the scale of blast furnaces. Led by Japanese technological innovations, blast furnaces increased in capacity from 1,500 tons per day in 1950 to 4,000 tons per day by the late 1960s (Manners, 1971, p. 27, 34) and to 22,000 tons per day by the early 1990s (McGraw-Hill, 1992, pp. 425–426). The Japanese adopted, perfected and diffused a second significant improvement in blast furnace operation, the improvement in the quality of the burden (the charge of raw materials into the furnace) through the sizing, agglomeration, and beneﬁciation of iron ore (Manners, 1971, p. 160). A similar process of blending varieties of metallurgical coal has had similar impacts on the costs of blast furnace operations. The premium prices commanded by metallurgical coals because of their useful properties have been reduced by Pulverized Coal Injection (PCI) that allows partial replacement of metallurgical coal with a wider variety of grades of coal. Japanese steel ﬁrms thus reduced production costs by escaping the ‘‘tyranny of metallurgical coal’’ by substituting less expensive steam coal. Overall, Japan’s global sourcing of iron and coal facilitated blending and allowed the Japanese significant raw materials cost savings (see Ciccantell & Bunker, 2002; Bunker & Ciccantell, 1995a, b).

Japan’s Economic Ascent

195

As Manners concludes, ‘‘all iron- and steel-producing countries beneﬁtted from the improvements of blast-furnace technology, but none perhaps quite so rapidly as Japan’’ (Manners, 1971, p. 38). This rapidity resulted in large measure from the role of MITI and other agencies in providing both capital and raw materials access support and imposing critical regulations. By combining increasing scale of the blast furnace with careful control of blending multiple ores and controlling the size of feed, Japanese steel mills became the largest and most efﬁcient in the world by the mid-1960s. These efﬁciencies and reduced transport costs, by reducing the amounts of iron ore and coking coal required to produce each ton of pig iron in Japan, were critical components of Japan’s competitive advantage in steel production since the late 1950s. The major fuel economies in steel production were driven by the size of the blast furnace that produces pig iron, but the scale of the second steel making stage limited the potential for scale increase, and therefore fuel economy, as the blast furnace and the second stage had to be made compatible. Fuel economy depends on scale of processing, and the scale of processing advances through myriad technical discoveries that progressively cheapen steel making, but does it through scale increases that progressively accelerate the consumption of raw materials. This increases the cost of space across which raw materials must be transported and within which there is an ever smaller number of deposits large enough to support the consumption of ever larger integrated smelters. The Japanese steel ﬁrms led the way in adopting a new technology, the basic oxygen furnace, that reduces the time required to produce one batch of steel to half an hour from the 4 to 5 h per heat required in an open hearth furnace by injecting pure oxygen under high pressure (Ohashi, 1992, p. 542). Japan was by far the most rapid adopter of the basic oxygen furnace (Whitman, 1965, pp. 853–855). This innovation dramatically increased the scale of production, since a basic oxygen furnace in Japan could produce eight to ten times as much steel in a given length of time relative to a U.S. open hearth furnace. The capital barriers to establishing this scale of smelting and the potential for crippling overcapacity in the still underdeveloped Japanese and Asian markets set the stage for state–sector–ﬁrm cooperation, regulation, and resolution of disputes. On this stage, various agencies of the Japanese state, most notably MITI and the ExIm Bank, took the lead in the creation of a tightly coupled relationship between capital and the state (Johnson, 1982). MITI learned technical and political skills that made it essential to the steel, shipbuilding, and shipping industries. MITI’s competence and power

196

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

allowed it to combat and restrain the self-interests of particular steel companies in the interests of the sector and those of the national economy as a whole. Much of what MITI essentially became originated in its critical role in the adoption and improvement of the Basic Oxygen Furnace. Rapid growth in the Japanese steel industry necessarily meant construction of greenﬁeld projects, which do not suffer the innovation-retarding drag of capital vested in obsolete plants, depleted sources, and restrictive distribution networks. As a result, Manners argues, ‘‘there is a good deal of evidenceyto suggest that, on average, Japanese steelmaking costs in 1965 were substantially below those of the United States iron and steel industry, and that Western European costs by and large lay somewhere between the two’’ (Manners, 1971, p. 116). Savings in raw materials and transport costs, combined with technological innovations and adoptions, had in less than 20 years transformed the Japanese steel industry into the world’s lowest cost, fastest growing steel industry. As a consequence: today, for a wide variety of steel products, a Japanese manufacturer of steel products can buy Japanese steel at prices ranging from 15 to 30 percent lower, depending on the gauge, than his American counterpart can buy it in the United States. This handily gives the Japanese manufacturer a cost advantage of 5 to 8 percent less over his U.S. competitor for products such as forklift trucks, construction equipment, automobiles, and ball bearings (Abegglen & Stalk, 1985, pp. 77–78).

This cost advantage for domestic steel consumers in Japan also translated into international competitiveness in steel exports. As a result, Japanese steel ﬁrms have dominated world steel trade since the early 1960s. Japanese steel exports rose from 8.8% of the total world exports in 1960 to a peak of 40.8% in 1976; in volume terms, the increase was from 2.3 to 36 million tons over the same period. Japanese steel exports have ranged from 19 to 33 million tons per year over the last 25 years, constituting 20–30% of the total world steel exports during this period (data calculated from OECD 1985 and USBM/USGS Various Years). In short, as the U.S. government and Japanese development planners foresaw in the late 1940s, the steel industry became the linchpin of a number of linked industries that complemented one another in a ‘‘virtuous cycle’’ of economic development. The steel industry was a generative sector providing the raw materials foundation for reducing production costs for many sectors of the Japanese economy, transforming Japan into the world’s second largest economy and the United States’ most formidable economic competitor.

Japan’s Economic Ascent

197

4. SHIPS, SHIPBUILDING, AND THE MIDAS All of these innovations in steel production have led to larger scale and increased distance across the needed spaces. The cost of distance offsets the savings of processing steel. This contradiction generates ﬁrm–sector–state collaboration to design and implement more efﬁcient transport technologies, more effective transport infrastructures, and the incorporation of ever broader spaces as potential sources. These transport technologies have required and promoted larger transport scale (Ciccantell & Bunker, 2002). The technical achievements and scale increases in steel production and their development over time have both required and provided the means for increased fuel efﬁciency of transport. For different but complementary physical causes (relating to inertia, momentum, and hydraulics), size of ship, or scale of transport, is directly associated with fuel economy. The advances in steel quality that are associated with scale-dependent fuel economies of scale in smelting contribute to the tensile strength requirements of the hulls of larger ships and to the capacity of the boilers and engines to withstand the temperatures and the pressures that are directly associated with efﬁciency of transfer from heat energy to mechanical energy. The transport strategy developed via the coordinated efforts of MITI, the ExIm Bank, and the Japanese shipping, shipbuilding, and steel ﬁrms allowed Japanese steel ﬁrms to take advantage of the tremendous economies of scale available in bulk shipping to dramatically reduce production costs of steel in Japan and to capture all of these beneﬁts for themselves, rather than sharing them with coal and iron ore producing ﬁrms and exporting regions. One of the key elements of transport as a Japanese raw materials access strategy was large government subsidies to shipbuilding and shipping ﬁrms via the Programmed Shipbuilding Scheme (ﬁrst introduced in 1947) that carefully controlled the number and types of ships built to suit the importing and exporting needs of the nation (Chida & Davies, 1990, pp. 66–90). The Japanese government also arranged an innovative lease of a former naval shipyard to a U.S. shipbuilding ﬁrm, in return for allowing unlimited access to Japanese engineers, managers, and workers. This provided the foundation for research and development on the construction of larger petroleum tankers and bulk carriers and the construction of large shipyards capable of building such large ships, as well as a variety of other innovations that reduced the cost and labor requirements of shipbuilding (Chida & Davies, 1990, pp. 98–112; Todd, 1991, p. 13). For example, during the 1980s, Japanese labor requirements for building a

198

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

62,000 dwt bulk carrier fell from 380,000 h in 1981 to 170,000 h in 1991 (UNCTAD, 1992, p. 39). Labor saving was of critical importance in Japan because of the high wages of shipyard workers. During most of the post-World War II era, shipyard workers were the highest paid workers in Japan (Chida & Davies, 1990). Just as was the case in the steel industry, heavy industrialization based on raw materials was a major component of domestic consumption in Japan both directly and indirectly through workers’ wages, providing a market for the industrial products that utilized steel and other processed raw materials. This government direction and subsidization reﬂects the importance of transport as a strategic component of both raw materials access and economic development efforts on the part of the Japanese government. This tightly coordinated system of state–ﬁrm relations explicitly sought to balance the interests of shipbuilding, shipping, and steel ﬁrms, and broader industrial and societal interest in low-cost raw materials imports and industrial export transport, with the state as arbiter. This system allocated scarce Japanese capital resources to supply low-cost transport without wasting resources on the notoriously cyclical shipbuilding industry, a delicate balancing act of restricting capacity and output in pursuit of broader state developmental goals. Although the Japanese steel ﬁrms initially intended to build their own ﬂeets of ore carriers, the Japanese government’s control over concessionary ﬁnancing and refusal to provide ﬁnancing to ﬁrms other than the major shipping lines forced the Japanese steel ﬁrms to invest in shipping ﬁrms in order to secure partial ownership of bulk shipping (Chida & Davies, 1990, p. 119). State–sector–ﬁrm coordination served to control potential competition in a capital intensive and cyclical industry, balancing the interests of steel, shipbuilding, and shipping ﬁrms with broader societal interests in conserving scarce capital and ensuring low-cost supplies of raw materials. The Japanese government also provided ﬁnancing on concessionary terms for the export of Japanese-built ships through the ExIm Bank of Japan. Additionally, the government during the 1950s provided funding for the modernization of the Japanese steel industry, with one important result being the reduction in the cost of steel plate used in shipbuilding, making Japanese ship exports much more cost competitive (Chida & Davies, 1990, pp. 108–109). With this support from the Japanese government, the Japanese shipbuilding industry increased from only 15.6% of world production (829,000 tons) to 43.9% in 1965 (5.4 million tons) and peaking at

Japan’s Economic Ascent

199

18 million tons in 1975, 50.1% of world production. The Japanese shipbuilding industry became the largest in the world in 1965 and produced about half of total world output from the mid-1960s through the mid-1980s (Nagatsuka, 1991, p. 17). Shipbuilding was also Japan’s most important export industry between 1956 and 1960 (when the steel industry surpassed it in exports) and has remained as one of Japan’s three major export industries until recently (Chida & Davies, 1990, p. 106). Japanese industrial groups control ocean shipping of raw materials on an FOB raw materials exporting port basis so that any reductions in transport costs caused by technological improvements or changes in world shipping market conditions are captured by Japanese importers. The shipping of Japanese raw materials imports is typically arranged by the ﬁrms which consume the raw materials (Drewry, 1978, p. 58). This is an innovative alternative to the U.S. model of raw materials ﬁrms’ vertical integration into shipping, a model the Japanese steel ﬁrms initially sought to replicate. Longterm contracts for the majority of coal and iron ore transport are supplemented by short-term arrangements, allowing Japanese steel ﬁrms to take advantage of cyclical downturns in freight rates to charter additional required capacity at even lower cost as shipping ﬁrms operate bulk carriers at operating cost or less in order to earn revenues to try to survive until the next boom market. Shipbuilding was thus a key generative sector in the Japanese government’s post-World War II development plans. The industry consumed huge quantities of Japanese steel and other inputs, was one of the most important export industries in Japan for many years, served as a model for the import, adaptation and improvement of industrial and organizational technologies, and provided the increasing scale of ships needed by Japanese shipping ﬁrms to import huge volumes of raw materials for the Japanese steel mills. The only way to link massive exporting and importing systems in the steel industry with their rapidly increasing scales of operation was to develop shipbuilding and shipping technologies and organizations that could supply bulk carriers of similarly increasing scales. Just as had been the case in earlier periods in Holland, England, and the U.S. (Ciccantell & Bunker, 2002), shipbuilding created a tremendous range of material, economic, organizational, and technological linkages to other industries. These Japanese transport strategies for raw materials access have been strikingly successful at guaranteeing long-term access at low cost to huge volumes of imported raw materials. Just during the 1960s, Sasaki (1976) estimated that Japanese government efforts to reduce the 20–30% share of freight charges in the total cost of imported raw materials through transport

200

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

subsidies resulted in ‘‘during the ten years beginning in 1961, the freight costs for both crude oil and iron ore were reduced by 40 per centyThe effects of this reduction were significant and the consequent reductions in the price of electricity, petrol, iron and steel and many other products have made an immeasurable contribution to the national economy’’ (Sasaki, 1976, p. 7). Ocean transport closes the ‘‘virtuous circle’’ of generative sectors in steel, shipbuilding, and shipping. Japanese strategies have restructured these global industries and turned remote ecosystems in Australia, the Brazilian Amazon, and western Canada into raw materials peripheries that have provided the material ingredients for Japan’s economic ascent. Equally important, the processes of technological, organizational and institutional innovation and learning inseparably linked these international relations with Japanese internal development. Economic ascent was the product of complex coordination between ﬁrms and nature, between ﬁrms in Japan and in its emerging raw materials peripheries, between the Japanese state and states in these raw materials peripheries and the U.S., and between the Japanese state, ﬁrm and sectors. In overcoming complex challenges presented by material, economic and political processes across multiple ecosystems and multiple political boundaries in the shadow of a powerful existing hegemon, the Japanese state and Japanese ﬁrms learned and innovated repeatedly to make Japan the global competitive leader in a wide range of industries.

5. CREATING AND EXTRACTING WEALTH FROM JAPAN’S RAW MATERIALS PERIPHERIES The other, inextricably linked dimension of Japan’s dramatic economic ascent after World War II was the creation by this same pattern of state– sector–ﬁrm relations of Japan’s raw materials peripheries, most notably in Australia, Brazil and Canada. All three of these nations became major raw materials exporters to Japan during the 1960s as the result of Japanese raw materials access strategies, experiencing tremendous booms in investment and exports in coal in Australia and Canada and in iron ore in Australia and Brazil. This boom, however, did not create the sorts of long-term economic development and profits that states and ﬁrms in these three nations envisioned during the 1950s and 1960s. In fact, the same system of state–sector–ﬁrm coordination and transport cost reduction that was so beneﬁcial to Japan

Japan’s Economic Ascent

201

progressively drove down prices and rents for coal and iron ore and transferred billions of dollars in investment and subsidization costs to its raw materials peripheries. In coal, this transfer of costs and risks has proven to be quite deleterious to exporting ﬁrms’ and nations’ interests in the long term. For example, Koerner (1993) found that ‘‘Pacific metallurgical coal markets have suffered significant distortion as a result of the resource procurement strategies of the Japanese steel industry establishment’’ (Koerner, 1993, p. 79). On the demand side, the Japanese steel mills’ joint negotiating strategy resulted in a bilateral monopoly, precluding competition on the demand side, while on the import side Japan’s diversiﬁcation strategy led to destructive competition between ﬁrms, state governments and coalexporting nations. Additionally, ‘‘the substantial transport component of delivered cost creates a situation of bilateral monopoly bargaining over the distribution of locational rents’’ (Koerner, 1993, p. 79), while the knowledge asymmetry between Japanese and suppliers’ negotiators has similarly favored Japanese interests. The sum total of these advantages, Koerner argues, can be seen in the producer surplus lost to Australian coal producers on the 365 million tons of metallurgical coal exported to Japan from the early 1960s to the end of the 1980s at US$3.6 billion in 1987 dollars (Koerner, 1993, p. 79). As another analysis of the Australia–Japan coal supply relationship concludes, ‘‘as long as Australia fails to receive a fair price for the energy resource that made the ‘Japanese miracle’ possible, it will never have a relationship of maturity and respect with Japan’’ (Botsman, Brett, & Roger, 1994, p. 4). In our own calculation of the rent foregone by Australian coal producers and the Australian state and national governments, we have compared Australian coal prices CIF in Japan with those for coal from the U.S. in order to estimate how much it would have cost the Japanese steel mills to have continued to rely on the U.S. for its coal imports, as they had done in the 1940s and 1950s. On this basis, the loss to Australia in current dollars totaled more than US$10.8 billion between 1963 and 2000; in constant 1992 dollars, the total is more than US$18.8 billion. This foregone rent accrued to the Japanese steel mills, steel consumers, and purchasers of products made from Japanese steel over the past four decades, conferring an important increment to ﬁrm profits and Japan’s global competitiveness and representing a significant loss of mining ﬁrm profits and state revenues in Australia. In iron ore in Australia, Australian analysts and state and national governments in the late 1950s and early 1960s saw iron ore exports as a means to increase ore supplies for and provide funds for the development of the domestic steel industry. Australian analyses at the time saw Japan as a

202

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

competitor in the steel industry and sought strategies to increase Australian steel exports, based on their raw materials cost advantage over Japan (Hughes, 1963, p. 151). Hughes’ conclusion is correct: the combination of immense, high grade metallurgical coal and iron ore resources in Australia has been the essential ingredient for the world’s most competitive steel industry; however, that steel industry was built in Japan, not in Australia. As had been the case in metallurgical coal in Australia and Canada, exports to Japan of raw materials were intended to support domestic industrialization of raw materials via export earnings, rather than becoming the goal and main economic activity of the region. Japanese raw materials strategies effectively trapped these regions in the roles of quarries for Japan, choking off further domestic raw materials-based industrialization in favor of exports because of what during the 1960s and early 1970s appeared to be extremely lucrative long-term contracts for exports to Japan that provided a seemingly secure source of profits for mining ﬁrms and sources of employment, local development, export earnings, and tax revenues to regional and national governments. Both Australian and Brazilian iron ore exports to Japan grew rapidly beginning in the mid-1960s. However, these two nations, the world’s lowest cost iron ore producers, have sacriﬁced billions of dollars in rent from their naturally favorable mining conditions and their investments in mining and transport infrastructure over the past four decades. In comparison with the average cost of iron ore exports from the Philippines, a significant supplier to Japan but with much higher costs, Australian mining ﬁrms have foregone a total of US$16.2 billion in current terms and US$23.8 billion in constant 1998 dollars between 1976 and 2000, based on CIF prices in Japan. Similarly, Brazilian iron ore mining ﬁrms have foregone a total of US$5.8 billion in current dollars and US$8.5 billion in 1998 dollars between 1976 and 2000. This US$22 billion subsidy over 25 years explains a great deal of Japan’s global competitiveness in steel production and steel-consuming industries, especially during the stagnation of the past decade in Japan. These massive subsidies from Japan’s two largest iron ore peripheries, combined with the subsidies from Japan’s coal peripheries, have sustained Japan’s steel and broader economic competitiveness for years by lowering steel production costs and thereby production costs for many other Japanese industries.

6. CONCLUSION These economies of scale in production and transport support and reinforce each other. If we consider them within an adequately spatialized

Japan’s Economic Ascent

203

understanding of the world economy, we note: (1) that scale increase of transformative processes leads to (a) increased heavy industrial consumption of raw materials, (b) accelerated depletion of the most accessible sources, and thus (c) cost-increasing distance to raw material sources and (2) that economies of scale in transport are required to resolve the contradiction between savings in processing driven by economies of scale and the increased cost of transport across ever greater spaces. The resolution of this contradiction has led to an ever greater volume of increasingly cheap steel in the world economy and to an ever expanding space in which coal and iron are mined and transported. This increase in space in which the iron and coal are provisioned and transported, however, have also eliminated the ‘‘natural tariffs of distance’’ (cf. Innis, 1956, ‘Mandel, 1975,’ Harvey, 1982). This consequence of cheaper transport has usually been addressed in terms of the increased competitiveness of imports against locally manufactured goods, but extractive industries suffer similar impacts. As the technology of transport globalizes raw materials sources, the importing countries are no longer constrained to more proximate sources. As the cost of transport is diminished, all mines or sources compete in the establishment of rent, which tends to lower prices. Additionally, in a situation where there are relatively few deposits of the size required to realize economies of scale, this kind of global competition reduces any element of monopoly rents. The Japanese state and ﬁrms have enhanced this technological assault on rent and thus on the prices they pay to exporting countries by using joint ventures, long-term contracts, and various kinds of aid to create excess capacity in both coal and iron. The Japanese have been extremely aggressive in their strategies to play off North American, South American, Asian, and African sources of raw materials against each other, and their success has involved considerable learning and manipulation of information, institutional relations, and contract arrangements (Anderson, 1987; Bunker & Ciccantell, 1995a, b; Graham, Thorpe, & Hogan, 1999; Koerner, 1993; Swan, Thorpe, & Hogan, 1999), but the effects of these social and political strategies would be far less without the globalization of raw materials sources that the Japanese-led revolution in maritime transport technology made possible. The revolution in transport technology would not have occurred without the massive increase of scale and fuel efﬁciency that the Japanese achieved with the Basic Oxygen Furnace. The consequences of the intersection of two different mechanisms of scale-dependent fuel saving and precision-enhancing technologies, one in smelting, the other in transport, by setting in motion multiple mechanisms to reduce raw materials rents,

204

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

directly accelerated the globalization of raw materials markets and exacerbated inequalities between the exporters and the importers of raw materials. The Japanese model of capital accumulation that emerged out of the conﬂictual efforts to promote coordination between the Japanese steel ﬁrms, MITI, and the ExIm Bank and directly linked to state–ﬁrm–sector coordination in shipbuilding and shipping (involving the parent ﬁrms of these steel ﬁrms and the same state agencies) manifested physically in the MIDAs that simultaneously served as ports to unload ever larger ships carrying raw materials from around the world at the new steel mills using the latest technology, which in turn supplied shipyards, automobile factories, and other major steel consumers located in the very same industrial park built with state subsidies on land reclaimed from the ocean. Steel as a generative sector provided a direct cost and quality advantage for domestic and export industries from autos to ships, served as a template for the pattern of state– sector–ﬁrm relations that underlay Japan’s rapid economic growth, and restructured remote regions of Australia, Brazil, Canada, and other nations into raw materials peripheries tightly linked to and in a decidedly weak bargaining position relative to the coordinated Japanese buyers. The cumulative impacts on Japan and on the rest of the world economy of the Japanese steel and linked industries have been profound.

REFERENCES Abegglen, J., & Stalk, G. (1985). Kaisha, the Japanese corporation. New York: Basic Books. Amin, S. (1996). The challenge of globalisation. Review of International Political Economy, 3, 216–259. Anderson, D. (1987). Japan’s coking coal procurement system: An evaluation. Materials and Society, 11(1), 23–36. Arrighi, G. (1994). The long twentieth century: Money, power, and the origins of our times. London: Verso. Arrighi, G. (1998). Globalisation and the rise of East Asia: Lessons from the past, prospects for the future. International Sociology, 13, 59–77. Ball, W. M. (1949). Japan-enemy or ally? New York: John Day Company. Biersteker, T. (1998). Globalisation and the modes of operation of major institutional actors. Oxford Development Studies, 26, 15–31. Bisson, T. A. (1949). Prospects for democracy in Japan. New York: Macmillan. Bisson, T. A. (1954). Zaibatsu dissolution in Japan. Berkeley: University of California Press. Borden, W. (1984). The pacific alliance: United States foreign economic policy and Japanese trade recovery, 1947–1955. Madison: University of Wisconsin Press.

Japan’s Economic Ascent

205

Botsman, P., Brett, E., & Roger, M. (1994). Respect and maturity in the Australia – Japan coal trade. Sydney: Evatt Foundation. Boxill, I. (1994). Globalisation, sustainable development, and post-modernism: The new ideology of imperialism. Humanity and Society, 13, 3–18. Bunker, S. G., & Ciccantell, P. (1995a). A rising hegemon and raw materials access: Japan in the post-world war II era. Journal of World-Systems Research, 1(10), 1–31. Bunker, S. G., & Ciccantell, P. (1995b). Restructuring space, time, and competitive advantage in the world-economy: Japan and raw materials transport after world war II. In: D. Smith & J. Borocz (Eds), A new world order? Global transformations in the late twentieth century. Westport, CT: Greenwood Press. Cargo Systems Research Consultants (CSR). (1982). Large bulk carrier employment in the eighties(50,000 DWT+). Surrey: Cargo Systems Research Consultants. Chandler, A. (Ed.) (1965). The railroads: The nation’s first big business. New York: Harcourt, Brace and World. Chandler, A. (1977). The visible hand: The managerial revolution in American business. Cambridge, MA: The Belknap Press of the Harvard University Press. Chida, T., & Davies, P. (1990). The Japanese shipping and shipbuilding industries. London: The Athlone Press. Ciccantell, P. (2000). Globalisation and raw materials-based development: The case of the aluminum industry. Competition and Change, 4, 273–323. Ciccantell, P., & Bunker, S. G. (2002). International inequality in the age of globalization: Japanese economic ascent and the restructuring of the capitalist world-economy. Journal of World-Systems Research, 8(1), 62–98. Drewry Shipping Consultants. (1978). Trends in Japanese dry bulk shipping and trade. London: Drewry Shipping Consultants. Garrett, G. (1998). Shrinking states? Globalisation and national autonomy in the OECD. Oxford Development Studies, 26, 71–97. Graham, P., Thorpe, S., & Hogan, L. (1999). Non-competitive market behaviour in the international coking coal market. Energy Economics, 21(3), 195–212. Hadley, E. (1970). Antitrust in Japan. Princeton: Princeton University Press. Harvey, D. (1982). The limits to capital. Chicago: University of Chicago Press. Harvey, D. (1995). Globalisation in question. Rethinking Marxism, 8, 1–17. Hughes, H. (1963). The Australian Iron and Steel Industry, 1848–1962. Melbourne: Melbourne University Press. Innis, H. (1956). Essays in canadian economic history. Toronto: University of Toronto Press. Johnson, C. (1982). MITI and the Japanese miracle: The growth of industrial policy, 1925–1975. Stanford: Stanford University Press. Kiely, R. (1998). Globalisation, Post-Fordism and the contemporary context of development. International Sociology, 13, 95–115. Koerner, R. (1993). The behaviour of pacific metallurgical coal markets: The impact of Japan’s acquisition strategy on market price. Resources Policy, March 1993, 66–79. Lloyd’s of London. (1990). Lloyd’s ports of the world 1990. Colchester: Lloyd’s of London Press Ltd. Mandel, E. (1975). Late capitalism. London: Verso. Manners, G. (1971). The changing world market for iron ore 1950–1980: An economic geography. Baltimore: The Johns Hopkins University Press for Resources for the Future.

206

PAUL S. CICCANTELL AND STEPHEN G. BUNKER

Marshall, J. (1995). To have and have not: Southeast Asian raw materials and the origins of the pacific war. Berkeley: University of California Press. McDougall, W. (1993). Let the sea make a noise: Four hundred years of cataclysm, conquest, war and folly in the North Pacific. New York: Avon Books. McGraw-Hill. (1992). McGraw-Hill dictionary of science and technology. New York: McGrawHill. Nafziger, E. W. (1995). Learning from the Japanese: Japan’s pre-war development and the third world. Armonk, NY: M.E. Sharpe. Nagatsuka, S. (1991). Changes in the world’s shipbuilding facilities for large size vessels and future prospects thereof. Tokyo: Japan Maritime Research Institute. O’Brien, P. (1992). Industry structure as a competitive advantage: The history of Japan’s postwar steel industry. Business History, 34(1), 128–159. OECD. (1985). World steel trade developments, 1960–83. Paris: OECD. Ohashi, N. (1992). Modern steelmaking. American Scientist, 80(6), 540–555. Osaka, P., & Bureau, H. (1987). Port of Osaka. Osaka: Osaka Port and Harbor Bureau. Osaka, P., & Bureau, H. (1999). Port of Osaka: 1999–2000. Osaka: Osaka Port and Harbor Bureau. Panda, R. (1982). Pacific partnership: Japan-Australia resource diplomacy. Rohtak, India: Manthan Publications. Pauley, E. (1945). Letter to General Douglas MacArthur and President Truman. Foreign Relations of the United States, 1945: The Far East (Vol. VI). Washington: United States Government Printing Ofﬁce. Penfold, A. (1984). Development aid funds terminal construction. Bulk Systems International, August 1984, 26–29. Prigogine, I. (1984). Order out of chaos: Man’s new dialogue with nature. New York: Bantam Books. Reading, B. (1992). Japan: The coming collapse. New York: HarperBusiness. Sasaki, H. (1976). The shipping industry in Japan. London: International Institute for Labour Studies. Shaw, M. (1997). The state of globalisation: Towards of theory of sate transformation. Review of International Political Economy, 4, 497–513. Sklair, L. (1998). Globalisation and the corporations: The case of the California fortune global 500. International Journal of Urban and Regional Science, 22, 195–215. So, A., & Chiu, S. (1995). East Asia and the world economy. Thousand Oaks, CA: Sage Publications. Swan, A., Thorpe, S., & Hogan, L. (1999). Australia-Japan coking coal trade: A hedonic analysis under benchmark and fair treatment pricing. Resources Policy, 25(1), 15–25. Todd, D. (1991). Industrial dislocation: The case of global shipbuilding. London: Routledge. UNCTAD. (1992). Review of maritime transport. New York: United Nations. United States Bureau of Mines/United States Geological Survey (USBM/USGS). (Various Years). Minerals Yearbook. Washington, DC: USBM/USGS. United States State Department. (1949). Basic initial post-surrender directive to supreme commander for the allied powers for the occupation and control of Japan. Political Reorientation of Japan, September 1945 to September 1948. Washington: United States Government Printing Ofﬁce. Whitman, R. (1965). Steel. In: United States bureau of Mines (Ed.), Mineral Facts and Problems 1965 Edition (pp. 847–873). Washington: United States Bureau of Mines Bulletin 630.

Japan’s Economic Ascent

207

Yaghmaian, B. (1998). Globalisation and the state: The political economy of global accumulation and its emerging mode of regulation. Science and Society, 62, 241–265. Yonekura, S. (1994). The Japanese iron and steel industry, 1850–1990: Continuity and discontinuity. New York: St. Martin’s Press. Yuzo, K. (Ed.) (1990). Yokohama past and present. Yokohama: Yokohama City University.

A PERPETUAL EXTRACTIVE FRONTIER? THE HISTORY OF OFFSHORE PETROLEUM IN THE GULF OF MEXICO Tyler Priest ABSTRACT For the first time since the ‘‘limits to growth’’ debate of the 1970s, we hear serious talk about the prospect of the world running out of oil. In the United States, concerns about reducing dependence on foreign oil have incited debate over the viability of alternative energy sources versus the oil industry’s search for new oil ‘‘frontiers.’’ The rancorous dispute over drilling in the Arctic National Wildlife Refuge (ANWAR) has captured the spotlight in this debate. Less controversial, but more significant for the future of U.S. oil production, are the bountiful ‘‘deepwater’’ reserves of the Gulf of Mexico (GOM). Offshore is central to the history of the petroleum industry over the last 50 years, and the GOM is the most explored, drilled, and developed offshore petroleum province in the world. In recent decades, revenue from offshore leasing has been second only to federal income taxes in value to the U.S. treasury. During the last 30 years, the search for oil and gas has continually moved into deeper waters and into new offshore environments. Still, the GOM remains the primary laboratory for technological innovation and regulatory practices. The Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 209–229 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10010-9

209

210

TYLER PRIEST

recent and spectacular revival in production there thanks to deepwater discoveries has strongly reinforced this demonstration effect. As offshore oil assumes a high profile in national development strategies around the world, any effort to analyze the political, social, and economic aspects of offshore exploration and development must use the GOM as a historical precedent or basis of comparison.

1. INTRODUCTION For the ﬁrst time since the ‘‘limits to growth’’ debate of the 1970s, we hear serious talk about the prospect of the world running out of oil. Prominent geologists and energy economists argue that the world faces a permanent peak in oil production in the coming decade or two and a precipitous decline thereafter (Campbell, 1997; Deffeyes, 2001). These dire forecasts appear to be conﬁrmed by the aggressive actions of oil ﬁrms scrambling to ﬁnd new sources of equity oil, as well as by the foreign policy of the United States as it conquers oil-rich Iraq and expands its inﬂuence in Central Asia where new reserves have been discovered (Klare, 2001). Domestically in the United States, meanwhile, concerns about reducing dependence on foreign oil have incited debate over the viability of alternative energy sources versus the oil industry’s search for new oil ‘‘frontiers.’’ The rancorous dispute over drilling in the Arctic National Wildlife Refuge (ANWAR) has captured the spotlight in this debate. Less controversial, but more significant for the future of U.S. oil production, are the bountiful ‘‘deepwater’’ reserves of the Gulf of Mexico (GOM).1 Offshore is central to the history of the petroleum industry over the last 50 years, and the GOM is the most explored, drilled, and developed offshore petroleum province in the world. From negligible production in 1947, the worldwide production of offshore oil reached 26 million barrels/day in 2003, representing 34 percent of total world crude production. Over the same period, worldwide natural gas production rose to around 685 billion cubic meters per day, with 25 percent of this total accounted for by offshore gas (Sniekus, 2004). Today in the continental shelf waters off Louisiana and Texas, there are nearly 4,000 active platforms servicing 35,000 wells, and 29,000 miles of pipelines. On an oil equivalent basis, output from the GOM, providing close to 25 percent of U.S. oil and gas production, already exceeds Texas and will soon surpass Alaska. Even conservative estimates see this percentage rising to one-third of the U.S. total by 2010 (Minerals

A Perpetual Extractive Frontier?

211

Management Service, 2002). In recent decades, revenue from offshore leasing has been second only to federal income taxes in value to the U.S. treasury. During the last 30 years, the search for oil and gas has continually moved into deeper waters (now routinely exploring in 10,000 ft and producing in 5,000 ft) and into new offshore environments (from the GOM and the North Sea to Brazil, West Africa, and elsewhere). Still, the GOM remains the primary laboratory for technological innovation and regulatory practices. The recent and spectacular revival in production there, thanks to deepwater discoveries, has strongly reinforced this demonstration effect. As offshore oil assumes a high proﬁle in national development strategies around the world, any effort to analyze the political, social, and economic aspects of offshore exploration and development must use the GOM as a historical precedent or basis of comparison.

2. THE BUNKERIAN FRAMEWORK Academic investigations of offshore oil in the Gulf have focused almost exclusively on the socioeconomic impacts on onshore support communities. Most studies tend to apply a classic boomtown model, which identiﬁes timeand space-compressed changes in the demand for labor as the primary cause for negative impacts (Seydlitz & Laska, 1994). The problem with this model is that it was developed to analyze the impact of extractive development on small, isolated rural towns from an earlier era in the American West. Since World War II, however, oil-dependent boomtowns in the United States have been rare. Technological innovation has extended the life of oil ﬁelds and reduced labor forces, while transportation improvements, management practices, and regional growth have reduced job-related migration (Olien & Olien, 1982). The need for platform workers to live onshore and commute long distances has created a unique kind of labor market offshore; concentrated work schedules (12-hour shifts for 7, 14, or 21 days, and then an equal number of days off) have functioned to stabilize the residences of the industry’s workforce. The pace of leasing, regulated by the federal government, checked the tendency for wild and uncontrolled growth, especially after the 1950s. Moreover, the offshore petroleum industry is not rural and isolated, but regionally expansive and functionally complex. It is segmented into oil companies, drilling companies, engineering and construction companies, service companies, supply companies, diving companies, and even catering companies. The boomtown model overemphasizes construction, to

212

TYLER PRIEST

the neglect of other segments of the industry, and tends to deduce socioeconomic effects from projected labor demand, when in fact larger social, economic, and political processes are at work, especially those relating to technological innovation, leasing, and strategic decisions and investments (Luton & Cluck, 2003). What is needed for both policy and historical analysis of the petroleum industry in the Gulf is a conceptual framework that explicitly links industry or regional-level effects to community or local-level effects. Stephen Bunker’s work on extractive economies is a good place to begin constructing one. Dissatisﬁed with the way in which economic and social models derived from industrial or manufacturing processes collapse the particular into the general in arriving at theories of regional development, Bunker revisited the ‘‘staples’’ studies of Canadian economic historian Harold Innis for greater insight into extractive economies. He discerned through Innis how specific commodities provide points of common reference for both external exchange systems and internal production systems, and how they might serve as a bridge between different levels of analysis, local and global, social and environmental, or ﬁrm and state. ‘‘Awkward, particular, and overdetailed they may be,’’ Bunker writes, ‘‘continued studies of the extractive process around specific staples will remain an essential counterbalance to our need for general statement as long as global industrial expansion continues to increase rates of extraction’’ (Bunker, 1989, p. 608). Using empirical and inductive methods of research and analysis – in contrast to the deductive method characteristic of the boomtown model of offshore petroleum – Bunker examined how local and environmental conditions constrained various extractive economies from the Brazilian Amazon to Western Australia. Writing at a time when notions of ‘‘postmodernity,’’ ‘‘postindustrial society,’’ and the ‘‘social construction of nature and space’’ were dominating academic discourse, he argued that the physical environment is, in many ways, irreducible. It actually shapes social organization, restructures markets, or at least forces a deeper adaptation by societies to the material world than is generally recognized. An overarching theme coming out of Bunker’s work is that there is a fundamental difference in the development trajectories of extractive economies and productive/ industrial economies. The expanding scale of capital and technology in industry has exhausted proximate sources of raw materials and required the exploitation of larger and more remote deposits, which are ever more highly constrained by the environment and by the physical characteristics of the raw materials themselves. Extraction, which involves taking raw materials from the earth – whether by mining, logging, or drilling – is much less

A Perpetual Extractive Frontier?

213

capable of manipulating time and space than production, which involves recombining and processing those raw materials. Bunker argues that a contradiction eventually arises between the capitalist logic of economies of scale and the ‘‘diseconomies of space’’ or ‘‘natural logic of matter.’’ These contradictions are resolved either through technological innovation or, to borrow a concept from David Harvey, a ‘‘spatial ﬁx’’ involving politically negotiated changes in property or market relations. Each cycle of contradiction and resolution, however, creates a new cycle of even greater economies of scale in production, expanding the material and spatial problems of the previous cycle. Ultimately, these problems reach global and unsustainable proportions, with the most adverse effects falling on the extractive regions of the world economy (Bunker, 1985, 1989, 1994; Bunker & Ciccantell, 2003; Harvey, 1982).

3. THE PHYSICAL ASPECTS OF OFFSHORE PETROLEUM EXTRACTION IN THE GOM The environmental constraints on exploring and extracting petroleum offshore are indeed formidable. The industry must design and build massive structures to withstand extreme wave forces, soil movements, and storms and hurricanes. Infrastructure built for this environment – platforms, pipelines, drilling and support vessels, onshore support centers – is highly specialized and requires staggering ﬁxed costs and long lead times (the ﬁrst generation of tension-leg platforms and semi-submersible production platforms installed in the 1990s have cost about $1 billion each). Along the Gulf Coast of Louisiana and parts of Texas, the dredged marshes allow entry to thousands of miles of pipelines from the ocean, and the organization of entire communities (shipyards, ports, depots, administrative centers) is geared to support offshore oil and gas. By cultivating a competitively organized offshore drilling and supply industry, the oil companies have protected themselves somewhat against the high infrastructure costs and irregular pace of exploration and development. The subsidiary parts of the industry and the parish communities in Louisiana have been much more vulnerable, though they have often restructured and adapted to keep the system functioning for more than half a century. The story the petroleum industry and the business press tells about offshore oil highlights the ability of innovative engineering to overcome challenges and liberate extractive enterprise from the limits imposed by the

214

TYLER PRIEST

material world. The subtext for this story is the current debate over the United States’ dangerous dependence on foreign oil. The message is: faith in technology will save us. Not only can it reduce our dependence on overseas sources, it can even emancipate us from the shackles of Mother Nature. The GOM has been written off as a ‘‘dead sea’’ time and again, the refrain goes, yet technology has always brought it back to life by enabling the industry to ﬁnd and produce oil and gas more cheaply. The latest revival, beginning in the mid-1990s, seemed to discredit much of the boomtown literature. No need to worry about an immanent peak in world oil production, or hurry to conserve petroleum and embrace alternative energy, says the corporate voice triumphantly. The industry, if left alone, will ﬁnd some way to pull more oil out of the earth than we can possibly imagine (Rist, 1999; Rauch, 2001). It is easy to be seduced by marine and geophysical technology. The scale, ingenuity, and power of offshore structures are almost sublime. If the Gulf could be drained of water, one would behold on the Louisiana–Texas continental shelf a sprawling, futuristic landscape of steel that would dwarf in height and extent the skylines of all the major American cities combined. Multiple varieties of awe-inspiring, dynamically positioned and semisubmersible drilling vessels are now working in water depths unimaginable not long ago. To produce oil from these new depths, the industry has moved beyond the towering steel-jacket platforms ranging over a thousand feet in height to develop mammoth tension-leg platforms, tethered to the bottom, and ﬂoating ‘‘spars’’ that resemble monstrous buoys in the water. Directional drilling and subsea wellhead technologies have greatly increased the development range of any given platform. The new geophysical ‘‘visualization’’ rooms with their brilliant 3-D and 4-D (multi-component, timelapse) seismic displays are a dramatic example of how the industry has learned to manipulate space and time in trying to pinpoint oil and gas deposits in the costly offshore environment. Advances in seismic technology allow geoscientists to limit exploration wells in search of new ﬁelds and extend output from old ﬁelds. Geoscientists even refer to their ability to ‘‘create’’ new reserves in these ‘‘digital oil ﬁelds.’’ ‘‘Intensivity’’ has now replaced ‘‘extensivity’’ in the search for raw materials, writes one enamored observer. This is the ‘‘new old economy’’ (Rauch, 2001). In this sense, the seemingly limitless capabilities to ﬁnd virgin sources in deeper waters and increase output from oil ﬁelds have made offshore oil a perpetual extractive frontier. Throughout the twentieth century, as ‘‘immanent peak’’ skeptics often point out, technological innovation has repeatedly created new supplies of petroleum and silenced doomsayers who warned of impending and

A Perpetual Extractive Frontier?

215

permanent shortages (Williams, 2003). But even if today’s ‘‘depletionists,’’ who are backed by more data and rigorous scientific analysis than their predecessors, are unduly alarmist, we still need to develop an alternative narrative of offshore oil that does not fetishize technology, or one that at least better explains the spatial, environmental, and socioeconomic development of offshore technology. We also should pay attention to the role of technology in the interaction between states and ﬁrms as well as the interaction between development in the GOM and in other parts of the world. Furthermore, petroleum is still a ﬁnite resource. Technology cannot expand frontiers and increase reserves indefinitely. Sooner or later, the Gulf will cease producing petroleum. Public policymaking toward offshore oil will beneﬁt from a better historical understanding of both the constraints and the liberating possibilities of offshore technology. The place to start, using the Bunker approach, is with the particular characteristics of marine resource development in the GOM. The physical environment in the Gulf created opportunities as well as challenges for the oil industry. Unlike most petroleum provinces, in which discoveries have been concentrated in a short span of one to three decades, substantial discoveries have been made in the GOM basin for the past nine decades. In contrast to the major provinces of the world, where hydrocarbons are concentrated in a small number of world-class ‘‘giant’’ ﬁelds (ﬁelds with a known recovery of 500 million barrels of oil equivalent [boe] or more), the GOM basin has yielded thousands of smaller ﬁelds as well as numerous giants and ‘‘large’’ ﬁelds of 50–500 million boe (Nehring, 1991). This unique geology has created opportunities for a wide range of companies and oil hunters, and an even greater number of subsidiary businesses. The Gulf’s gradually sloping, deltaic plain permitted trial-and-experimentation with building freestanding structures in the open water, something that had never been tried before the 1940s. The sedimentary layers of the Gulf’s ocean bed are relatively soft, making them easier to drill than hard-rock layers in other regions, onshore or offshore. The water is shallow and the conditions are mild, except for hurricanes. In the 1950s, however, the main areas of activity in the Gulf were largely hurricane-free, with the exception of Hurricane Audrey in 1957, which completely destroyed the onshore support community of Cameron, Louisiana. There are few barriers to gathering seismic and geological information offshore. Companies did not have to contend with individual property holders or imposing topography, as they did onshore. The achievement of mobility on the water, ﬁrst in geophysical prospecting (offshore crews could collect much more seismic data than onshore at a lower cost-per-proﬁle) and then in drilling with the successive development of submersible, jack-up, and

216

TYLER PRIEST

semi-submersible vessels, made offshore oil viable by eliminating the need to build platforms for exploratory drilling (in the event of a dry hole, exploratory drilling platforms involved huge, stranded costs). Offshore ﬁelds were close enough to established reﬁning centers along the Gulf Coast that early output could be barged to supply depots, and the laying of the ﬁrst underwater pipelines was not prohibitively expensive or technically challenging (Pratt et al., 1997; Gramling, 1996).

4. FROM WETLANDS TO OPEN WATER Before the industry truly developed in the 1950s, early drilling success along the coast of the GOM was due mostly to geologic good fortune and corrupt deals with Louisiana ofﬁcials. Salt dome discoveries, ﬁrst in the marshes, bays, and swamps of south Louisiana in the 1920s and 1930s, and then in the shallow waters offshore in the 1940s and 1950s, were numerous and significant by U.S. standards. Seismic technology and gravity measurements proved useful in crudely identifying the location of oil prospects related to salt domes in south Louisiana, but early instruments and techniques were not entirely accurate. Moreover, the soft, unconsolidated sands in many places on the Gulf Coast frequently did not yield workable results with the revolutionary reﬂection seismograph. The drill was often the only way to obtain reliable geologic information (Owen, 1975, pp. 454–457, 759–769). Consequently, oil companies desired large and cheap leases to allow for wider ranging drilling programs, a ‘‘spatial ﬁx’’ to compensate for imperfect technology and the relatively higher costs of operating in the wetlands environment. Louisiana ofﬁcials were happy to oblige. In the late 1920s and early 1930s, the state and local levee districts leased millions of acres to oil interests through widely known kick-back schemes involving Huey Long, his machine cronies, and other local political bosses such as the infamous Plaquemines Parish district attorney, Leander Perez (Jeansonne, 1977; Banta, 1981; Dodd, 1991). Some single leases, many of which ended up being held by Texaco (Gulf Reﬁning, Humble Oil, and Shell Oil were the other major players) covered hundreds of thousands of acres (State Mineral Board of Louisiana, 1946). One of Texaco’s leases, the infamous State Lease No. 340, encompassed Vermillion Bay, West Cote Blanche Bay and extended seaward indefinitely into the GOM.2 The reform of the leasing system in 1936, limiting leases to a maximum of 10,000 acres and transferring authority for leasing from the governor to a State Mineral Board, ended the most egregious

A Perpetual Extractive Frontier?

217

abuses. The board nevertheless remained a tool of the governor, who sat on the board and appointed its members, and who effectively retained wide authority over leasing. The price of lease bonuses (payments for the right to obtain a lease, usually awarded at an auction), a key measurement of the cost of access and exploration risk, rose after the reform but remained quite cheap, an indication that money was likely still being paid through informal channels to Long machine politicians and intermediaries. Before the reform, Texaco paid a 10 cents/acre bonus for its 500,000 acre lease and 30 cents/acre for its 250,000 land acres in State Lease 340 (State Mineral Board of Louisiana, 1946). After the reform, the price of bonuses typically ran in the $1–5/acre range and did not rise much with the leasing of offshore tracts in the late 1940s and early 1950s. Those generally went for $3–6/acre, compared to the $200–300/acre bonus bids in early federal lease sales just a few years later, which were more competitive and less politicized.3 Corrupt though they were, leasing practices in Louisiana provided a great stimulus to oil development in marine locations onshore and on into the open sea. Hugely profitable oil and gas ﬁelds in south Louisiana both generated interest in the adjacent offshore domain and ﬁnanced a new wave of exploration and drilling by the established ﬁrms. Meanwhile, other oil ﬁrms that lacked the political connections and had missed out on the action in south Louisiana also cast their sights into shallow open water. In the traditional narrative of offshore oil, the early pioneers took supreme risks and overcame daunting challenges to open up the offshore frontier. However, the risks had to be measured by the tremendous opportunities afforded by the GOM’s location, topography, and geology, which shaped the politics and policies offering incentives to oil extraction. Had the GOM shelf dropped off more steeply, had the coastal plain not produced such bountiful quantities of oil, or had it been leased in a more responsible manner, few would have bothered looking offshore, at least for many years. The opportunism of oilmen was rewarded. The increase to U.S. domestic reserves from offshore development during 1949–1956 was nine times the average for onshore wells (U.S. Department of Interior, 1969). Technological innovations in geophysical exploration, drilling, and marine design and construction held the keys to this success. Still, in a truly open oil market, the costs and risks of developing reserves in increasing water depths offshore might have outweighed the returns. State aid and support for the offshore industry, both direct and indirect, were crucial to its ongoing viability. The governance arrangements in the U.S. oil industry allowed for the shifting of differential oil rents from lower cost producing areas overseas, which were radically changing the world oil market in the 1950s, to higher cost U.S.

218

TYLER PRIEST

prospects, such as those offshore (Baxter, 1997). These arrangements included not only all the well-known tax beneﬁts and incentives that the industry as a whole received in the United States – such as the percentage depletion allowance and import quotas, which were imposed on a mandatory basis in 1959, just after the major offshore discoveries of the mid-1950s came on-stream – but specific assistance in the form of government technology transfers. Sonar and radio-positioning systems developed for warfare at sea proved essential for oil exploration offshore. The Navy Experimental Diving Unit trained schools of divers in underwater salvage operations and developed mixed-gas and saturation diving techniques, jump-starting the postwar commercial diving business, which became a vital adjunct to the offshore industry. Gulf Coast construction companies acquired war-surplus landing craft for pennies on the dollar and converted them to drilling tenders, supply and crew boats, and construction and pipelaying vessels (Pratt et al., 1997). Under the interstate prorationing system, generous production ‘‘allowables’’ for offshore wells set by the state of Louisiana and the federal government compensated for higher ﬁxed costs compared to onshore.4 Little or no safety and environmental regulations at ﬁrst also minimized operating costs – but not worker fatalities, which were all-too-frequent – and encouraged trial-and-error experimentation with engineering and construction. And despite oil company fears that the triumph of the federal government in the legal showdown with Louisiana and other states over control of the ‘‘tidelands’’ in the 1950s would restrict development, federal leasing introduced an orderly method for cheaply and efﬁciently transferring huge tracts of the public offshore domain to industry.

5. THE CONTRADICTIONS OF OFFSHORE EXPANSION IN THE 1960S Once offshore oil became a viable and growing industry, the federal government did adopt an explicit policy to ration the leasing of tracts in the GOM. In the historic March 1962 offshore lease sale, the federal government collected $445 million in bonus bids. Department of Interior (DOI) ofﬁcials were awakened to the fact that this program, with only about 30 people total, took in more money in a single sale (and in later years a single offshore tract) than all the timber sales in Oregon and California and onshore mineral leasing for the year combined. The next general sale was not held until 5 years later, in 1967, and in the intervening years, the DOI developed a new system referred

A Perpetual Extractive Frontier?

219

to as ‘‘tract selection,’’ which imposed acreage limitations on sales to increase cash bonuses. In a typical lease sale, companies would ‘‘nominate’’ tracts they felt were prospective, as well as cosmetic ones they did not like, in order to hide their true interests and misdirect competitors. The DOI’s Bureau of Land Management (BLM) would pick and choose from the set of nominated tracts to offer at a sealed-bid auction. After the 1962 sale, in which nearly every tract nominated was put up for sale, the BLM became much more selective about the tracts it offered. The government recognized offshore leasing as a significant source of revenue, and as the costs of the Vietnam War escalated, Johnson Administration ofﬁcials pressured the BLM to increase its take from bonuses and search for a more scientific estimation of ‘‘fair market value’’ for the public lands being offered (U.S. Department of Interior, 1969; Rankin, 1986). The main problem with this system was that the grid pattern of the leasing maps, in which the offshore public domain was divided into 5,670-acre square (9 sq. miles) tracts and subdivided blocks, did not correspond to the geology of oil and gas deposits and often frustrated or complicated the interpretation of the geology. Tract selection offered tracts or blocks in a piecemeal fashion, which hindered more efﬁcient exploration strategies involving basin-wide assessments or the pursuit of structural trends or ‘‘plays’’ (a group of geologically related prospects) that transcended tract boundaries (Lester, 1992, pp. 91–93). Even individual prospects within a given trend or play might be tricky to value and acquire. A given tract covering the crest of a salt dome, where oil and gas was unlikely to be found, would invite much lower bids than ones for nearby or adjoining tracts on the ﬂanks of the same salt dome, where the probability of a discovery was much higher. A company in possession of a ﬂank tract would likely bid more for the crest tract to unify its control over the ﬁeld. For all tracts, bonus bids could vary greatly between bidders who all had staffs using sophisticated seismic data and geological analyses unavailable to the BLM and the U.S. Geological Survey (USGS), which were charged with an impossible task of estimating the market value for leases (Rankin, 1986). The government often ended up offering tracts not desired by the leading oil companies, and at times the BLM and USGS could not even agree on the proper ones to offer. Concerned with protecting their own proprietary data, companies would not share their specific interests with the government in the nomination process. This secrecy obscured an oil company’s specific interest not only from their competitors but also from leasing ofﬁcials, who often did not fully understand what the industry’s objectives were until after the sale (Lester, 1992).

220

TYLER PRIEST

Fair market value, then, proved to be an elusive concept, and the industry often cried foul at the ‘‘checkerboarding’’ of tracts offered under the tract selection system. The system brought in substantial revenue for the government, as many unproductive tracts were leased. It kept demand for offshore leases high, but frustrated the efforts of offshore operators to ﬁnd larger reservoirs with greater production in progressively deeper waters. A boom in the drilling and service industries in the mid-1960s, as oil companies drilled their large inventory of leases, disguised impending troubles. Although enough important discoveries were made to hold the industry’s interest, many of the leases proved to be unproductive and the cost of bringing in the productive ones began to outrun the price of oil, which had stubbornly remained in the $2–3 barrel range since the end of the Second World War (Wilson, 1966). Furthermore, three major hurricanes in 1964, 1965, and 1969 damaged numerous producing platforms and forced a reevaluation of the GOM within the oil industry (Dunn, 1994).

6. THE ‘‘BRIGHT SPOT’’ ERA In the early 1970s, the major contradiction of the early offshore system in the Gulf was temporarily resolved. Most obviously, the price spike caused by the Oil Producing and Exporting Countries (OPEC) oil embargo in 1973 made offshore a much more profitable endeavor. Companies could afford a higher ratio of dry holes and unproductive leases to discoveries, even in progressively deeper water. However, lease bonuses began rising sharply before 1973, so this trend cannot be solely attributed to rising oil prices or to the federal policy of tract selection. An unappreciated factor in this trend was the discovery and adoption of the ‘‘bright spot’’ method of interpreting seismic data. Advanced digital recording and processing of seismic data, which had made quantum leaps in the mid-1960s, allowed geophysicists to measure the relative wave amplitudes between seismic traces for the ﬁrst time. Up to that point, seismic techniques only helped map subsurface structures and identify possible oil traps. Operators still had to take the risk of drilling to ﬁnd oil and gas. But the new digital seismic data enabled the ‘‘direct detection’’ of hydrocarbons. The acoustic impedance of a loosely cemented rock ﬁlled with hydrocarbons was different from that of a similar water-ﬁlled rock, and with advanced digital methods, this difference could often be detected as a socalled bright spot on the seismic record (Forrest, 2000a, b). Bright spots greatly diminished the dry hole factor in the risk equation. For example, if a bright spot scan reduced the potential for drilling a dry hole from

A Perpetual Extractive Frontier?

221

50 percent to 10 percent, then on a risk-weighted basis, an oil company could put a lot more money into its lease bids and more than make up for it in decreased drilling costs. Once the technology was developed and embraced, it had a tremendous impact on offshore exploration in the Gulf and helped pioneer a new deeper water hydrocarbon trend. An important qualifying point about direct detection or bright spots, however, is that it worked only for select kinds of geology. The highly porous, clastic, deltaic rocks of the Gulf Coast were highly amenable to this technique. Most areas of the United States and around the world are not. Regions where bright spot technology has worked include the Campos Basin off Brazil and the delta regions in the Gulf of Guinea off West Africa. Not surprisingly, these are the two hottest oil plays in the world right now outside the GOM and Caspian basin. Bright spots and the digital seismic revolution provided the technological innovations needed to resolve the contradiction between economies of scale and diseconomies of space in the earlier phase of offshore development in the GOM, through a process of adapting to the particular the characteristics of the region’s environment and resource base. By the early 1980s, however, this adaptation appeared to give the industry only a short lease on life in the Gulf. The price of lease bonuses had risen to astronomical levels. Companies were desperate for new oil sources after the nationalization of foreign holdings the 1970s and the shrinking of virgin exploration frontiers in North America. Offshore GOM remained one of the few promising, largely unexplored areas, leading the industry to spend more than $1 billion at each GOM sale. In September 1980, oil companies paid $2.8 billion in winning bids, dwarfing the previous record of $2.1 billion set in 1974 (‘‘Oil Firms Spend Record Amount,’’ 1980). Even the largest ﬁrms could not afford to bid alone, and had to bring in partners to offload some of the capital risk. Despite the price spike of the ‘‘second oil shock’’ and the application of new technologies, offshore leases were becoming prohibitively expensive. So was the cost of development as the deepwater frontier migrated out to the edge of the continental shelf in 1,000-ft-plus depths. In 1981, oil prices peaked and began to plunge. Industry leaders complained that the tract selection system of leasing was creating a shortage of exploration opportunities in the declining oil province of the United States, especially as environmental concerns blocked leasing off the Atlantic and Pacific coasts (Farrow, 1990, pp. 137–138; Nanz, 1975). They lobbied for a reform that would allow companies access to much larger acreage, both to bring down the price of bonuses and to enable them to implement more creative exploration strategies and acquire a greater number of contiguous blocks around a given prospect, with which they could gain control over a

222

TYLER PRIEST

greater share of production from such a prospect to offset the soaring ﬁxed costs of deepwater development (‘‘At Issue: Land Access,’’ 1981).

7. THE AREA-WIDE LEASING SYSTEM In 1981, the companies got their wish with the controversial appointment of James Watt as secretary of the interior under Ronald Reagan. In short time, Watt instituted a new ‘‘area-wide leasing’’ (AWL) system offshore, which put into play entire planning areas (e.g., the central GOM) up to 50 million acres, as opposed to tracts specifically nominated and offered under the tract selection system. In other words, companies could bid on any tract they wanted in the planning area, rather than have to choose from a limited number of carefully selected ones, and they would be more likely to acquire them in bunches, giving them greater control over large prospects. Beginning in 1983, major offshore acreage was leased in the GOM planning areas at sharply reduced bonus prices. Unlike under tract selection where high bids could be rejected if they did not meet fair market value criteria, procedures under the new system led to the acceptance of practically every high bid, no matter how low (Boue´ & Luyando, 2002). Varying criticism has been leveled at the AWL system. Opponents argue that it is a giveaway to the major oil companies, who are permitted to pick up vast acreage for ‘‘low ball’’ bids. Total Outer Continental Shelf (OCS) leasing revenues for the federal government, as a consequence, has declined significantly since the early 1980s. Critics also point out that AWL sharply reduced competition between oil companies for offshore acreage, or at least over control of deepwater development. The superior capital and technological capabilities for exploring deepwater gave the majors and larger independents a substantial edge under this system. Furthermore, the political controversies over AWL led to the moratoria on drilling in other vast areas of the U.S. OCS, such as the Atlantic and Pacific coasts. Finally, one can argue that AWL has been given too much credit for the deepwater boom of recent years. The link, according to an Oxford Energy Institute study, is ‘‘sequential and not consequential, because the real driving force behind the renaissance in GOM production has been technological progress, rather than ease of access to prospective acreage’’ (Boue´ & Luyando, 2002, p. 57). These criticisms are not entirely valid. AWL cannot be completely discounted as an important factor in the move into deepwater. Certainly, technological innovation – especially bright spot and 3-D seismic, tension-leg platforms, directional drilling, and subsea wells – have brought

A Perpetual Extractive Frontier?

223

these ﬁelds into play. But not until companies possessed cheap and extensive acreage did they have the incentive to develop and reﬁne these technologies. Competition was indeed muted early in the game. Deepwater is too expensive, risky, and dangerous for small or undercapitalized companies. Shell Oil acquired a vast majority of the early deepwater leases under AWL and pioneered the technologies needed to operate on them. But after Shell brought in major production at its Auger prospect in 1994, and after Congress passed the Royal Relief initiative passed in 1995, which suspended royalties on portions of production from deepwater ﬁelds, numerous companies, large and small, entered the game in force. As of 2002, at least 40 different operators had drilled deepwater wells in the Gulf (Godec, Kuuskraa, & Kuck, 2002). It is too soon to tell how competitive this game will become. Established ‘‘basin masters’’ such as Shell and other ‘‘super majors’’ such as British Petroleum (BP) might continue to maintain an advantage through their control of platform and pipeline infrastructure. The point is, however, that the relaxed terms of access under AWL and Royalty Relief – a new spatial ﬁx – has undeniably ushered in a new cycle of extractive development in the Gulf. Moreover, this new regime does not subvert the purpose of the U.S. federal leasing program, as critics imply. The statutory mission of the leasing program has never been to maximize federal revenues, but broadly to promote the ‘‘expeditious development’’ of oil and gas resources on the OCS in accordance with other federal policies. In a globalized world economy, in which the United States increasingly competes with other nations for the investment capital of oil ﬁrms, federal leasing administrators have felt compelled to provide incentives to deepwater exploration, which have led to some spectacular developments during the 1980s and 1990s. Between 1985 and 2001, 192 ﬁelds were discovered in deepwater, 38 of which contained more than 100 million boe, including ﬁve giant ﬁelds with more than 400 million boe (Godec et al., 2002). All this happened just as many people were writing the Gulf’s obituary as an exhausted, overadapted extractive economy (Freudenberg & Gramling, 1994; Gramling, 1996).

8. THE LIMITS AND CONSTRAINTS ON OFFSHORE DEVELOPMENT IN THE GOM During the past 50 years, the GOM has yielded great wealth and significant but uneven economic development for the Gulf Coast region. The tremendous new oil and gas reserves being discovered and developed in deepwater

224

TYLER PRIEST

have reestablished this province as one of the world’s hottest frontier regions and a proving ground for the next generation of innovations in offshore technology. Still, those who have predicted the demise of offshore oil in the GOM may not have been incorrect. Only their timing may have been unfortunate. Before getting too excited about the future of deepwater for the United States or for the Gulf Coast region, we should understand the nature of this potential, as well as its limits and constraints. First, the deepwater boom in the GOM owes its success not only to technology and easier access, or even the price of oil, but, once again, most of all to the particular characteristics of the resource. The deepwater reservoirs beyond the continental shelf occur in unique geological conditions, in turbidite sandstones which are capable of producing as much as 40,000 barrels/day from a single well, compared to 500–1,000 barrels/day from a good well on the shelf. Although the ﬁelds do not compare in size to the Middle East, they generate Middle Eastern rates of production. The surprising discovery of how productive these reservoirs were, ﬁrst at Shell Oil’s Auger platform in 1994, entirely changed the cost structure of the deepwater play (Ryser, 1995; Thorpe, 1996; Shell Oil Company, 1999). An operator could drastically reduce the number of expensive wells on a given platform and still produce at a greater rate than original estimates required for making the ﬁeld profitable. All the wonderful technology developed to produce deepwater hydrocarbons would not have been economical, at least in the near- to medium-term, had it not been for the productivity of those reservoirs. Offshore West Africa and Brazil have similar geology and similar well production rates, but these areas and the GOM are special cases and should not serve as models of extractive development that can be applied to most offshore environments. Even in the GOM, the incredible productivity of deepwater wells may have only postponed an inevitable day of reckoning for the Gulf Coast economy, which is both highly dependent on, yet in crucial ways increasingly ambivalent about, offshore oil and gas. Costs will continue to rise as development moves toward ever deeper waters and deeper formations. Deepwater ﬁelds are likely to have relatively short life-cycles, as high rates of production are bound to deplete them faster than production from equalsize ﬁelds on the continental shelf. Furthermore, deepwater GOM, so far, has turned out to be mainly an oil province. It has contributed little to slowing the steep decline in U.S. natural gas reserves, which is, at the moment, the most serious threat to U.S. energy security. Gas reserves additions have not reached a level where they have replaced declining production from the shelf. Although more oil reserves have been added in the late 1990s than

A Perpetual Extractive Frontier?

225

were produced offshore in the Gulf, the industry will have to continue ﬁnding major ﬁelds each year in order to sustain this performance (Godec et al., 2002). The second point, related to the ﬁrst, is that the current offshore ‘‘boom’’ is not like past booms in the GOM. Regional economic gains, other than to the major oil companies and larger independents, may not be as substantial. Many ﬁeld developments are tied into existing infrastructure, and more of the infrastructure that is added often is built overseas, usually in East Asian shipyards, and transported to the Gulf (‘‘Na Kika Topsides on the Move,’’ 2003). Smaller oil companies do not have the capital to participate to the extent that larger ones do, and fewer companies engaged in various offshore activities are local. As activity gets pulled further offshore, technology becomes more specialized and employment stability in the region becomes increasingly compromised. Although Louisiana is still more closely linked to offshore developments than anyplace else, deepwater does not promise to restore the state’s ﬁscal health, which has not really recovered from the bust of the 1980s after decades during which the state based its revenue almost entirely on oil severance taxes, royalties, and leases (Kurtz & Peoples, 1990). With yearly production in permanent decline, Louisiana now derives less than 30 percent of its revenues from oil and gas. Without any provisions for federal–state revenue sharing, coastal states like Louisiana that assume risks for offshore petroleum activities do not see ﬁscal returns from development in federal waters. In recent years, resentment toward this situation in Louisiana and other coastal areas has amounted to a veritable ‘‘Seaweed Rebellion’’ (Fitzgerald, 2001). Third, another manifestation of the physical constraints imposed by the environment is the accumulating damage to the Louisiana coastal wetlands caused by offshore petroleum development (not to mention the dangerous exposure of offshore infrastructure to damage by hurricanes). In the United States and in the Gulf, the offshore industry’s record on oil spills and pollution generated from offshore activities has been quite good since at least 1970, when the Santa Barbara blowout and two other major blowouts in the GOM alerted everyone to the potential hazards. However, the thousands of miles of canals dredged and laid with pipelines in a spaghetti-like maze through the marshes and swamps of south Louisiana have contributed to what is increasingly recognized as an environmental catastrophe. The canals break up natural barriers and provide easy conduits for salt-water intrusion and tidal scouring, leading to massive erosion and drowning of the marshes (Hallowell, 2001; Tidwell, 2003). Each year, Louisiana loses 25–35 sq. miles of coast, a land area larger than Manhattan, destroying the ecosystem that

226

TYLER PRIEST

supports the nation’s largest commercial ﬁshing industry. Indeed, the receding coast threatens whole communities as well as the survival of Cajun culture. The greatest factor in this tragedy is probably the containment of the Mississippi River by levees, which prevents soil replenishment by periodic ﬂooding and the spreading of estuaries along Louisiana’s coast. But scientists at Louisiana State University believe that canals are responsible for no less than one-third of the total coastal-zone degradation, and that once dredged, those canals tend to double in width every 14 years (Tidwell, 2003, p. 118). To be sure, technological innovation has been, and will continue to be, the driving force in the expansion of oil and gas into ever deeper waters in the GOM. But we must rigorously examine and contextualize this story, rather than merely celebrate it. Our fascination with technological achievement should not obscure other material factors in the industry’s longevity, uneven dynamism, and future. The GOM petroleum province has presented unique challenges and opportunities. Interactions between technology, capital, geology, and the political structure of access in the GOM have generated a functionally and regionally complex extractive industry with a history of resolving the contradiction between economies of scale and diseconomies of space, defying critics who have predicted its boom-and-bust demise. Nevertheless, as the Gulf matures, the limits on its potential become more apparent, even with the revival of activity in deepwater. It might be premature to suggest that rising costs and risks offshore will eventually expose the GOM offshore industry to the volatility of international oil prices and once again force painful adjustments. But as Stephen Bunker would remind us, this is still an extractive economy, characterized by extensivity and increasingly subject to material constraints.

NOTES 1. The definition of deepwater varies among companies and countries, but the most commonly accepted measurement is 1,500 ft of water or deeper. 2. In the 1960s, after the federal government gained ﬁrm legal jurisdiction beyond a 3-mile line from the coast, Texaco sued to have the boundaries of this lease recognized and excluded from federal offshore lease sales. A settlement was eventually reached limiting the size of State Lease 340. Nevertheless, it remained the largest offshore lease by far, and a productive one with many oil and gas ﬁelds. 3. The ﬁrst federal lease sale took place in October 1954. Per-acre bonus bids rose to as high as $5,000 in the early 1980s, until ‘‘area-wide’’ leasing brought them down to the $100–150 range (Minerals Management Service, 2002).

A Perpetual Extractive Frontier?

227

4. Prorationing was instituted in the 1930s to prevent ‘‘wasteful’’ competition by restricting overall output from individual ﬁelds. In the mid-1950s, the production allowable for a 10,000-ft.-deep well offshore (242 barrels/day) was nearly double the allowable for a comparable well onshore. Encouraging a greater spacing of development wells (up to 40 acres offshore compared to 30 acres onshore), the higher allowable lowered ﬁeld development costs to compensate for higher individual well costs. Allowables in federal waters usually followed those set by Louisiana (Pratt et al., 1997, p. 40).

REFERENCES At Issue: Land Access. (1981). Shell News, 5, 18–19. Banta, B.M. (1981). The regulation and conservation of petroleum resources in Louisiana, 1901–1940. Ph.D. dissertation, Louisiana State University. Baxter, V. (1997). The effects of industry governance on offshore oil development in the Gulf of Mexico. International Journal of Urban and Regional Research, 21(2), 238–258. Boue´, J. C., & Luyando, G. (2002). U.S. Gulf offshore oil: Petroleum leasing and taxation and their impact on industry structure, competition, production, and fiscal revenues. Oxford: Oxford Institute for Energy Studies. Bunker, S. G. (1985). Underdeveloping the Amazon: Extraction, unequal exchange, and the failure of the modern state. Urbana: University of Illinois Press. Bunker, S. G. (1989). Staples, links, and poles in the construction of regional development theories. Sociological Forum, 4(4), 589–610. Bunker, S. G. (1994). Flimsy joint ventures in fragile environments. In: B. Barham, S. G. Bunker & D. O’Hearn (Eds), States, firms, and raw materials: The world economy and ecology of aluminum (pp. 261–296). Madison, WI: The University of Wisconsin Press. Bunker, S. G., & Ciccantell, P. S. (2003). Generative sectors and the new historical materialism: Economic ascent and the cumulatively sequential restructuring of the world economy. Studies in Comparative International Development, 37(4), 3–30. Campbell, C. J. (1997). The coming oil crisis. Brentwood, England: Multi-Science Publishing and Petroconsultants. Deffeyes, K. S. (2001). Hubbert’s peak: The impending world oil shortage. Princeton: Princeton University Press. Dodd, W. J. (1991). Peapatch politics: The Earl long era in Louisiana politics. Baton Rouge, LA: Claitor’s Publishing Division. Dunn, F. P. (1994). Deepwater production: 1950–2000. OTC 7627. Proceedings of the offshore technology conference, Houston, TX. Farrow, R. S. (1990). Managing the outer continental shelf lands: Oceans of controversy. New York: Taylor & Francis. Fitzgerald, E. A. (2001). The seaweed rebellion: Federal–State conflict over offshore energy. Lanham, MA: Lexington Books. Forrest, M. (2000a). Bright idea still needed persistence. AAPG Explorer On Line. May. Forrest, M. (2000b). Toast was on the breakfast menu. AAPG Explorer On Line. June. Freudenberg, W. R., & Gramling, R. (1994). Oil in troubled waters: Perceptions, politics, and the battle over offshore drilling. Albany, NY: State University of New York Press.

228

TYLER PRIEST

Godec, M. L., Kuuskraa, V. A., & Kuck, B. T. (2002). How U.S. Gulf of Mexico development, ﬁnding, cost trends have evolved. Oil & Gas Journal, May 6, 52–60. Gramling, R. (1996). Oil on the edge: Offshore development, conflict, gridlock. Albany, NY: State University of New York Press. Hallowell, C. (2001). Holding back the sea: The struggle for America’s natural legacy on the Gulf coast. New York: HarperCollins. Harvey, D. (1982). The limits to capital. Chicago: University of Chicago Press. Jeansonne, G. (1977). Leander Perez: Boss of the delta. Baton Rouge, LA: Louisiana State University Press. Klare, M. T. (2001). Resource wars: The new landscape of global conflict. New York: Metropolitan Books. Kurtz, M. L., & Peoples, M. D. (1990). Earl K. Long: The saga of uncle Earl and Louisiana politics. Baton Rouge: Louisiana State University Press. Lester, C.F. (1992). The search for dialogue in the administrative state: The politics, policy, and law of offshore oil development. Ph.D. dissertation, University of California, Berkeley. Luton, H., & Cluck, R. E. (2003). Social impact assessment and offshore oil and gas in the Gulf of Mexico. Draft paper. U.S. Department of Interior, Minerals Management Service, Gulf of Mexico OCS Region, New Orleans, Louisiana. Minerals Management Service. (2002). Website: Offshore Stats and Facts, http:// www.mms.gov/stats/OCSproduction.htm. Na Kika Topsides on the Move in the Gulf of Mexico. (2003). Go Gulf Magazine, May/June, 24–25. Nanz, R. H., Vice President, Exploration, Shell Oil Company. (1975). The offshore imperative: The need for and potential of offshore exploration. Paper presented to colloquium on conventional energy sources and the environment, University of Delaware, Newark, Delaware, April 30. Nehring, R. (1991). Oil and gas resources. In: A. Salvador (Ed.), The Gulf of Mexico basin (pp. 445–494). Boulder, CO: The Geological Society of America. Oil Firms Spend Record Amount for Gulf Leases. (1980). Wall Street Journal, October 1, pp. 4, 15. Olien, R. M., & Olien, D. D. (1982). Oil booms: Social change in five Texas towns. Lincoln, NE: University of Nebraska Press. Owen, E. W. (1975). Trek of the oil finders: A history of exploration for petroleum. Tulsa, OK: The American Association of Petroleum Geologists. Pratt, J. A., Priest, T., & Castaneda, C. (1997). Offshore pioneers: Brown & root and the history of offshore oil and gas. Houston: Gulf Publishing. Rankin, J. (1986). History of federal OCS leasing. Untitled Manuscript in author’s possession. Rauch, J. (2001). The new old economy: Oil, computers, and the reinvention of the earth. The Atlantic Monthly, January, 35–49. Rist, C. (1999). Why we’ll never run out of oil. Discover, 20(6, June), 80–87. Ryser, J. (1995). Hot play in the Gulf. Texas Business, August, 33. Seydlitz, R., & Laska, S.B. (1994). Social and economic impacts of petroleum ‘boom and bust’ cycles. U.S. Department of Interior, Minerals Management Service, Gulf of Mexico OCS Region, New Orleans, Louisiana. Shell Oil Company. (1999). Shell in the U.S.: 1999 annual review. Houston: Shell Oil Company. State Mineral Board of Louisiana. (1946). Biennial report. Baton Rouge: State of Louisiana.

A Perpetual Extractive Frontier?

229

Sniekus, D., (2004). Offshore production rocket on the launchpad. Offshore Engineer, April 1, www.oilonline.com/news/features/oe/20040401. Thorpe, H. (1996). Oil and Water. Texas Monthly, February, 90–93, 139–145. Tidwell, M. (2003). Bayou farewell: The rich life and tragic death of Louisiana’s Cajun coast. New York: Pantheon. U.S. Department of Interior. (1969). Petroleum and sulfur on the U.S. continental shelf. Box 134, CCF, 1969–1972, Record Group 48, Records of the Secretary of the Interior, National Archives and Records Administration, II, Washington, DC. Williams, B. (2003). Debate over peak-oil issue boiling over, with major implications for industry, society. Oil & Gas Journal, July 14, 18–29. Wilson, H. M. (1966). Drillers face offshore deadline with 40 leases to test. Oil & Gas Journal, April 11, 48–51.

COMMODITY FRONTIER AS CONTESTED PERIPHERY: THE FUR TRADE IN IROQUOIA, NEW YORK AND CANADA, 1664–1754 Jonathan Leitner ABSTRACT This chapter examines the fur trade commodity frontier in northeastern North America as a contested periphery, involving an evolving process of conflict and cooperation between North American indigenous groups and European powers. Native people used European powers for help in their battles with other native groups, and European colonial authorities attempted to use native people as proxies in their attempts to make up for often low European populations in the various North American colonies. Within the colonies there were also splits between commercial/trading interests and more purely geostrategic concerns. This chapter will explore these various conflicts involving the Iroquois, English and French, and will consider how the trade’s fundamental material, environmental and geographical structure shaped the evolution of this peripheral extractive political economy and the efforts of those in the core seeking to exploit the area’s resources.

Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 231–252 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10011-0

231

232

JONATHAN LEITNER

1. INTRODUCTION: FUR AS COMMODITY FRONTIER, COMMODITY FRONTIER AS CONTESTED PERIPHERY ‘‘The development of the fur trade runs like a thread of blood and gold through the history of North America’’ (Wolf, 1997, p. 409).

Moore (2000, p. 428) suggests two basic types of ‘‘capitalist spatial expansion’’ in the early modern world-economy: the frontier mode, or ‘‘commodity frontier,’’ primarily in the Americas; and ‘‘‘trading post’ imperialism, which operated in Africa and Asia.’’ The latter mode also existed in the Americas, via the fur trade between Europeans and indigenous people. Though ‘‘redistributive and trade based’’ as Moore (2000, p. 428) generalizes, the fur trade also had serious ecological impacts, albeit not to the same extent that sugar or any other agricultural and/or industrial commodity production had; indeed, the beaver (Castor canadensis) in the northeastern U.S. has recovered over the centuries to the point that many humans now consider it a major pest species (Stevenson & Abruzzi, 2001). But during the period of serious fur extraction in northeastern North America, the localized impacts on fur-bearing mammal populations did in turn prompt the geographic expansion of the trade, the same way that, e.g., soil degradation prompted the geographic expansion of sugarcane, which made it a frontier, i.e. ‘‘a zone beyond which further expansion is possible in a way that is limited primarily by physical geography and the contradictions of capitalism rather than the opposition of powerful world-empires’’ (Moore, 2000, p. 412; his italics). Geographic expansion of a regional fur trade due to fur-bearing mammals’ localized extermination is an old story (or rather, analytic/narrative strategy) for North American authors following the generally westward course of ethnic European conquest and resettlement, and with a particular focus on the beaver (Castor canadensis) ‘‘[t]he main target of the North American [fur] tradeyespecially after the decline of the animal in Europe by the end of the sixteenth century’’ (Wolf, 1997, p. 159). The European beaver, Castor fiber, is the only other species in the genus Castor (Stevenson & Abruzzi, 2001) hence a close relative of C. canadensis. Both species’ inner layer of fur was good for making the men’s felt hats fashionable in western Europe during 1550–1850, with a barbed structure that let the hairs interlock and make the felt fabric comprising the hat, thereby facilitating manufacture (Stevenson & Abruzzi, 2001; Kardulias, 2002, p. 4).

Commodity Frontier as Contested Periphery

233

Writing in the 1920s, Innis (1970, p. 5) explicated the geographic expansionary dynamic of the North American beaver fur trade, suggesting that the beaver’s ‘‘non-migratory tendencies and elaborate housing facilities made destruction certain.’’ From local destruction ‘‘came the necessity of pushing to the westward and northwestward to tap new areas’’ (Innis, 1970, p. 6). There is certainly much more to Innis’ study than just that, but in conclusion he still comes back to that basic cause: ‘‘[t]he animal was not highly reproductive and it was not a migrant. Its destruction in any locality necessitated the movement of hunters to new areas’’ (Innis, 1970, p. 387). For Wolf (1997, p. 161), this is the ‘‘signal feature’’ of the North American fur trade, ‘‘its rapid movement westward as one beaver population after another was hunted out, and the fur hunters had to push farther inland in search of untapped beaver grounds’’ (also see Cronon, 1983, p. 99 on the phenomenon as south-to-north in 17th century New England; and Carlson, 2002 on the Pacific Coast fur trade as independent of overland expansion from eastern North America). Indeed, Wolf (1997, p. 159) recognizes that the North American fur trade resulted from Eurasian scarcity, with, e.g., the Russian fur trade’s own ‘‘speed of expansion being determined by the exhaustion of fur-bearing animals in each successive [river] basin.’’ Though some animal skins were used as industrial inputs in the core, such as deerskin (Dunaway, 1994, pp. 226–227), beaver fur was basically a luxury good, at least in terms of its relatively high value-to-volume ratio, although Price (1970, p. 850) divides fur into ‘‘luxury furs used on the pelt for garments’’ and ‘‘staple furs used by hatters or felt makers,’’ including, most importantly, beaver (also see Wallerstein, 1980, p. 273, n. 200, who notes K.G. Davies’ view that the beaver hat helped make fur less of an exclusive luxury). More precisely then, beaver pelts were largely a raw material input into the craft industry of beaver hatmaking, still an artisanal process that was ‘‘long and complicated’’ with many distinct steps, mostly performed by hand, not machine (White Oak Society, 2001). Yet because beaver fur, like fur in general, was basically more of a luxury good in this era (Hall, 1989, p. 18), at least in terms of having a relatively high value-to-volume, its extraction was not ecologically transformative to the degree that Moore’s agro-industrial commodity frontiers were. But the ‘‘commodity trading post frontier’’ (to combine Moore’s two formulations) for fur in North America had enough ecological impact to entail its geographic expansion by those involved, both indigenous and European. Some would further posit that even low-volume luxury goods can have major political–economic impacts; as Carlson (2002, pp. 410, 433) points out, luxury goods’ profit potentials promote expansion at the systemic level

234

JONATHAN LEITNER

and also help ‘‘[offset] otherwise prohibitive ‘start-up’ costs associated with colonial expansion’’ – though per Innis, luxury goods, due to their usual low value-vs.-volume can actually hamper colonial expansion (see below). Following from Wolf (1997 [1982]), Hall (1989, p. 18) more cautiously points out that fur trading did not ‘‘make or break empires,’’ though it did have ‘‘profound consequences for the groups that participated,’’ in particular the indigenous people who provided much of the labor and whose ways of life were dramatically altered, usually for the worse (also Dunaway, 1994, pp. 215–216). With the beaver trade in what is now the northeastern U.S. and eastern Canada, representatives of core European states in hegemonic competition were also involved. These European struggles articulated with those of Native American nations struggling for hegemony within their own world-systems (see Chase-Dunn & Hall, 1993) and autonomy on the European world-economy’s periphery, in large part due to struggles over fur supplies to meet European demand, to in turn meet their own desire (and increasingly need) for European trade goods – especially ﬁrearms, which was driven by international indigenous conﬂicts. These various indigenous people would face eventual incorporation into the world-economy – if not necessarily by dint of military force than via economic and political ‘‘transformation’’ (Hopkins & Wallerstein, 1987, p. 776) – vis a` vis their former status as part of a ‘‘contact’’ or ‘‘marginal’’ periphery that was less articulated with the European world-economy as well as more politically autonomous (Hall, 1989, pp. 19–20). A larger point is that incorporation is an ‘‘interactive’’ process between natives and colonizers, with native people having a great deal more agency than might appear after formal incorporation into the world-economy (Dunaway, 1996, pp. 454–458). The fur trade commodity frontier in northeastern North American was therefore also a contested periphery, ‘‘a region in the interstices between major states over which the latter ﬁght for control’’ (Kardulias, 2002, p. 2). In particular, Kardulias (2002, p. 2) sees early incorporation in the fur trade ‘‘as much a matter of Europeans being incorporated into the native American system as vice versa’’ – i.e., native skill in both gathering and producing fur required Europeans engaged to deal with them on at least somewhat equal terms, with native people in turn having ‘‘some level of control in managing the relationship’’ (Kardulias, 2002, p. 11). The relationship only began changing after native societies became dependent on European trade goods (Kardulias, 2002, pp. 8–9; more generally Wolf, 1997, p. 306). Kardulias (2002, p. 10) also briefly mentions that European powers – specifically the French, Dutch, and English – were drawn into the rivalries of their ‘‘native trading partners,’’ by way of keeping their fur supplies

Commodity Frontier as Contested Periphery

235

secured; other authors observe similar phenomena (e.g. Rotstein, 1977, p. 15; Gitlin, 1994, p. 103; Jennings, 1984, p. 72; Wolf, 1997, p. 169). At the same time, European colonial authorities attempted to use native people as proxies in their attempts to make up for often low European populations in the various North American colonies, while within the colonies there were splits between commercial/trading interests and more purely geostrategic concerns – though at least in New France there was a certain concurrence. This chapter will explore these various conﬂicts involving the Iroquois, English and French, following Carlson’s (2002, p. 390) ‘‘middle ground’’ approach to regional studies of world-systemic expansion (i.e. between ‘‘inside-out’’ Euro-/state-centered and ‘‘outside-in’’ indigenous-centered) – itself a sort of cross-ethnic incorporated comparison (McMichael, 1990) – and in consideration of the trade’s fundamental material, environmental and geographical structure, an underlying necessity when examining the relations between peripheral extractive political economies and those in the core seeking to exploit their resources (Bunker & Ciccantell, 2003, p. 27).

2. GEOPOLITICS VS. MARKETS IN COLONIAL NEW YORK’S FUR TRADE, 1664–1754 According to Innis (1970), beaver fur’s relatively high value-vs.-volume and ostensibly high profitability (also Kardulias, 2002, p. 4) had serious consequences for early Canadian development, due to its non-promotion (if not discouragement) of European agricultural resettlement. Innis’ interpretation was based on fur being lightweight and valuable, thereby easily and economically exported in its raw form. While the fur trade was the major economic activity in Canada, ships coming from Europe to North America were loaded with large amounts of supplies needed for the trade, in particular manufactured goods desired by indigenous people. As Innis (1970, p. 58) explains it, ‘‘Beaver skins were light, and the goods which were brought out from France bulky and heavy. ‘Most of the ships go laden to Canada and return light or empty.’’’ In other words, there was very little room for prospective European settlers on one of these transatlantic supply vessels, unlike with timber, in which timber ships sailing from Canada to Europe were full of that cheaper, bulkier commodity, but then had to ﬁnd paying cargo for the return trip. Once it became a British possession, emigrants were actively recruited to sail to Canada in the holds of timber boats in often horrific conditions, as opposed to the generally better conditions for

236

JONATHAN LEITNER

migrants to Canada during the French regime, who traveled with the crew rather than in steerage (Moogk, 1989, p. 468). This allowed boat owners to earn some income (or at least reduce their losses) on the return trip. Timber thus promoted settlement, while fur ‘‘retarded’’ it (Innis, 1956, pp. 67, 75, 359; cf. Eccles, 1979). Following from this physically derived dynamic, European-Amerindian relations were in turn marked by a mutual dependence, with Amerindians at least coming to appreciate the efﬁcacy of European manufactured goods, while Eurocanadians came to rely on certain native nations as both fur source and proxy force. For over a half-century before the English came to possess what they called New York, the Dutch colony of New Netherland (1609–1664) was subject to a similar dynamic (see Rink, 1986), with low Dutch population leading colonial authorities to foster a proxy relationship with indigenous people, in particular the Mohawks, easternmost nation of the ﬁve-nation Iroquois League. As the commodity frontier moved up the Hudson and then west, the Mohawks eventually became the main indigenous trading partners to New Netherland, and would expand their own power vis a` vis other indigenous nations in large part thanks to Dutch ﬁrearms (Jennings, 1984, p. 71; Trelease, 1997, p. 136; Rink, 1986, pp. 221–222). Post-English conquest, however, the fur trade would lose importance relative to other activities in the steadily diversifying New York economy, a trend which continued into the 18th century (Richter, 1992, p. 270) as depletion in Iroquoia (the area covering what is now upstate New York and southeastern Ontario) took hold and conﬂicts over hunting territory took precedence over actual hunting and trade. Nor, as with the Dutch economy (Leitner, 2003), and the French for that matter, would fur imports ever be significant to England’s economy (Eccles, 1979, p. 421, n. 8); fur, as an essentially high value–low volume good was certainly not a generative or even a leading sector for either the Dutch or English economies (Bunker & Ciccantell, 2003, pp. 15–20). Moreover, the European fur market was oversupplied by the latter 17th century, thanks to the expense of high-quality Canadian beaver, which in turn led to the use of poor quality fur or outright substitutions. Changing tastes in men’s fashion, including smaller hats, also meant less demand for beaver, and a general falling demand for beaver hats and collars by the 1680s, with the London price of beaver pelts dropping from 14s./lb in 1664 to 5s./lb in 1700, a decline of over 64% (Matson, 1994, p. 395; Innis, 1970, pp. 74–75). As a result, and perhaps in part due to the nature of their initial praxis of core ascent (which emphasized piracy and military force as much as

Commodity Frontier as Contested Periphery

237

trade; Bunker & Ciccantell, 1999, p. 112, 2003, p. 18), English/British authorities eventually emphasized the fur trade’s geopolitical aspect, they used it to block French expansion in the Great Lakes basin (much as the French were doing vis a` vis the English), rather than emphasizing profitability, as the mostly ethnic Dutch traders of Albany (including about 50 major and 80 minor traders in 1664; Bielinski, 1991, p. 122) continued to do, along with their fellow colonial traders in Montreal. These mercantile interests continued a highly profitable (albeit illegal) trade during the decades of competition and occasional warfare between the core powers. Such was the basic policy split in New York, as well as Canada and Iroquoia (Jennings, 1984, p. 173): north–south trade between Albany (which had generally better and cheaper trade goods) and Montreal (which had generally better and cheaper furs) or westward expansion, as much for reasons of imperial prerogative as fur supply access in and of itself.

3. FRENCH AND IROQUOIS EXPANSION, ENGLISH RESTRUCTURING, 1664–1689 While New York’s new English rulers let the Iroquois-(ethnic) Dutch fur trade alone initially (Bufﬁnton, 1922, pp. 334–335; Trelease, 1997, pp. 215, 253), the previous year (1663), the French Crown revoked the failing Canadian Company of One Hundred Associates’ charter and took control of New France for strategic purposes (Eccles, 1972, p. 56, 1983b, p. 51; Richter, 1992, p. 102; Diamond, 1961, pp. 4–5), similar to what the States General had threatened to do to the West India Company and New Netherland in the 1630s, thanks to the perpetually low Dutch/European population in that colony (Rink, 1986, p. 134; Kammen, 1975, p. 38). Because 20–25% (and sometimes over 30%) of Canada’s adult male labor force was consistently engaged in the fur trade during the 17th century (Diamond, 1961, p. 30), Colbert instead wanted the small French population (and its capital and labor) concentrated around Montreal, which he ‘‘expectedyto grow by natural increase, once a nucleus of colonists had been established’’ (Moogk, 1989, p. 468). Instead of westward fur-trading journeys, Amerindians would be encouraged to come down the St. Lawrence to trade furs, while les habitants were to focus on agriculture and develop industries (e.g. lumbering, shipbuilding, ﬁshing) and trade with the West Indies colonies and France (Eccles, 1983b, p. 51). As part of his general economic plan to best the Dutch and English (Wallerstein, 1980, pp. 116–118), Colbert ‘‘wanted New France

238

JONATHAN LEITNER

to play the same mercantilist role in the French empire’’ as England’s North American colonies did for their metropole (Eccles, 1983b, p. 51). However, France’s colonies were actually doing this for the United Provinces, thanks to the still-dominant Dutch merchant ﬂeet (Eccles, 1972, p. 60). Colbert’s policies actually promoted the fur trade, because he kept fur marketing in France as a monopoly under the Compagnie de l’Occident, which paid ﬁxed prices for pelts, regardless of source. However, trade with indigenous peoples, ‘‘the more lucrative of the two operations’’ (Choquette, 2002, p. 196), was not monopolized; Colbert’s marketing monopoly would therefore prove highly profitable for independent Canadian traders who were also freed of having to market the fur (even after la Compagnie failed in 1674 and its functions taken over by a tax-farming syndicate). ‘‘A tremendous expansion in the fur trade’’ resulted, with consequent reduction in Colbert’s desired labor supply (Eccles, 1972, pp. 84–85). The trade’s expansion and growing importance to New France soon led to an increased French military presence in the west and active French aggression toward the Iroquois League. French colonial authorities also wanted to remove Anglo-Dutch involvement in the trade, because Amerindian trappers/hunters could potentially get more, and putatively better quality, European manufactured goods for a given quantity of their furs at Albany or other English trading centers than at any French trading center (Jennings, 1984, p. 79; Bufﬁnton, 1922, pp. 336–337; Innis, 1970, pp. 52–53, 84–85; Trelease, 1997, pp. 217, 246; Norton, 1974, pp. 90–91; cf. Eccles, 1972, pp. 100–101, n. 12, 1979, pp. 425, 430–432, 1983a, p. 349). English trading advantages were due in part to factors contributing to the overall trend toward English hegemony: more advanced manufacturing industry (or at least lower manufacturing costs) and domination of the sea lanes (ergo lower transport costs). More specifically, the English had West Indies rum while the French had a colonial policy of high import tariffs on trade goods; the St. Lawrence River tended to remain frozen longer than the Hudson; and the journey from France to Quebec was longer than between Britain and its North American colonies (Innis, 1970, pp. 52–53; Trelease, 1997, pp. 217–218; Wolf, 1997, p. 169; Moogk, 1989, p. 468). As Wallerstein (1980, pp. 102–104) explains, England’s North American colonies helped provide it with the increased markets that in turn helped its capitalists to outcompete their French counterparts, who with their larger domestic market arguably had less need to promote settler colonization (also see Moogk, 1989), an additional reason for New France’s chronically low European population. However, England’s main foe at this point was still the Netherlands (which in fact reconquered New York in 1673, only to relinquish it via treaty

Commodity Frontier as Contested Periphery

239

in 1674), so ‘‘Anglo-French rivalry along the New York frontier developed only gradually,’’ continuing well into the 1670s ‘‘an essentially commercial rivalry with Canada for the good will of the Indians and the proceeds of their hunting’’ (Trelease, 1997, p. 241). Less than two years after treating with the English in 1664, the Iroquois League also did so with the French and their indigenous allies, the Wyandots (a refugee nation of non-league Iroquoian-speakers, primarily Hurons, Eries, Petuns, and Neutrals, all defeated by the Iroquois League in the 1640s and 1650s; Richter, 1992, p. 62) and northern Algonkians (especially the Ottawas), followed by a further Iroquois–French peace treaty in 1667 (Richter, 1992, p. 102). The English were of course opposed; New York Governor Richard Nicolls saw that ‘‘a primary French objective was ‘ingrossing the Beaver trade by destroying and interrupting ours at Albany’’’ (in Trelease, 1997, p. 245). But peace ca. 1670 was more immediately problematic to the French, allowing the Iroquois to freely hunt north of Lake Ontario, and even settle on the north shore (Konrad, 1981). Indeed, the territory north of the Hurons’ former home in southern Ontario became the Iroquois’ primary hunting ground in the early 1670s (Richter, 1983, p. 542), in particular around Lake Simcoe, Georgian Bay, and points north and west (Konrad, 1981, p. 133). This was a ‘‘direct challenge to French interests,’’ with Iroquois hunters taking fur from lands French colonial ofﬁcialdom regarded as its own (or at least its indigenous allies’) to the Anglo-Dutch traders at Albany (Richter, 1992, pp. 129–130). The Iroquois also began bringing pelts captured from the Ottawas to Albany for its higher prices (Trelease, 1997, p. 246; Bufﬁnton, 1922, p. 339). French colonial authorities in turn threatened force against the Iroquois ‘‘to safeguard the ﬂow of Great Lakes furs to the St. Lawrence,’’ but a treaty resulted in 1672 (Richter, 1992, p. 130), thanks in part to the temporary Anglo-French alliance during the third Anglo-Dutch War (Jennings, 1984, p. 134). Yet the price differential also impelled independent French fur traders, the coureurs de bois, to illegally bring furs to Albany, regardless of ‘‘considerations of national policy’’ (Trelease, 1997, pp. 246–247; also Bufﬁnton, 1922, pp. 340–341). Peace was but temporary: during the third Anglo-Dutch war the Netherlands dropped behind England and France in their three-way hegemonic competition; the latter two ‘‘now turned primarily on each other’’ (Wallerstein, 1980, p. 80). To that end, and more immediately to protect the fur supplies ‘‘on which the life of Canada depended’’ (Bufﬁnton, 1922, p. 342), Canada’s Governor General Frontenac ordered forts built at key western transit and trade points (contra Colbert’s centralizing preference; Eccles, 1972, pp. 85–86), particularly along the Great Lakes (Trelease, 1997,

240

JONATHAN LEITNER

p. 247; Richter, 1992, pp. 130–131). Ultimately, the French built a series of forts from the Great Lakes to the Illinois and Mississippi rivers to draw trade from those watersheds, and also to ship furs directly to Europe via the Mississippi and Gulf of Mexico; for the Iroquois, these represented a French ‘‘noose around vital beaver-hunting territories and trade routes while also menacing [their] northern and western ﬂanks’’ (Richter, 1992, p. 139; also Jennings, 1984, pp. 173–175; Branda˜o, 1997, p. 120). Yet while a French military presence in and around Iroquoia prevented the League nations from ‘‘mediat[ing] trade’’ between the western nations and Albany, it also allowed easier and safer access to Montreal, which had become a more attractive trading post to many among the League, thanks to the inﬂuence of Jesuit missionaries in Iroquoia (Richter, 1992, p. 131). In comparison to their Canadian counterparts, for whom fur supply was vital, New York colonial authorities were trying to meet the demands of the empire by organizing the colony’s economy as a regional core-periphery structure; or rather, as had the Dutch, one of large entrepoˆt–small entrepoˆt (Leitner, 2003). This had much to do with the increasing importance of New York food exports (especially grain) to England’s Caribbean plantation colonies, particularly for Manhattan merchants (Burrows & Wallace, 1999, p. 84; Matson, 1994, pp. 399–400), and soon for the previously fur trade-dependent Albany merchants as well (Bielinski, 1992, p. 266). In the 1670s and 1680s, New York Governors gave Manhattan merchants trade preferences that made New York City the colony’s sole port of entry for foreign trade (Kammen, 1975, p. 106; Burrows & Wallace, 1999, p. 84; Matson, 1994, p. 403; Norton, 1974, p. 84). These measures reduced Albany’s stature in New York’s economy (or arguably were reﬂective of Albany’s reduced stature with the decline of the fur trade), with the result that ‘‘[a]fter 1680 [Manhattan’s] merchants enjoyed a virtual stranglehold on the trade of the entire colony’’ (Burrows & Wallace, 1999, p. 85). Albany merchants received a gubernatorial quid pro quo in 1678, with a legal monopoly over the fur trade (Norton, 1974, p. 46), just as the Iroquois were ‘‘refocus[ing] westward their quest for furs and captives’’ (Richter, 1983, p. 544), having recently made peace with the Mahicans of the central Hudson Valley and defeated the Susquehannocks from what is now Pennsylvania. Peace on their eastern and southern boundaries in turn gave Iroquois traders safer access1 to their Euroamerican counterparts in Albany, who at this point were reportedly paying over one-third more for furs than traders in Canada (Richter, 1992, p. 137). However, the Albany merchants were soon politically overshadowed by gubernatorial prerogative in Manhattan (along with being economically overshadowed by Manhattan

Commodity Frontier as Contested Periphery

241

merchants). While the Albanians (still largely ethnic Dutch) had come to prefer peace between all parties in the interest of trade (Bufﬁnton, 1922, p. 343; Trelease, 1997, p. 260), a succession of English/British colonial governors sought to use the Iroquois League as a tool of empire against the French and their indigenous allies (Trelease, 1997, p. 214). This policy harmonized with Iroquois goals, which continued to stress access to hunting territories (as well as captives for population replacement), leading to warfare in the late 1670s and 1680s with the nations around the Great Lakes. However, with Iroquois warfare then predominant over hunting, Albany’s fur trafﬁc actually decreased, and also contributed to further Iroquois population losses during 1679–1685 (Richter, 1992, pp. 144, 148). In exasperation of continued Iroquois western hunting and trade, the French invaded western Iroquoia’s Seneca country twice in the 1680s; as an Iroquois spokesmen put it to the Albanians: ‘‘The French will have all the Bevers, and are angry with us for bringing any to you’’ (in Trelease, 1962, p. 43; in Richter, 1992, p. 149). In response, New York Governor Dongan launched westward military-led trading expeditions that largely failed (Richter, 1983, p. 545, 1992, pp. 150–152; Bufﬁnton, 1922, pp. 344–345; Norton, 1974, pp. 152–154; Jennings, 1984, pp. 187–189). Though he had hoped to ‘‘divert the major part of Canada’s western trade to Albany,’’ the French ‘‘had too much at stake in the west to permit this dangerous experiment to succeed’’ (Trelease, 1997, pp. 268–269, 271). Per Trelease (1997, p. 293), Dongan perhaps would have been better off with ‘‘letting the trade take care of itself,’’ i.e. waiting for the fur to come to Albany, which ‘‘could probably have encouraged a greater ﬂow of peltry than by enmeshing it in power politics.’’ The Iroquois League’s response was more direct: after the French destroyed the Senecas’ six villages and 400,000 bushels of mostly standing corn in 1687, Iroquois League warriors ‘‘relentlessly assaulted’’ New France throughout the rest of 1687 and into 1688, allegedly cutting off all fur shipments to Canada from the French post at Michilimackinac (the northern tip of lower Michigan) for two years (Jennings, 1984, pp. 189–190; Richter, 1983, p. 546, 1992, pp. 156–159; Trelease, 1997, p. 283). Seeking advantage despite military weakness, New York’s Dongan prompted the Iroquois to make peace with Amerindian nations in the French-claimed Ohio Valley, thereby including them in the ‘‘Covenant Chain’’ (the alliance system of the Iroquois League, those they conquered, and the English Crown; Jennings, 1984). Based on English acceptance of Iroquois ‘‘submission’’ to the English Crown as a legal ﬁction to claim sovereignty over Iroquois territory (Richter, 1992, p. 151), this strategem was meant to therefore bring the Ohio Valley nations under de jure English

242

JONATHAN LEITNER

suzerainty and divert their furs to Albany. Though League leaders generally agreed with the plan, they did not see themselves as English subjects per se´, especially as the English needed them as much or more than they needed the English; as a Mohawk chief pointed out to Dongan, ‘‘the French would faine kill us all and when that is done, they would carry all the Bever trade to Canida, and the great King of England would loose the land likewise’’ (in Jennings, 1984, pp. 191–192).

4. CORE CONFLICT, IROQUOIS DEFEAT, AND BEAVER SHORTAGE/GLUT, 1689–1701 Soon enough, direct interstate conﬂict (War of the League of Augsburg, a.k.a. King William’s War in North America) broke out again in the 1690s, following the Glorious Revolution, that fairly desperate gambit by the States General to turn England against the militarily powerful French, with whom the United Provinces were engaged in a guerre de commerce about to turn violent (Israel, 1991). Emboldened by changed political circumstances that promised greater help from their English allies, Iroquois League assaults in Summer 1689 forced the ‘‘temporary abandonment’’ of several French Great Lakes posts (Richter, 1992, pp. 159–160; Jennings, 1984, pp. 195–196; Trelease, 1997, pp. 297–298; Eccles, 1972, p. 96). But the English colonial leadership’s military and logistical incompetence meant the Iroquois ‘‘bore the brunt of the ﬁghting on the Canadian frontier’’ for most of King William’s War (Eccles, 1972, p. 97; also Richter, 1992, p. 166; Trelease, 1997, pp. 303–306). French colonial authorities soon induced the Ottawas and Wyandots to attack the Iroquois League, which suffered major losses as a result (due also to the Canadian militia’s adoption of Amerindian combat tactics; Branda˜o, 1997, p. 126); ﬁghting continued even after the 1697 Treaty of Ryswick ended the Europeans’ war (Richter, 1992, pp. 185–188, 195; Jennings, 1984, pp. 206–209). By 1701, their warrior manpower cut in half, overall population reduced by at least a quarter,2 and following massive physical destruction of their settlements and crops (Richter, 1992, p. 188), the Iroquois accepted French terms forcing them into neutrality. The French intended to use the League nations ‘‘as a means of channeling trade to Canada’’ and to obviate English westward expansion by Iroquois proxy (Jennings, 1984, pp. 210–211). As well, French authorities wanted the Iroquois as a barrier between their western allies and Albany’s cheaper trade goods (Eccles, 1972,

Commodity Frontier as Contested Periphery

243

p. 100, 1983a, p. 343; Jennings, 1984, pp. 210–211). At the very least, the French wanted Iroquois neutrality in any future Anglo-French conﬂict (Richter, 1983, p. 553; Eccles, 1983a, p. 344), a policy that already had many Iroquois proponents, and ultimately became their main political philosophy for dealing with the Europeans (Richter, 1992, pp. 236–254; and below). Given their weakened condition, the Iroquois feared ‘‘they could no longer hold their [hunting] lands by force’’ (Branda˜o, 1997, p. 127; also Trelease, 1962, pp. 47–48; Richter, 1992, p. 205). Concurrent to their French treaty, they deeded their 17th century western conquests (comprising what is now southern Ontario between Lakes Ontario, Erie, and Huron, and much of lower Michigan; Trelease, 1962, p. 47) to the English, thereby formalizing the legal ﬁction that the English had the right to those lands via the Iroquois’ status as English ‘‘subjects.’’ Of course, the deed overlooked that these territories had just been lost to western indigenous nations and the French; rather, its main intention was ‘‘a challenge for the English to ﬁght on behalf of the Iroquois for a change, instead of the Iroquois ﬁghting for the English’’ (Jennings, 1984, p. 212; also Richter, 1992, p. 212; Branda˜o & Starna, 1996, p. 228). Per Iroquois intentions, the English in turn used the deed to justify their western territorial claims against the French (Jennings, 1984, p. 212). At least one interpretation therefore sees both 1701 treaties as driven by the Iroquois’ goals and diplomatic initiatives, and ultimately very much in their favor since it helped secure some access to hunting territories and ended a losing conﬂict (Branda˜o & Starna, 1996; also Kammen, 1975, p. 197). As the Iroquois were their main fur source, Albany interests still in the trade were compelled to adopt a de facto neutrality (Bufﬁnton, 1922, p. 350), though this was perhaps as much about maintaining their military security (Norton, 1974, pp. 8, 176). Albany’s economy had been severly hurt by King William’s War, with a disrupted fur trade and ever-present threat of attack prompting a 25–30% population decline, mostly by out-migration, from 2,142 in 1686 to 2,016 in 1689 and then 1,449 in 1698 (Kammen, 1975, p. 145; Trelease, 1997, pp. 204, 323; Jennings, 1984, p. 206, n. 53). By 1693, the war effectively halted Albany’s fur trade, which had also suffered during the 1680s Franco-Iroquois conﬂict; according to a colonial ofﬁcial in July 1689, there had been ‘‘little or no trade’’ at Albany for three years, ‘‘which occasions great poverty, by most of the inhabitants’’ (in O’Callaghan, 1853, p. 599, also see pp. 693, 705; & Bufﬁnton, 1922, p. 348). That same year, New York reportedly exported only 9,000 beaver pelts, compared to 35–40,000 previously (Matson, 1994, p. 395; also Trelease, 1997, p. 216). Even after the Europeans’ war ended, the continuing Iroquois-Ottawa/

244

JONATHAN LEITNER

Wyandot conﬂict kept furs from Albany and ‘‘combined with postwar shortages of goodsycreated serious economic hardship’’ for Albanians (Richter, 1992, p. 188). The converse of New York’s beaver pelt shortage was a Canadian beaver pelt glut; furs not going to Albany via Iroquois traders or from Iroquois raids on Ottawa or Wyandot fur shipments were instead getting to Montreal in great quantities during the 1690s, at ‘‘three times the annual volume it had before 1675’’ (Jennings, 1984, p. 210). Combined with the furs coming from French posts in the Illinois country, there had already been a tendency toward oversupply (Eccles, 1972, pp. 86–87). Beaver pelt supply was four times annual French market demand by 1695, with the Crown’s pelt purchasing contractor (paying ﬁxed prices for furs) stuck ‘‘with a surplus of 3.5 million livres’ weight, with more ﬂooding in’’ (Eccles, 1972, p. 98). Though ofﬁcial prices were cut for traders’ beaver pelts, the western posts remained open, thanks to French ofﬁcials in Canada pointing out their strategic importance to the Minister of Marine: they helped maintain alliances with western Amerindians, whose peltry and manpower were key to New France’s survival. And despite a large oversupply of fur – by 1698, still about one million excess livres’ weight in French warehouses, ‘‘enough to manufacture half a million hats’’ (Eccles, 1972, p. 99) – as Eccles (1972, p. 99) puts it, though ‘‘Economics dictated the abandonment of the west; politics dictated its retention’’ (also see Eccles, 1983a, pp. 341–342, 1983b, p. 55).

5. FRENCH ‘‘FUR TRADE IMPERIALISM,’’ ENGLISH FUR MERCANTILISM, IROQUOIS NEUTRALISM, 1701–1730S The French then extended this strategy. Post-1701, with Iroquois neutrality secured, Louis XIV created the colony of Louisiana and ordered a new set of western settlements in and around the Great Lakes, most notably Detroit, to further contain the English colonies; for the French, ‘‘[t]he fur trade was now definitely subordinated to a political end,’’ and was also put under military control (Eccles, 1983a, pp. 345–346). Colonial ofﬁcials in Quebec City were chagrined, as were Canadian merchants, who feared Detroit could allow the now-neutral Iroquois (and by extension Albany traders) a way to trade with western Amerindians (Eccles, 1983a, pp. 344–345; also Branda˜o & Starna, 1996, p. 232). Indeed, the Iroquois were soon allowing

Commodity Frontier as Contested Periphery

245

‘‘substantial numbers of Mississaugas, Ottawas, Wyandots, and Miamis’’ to trade at Albany after 1703. By 1708, ‘‘a mere sixty western Indians bartered at Montreal’’ (Richter, 1992, pp. 223–224), presumably due as well to Queen Anne’s War (a.k.a. War of the Spanish Succession). Conversely, New York authorities passed laws in 1714 and 1717 to encourage western nations to trade at Albany (Bufﬁnton, 1922, p. 357), which resulted in another dramatic increase in fur volume from the westerners, up from only 30 canoeloads of fur brought to Albany in 1716–1720, to 323 canoeloads in 1720–1724 (Richter, 1992, p. 249). New York’s assembly and new governor then banned trade with Canada in 1720, specifically that of Albany–Montreal, engaged in since at least the 1680s and especially post1701 (albeit with occasional interruptions; Norton, 1974, p. 128). Albany–Montreal trade was driven by a compelling economic–geographic logic, ‘‘in deﬁance of imperial policies and conﬂicts’’: with the Iroquois driven out of their western hunting lands in Michigan and Ontario, the French had access to furs that would otherwise have mostly gone to Albany, while the Anglo-Dutch Albanians had access to generally cheaper trade goods, thanks to generally better English manufacturing and year-round Atlantic access via the mostly ice-free Hudson (Jennings, 1984, p. 284; Norton, 1974, p. 122). Ergo, ‘‘Montreal’s merchants swapped to Albany’s the furs that came cheaply to Montreal for the trade goods that came cheaply to Albany, and Indian porters carried the cargoes both ways,’’ aided by the nearly all-water Hudson–Champlain linkage (Jennings, 1984, p. 82), further prompted by fur’s low bulk and high value, making it easy and profitable to smuggle. Indeed, these same basic conditions also led certain ‘‘renegade’’ French coureurs de bois to bring fur directly from the west to both Albany and Philadelphia (Jennings, 1984, p. 82), as they had in the 1670s. By 1720, when the ban was enacted, the Albany–Montreal trade was valued at £10–12,000 per year (Bufﬁnton, 1922, p. 356; Jennings, 1984, p. 284). And though it ‘‘was easy, profitable, without risk’’ for the Albany traders, since their Montreal counterparts had to ﬁrst procure the fur from western Amerindian traders, British colonial ofﬁcials recognized that the Albany–Montreal trade allowed the French to ‘‘maintain their inﬂuence’’ with the western nations, whereas the British would gain inﬂuence if western Amerindians traded at Albany (Bufﬁnton, 1922, p. 358; also Eccles, 1983a, p. 352). Conversely, French colonial ofﬁcials understood their advantage in allowing the Albany–Montreal trade to continue: though they claimed as well to want it ended, ‘‘their unenthusiastic efforts were less than efﬁcacious’’ because they ‘‘grasp[ed] how closely intertwined were the economics and politics of the situation’’ (Eccles, 1983a, pp. 351–352; also Choquette, 2002, p. 198).

246

JONATHAN LEITNER

New York’s governor, on the other hand, was prompted by the antiFrench/pro-war ‘‘landed interests’’ in the colonial assembly (as opposed to the generally anti-war ‘‘mercantile interests’’ who either traded with the French Canadians or for whom war was otherwise bad for business). Despite the 1720 ban (and follow-up acts in 1722 and 1724; Norton, 1974, pp. 87, 131, 137ff; Jennings, 1984, pp. 284, 298–299), Albany–Montreal trade continued at apparently the same rate, albeit ‘‘less pleasant and more expensive’’ for those involved (Norton, 1974, p. 141). Following New York Governor William Burnet’s transfer in 1728, these colonial trading interests then prompted Privy Council vetoes of the ban, via lobbying of the Board of Trade in London by their corresponding English merchants (Norton, 1974, p. 142; Bufﬁnton, 1922, p. 362). As Bufﬁnton (1922, p. 362) observes, the ruling ‘‘was quite in keeping with the general trend of British policy in the age of Walpole that economic considerations triumphed over political.’’ A legal Albany–Montreal trade continued until 1744 (when the next war broke out; Kammen, 1975, p. 19), but with an equal tax on trade goods shipped either to Montreal or the new British post at Oswego, which was funded by the proceeds (Bufﬁnton, 1922, p. 362). Somewhat ironically, both Oswego and Niagara became ‘‘parallel retail branches of the same wholesale system,’’ since furs acquired by the French at Niagara ‘‘were likely to end up at Albany,’’ traded for British manufactured goods that Montreal merchants then shipped to Niagara to be traded to the western natives for more fur (Richter, 1992, p. 269). Established in 1725 near Lake Ontario’s east end, Fort Oswego swamped Montreal as an Albany fur source; that ﬁrst season, nearly 1,000 bundles came to Albany directly from the west (about 80% of which was directly from Oswego), versus only 176 from Canada (Bufﬁnton, 1922, p. 358; Norton, 1974, p. 185). But Oswego ‘‘infuriated’’ the French, who saw a threat to both their fur supply and political relations with the westerners, and thereby their geopolitical hold over interior North America (Kammen, 1975, pp. 194, 196–197; also Innis, 1970, pp. 86–87; Bufﬁnton, 1922, p. 363). A military solution forestalled by the Fleury–Walpole entente cordiale (Eccles, 1983a, p. 354), French colonial authorities instead blocked the Albany–Montreal trade (Norton, 1974, pp. 148–149), as well as kept the crown monopoly on fur trading at its three neighboring forts around Lake Ontario ‘‘so that prices could be kept competitive with those at Oswego by selling at a reduced profit or even a loss if necessary’’ (Eccles, 1983a, p. 355). These policies cut New York’s fur shipments to England from £6,952 in ‘‘the boom year’’ of 1725 to £2,610 in 1730, about where it stayed until the mid1730s (Norton, 1974, pp. 149, 164, 166–168), and this at a time when the

Commodity Frontier as Contested Periphery

247

average price of a beaver pelt in Britain was increasing, from 5.88s in 1725 to 9.56s in 1729 and 8.71s in 1730 (Carlos & Lewis, 1999, p. 712), which implies a drop in pelt exports from over 23,600 in 1725 to just under 4,000 in 1730.

6. FROM CONTESTED PERIPHERY TOWARD PERIPHERAL PEOPLE BETWEEN HEGEMONIC RIVALS, 1730S–1754 Oswego also challenged the Iroquois, who resented the territorial incursion and the competition from Euroamerican traders for the westerners’ fur (Richter, 1992, p. 250; Jennings, 1984, p. 353). Following the 1701 treaties making them ofﬁcially neutral vis a` vis the French and British, ‘‘neutralism’’ had gradually become the Iroquois’ predominant (if not unanimous) European policy by the late 1710s (Richter, 1992, pp. 234–235). Neutralism entailed avoiding political and cultural dependence on either European power (albeit not economic dependence, a growing and generally acknowledged reality); and in terms of diplomacy, playing both British and French off one another, and thereby perchance gain some space between them (Richter, 1992, pp. 180, 206–207, 235). Unfortunately, neutralism did not quite work as intended, because British–French competition also undercut Iroquois autonomy. In particular, the French fort at Niagara on Seneca land was allowed by the Onondagas (much to the Senecas’ chagrin) by way of balancing the British fort at Oswego on Onodaga land (Richter, 1992, p. 251). Though the Senecas convinced the Onondagas ‘‘to withdraw their permission, the damage was done because the French fort was by that time established and the French refused to tear it down’’ (Branda˜o, 1997, p. 25; also Jennings, 1984, p. 300). The French then greatly fortiﬁed their Niagara post, apparently uniting the Iroquois League in alarm, to the extent they seriously discussed making war on both the French and British in 1726 with their nominal tributary nations, all of whom rejected the idea (Jennings, 1984, pp. 300–301). It was likely a wise rejection under the circumstances, since the balance of power had by then moved to the Europeans; as Richter (1992, pp. 255–256) sums up, by ca. 1730 ‘‘the era of European colonization was over; the Six Nations [the ﬁve having been joined by the Tuscaroras in the 1720s] now ranked among the colonized [his italics].’’ This included certain Iroquois individuals’ entry into the cash economy (e.g. with Onondagas and Senecas

248

JONATHAN LEITNER

working as wage laborers at Forts Oswego and Niagara, with the posts serving as ‘‘mechanisms of incorporation’’ as they would elsewhere on the continent; Carlson, 2002, p. 430), which became more common for the Iroquois overall by the 1740s (Richter, 1992, p. 262). As well, there was a general dependence on European-manufactured tools, textiles, and especially weapons: ‘‘apart from food and shelter, virtually every aspect of Iroquois material life seems to have been dependent upon the trade Oswego and Niagara represented’’ by 1740 (Richter, 1992, pp. 264–268). By the mid-1730s, other Iroquois had been selling off land for cash, though traditional utilitarianism was still prevalent enough to allow ‘‘redistribution and a multiheaded political system [that] narrowed the gap between haves and have-nots’’ (Richter, 1992, pp. 262–263). Formal political incorporation would wait until later in the century as well; even as late as the mid-1750s, while the Mohawk Valley between Albany and Oswego was apparently swimming in British-made consumer goods, it was not yet controlled by British soldiers or settler-colonists (Shannon, 1996, p. 32). However, though still a relatively autonomous, unincorporated people on a contested contact periphery, the Iroquois had at very least been ‘‘groomed’’ (Carlson, 2002, p. 432) by their involvement in the fur trade, and were well on their way to being incorporated into the world-economy, ultimately as a dependent, peripheral social formation, which would face formal political incorporation into the newly created U.S.A. later in the 18th century, albeit with some vague and perpetually shifting degree of sovereignty (Venables, 1980). On the Euroamerican side, Oswego pulled the Albany traders westward (at least those who could afford it), and also served to eliminate those few involved Manhattan-based traders from the business, due to the increased transport and labor expenses so entailed. These two factors also decreased control by either Albany- or Manhattan-based parties over the actual transactions with their native counterparts (Norton, 1974, p. 94), and ‘‘tended to lessen the importance of Albany,’’ which was further lessened by increased Euroamerican settlement and trade with Amerindians in the Mohawk Valley during the 1730s and early 40s (Bufﬁnton, 1922, pp. 363–364; Norton, 1974, pp. 59, 99; Jennings, 1984, pp. 353–354; Shannon, 1996). By 1750 only 10 Albany merchants were explicitly identiﬁable as ‘‘Indian traders’’ and were ‘‘a highly specialized minority within the Albany business community’’ at that (Bielinski, 1991, p. 122). A related result was land speculation by certain larger Albany merchants, as well as general encroachment on Iroquois land, initially the Mohawks’,

Commodity Frontier as Contested Periphery

249

who had to deal with fraudulent or at least overly aggressive claims by their Euroamerican neighbors (Bufﬁnton, 1922, p. 364; Richter, 1992, pp. 271–272; Nammack, 1969, pp. 22–38). Pressure on the Mohawks was somewhat relieved by the 1754 Albany Congress (Norton, 1974, p. 196; Shannon, 1996, p. 13), which also established colonial military coordination as a preface to the renewed violent turn in the ongoing British–French hegemonic contest. The subsequent Seven Years’ War marked Britain’s ‘‘deﬁnitive achievement of superiority’’ over France (Wallerstein, 1980, pp. 257–258), whose Crown (if not colonists) was mostly ejected from eastern North America. As members of the emergent hegemon, with its peculiarly ‘‘commercial and warlike’’ character (Seeley in Wallerstein, 1980, p. 256, n. 62), the Albany merchants shed their pro-trade/anti-war reputation for a share of the ‘‘lucrative business’’ of military provisioning, which more than offset the temporary end of an already reduced fur trade. This trade would be reduced even more, at least in terms of its contribution to the imperial economy once Canada was formally incorporated into Britian’s empire (Norton, 1974, pp. 197, 199), obviating any advantages Albany merchants had in trading with Montreal. Ironically, Albany traders lost their competition with Montreal’s in the context of a victory for their metropole, the new hegemon, which put both groups in the same imperial sphere, with Montreal merchants able to take advantage of better access to the commodity frontier for fur even (and especially) after U.S. independence (see Innis, 1970, pp. 149–280; more generally Choquette, 2002, pp. 202). Perhaps ironically as well, the Iroquois’ ultimate peripheralization (or perhaps worse) brought them full circle; having been peripheral to pre-contact North America’s world-system, or more precisely ‘‘at the outer margins of an arena of exchange centered in Cahokia’’ (Salisbury, 1996, p. 453, citing Dincauze and Hasenstab), European contact allowed the Iroquois League to have an ‘‘ambiguous’’ sort of hegemony (per Jennings, 1984), but very much in the context of a contested contact periphery and as providers of a raw material to a fairly limited set of core interests, making it a rather temporary proposition. The dynamic would be played out among certain groups of indigenous people and Euroamericans elsewhere on the continent over the ensuing 120 years or so (see Wolf, 1997, pp. 170–194).

NOTES 1. Also thanks to increased regulation of the trade at Albany, to prevent Amerindian traders from being cheated or even assaulted by Euroamerican traders (Richter, 1992, p. 137; Trelease, 1997, pp. 221–222; Norton, 1974, pp. 60–62).

250

JONATHAN LEITNER

2. Warrior numbers declined from an estimated 2,570 in 1689 to an estimated 1,230 in 1700 (Jennings, 1984, p. 206; Trelease, 1997, p. 323; Branda˜o, 1997, p. 126). Overall losses were perhaps 2,000 persons from an 8,600–9,000 total population (Richter, 1992, pp. 188, 355–356, n. 60; Mancall, Rosenbloom, & Weiss, 2002, p. 35). These losses followed earlier declines in the mid-17th century, with the ﬁve League nations’ collective population estimated at 10,000–12,000 in 1640, and perhaps 22,000 in 1630, with pre-contact totals perhaps much higher (Richter, 1983, p. 543; Mancall et al., 2002, p. 35; cf. Branda˜o, 1997, pp. 158ff).

REFERENCES Bielinski, S. (1991). How a city worked: Occupations in colonial Albany. In: N. A. M. Zeller (Ed.), A beautiful and fruitful place: Selected rensselaerswijck seminar papers (pp. 119–136). Albany, NY: New Netherland Publishing. Bielinski, S. (1992). A middling sort: Artisans and tradesmen in colonial Albany. New York History, 73(3), 261–290. Branda˜o, J. A. (1997). Your Fyre Shall Burn No More: Iroquois policy toward New France and its native allies to 1701. Lincoln: University of Nebraska Press. Branda˜o, J. A., & Starna, W. A. (1996). The treaties of 1701: A triumph of Iroquois diplomacy. Ethnohistory, 43(2), 210–244. Bufﬁnton, A. H. (1922). The policy of Albany and English westward expansion. Mississippi Valley Historical Review, 8(4), 327–366. Bunker, S. G., & Ciccantell, P. S. (1999). Economic ascent and the global environment: Worldsystems theory and the new historical materialism. In: W. L. Goldfrank, D. Goodman & A. Szasz (Eds), Ecology and the world-system (pp. 107–122). Westport, CT: Greenwood. Bunker, S. G., & Ciccantell, P. S. (2003). Generative sectors and the new historical materialism: Economic ascent and the cumulatively sequential restructuring of the world economy. Studies in Comparative International Development, 37(4), 3–30. Burrows, E. G., & Wallace, M. (1999). Gotham: A history of New York city to 1898. New York: Oxford University Press. Carlos, A. M., & Lewis, F. D. (1999). Property rights, competition, and depletion in the eighteenth-century Canadian fur trade: The role of the European market. Canadian Journal of Economics, 32(3), 705–728. Carlson, J. D. (2002). The ‘Otter-Man’ empires: The Pacific fur trade, incorporation and the zone of ignorance. Journal of World-Systems Research, 8(3), 390–442. Chase-Dunn, C., & Hall, T. D. (1993). Comparing world-systems: Concepts and working hypotheses. Social Forces, 71(4), 851–886. Choquette, L. (2002). Center and periphery in French North America. In: C. Daniels & M. V. Kennedy (Eds), Negotiated empires: Centers and peripheries in the Americas, 1500–1820 (pp. 193–206). New York and London: Routledge. Cronon, W. (1983). Changes in the land: Indians, colonists, and the ecology of New England. New York: Hill and Wang. Diamond, S. (1961). An experiment in ‘Feudalism’: French Canada in the seventeenth century. William and Mary Quarterly (3rd ser.), 18(1), 3–34.

Commodity Frontier as Contested Periphery

251

Dunaway, W. A. (1994). The southern fur trade and the incorporation of Southern Appalachia into the world-economy, 1690–1763. Review: Fernand Braudel Center, 17(2), 215–242. Dunaway, W. A. (1996). Incorporation as an interactive process: Cherokee resistance to expansion of the capitalist world-system, 1560–1763. Sociological Inquiry, 66(4), 455–470. Eccles, W. J. (1972). France in America. New York: Harper & Row. Eccles, W. J. (1979). A belated review of Harold Adams Innis, The fur trade in Canada. Canadian Historical Review, 60(4), 419–441. Eccles, W. J. (1983a). The fur trade and eighteenth-century imperialism. William and Mary Quarterly (3rd ser.), 40(3), 341–362. Eccles, W. J. (1983b). The frontiers of New France. In: G. Wolfskill & S. Palmer (Eds), Essays on frontiers in world history (pp. 42–70). College Station: Texas A&M University Press. Gitlin, J. (1994). Empires of trade, hinterlands of settlement. In: C. A. Milner II, C. A. O’Connor & M. A. Sandweiss (Eds), The Oxford history of the American West (pp. 79–113). New York: Oxford University Press. Hall, T. D. (1989). Social change in the southwest, 1350–1880. Lawrence: University Press of Kansas. Hopkins, T. K., & Wallerstein, I. (1987). Capitalism and the incorporation of new zones into the world-economy. Review: Fernand Braudel Center, 10(5–6), 763–779. Innis, H. A. (1956). Essays in Canadian economic history, M.Q. Innis (Ed.). Toronto: University of Toronto Press. Innis, H. A. (1970 [1956]). The fur trade in Canada: An introduction to Canadian economic history ðrev:ed:Þ. Toronto: University of Toronto Press. Israel, J. I. (1991). The Dutch role in the glorious revolution. In: J. I. Israel (Ed.), The AngloDutch moment: Essays on the glorious revolution and its world impact (pp. 105–162). Cambridge: Cambridge University Press. Jennings, F. (1984). The ambiguous Iroquois empire: The covenant chain confederation of Indian tribes with English colonies from its beginnings to the Lancaster Treaty of 1744. New York: Norton. Kammen, M. (1975). Colonial New York: A history. New York: Charles Scribner. Kardulias, P. N. (2002). Negotiation in a contested periphery: Indians in the fur trade. Paper presented at the annual meeting of the American Sociological Association section on the political economy of the world-system, UC-Riverside, May 2–4. Konrad, V. (1981). An Iroquois frontier: The north shore of Lake Ontario during the late seventeenth century. Journal of Historical Geography, 7(2), 129–144. Leitner, J. A. (2003). Sideshow at the hegemonic carnival: The New Netherland fur trade, 1609–1664. Unpublished manuscript. McMichael, P. (1990). Incorporating comparison within a world-historical perspective: An alternative comparative method. American Sociological Review, 55(3), 385–397. Mancall, P. C., Rosenbloom, J. L., & Weiss, T. (2002). The economic activity of Native Americans in British North America. Paper presented at the 13th world congress in Economic History, Buenos Aires, July. Matson, C. (1994). ‘Damned Scoundrels’ and ‘Libertisme of Trade’: Freedom and regulation in colonial New York’s fur and grain trades. William and Mary Quarterly (3rd ser.), 51(3), 389–418. Moogk, P. N. (1989). Reluctant exiles: Emigrants from France in Canada before 1760. William and Mary Quarterly (3rd ser.), 46(3), 463–505.

252

JONATHAN LEITNER

Moore, J. W. (2000). Sugar and the expansion of the early modern world-economy: Commodity frontiers, ecological transformation, and industrialization. Review: Fernand Braudel Center, 23(3), 409–433. Nammack, G. C. (1969). Fraud, politics, and the dispossession of the Indians: The Iroquois land frontier in the colonial period. Norman: University of Oklahoma Press. Norton, T. E. (1974). The fur trade in colonial New York, 1686–1776. Madison: University of Wisconsin Press. O’Callaghan, E. B. (Ed.) (1853). Documents relative to the colonial history of the State of New York (Vol. 3). Albany, NY: Weed, Parsons. Price, J. M. (1970). Economic activity: I The map of commerce, 1683–1721, In: J. S. Bromley, (Ed.), The New Cambridge modern history: The rise of Great Britain and Russia, 1688–1715/25, (Vol. 6, pp. 834–874). Cambridge: Cambridge University Press. Richter, D. K. (1983). War and culture: The Iroquois experience. William and Mary Quarterly (3rd ser.), 40(4), 528–559. Richter, D. K. (1992). The Ordeal of the Longhouse: The peoples of the Iroquois League in the era of European colonization. Chapel Hill: University of North Carolina Press. Rink, O. A. (1986). Holland on the Hudson: An economic and social history of Dutch New York. Ithaca, NY: Cornell University Press. Rotstein, A. (1977). Innis: The alchemy of fur and wheat. Journal of Canadian Studies, 12(5), 6–31. Salisbury, N. (1996). The Indians’ old world: Native Americans and the coming of Europeans. William and Mary Quarterly (3rd ser.), 53(3), 435–458. Shannon, T. J. (1996). Dressing for success on the Mohawk frontier: Hendrick, William Johnson, and the Indian fashion. William and Mary Quarterly (3rd ser.), 53(1), 13–42. Stevenson, R. D., & Abruzzi, S. (2001). Beavers challenge human efforts to control the landscape. Conservation perspectives: The On-line Journal of MassSCB, Spring, http://www.massscb.org/epublications/spring2001/beaver.html Trelease, A. W. (1962). The Iroquois and the Western fur trade: A problem in interpretation. Mississippi Valley Historical Review, 49(1), 32–51. Trelease, A. W. (1997 [1960]). Indian affairs in colonial New York: The seventeenth century. Lincoln: University of Nebraska Press. Venables, R. W. (1980). Iroquois environments and ‘We the People of the United States’: Gemeinschaft and gesellshaft in the apposition of Iroquois, Federal, and New York State sovereignties. In: C. Vecsey & R. W. Venables (Eds), American Indian environments: Ecological issues in native American history (pp. 81–127). Syracuse, NY: Syracuse University Press. Wallerstein, I. (1980). The modern world-system II: Mercantilism and the consolidation of the European world-economy, 1600–1750. New York: Academic Press. White Oak Society. (2001). The beaver fur hat. http://www.whiteoak.org/learning/furhat.htm Wolf, E. R. (1997 [1982]). Europe and the people without history. Berkeley: University of California Press.

EXTRACTION, GENDER AND NEOLIBERALISM IN THE WESTERN AMAZON Susanna Hecht ABSTRACT This chapter explores the work of Stephen Bunker through a review of the role of women in Amazonian extractive economies and how shifting ideas of development affected them in the Western Amazon. While the initial development programs for Extractive Reserves focused on green marketing, consumer coops and value added through processing carried out in an urban factory in the village of Xapuri, as structural adjustment programs gained importance, the development emphasis shifted to decentralized processing and piecemeal contracts on the individual seringal (rubber tapping estate) or in mini factories in forests. While this was an appealing approach given the kinds of development concerns at the time; non-timber forest products, income generation for women in the forest itself, and neoliberal ideologies of economic and labor decentralization, it failed to appreciate the demands and the opportunity costs on women’s time in rural areas and underestimated the importance of formal employment in urban areas. The logics on which the shift was justified, enhanced production, efficiency and lower costs, did not materialize.

Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 253–285 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10012-2

253

254

SUSANNA HECHT

The man of the house returned from the forest about noon, bringing nearly two gallons of (rubber) milk that he had been collecting since about daybreakyWhen he himself attended to the 120 trees he could collect the same amount for several months. But his girls could only collect from 70 treesyIn making the shoes, two girls were the artistes in the little thatched hut which had no opening but the doorythe (shoe )last was dipped in rubber milk and immediately held over the smokey(where it) dried at once. It was then redipped and the process repeated until the shoe was of sufﬁcient thicknessythe shoe is now cut from the last and is ready for sale, bringing in about 12 cents. Edwards, H.E. 1847. The River Amazon

1. INTRODUCTION: THE EXTRACTIVE QUESTION AND AMAZON DEVELOPMENT Few scholars have given greater thought to the nature of extraction than Stephen Bunker. His paper on the Modes of Extraction (1984) and his book Underdeveloping the Amazon (1985) sought to recast the labor theory of value through an analysis of extractive economies at extreme peripheries. His research has moved into new directions and regions, but I will focus on his contributions to Amazonian extraction and use it as a platform for exploring the questions raised by modern extraction through a case-study in the Western Amazon. Bunker’s work was especially prescient in three areas. First, it illuminated the problems of ecosystem and social valuation, which is now a growth industry in Ecological Economics. As he points out, ‘‘production models cannot explain the dynamics of extractive economies because the exploitation of natural resources uses and destroys values and social formations whose worth cannot be uniquely calculated in terms of labor or capital.’’ His ideas were used by many colleagues as a starting point for the understanding of a ‘‘subsidy from nature’’ in the creation of wealth and livelihoods in Amazonian frontiers and to emphasize the role of gathered forest products for export and subsistence in rural income formation (Hecht, Anderson, & May, 1988). Next, in his clear elucidation of the differences between extractive and productive economies, he points out that expansion of scales in productive economies usually results in a decrease in unit price as infrastructure, production techniques and institutional structures change. This is an outcome of technical change and an increase in relative surplus value. Expanding the scale in extractive economies tends to increase unit costs as search, support, control and transport costs rise as more distant sources are brought into play. There is a deﬂection ‘‘price point’’ in capitalist economies where

Extraction, Gender and Neoliberalism in the Western Amazon

255

distance starts to preclude extraction in many economic contexts because of costs of transport, labor and control. But this deﬂection point does not hold where labor deployment in extensive extraction systems is maintained by non-economic forces, such as terror and slavery on the Putumayo and the Congo, debt peonage, or by family effort (Hecht, 2005; Nugent, 1994, 2002; Rafﬂes, 2002; Taussig, 1987; Weinstein, 1983). Thus absolute rather than relative surplus value characterizes accumulation in extractive economies, especially those that are based on forest products, and thus it is the extension rather than labor productivity that underpins extraction. Bunker’s extraction work also goes to great lengths to explore the effects of unequal exchange, a topic now also taken up by resource and conservation economists and development practitioners under the more comfortable heading of ‘‘getting the prices right’’ and marketing strategies that became part of the fair trade/green trade efforts throughout the tropics. His central point was that a given mode of extraction, by depleting the environment and deeply disrupting local societies, limits the development capacity in the next mode. Stagnation and boom thus become the deﬁning features of extractive economies in extreme peripheries. This position is perhaps more open to contestation since technologies and markets can change what is considered a periphery in the long life of a region. Moreover, in the case of rubber, the resources were not depleted. It was labor control and the quantity of labor that could be pressed into service that ultimately was the problem, rather than environmental resources. As a theorist, his contributions to understanding the underlying causes of economic stagnation in political economies of extraction stimulated a great deal of exploration of this topic. While Bunker has been widely critiqued, his work provided the groundwork and counterpoint for later research. His efforts are now especially germane in light of the emergence of Extractive Reserves (ER) as a regional development model for Amazonia. How ecologically sustainable is the extraction of non-timber forest products? How much do they reﬂect human manipulation of the environment (cf. Schroth et al., 2003)? What are the conservation and management dimensions of extraction (see, for example, Browder, 1992; Cardoso, 2002; Freese, 1997; Salafsky, Dugelby, & Terborgh 1993; Nepstad, & Schwartzman, 1992)? In the meantime, the social dynamics within extractive economies still remain relatively unresearched, (but see Almeida, 1992; Anderson & Ioris, 1992; Brown & Rosendo, 2000; Campbell, 1996; Gomes, 2001; Hecht et al., 1988; Kainer, Schmink, Leite, & Fadell, 2003). Finally, if his arguments hold, can modern regional economies be developed within the framework of extraction of non-timber forest products (NTFPs)?

256

SUSANNA HECHT

This chapter has four central concerns: to review the narratives about extraction and stagnation in the past, to explore the narratives of extraction and development in the current Amazonian context, and to examine the ‘‘black box’’ of livelihood production in extractive economies today. The chapter then analyzes how current development approaches have shaped Brazil nut extraction in an ER in Acre. The study explores these questions through the political economy of gender.

2. NARRATIVES OF EXTRACTION, DEVELOPMENT AND ENVIRONMENT IN AMAZONIA 2.1. Extraction and Stagnation: Revanchist Drag or Sustainable Development? Extractive activities have been part of an intense polemic about tropical development in the Amazon Basin for hundreds of years. What has remained a key question is the failure of the immense wealth of the rubber era to translate into capital and social formation necessary for the next phases of regional development. Why did highly profitable precapitalist structures not give way to capitalist production regimes and dynamic regional growth, as occurred with the coffee economy of Sao Paulo? What might this suggest about development and extraction today? This is a significant issue because ERs have been increasingly proposed as a development model today, and the National Rubber Tappers Council dreams of 10% of Amazonia in this form of tenure. How and why extraction constrained economic transformation earlier, thus becomes more than an abstract historical question. The issue of stagnation and extraction merits some review as a means of illuminating some of the features of the modern dynamics of extensive extraction. There are several ways to approach this question, including those that focus mainly on macro-level dynamics, those emphasizing more ‘‘meso’’ processes, and those that explore the social relations of extraction itself. The most well-known model might be considered that of ‘‘Competition and Ecologies.’’ Ecology constrained developing plantations in Amazonia, while efﬁcient Asian plantations based on waged labor could deliver rubber to industrial markets at a better price. This rural plantation proletariat was increasingly characteristic of the production systems for valuable commodities throughout the tropics during much of the colonial period. Warren Dean’s (1987) environmental history of rubber concentrates especially on

Extraction, Gender and Neoliberalism in the Western Amazon

257

the local ecological conditions that thwarted plantation development and that made the region ultimately unable to transform itself into a more modern pan-tropical economic production formation on the basis of one of the most valuable commodities on the planet at the time. The thesis of Alfredo Homma (1993) then comes into play: as commodity prices rise, there are several outcomes, all of which produce stagnation in or undermine the initial extensive extraction systems. Overexploitation, domestication (and plantation development away from its center of origin), and the search for synthetic replacements made the natural product largely irrelevant. Rubber, of course, experienced all three of these processes. Bunker, focusing on the ecological and social dimensions of unequal exchange of extraction, viewed stagnation as an outcome of the siphoning of energy and value out of the system and the disorganization of Amazonian ecologies and societies to the detriment of later productive activities (Bunker, 1984, 1985). For Bunker, the rubber economy was rooted in a world system where the dynamics of unequal exchange prevailed and where the distribution of the resource itself made it quite problematic. This, combined with an impotent state, one highly dependent on rubber revenues and the oligarchs who controlled them, produced such distortion that when the regime collapsed (as an outcome of the ecology/competition dynamic), it left the region largely starved of capital due to the very limited linkages to any local sectors, and with labor ‘‘stranded’’ by peonage on rubber estates. Bunker works out his ideas from the classic framework of the ‘‘blood rubber model’’ that involved slavery and terror on remote rubber estates in the upper Amazon (Taussig, 1987; Hardenburg, 1905; Stanﬁeld, 1998). A somewhat different analysis emerges from studies that focus more on the central and lower Amazon where simple commodity producers, whose production was based on family labor, were far more prominent.

2.2. The Labor Question in Extraction Stephen Nugent (1994, 2002), an anthropologist, also takes on the question of extractive economies, caboclos and stagnation largely through a discussion of mechanics of merchant capital. In this sense, he works at the meso-level to understand why extraction remained undynamic. Nugent, whose work on contemporary caboclos is informed by a historical consideration of post-rubber boom stagnation, argues that Amazonian mercantile capital was able to profit without disenfranchising populations or changing the social relations of production because it enjoyed a monopoly position in

258

SUSANNA HECHT

the world market for most of a century, and because the ecological conditions of the production of Amazonian rubber did not really permit intensiﬁcation. The Amazonian mercantile economy remained basically unchallenged by capitalist production regimes until the 1960s. Unlike many other areas in the tropics where simple commodity producers began to compete with plantations or other capitalist forms of production, in Amazonia extractors were insulated from this transformation by the nature of the mercantilism (aviamento – a clientistic form that advanced trade goods against products) that trafﬁcked only in commodities, the isolation of producers, and a more general lack of regional economic dynamism. Thus, surplus extraction remained absolute rather than relative, realized through the mechanisms of unequal exchange. Nugent focuses on the nature of the caboclo peasantry, pointing out that, like Weinstein (1983), there was no need to change the social relations of production as long as the commodity was delivered. Nugent emphasizes the inability of merchant capital to control and discipline workers on the lower Amazon, or even to penetrate its labor processes. As Henry Pearson1 points out in his ironically optimistic panorama of the Rubber Economy on the eve of its collapse, Rubber Country of the Amazon (1911), ‘‘the tapper will tap tomorrow or the next day, or whenever he likes.’’ Nugents work builds on and extends the insights of the historian, Barbara Weinstein (1983), who brought attention to the array of forms that extraction could take as she focused on the petty commodity producers/extractors. Arguing from the idea of modes of production, Weinstein focused on the social relations of production in simple commodity production/extraction. While producing for markets, labor mobilization depended on family and kinship, as well as debt-based forms of recruitment, but Weinstein also emphasized the limited control over the labor process in these lower Amazon economies. Her research revealed the importance of autonomy to tappers, a widely bruited complaint among those who wished for closer management of Amazonian rubber production. Weinstein’s and Nugent’s research focused on the lower Amazon, where several conditions prevailed that contrasted with those of the upper Amazon: ﬁrst, the impacts of drastic Cabanagem rebellion had undermined elite control and dramatically altered local power relations over forest workers (cf. Cleary, 1998; Hecht & Cockburn, 1989). Next, populations were often organized in quilombos2 of runaway slaves as well as detribalized Indian settlements. These producers were part of communities rather than isolated households and so had both more product and more negotiating power. These groups were not ‘‘prisoners of the landscape’’ in the way that migrant

Extraction, Gender and Neoliberalism in the Western Amazon

259

recruits/debt peons from the Northeast of Brazil were in the upper Amazon, because, as local ‘‘caboclos’’, they had kin and afﬁnes in the area, and they were natives of, rather than exiles to, Amazonia. Third, producers were part of families who relied on a portfolio of activities for their livelihoods, not just rubber. Their families also participated in the varied activities of agriculture and of extraction (see Edwards, 1847). Finally, aviamento certainly operated in these lower Amazon communities, but the system was more complicated and complex in lower Amazonia where proximity to many small-scale traders implied some competition among them. Pearson (1911) mentions the number of aviadores, and the relative ease with which one could become one. Pearson remarked that many aviadores were often seringueiros themselves, and thus the patron–client relation was often afﬁnal and infused with a great deal of social meaning and mutual obligation, and not necessarily a strictly commercial transaction. It was the persistence of aviamento, which was based on the exchange/barter of goods and selfprovisioning by local peasantries, that stunted internal demand and effectively stiﬂed local commercial agriculture in much of the lower Amazon, to the regular howls of the regional elites. Weinstein’s work presages elements of moral economies a la Scott (1976), particularly in light of contemporary research on Amazonian middle men, who are often themselves tappers. Thus, the relations may have been more nuanced, with clear beneﬁts in stressful times (cf. Padoch, 1992; Nugent, 1994; Rafﬂes, 2002). Scott’s (1997) ‘‘everyday resistance’’ also has relevance for this case. The deﬁance of some simple commodity producers/ extractors toward waged activities was possible in part because of the feeble development of monetized capitalism in the Amazon economy, the difﬁculty of labor control in the Amazon, and the relatively weak presence of the state in frontier affairs, but especially because they maintained control over their agriculture and had access to the rich array of ‘‘use value’’ as well as exchange value extractive products. This permitted them to withdraw from some patrons, and helps explain why extra-economic means of consolidating this relationship have been so important, whether through afﬁnal means or coercion. In addition, the various types of adulterations of rubber can certainly be seen as forms of evasion of control as well as fraud. This resistance to incorporation into capitalist social forms produced an ‘‘uncaptured’’ peasantry, with significant autonomy in some regions. It was, after all, labor that was the element of production in shortest supply. As Nugent (1994) points out, mercantile capital is ‘‘agnostic’’ in terms of labor deployment as long as the commodity is delivered, and thus

260

SUSANNA HECHT

simple commodity production/extraction (based on family effort) could co-exist with many other servile forms such as the corvee of terror, debt peonage in the more isolated reaches of the Basin, or the commercialization of tribute from native groups. But it was only the simple commodity form predicated on families that could physically and economically reproduce itself after the collapse of the rubber economy. But, of course, the future of Amazonia did not necessarily have to reside in rubber. This concern is that of Barham and Coomes (1994), who point to the distortions of ‘‘Dutch Disease’’ on regional economies as capital was increasingly sunk into infrastructure for extraction that could not be easily or effectively released from or realized outside of the boom sector, and on spending effects located largely outside the region that engendered significant dislocations in the economy. The ‘‘sunkness’’ of capital precluded its movement into other sectors in the post-boom period. Thus the effects of unequal exchange, social relations, ecology and form of capital (mercantile) or ‘‘sunkness’’ are all invoked to explain why more complex economies did not evolve during and after the boom. In Amazonia, what was certainly clear was that both labor and capital seemed immobilized in ways that caused extraction to be roundly reproached as a brake on Amazon regional development by locals and observers, a revanchist and despised activity condemning this tropical immensity to the status of permanent backwater. Demonized as the most regressive social conﬁguration imaginable, extraction, until the end of the 1980s, was literally off the map, an embarrassing residue of the previous century in its spatial form, tenurial regimes and social relations. This attitude was parroted in virtually every government development document on Amazonia from the 1960s and 1970s, resulting in a systematic bias against investment in or appreciation of the important economic role of extractive economies for most Amazonian environments. Extraction had persisted in its ‘‘portfolio’’ form of linked agriculture, use value and commercial extraction, undetectable to outsiders under the mantle of Amazonian forests. Its main conﬁguration might simply be called ‘‘simple commodity extraction/production’’, a hybrid of previous elements of Amazonian life and culture carrying out extraction of a vast array of products in a modiﬁed mercantile system. The omnipresence of this system was only matched by its invisibility and that of a large portion of its labor force: women and children. What came to dominate the perception of extraction, in spite of the literally hundreds of thousands of families that engage in it, was that of the isolated man, enmeshed in a pitiless production regime.

Extraction, Gender and Neoliberalism in the Western Amazon

261

3. THE GENDERED ICONOGRAPHY OF EXTRACTION The image of the lonely tapper setting out before dawn, disciplined by hunger, the patron and international capital, slogging on in endless repetitive cycles of tapping, curing and peonage in a pitiless regime of cruelty and isolation, may have been an accurate representation of some of the rubber producers at the ‘‘margins of history’’ during the peak of the rubber boom. But as evocative as this image was then and now, it fails to capture a forest reality that included families. The dominant image of rubber tappers (and by extension extraction) remains the solitary, enslaved man – a poignant vision, and one that makes the political triumph of the rubber tapper movement in the late 1980s all the more compelling. This image relies on descriptions by Alfred Rangel (1927), Euclides da Cunha (1906), and Roger Casement (1911, 1912) at the height of the boom as well as by more modern interpreters of the region such as Leandro Tocantins (1979). This perspective framed but one, albeit extremely brutal, incarnation of the rubber economy. 3.1. Invisibility and Women Often overlooked in this highly masculine model of extractive systems in the western Amazon is that women and families were present, even at the height of the boom of the early 1900s on the main water courses in the upper tributaries, like the Purus, where the commerce in rubber and Caucho was thriving. Euclides da Cunha (1906) noted robust yeomen communities, and indeed, in legal cases the profession of men is as often self-identiﬁed as farmer (agricultor) as rubber tapper (seringuiero) (Wolff, 1999). In one of da Cunha’s most dramatic evocations, ‘‘Judas Asvero’’, he describes a tapper making the efﬁgy of Judas, a poignant symbol of the isolation of tappers, as his children play around him (Da Cunha, 1906). Da Cunha notes the casual cruelty toward women: his boat passes a dead women in a ﬁeld, the off hand dismissal of loyal common law partners, and the harems of the Rubber Barons. Even ‘‘coffee table’’ books produced by local photographers during the boom showed women on the rubber frontier. The Paraense photographer Facao (1906) Album do Acre provided pictures of Acre’s revolutionary heroine, Angelina Goncalves Sousa, and of various seringais that included wives (even women riding side saddle on beautiful mules), female servants and children. The ﬁrst census of the Upper Jurua in 1904 notes that some

262

SUSANNA HECHT

27% of the population was female. More than 50% of the civil cases were brought to the courts by women (Wolff, 1999). A new historiography of women in the Amazon is emerging. Traveler commentaries such as that of Edwards (1847) show women and children collecting, processing and, in Edwards’ case, making rubber shoes for export. Elite observers, such as Mrs. Agassiz whose husband founded Harvard’s Natural History Museum and who traveled in Amazonia under the aegis of Dom Pedro II, reﬂected on the impact of the military draft for the Paraguayan War, which put all the burdens of agriculture, collecting and marketing on women and children (Agassiz, 1868). Studies of the historical role of women and their representation using an array of documents, ranging from court cases, travelers’ accounts, newspaper sources and oral histories, is actively excavating a gendered regional history that includes the image of black women (Rocha de Almeida, 1996), those of ‘‘low repute’’ (Trinidade, 1996), and rubber tappers on the Jurua (Wolff, 1999; Simonian, 1996). This new scholarship provides insight into the culture of the regional economy that has been almost entirely lacking. More recently, the inﬂuence of Brazil’s women’s movement in leftist political programs produced a self-conscious effort to include at least some rhetoric about women in virtually all the policy statements in Amazonia’s labor movements such as CUT (Centro Unico dos Trabalhadores), PT (Workers’ Party), STR (Sindicato de Trabalhadores Rurais), and CNS (Conselho Nacional de Seringueiros) and more recently the Movimento sem Terras (MST). Typically the documents focus on alliances3 and remain largely on the rhetorical and political plane (Deere, 2001). All worker movements in Amazonia claim women members. In rural Amazonia, women have taken part in the showdowns (empates) where extractors faced off against deforesters, in land occupations of the MST, and in public manifestations. Their political participation and inﬂuence remains overall fairly weak and poorly documented, although studies of emerging political mobilization emphasize women’s empowerment with the rise of more rural organization and syndicalism (cf. Campbell, 1996; Deere, 2001; Hecht, 1992; Maneshchy, Alencer, & Nasciamento, 1996). There are also spectacular exceptions to the ‘‘rhetorical’’ versus substantive role of women in Amazonian politics, such as Marina Silva, the rubber tapper and syndicalist who became the Senator from Acre, and then Minister of Environment under the Worker Party President, Ignacio ‘‘Lula’’ da Silva. Under increasing decentralization and local community development efforts, women’s political participation does seem to be increasing, even as the empirical analysis of the women’s role in regional economies remains relatively sparse.

Extraction, Gender and Neoliberalism in the Western Amazon

263

Women remain shadowy ﬁgures in the popular understanding of extractive economies, in spite of documented studies that clearly show women as actors in extractive economies in land management (Anderson & Ioris, 1992; Kainer & Duryea, 1992; Kainer, Schmink, Leite, & Fadell, 2003), collecting (Anderson, 1988; Anderson, Balick, & May, 1991; Campbell, 1996; Hecht et al., 1988), processing (Hecht et al., 1988; Clay, 1997), and marketing (Padoch, 1992). In Brazil’s North, the number of Babassu collectors (450,000 families), the Brazil nut processers and collectors throughout the Amazon (including Bolivia, Peru and Ecuador) number easily around 25,000, and intensive extractors of palm heart and palm fruits on the lower Amazon (20,000) include a vast female labor force within the forests.

3.2. From Demonization to Deliverance: The New Narrative of Extraction The practices of the modernist and conservation visions of the Amazon that began to be implemented in the mid-1960s relied on enclosure and rational planning and were carried out under the aegis and power of the authoritarian state. To this juggernaut of regional transformation, extraction emerged as a significant discourse of political resistance that evolved from a complex history of human rights, liberation theology and rural labor organizing. It also reﬂected an emergent populist conservation movement that viewed tropical landscapes as artifacts of human management, rather than the pristine nature evoked by most conservationists, and emphasized forms of sustainable forest use. The ideological foundations of modern extraction emerged as a counternarrative to both of the main models of land use on offer and contested both conservation zones empty of population and the blasted and deforested landscape of ephemeral cattle production. To compete with the other models, and to rehabilitate extraction as a viable and reasonable activity, it had to challenge the premises and the practices of the dominant modes of both conservation and development, within the dominant frameworks of the ideologies of science, economics and planning. It required a means of developing a ‘‘forest path’’ that could reconcile and recast the powerful forces that were bearing down on forest and traditional peoples, both bent on excluding them from their traditional lands. The rehabilitation of extraction required the idea that this economic formation, unlike the prediction of Bunker, could indeed transform itself and embrace the social relations of capitalist economies. That is, the rubber tappers were interested in the modernization of their economies by

264

SUSANNA HECHT

developing the institutional structures (cooperatives, markets, etc.) and (waged) social relations of modern capitalism in lieu of mercantile capital, but they would do so within an ‘‘ecological equilibrium’’ framework that did not require deforestation. This approach has been baptized by some analysts as ‘‘Neoextractivismo’’ (de Rego, 1999). At the mythical level, instead of El Dorados or Lost Edens, the rubber tappers and other traditional peoples evoked the powerful, romantic images of tropical Arcadias – a benign forest form of land occupation calling up visions of a more harmonious time when economy and ecology coexisted. In this view, ‘‘nature’’ was part of a dialectical ecosystem with people inserted into them, creating and protecting them through their stewardship. In this sense, forests were neither obstacle nor antidote to development (the positions embraced developers and conservationists) but rather the result of, and a new platform for, autochthonous versions of development. They simply and brilliantly were able to disarm the critiques of each side of the debate by presenting a forest model of modernization. In communicating this message they had considerable help, but it was an extremely powerful one as a means of unblocking the ever more rancorous Amazonian conservation and development impasses, presenting a reworked past as a viable future. While initially focused on the Amazon’s indigenous populations who were under massive and aggressive threat, the idea of the inhabited forest gained ground through an increasing amount of research on the region’s caboclo and traditional populations (cf. Nugent, 1994; Padoch, Ayers, & Pinedo, 1999; Rafﬂes & Winklerprinz, 2003). Extractors were thus able to argue that, rather than a tabula raza or a world without history, the region was deeply imbued with deep traces of its native and mestizo past, and that development approaches could build on these deeper historical structures, whether intellectual or tenurial. Thus, a hybrid science that melded the insights of local knowledge systems with the techniques and methods of modern sciences could be developed to improve management, production and sustainability. The emerging comparative research on indigenous knowledge systems began to suggest that the management approaches of traditional peoples were at least as good as the failing agricultural models that were being implemented throughout the Basin, and in many cases much better. As a political model, it envisioned forest peoples engaged in the extraction of non-timber forest products – rubber, nuts, palm products, etc. within ERs, areas that upheld traditional access rights but ratiﬁed in a modern cadastral way. These were presented as a land reform for forest peoples (CNS, 1988; Hecht & Cockburn 1989) and focused on a kind of forest-based development that was based in participatory political processes (as opposed

Extraction, Gender and Neoliberalism in the Western Amazon

265

to previous authoritarian practices), and a model of conservation with development, and one based in social justice (a weak point in the other models). Its cultural component included the ‘‘recovery’’ of local histories (Rafﬂes, 2002), the preserving of traditional life ways and the local and indigenous knowledge within it, and the elaboration of grassroots versions of social transformation. As an economic proposition the strategies for ER development in the 1990s were inﬂuenced by the important intellectual contribution supplied by the Workers’ Party, whose origins lay in the uprisings in the factories of Sao Paulo. Thus, the earlier conceptualizations of reserve development incorporated an essentially industrial model in their approach to region. The ER modernization programme argued that elements that inhibited development (aviamento, unequal exchange, precapitalist labor relations) would be transformed into recognizably modern forms within a framework of the new state and cadastral tenurial regime. While the architects of these ideas were inspired by egalitarian/socialist frameworks in the distribution of resources and political processes, the system itself would have to operate within the conﬁnes of the larger capitalist system. For the National Council of Rubber Tappers (CNS), economic proposals for ERs relied on a range of strategies. In terms of marketing – the site which many viewed as the key area of surplus extraction via aviamento and its inherent unequal exchange – the CNS argued for the elimination of the aviamento in favor of consumption as well as production cooperatives that would, in principle, increase producer incomes by eliminating middleman markups. Next, ‘‘green marketing’’ was a priority so that products could exceed average global prices because of the ecological and socially desirable context of the reserves. In addition, a kind of ‘‘branding’’ associated with the international appeal of the rubber tappers and tropical conservation would be possible. Third, increasing the value added via the export of more elaborated products was also seen as central economic strategy for enhancing rural incomes. Rubber factories would be too costly, but the industrialization of Brazil nuts, whose processing factories stretched from the Andes to Belem, were an important option, especially in light of rising Brazil nut prices throughout the 1990s. The main nut traders, the Belem-based Mutran family, processed most of the Brazilian Amazon production, and thus had monopsonistic control on the prices paid to producers. By creating a means of adding value by processing and exporting through their own factory and into niche market producers, it was argued, they could capture more of the income associated with their products. In the case of Brazil nuts, extractors would in theory enhance their return by a factor of 10 (IDE, 1991).

266

SUSANNA HECHT

Table 1. Name

Brazilian Federal Extractive Reserves.

State

Created

Area (ha)

Population

Activity Rubber Rubber, nut collection Rubber, nuts, copaiba, acai Rubber, nuts, acai, copaiba Babassu Fish Babassu, ﬁsh Babassu, ﬁsh Babassu, agriculture Rubber, ﬁsh

Alto Jurua Chico Mendes

Acre Acre

1990 1990

506,186 970,570

3,600 7,500

Rio Cajari

Amapa

1990

481,650

3,800

Ouro Preto

Roraima

1990

204,583

700

Ciriaco Pirajubae Tocantins Mata Grande Quilombo Do Frexal Medio Jurua

Maranhao Sta Caterina Tocantins Maranhao Maranhao

1992 1992 In process In process In process

7,050 1,444 9,280 10,450 9,542

1,150 600 800 500 900

Amazonas

In process

600

600

Beyond this modernization program, the reserves were imbued with a kind of post-modernist luster: the juxtaposition of archaic and pre-modern production with the progressive social movements, the use of high-tech technologies (computers, GPS, satellite mapping) with models of local knowledge, advanced environmental paradigms coupled with a sophisticated understanding of ‘‘Green’’ globalization (international alliances, marketing strategies and media). The politics of the period coupled with growing international concern resulted in the establishment of over 2 million as ERs. Table 1 outlines the main areas of ERs, their populations and main products.

3.3. From Modernization to Neoliberal Development The reasoning behind the transformation of extractive economies was compelling and led groups such as Ben and Jerry’s Ice Cream and Cultural Survival to provide seed funds for the development of a small-scale Brazil nut factory in the village of Xapuri, located on the Acre River. The start-up funds attracted several other donors who provided monies for technical support, working capital and machinery, as well as gifts in kind. The enormous publicity that had surrounded the rubber tapper movement, and the idea that they were actually implementing a small industrial project, caused much excitement. The emerging Brazil nut cooperative and the processing factory all associated with the ‘‘Projeto Castanha’’ or Brazil nut project

Extraction, Gender and Neoliberalism in the Western Amazon

267

captured everyone’s enthusiasm. Indeed, local politicians allied with the movement pointed to the ‘‘Projeto Castanha’’ as a model of Workers’ Party development which would bring economic beneﬁts with environmental sustainability to the most marginalized classes of the Amazon. What was especially interesting about the factory was its emphasis on women workers as the new ‘‘industrial’’ workforce in the heart of the Amazon. The factory with its constellation of funders was the palpable sign of the ability of the National Rubber Tappers Council to actually get things done through innovative coordination of public, private and NGO efforts with a progressive social agenda, and one that modernized at least a few of the parts of the production processes. Implemented at the height of rainforest ‘‘craze’’, the project seemed the incarnation and model for sustainable, equitable development. The factory represented the modernization of production within globalization’s niche economies. By changing the terms of trade through external commerce, cooperatives, and their unique capacity to ‘‘brand’’ their product, the distributional outcomes would beneﬁt local cooperatives rather than the historical mercantilists. This ‘‘modernization’’ effort seemed to transform the extractivist context in one rosy swoop. However, several processes intervened, most importantly the rise and diffusion of neoliberal ideas of decentralization, efﬁciency and market-based conservation approaches. Within 3 years, this dramatic and highly publicized ‘‘factory’’ story was on the wane, eclipsed by a new narrative, essentially neoliberal in its focus, that condemned the previous strategy as vulgarly centralized and classically industrial, contaminated with urban bias when compared to the new, forest-based decentralized development. Neither the old or the new decentralized ‘‘Projeto Castanha’’ paid much attention to the social context of workers. Both viewed the production processes and their local economies as ‘‘black boxes’’, and thus never assessed the relative impacts of the project on its rural or urban female labor forces. Infused with classical development theory ideologies of rural underemployment of women, enthusiasm for ‘‘in forest’’ employment, and the uncritical application of ‘‘efﬁciency’’ criteria, CAEX, 1993 (The AgroExtractive Cooperative of Xapuri, the export coop) dismantled the ‘‘urban’’ factory, ultimately exchanging an adult urban female labor force for a rural labor force of what ultimately became child piece workers.

3.4. The Castanha Projects: Trouble in Paradise #1: The Factory Amid great fanfare, the Chico Mendes Brazil nut-processing factory was inaugurated in November 1989. Seventy people were initially hired at the

268

SUSANNA HECHT

new factory, and it thus became the largest employer in Xapuri, and one of the largest in the whole state of Acre. Many of the women who were its main labor force were from families who had been pushed off the land during the very violent land conﬂicts of the 1980s. In addition, older men no longer able to work in the rubber forests also found employment in the factory. The factory thus was seen as having an important social as well as economic function for those who had been dispossessed. Besides the crackers, the factory included two managers, eight classiﬁers, two packagers, one accountant and two administrative assistants, as well as four janitors and two guards for a total factory staff of 70. In addition, the Castanha cooperative itself had a centralized staff housed with the Rubber and Consumption cooperatives where logistical, ﬁnancial and policy decisions were made. After a 3-month training period, workers were paid a minimum salary (about $50.00/month). The model of the new factory imitated technologies and production organization practices widely found in Brazil and Bolivia. Brazil nuts begin to come out of the forest in January–March until about mid-May. They are stockpiled and processed for the next 6–8 months. Nuts are ﬁrst dried and heated to release the moist nut from the shell. Then they are either autoclaved or placed in a water tank to soften the otherwise very hard shell. The nuts are cracked on work tables with simple individual mechanical cracking machines by a predominantly female labor force that carries out the repetitive task of cracking and the initial separation into rough categories (rotten, broken, ﬁrst category). Nuts are then classiﬁed into the commercial categories, further dried, and vacuum packaged for export. Various problems emerged right away. The ﬁrst was the structure of the Brazil nut harvest itself, and the fact that the factory was often idle from 2 to 4 months because of lack of raw materials. The next, more central issue had to do with Brazil’s labor laws: the employer must pay the social security costs of the worker, which is calculated at 65% of the salary. It is extremely difﬁcult to hire or lay off people periodically, so payrolls had to be met whether the factory was running or not. Productivity was not as high as hoped. By instituting a 10 kg/day quota (which produced about 3.3 kg of ﬁnished nuts), profitability could be maintained. But since the workers were waged, many did not come to work if their 20-day quota was reached prior to the end of the month. In addition, workers began to organize. Given the largely female nature of the workforce, there were demands from workers for child-care facilities at the factory, which they argued would reduce absenteeism. The only techniques for increasing productivity focused on ﬁnes and small ﬁnancial incentives as

Extraction, Gender and Neoliberalism in the Western Amazon

269

a way to discipline this labor force. No attempt was made to improve the technology, work tables or backless benches where the factory workers toiled. CAEX was often tardy in its payments to workers, which during Brazil’s hyperinﬂation in the mid-1990s of close to 600%, reduced the wages to pathetic sums, even as CAEX nuts received premium prices on the world market. However, CAEX also had to address the real question of global competition: Bolivian nuts arrive in the world market at a cost that is 42% less than that of the CAEX product using roughly the same techniques, paying lower wages but large state-supported social beneﬁts, and with much lower storage losses since nuts were stored in warehouses in chilly, arid La Paz where post-production losses were minimal. In Xapuri, the costs of the factory labor force with beneﬁts that produced nuts was seen as a key problem. Barely had the factory started up before attempts began with piecework subcontracts in Xapuri to supplant the nascent industrial enterprise. After the ﬁrst year of operation, CAEX began to explore a strategy to decentralize production as a way to avoid paying the 65% beneﬁt cost. To initiate contracting, the key legal mechanism would be to transform the waged workforce with its social costs into an ‘‘association’’ which could then crack nuts on a piece work, ‘‘putting out’’ basis. The cooperative began working with a labor lawyer in order to see how the factory could be transformed from a waged to a piecework outﬁt. This would require the dismissal of the entire cracking staff, and then sub-contracting to selected workers. Needless to say, factory employees were outraged, and this approach undermined not only their livelihoods but also an emerging political autonomy, which seemed at odds with the pro-labor rhetoric expounded by the Coop’s masters. This ill will further oriented the decentralization strategy away from the village,4 in favor of decentralized production located in the forests themselves. While some analysts might have viewed this as a ‘‘runaway shop’’, the language of decentralization, which was especially appealing after a generation of highly centralized, authoritarian regimes, coupled with several efﬁciency and social welfare arguments (mothers could be near their children), enhanced the desirability of this approach.

3.5. Trouble in Paradise #2: Decentralized Production The logic behind the shift to decentralized production was clearly oriented to reduce personnel costs of nut crackers, but the justiﬁcations were presented as part of an efﬁciency, decentralization, technology and rural

270

SUSANNA HECHT

development package. Infused with neoliberal ideas on production, the castanha project leaders embarked on the new approach. In principle, decentralized processing would: 1. Reduce transport costs (the heavy shell and water of the nuts account for some 60% of its unprocessed weight); 2. Reduce losses – nuts that sit in the shell for a long time mold. More rapid processing would improve the general quality of nuts and reduce storage losses; 3. Permit better utilization of ‘‘excess’’ labor residing in the forests; 4. Improve the income of forest peoples; and 5. Create jobs within ERs, not in their villages. Indeed, in the most delirious portions of the project literature on decentralization in ERs, advocates foresaw 180 direct beneﬁciaries, and that 900 collector families would gain improved incomes from the decentralized process (IDE, 1991). Thus, not only would nut producers receive a better price, but the cooperative itself would run a more profitable, sleeker operation. Rural employment would be generated and a form of decentralized development would emerge. This approach ﬁt well with the national eagerness for democratization and decentralization as a counterweight to the traditionally highly centralized and corrupt national economic structures. It was hoped that 50% of the production of the Brazil nut cooperative would be processed in four ‘‘microusinas’’ (small factories) with an estimated average productivity of 12 t a month. Forty people would be directly employed in cracking in these decentralized operations, with employment also generated in transport, supervision, etc. The upper tiers of factory management (accountants, classiﬁers, etc.) would be retained, as well as the central cooperative management personnel. The factory would still provide ﬁnal processing if necessary, grading, ﬁscal control, and export packaging of the ﬁnished product. (Fig. 1) The new forest processing would follow three main lines of organization. In the seringal, micro, mini and household cracking were possible. In the microusinas, there would be work stations for some 10 crackers, plus a manager and a helper. Nuts would come to the microusina, basically a storage shed and processing barracks consisting of tables and cracking machines. Then nuts would either be processed on site or distributed for mini and ‘‘household’’ crackers. In the miniusina, a complete system of dryer, wetting tank and ﬁnal drying oven would be installed on a multi-family seringal so that extended families could process their own nuts, giving the

Extraction, Gender and Neoliberalism in the Western Amazon

271

RURAL DECENTRALIZED PROCESSING

Mini Usina

Micro Usina

Domestic Process

Family labor (Compound)

Independent and Family Labor

Household Labor

Semi Finished product

Semi Finished

Final Drying Vacupak for final packaging at Chico Mendes Factory

URBAN PROCESSING Usina Chico Mendes Factory Processing

Decentralized Domestic

Waged Labor

Family Labor

Finished Product

Semi Finished Product Grading, packaging

EXPORT MARKET

Fig. 1.

Pathways of Processing Brazil Nuts.

ﬁnal processed nuts to the micro factory for transport to the Xapuri factory and ﬁnal sale. Household crackers were provided a water tank, drying platform and cracking machine. Processed nuts would be transported back to forest factory and then again to the main factory in Xapuri for export. All work was to be on a piece work basis. In urban areas, household cracking would predominate. This involved receiving nuts in the shell that had been

272

SUSANNA HECHT

dried and moistened, and thus were ready for cracking, but was never extensively developed. This optimistic scenario, however, was deeply at odds with the rural reality.

4. BRAZIL NUTS AND WOMEN IN DEVELOPMENT The late 20th century ‘‘discovery’’ of extraction economies focused on a masculine re-imagination of forest extraction at the height of the boom in the most remote parts of the basin, and thus was at odds with the everyday reality of these economies that regularly included families, women and children engaged in agriculture as well as extraction. This imagery has obscured the very intimate links between gathering and agriculture that has prevailed in Amazonian livelihoods for millennia. The regional economic volatility appears to result in ﬂexible emphasis in land uses between agriculture and extraction, both in the past and today. Most extractors are members of families. These families participate in every extractive activity and carry out the agricultural work, animal management and domestic processing that makes the reproduction of life as a forest extractor possible. Tapping, which also includes activities necessary for maintaining the trees and estradas (rubber paths) involves significant family effort. The image of the lonely, fever-racked male tapper, belies the fact that extractive activities of all kinds are part of the cash and subsistence contributions of women and children, as mentioned by the observant Mr. Edwards in 1847.

4.1. Structure of Income and Labor Deployment The emphasis given to rubber tapping reﬂects the historical origins of the occupation of the area at the end of the 19th century, the importance of the tappers in the forest peoples’ movements, and the reality that it is the activity most dominated by men. While rubber’s symbolic meaning in the region is tremendous, the economic role in regional economies is quite different. Surveys in the Acre Valley and in two ERs (Table 2) reveal a few important features: ﬁrst, the structure of income formation varies among sites (and within seringais themselves) in terms of the most significant source of income, but agriculture is the largest income source on average (about 38%), a ﬁgure similar to those among extractors in Babassu in Maranhao

Extraction, Gender and Neoliberalism in the Western Amazon

273

Table 2. Income Formation in the Vale do Acre (In Dollars and as a Percentage of Total Income). Product

Extractive Reserve a

Cachoeira

b

Chico Mendes

Mean c

Vale Do Acre

Return (US$/%) Rubber Brazil nuts Agriculture Mean total

288 (20%) 482 (34%) 648 (46%) 1,418

388 (44.7%) 124 (24.6%) 144 (29.5%) 656

221 (35%) 157 (25%) 240 (38%) 618

299 (33.3) 254 (28.3) 344 (38.4) 897

Sources: a UCLA /IEA/Funtac (1995). b Conselho Nacional de seringueiro (1992). c CIDA (1992).

(Hecht et al., 1988) and forest households in Ecuador (Thapa, Bilborrow, & Murphy, 1996), and Peru (Escobal & Aldana, 2003). Next, sites with more Brazil nuts have substantially more income. This income portfolio is significant, because it emphasizes the diversity of income-generating activities and suggests that, although there may be complementarity among the different activities, there is also the potential for competition for labor time.

4.2. The Gender Division of Labor on the Seringal The division of labor on the seringal reﬂects those of gender and age, but in reality the lines are a good deal more ﬂuid. The contribution of women in various productive activities associated with the reserves is largely underreported, and the contribution of child labor, especially that between the ages of 10 and 20, is almost completely invisible in the published statistics on labor deployment, even though families in the reserves were young, with 32.3% of the population under the age of eight, and 27.2% younger than 17. The average family size on the seringal is six people (CIDA, 1992). Families operate in a Chayanovian manner where the greater number of children permit one to exploit more completely the resources on the site. If one lacks children, or if they are too young to effectively work in extraction, share crop relations known locally as ‘‘a meia’’; contracters (empreiteiros), or wage workers can be employed. For extraction, meeiros and empreiteiros are preferred, while short-time waged worker can be hired if necessary in

274

SUSANNA HECHT

agriculture. In addition, worker families can also be aggregados – afﬁnes or ﬁctive kin who are given access to resources and are permitted to build a house or live in an existing building, but who also contribute to the main seringal family through labor corvees and products in kind. These may be newly married family members or families with very young children. While a detailed description and discussion of the gender division of labor are available in a different paper (Hecht, 2005) those results can be summarized as follows: Table 3 compares the labor requirements in the main extractive and agricultural activities. Agriculture, largely dominated by women and children, is the most demanding in terms of labor needs, followed by rubber and Brazil nuts. What the table also emphasizes is the large contribution of women’s and children’s work in these activities (70% overall), which does not include the activities associated with household reproduction and subsistence extraction (hunting, ﬁshing ﬁrewood collection, fruit collection, drawing water, etc.). As Deere and Leo´n (1987) have noted, the time demands of household reproduction are consistently under-represented. What is clear is that production of forest products and agriculture in extractive economies are highly dependent on family labor, with women and children supplying roughly half the labor in extraction and 91% of the effort in agriculture. The table reveals the large demands on female time, even as it excludes those activities associated with household reproduction. This was the context for the forest development of the various Brazil nut projects.

4.3. Decentralized Production and the Rural Economy Women viewed decentralized nut processing with considerable interest, because it appeared to provide a mechanism through which women and girls Table 3. Labor Time, Division of Labor and Returns to Labor. Task Days

Production Activities

% F+C

MD

Return TD

Return MD

Male MDs

Female MDs

Child MDs

4.96 1.85 – 2.48

2.4 1.13 – 1.52

98 117 41.5 40

28 12 37 140

75 96 63.5 245

296

217

479

Brazil nuts Rubber Land prep Agriculture

97 155 76 261

201 225 141 425

Total days

513

992

51 48 71 91

Extraction, Gender and Neoliberalism in the Western Amazon

275

could gain control of cash. Given the almost complete masculine control of funds from the sale of rubber and unprocessed nuts, and that most of their agriculture and household production labor is not compensated, this dimension of the domestic cracking was extremely attractive. However, things in the forest factories were not running smoothly. Productivity was much lower than that of the now closed central factory. Forest crackers were only able to process about 6 kg of nuts per day with an average work month of 11 days, rather than the predicted 20 days. This compared poorly with the 10 kg/day minimum over 20 days for the Xapuri factory. Attrition of workers was very high. Labor discipline was a real problem because of the range of demands on women’s time, the distances many had to walk to do the work and payment problems. The payment delays were a regular feature of the microusinas (as they were at Chico Mendes), which under the hyperinﬂation of 600% a year strongly devalued the effective purchasing power of the funds to those cracking, while lowering the costs of the processed nuts to the factory to practically zero – in essence reducing the labor value of the processed nuts to a fraction of its purported costs, an enormous site of surplus extraction. These payment delays could be extensive, in some cases as long as 3 months. The tables and seats were extremely uncomfortable for carrying out the tedious task, and backaches and neckaches were common. The central feature of the microusinas was their social impact. While viewed as a positive feature of the landscape at ﬁrst, and while interest in acquiring machines was initially high, the adult women soon ceased working in them. Intended for female labor, on the urban model, the usinas largely employ adolescents and children at the expense of their schooling. The poorly informed understanding of rural production economies and more specifically the demands on female labor completely underestimated the competing claims on women’s labor time. Ironically, it was the urban context where employment was most needed, because of the lack of subsistence and income alternatives.

4.4. The Opportunity Costs of Labor in Rural Areas While much conventional development theory is predicated on the idea of rural surplus labor, the Seringal economy operates in a context of labor scarcity. Women engage in extraction, agricultural production and household reproduction. What then are the comparative returns to the various activities compared with nut cracking? The miniusina was buying in

276

SUSANNA HECHT

unprocessed nuts at around $0.23/kg and selling the ﬁnished product to the cooperative at between Cr 950–1,500/kg. The payment to workers was $0.63/kg of ﬁnished nuts. The average production for the rural factory per worker was 6 kg of unprocessed nuts per day or of 2.1 kg of ﬁnished nuts per day, generating a return of between $1.32 and $2.00/day. Thus the monthly 11 days would generate a wage of less than $22/month, about half the minimum wage. What were the returns to the other activities that engaged women’s time? As a monthly return, cracking was the most poorly paid, followed by agriculture. Agricultural effort is essentially unpaid, so we determined its value by shadow pricing (Table 4). When one takes into consideration the domination of male labor in the ﬁrst three more lucrative activities, the late payment for the processed nuts, the fact that relatively less productive labor (young adolescents, physically incapacitated adults) is being siphoned into this activity when families are large is not surprising. Even so, household labor was regularly pulled into other tasks when needed, as the 11 workdays a month clearly reveal.

4.5. Mini and Domestic Nut Production The seringal family is hierarchical, with senior men and women given the power to organize production, to capture the returns that accrue to the family, and to make basic decisions for their allocation. In domestic production, the travel time to the main factory is avoided, and the processing can be more easily integrated into domestic tasks. For senior women able to deploy the labor of daughters-in-law and teenage girls, domestic and mini usina production – small factories of two or three cracking machines on a family compound – provided useful cash supplement; for the girls, of course,

Table 4.

Day labor Rubber Brazil nuts Agriculture Cracking a

Returns to Activities on the Seringal. Day (Per day)

Monthlya

2.06 1.13 2.40 1.52 1.32–2.00

41.00 (highly ephemeral) 24 40 54 11

This is calculated on the total returns to the activity using Table 3 as the baseline for extractive activities and agriculture, and CAEX data on decentralized cracking payments to the seringal.

Extraction, Gender and Neoliberalism in the Western Amazon

277

it was yet another set of activities they are obliged to do as household members, thus consolidating the wealth of the central family. In small families with young children, the added labor time in an already extremely burdened day made the domestic form of processing extremely oppressive, often adding as much as another 5 h to the very long workday. These were the effects for individual women and girls, but the larger program itself had implications for patterns of differentiation among the seringais themselves.

4.6. The Social Implications of Rural Decentralization and Differentiation The image of an unstratiﬁed seringal is a myth. During the rubber boom, the condition of workers was probably all roughly similar, but autonomous rubber-tapper communities are highly differentiated, reﬂecting patterns of life cycle as well as those of resource distribution and accumulation. While different household income strategies may evolve depending on resource endowments, available labor and other economic alternatives, there are also very sharp differences in wealth between households. This difference is expressed in the volume of products sold, control over the movement of commodities into and out of the seringal, and also in the number of share ‘‘croppers’’, contracted workers, and associated ‘‘agregados’’ and extended family members residing on the colocacao. In Cachoeira, 35% of the resident population on the colocacoes are agregados, meeiros and contract workers. That is, 35% of the population is functionally landless. In ER Cachoeira, there are sharp differences in production that are obscured by the use of averages. For example, the highest producing colocacao of rubber generated some 3,000 kg, while the other end of the spectrum includes one household which produced only 30 kg. About 10% of the producers sold 20% of the total production of the reserve. In the case of Brazil nuts, 43% of the production is derived from only 10 colocacoes. This lucrative activity is highly concentrated among a small number of households with the ability to capture this wealth and relies on the deployment of all forms of labor: meeiro, adjunto, and empreiteiro as well as able families are all applied to this activity. A quick comparison of the production data of extractive products at Fazendinha from households initially involved with the ﬁrst micro usina compared to the ER Cachoeira average is shown in Table 5. The table clearly documents the very high production of Brazil nuts in the mini- and micro usina households – double and triple the reserve average. The households of Nova Esperanca and Fazendinha included meeiros and

278

Table 5.

SUSANNA HECHT

Average Production of Extractive Goods at ER Cachoiera Compared with Beneﬁciaries of Projeto Castanha.

ER mean Fazendinha families (Microusina) Miniusina

Rubber

Nuts (latas)

668 558 575

351 603 957

Source: CAEX documents.

agregados. The mini usina households were characterized by agregados/ meeiros/resident relatives in all cases. The main beneﬁciaries were thus among the better off of the ER Cachoeira. There is a great deal of rationality to placing processing where there is product, but the net result of the decentralized intervention was to enhance differentiation of peasant households in the seringal and concentration of wealth. The location of decentralized processing followed a rationale that focused on areas which were deeply politically organized and loyal to CNS, and where some transport infrastructure already existed. Most of the workers at the processing plant were kin. The families who provided labor for the construction of the micro usina also came to dominate the ‘‘work stations’’ as well. Thus, in one rural factory, ﬁve of the 10 nut-cracking machines were initially dominated by one family, which also had considerable inﬂuence over who else could get access to the micro usina, and tended to distribute ‘‘position’’ along kin lines. While there is a great deal of consanguinity in the seringal, a cynic might argue that the micro usinas had a dimension of political patronage as well. More critically, the beneﬁts of the microusinas were highly concentrated in just two large, comparatively wealthy families, and the distribution of miniusinas followed a similar pattern. Thus, rather than broadly improving rural incomes, it tended to consolidate the gains very narrowly among comparatively wealthy families, most of whom were quite closely related.

4.7. The Decentralized Model and its Future Seringal life involves tremendous effort of all who live and work there to make ends meet and to assure the necessary income and subsistence. Women and children labor in all activities on the seringal, and in the activities that are not remunerated. For this reason the arrival of a source of cash income that could possibly be controlled by women excited a great deal of enthusiasm.

Extraction, Gender and Neoliberalism in the Western Amazon

279

This dimension of the cracking machines is very positive, and was the reason that the microusinas were initially viewed so favorably. However, the time demands and the pay rate do not compensate for the value of time and their obligations in other activities of production and household maintenance. As a consequence, women withdrew their labor. Decentralized production was promoted partially on the ideas of enhancing the use of the female rural labor force. As this chapter suggests, the strategy overburdens an already quite taxed workforce when the demands of all activities are taken into account. Participants in the project noted that payment was often late, transport of material irregular, and packing supplies were insufﬁcient. In addition, participants who are adults do understand that they are carrying out the same work as the factory but receiving no social beneﬁts. Workers in the project also had no security at all. Rural incomes were raised slightly, but these beneﬁts were concentrated. In terms of any broader linkages in the regional economy, the impact was very low. Ironically, this strategy was chosen in lieu of one where these activities had high positive impacts.

5. THE URBAN SOCIAL IMPACT The decentralization model that was promoted by the CAEX was predicated on the idea that the value of rural labor is zero and that there is a great deal of surplus labor, which was not the case. On the other hand, urban unemployment is extremely high throughout the Amazon (Browder & Godfrey, 1997). The reality is that, under the conditions of contemporary Amazonian development, rural to urban migration has been extremely high, and job creation in the formal as well as informal sectors is now stuttering to a halt. In villages such as Xapuri, jobs are given through the mayor’s ofﬁce, a few commercial stores, and some day work in agriculture as boias-frias and domestic work. For women, the economic opportunities were extremely limited. The Xapuri labor force was almost entirely composed of former forest dwellers forced by violence, abandonment or other personal catastrophe to the city, where they lived extremely modestly on the minimum salary, credit, garden produce, remittances from family members, taking in laundry, preparing snacks and pooling income. About 20% were single mothers, and there were also a few older men who had also ﬂed the forest during the land wars. Without question, this factory was seen as a major social good and became the largest employer in Xapuri. While its productivity was not

280

SUSANNA HECHT

competitive with factories in Bolivia, its working conditions and beneﬁts were not comparable either. However, this factory, while it operated, was capable of generating 100,000 t of Brazil nuts per year and became a symbol of progressive sustainable development with forest conservation. The social costs and lack of discipline were widely cited as the reason this unionized form of factory organization had to be ended. Barely had it begun to operate before schemes to dismantle the unionized labor force were afoot, a rather strange position for a group associated with the Workers’ Party to adopt. Finally, the capital for the purchase of Brazil nuts provided by the Austrian government was used to pay the social costs to the workers and they were all ﬁred. The cooperative maintained a ﬁction that, because they had spent a great deal of money buying nuts (and then, ironically, selling them to the Mutran family for processing), they were unable to pay for nut processing. Yet, they constantly led the workers on, explaining that they would open in a few weeks, that they would call every one to work next Monday, etc. This lack of honesty was ﬁnancially devastating to the female labor force, which waited to return to work. Accustomed to not working at the factory for 3 or 4 months a year, they had developed systems of credit with local merchants, and had devised numerous small-scale income earning strategies that were both transitory and poorly paid, but would maintain the household for 3 or 4 months until the cracking period started up. Keeping this labor force on a string had several very negative effects. First, no real economic plan at the level of the household could be sought, because it was thought that they would eventually return to work. Therefore, borrowing continued, purchasing ‘‘a ﬁado’’ proceeded, and ﬁnancial promises were made, all of which could not ultimately be kept. Second, the ability to hold livelihoods together between cracking periods was generally fairly limited. To do so for 3 or 4 months was possible, but any longer than that was quite difﬁcult. By May, 7 months after the golden handshake, some 15 workers had left Xapuri for points such as Rio Branco and Porto Velho. Four had returned to the Seringal. In essence, slightly more than 30% of its trained workforce had left the city. Since a 3-month training period had gone into worker preparation, this investment in human capital was lost. At the level of the economy of Xapuri, the nut crackers’ modest income was spread among the local micro enterprises which included the tiny barracks that provision the humble neighborhoods where these ladies resided, sellers of modest school supplies and children’s clothing, local seamstresses, babysitters, etc. The linkages to the regional economy were much wider than those characteristic of the seringal, where the returns to the cracking were largely spent on provisions from the Cooperative, thus concentrating

Extraction, Gender and Neoliberalism in the Western Amazon

281

expenditures rather narrowly. The monthly injection of U.S. $2,450 into the poorest strata of the economy had broad linkages with the modest businesses of the informal sector of Xapuri. These linkages were dispersed through the poorer sectors of Xapri’s economy and had a very positive impact among the poorest sectors of the urban economy, contributing to the beneficent image of the cooperative and CNS. The shift to the decentralized forest model had its allures, but were the women and the social beneﬁts paid to them really the source of the ﬁnancial problems with the cooperative, or were other factors at work? (Tables 4 and 5).

5.1. The Cost of Labor The promotion of the decentralized model was justiﬁed on three main grounds: worker productivity, quality and the factory costs. Table 6 reviews the comparative productivity of the workers. Much of the promotion of the decentralized model emphasized efﬁciency and mentioned that decentralized productivity exceeded that of the factory counterparts. Factory production per worker was 10.5 kg/day or 210 kg/ month. The decentralized processing operation in the forest, however, was far less productive than the urban crackers; total per worker productivity was only 28% of urban workers. Twenty three workers – roughly equal to half the CM workforce – produced only one-tenth the total commercial product. Of the data on worker productivity provided by ECOTEC (1993),

Table 6.

Worker Productivity in Decentralized and Centralized Processing.

No. of Workers Total Annual Production (ﬁnished nuts) (kg/worker) Dailyc Monthlyd (kg/worker) Annual Production per Worker (ﬁnished nuts) a

Microusinasa

Chico Mendesb

23 9,340 6,7 70 406

48 95,760 10–16 249 1,995

Data from Ferreira (1993). Data from CAEX Annual Report (1993). c Unshelled nuts. d Calculated by months per year operation. Microusina only operate for 6 months. CM is calculated on an 8-month base. b

282

SUSANNA HECHT

the consultant group that developed the decentralized processing system, only one worker in 1 month ever reached 10 kg/day, the production base of the factory.

5.2. Costs and Accountability In the decentralized system, the monthly cost of the managers and helpers was $823, while workers received $460, for an average monthly production of 1550 kg of ﬁnished nuts, calculated on a 6-month operation. This gives wage cost per kg of nut of 0.83 in the decentralized system. In the urban factory, the wage bill for crackers plus beneﬁts was generally around $4,000/ month, usually between 5 and 10% of the total factory cost. With a monthly production (calculated on an 8-month production cycle because of the great volume of nuts and storage) of some 11,970 kg, or a wage cost of 0.33/kg, the CM factory workers delivered ﬁnished nuts at 50 cents less than the per kg cost of the forest factories. More germane, however, was that the factory itself was extremely costly because wages and beneﬁts paid to cooperative staff were charged as overhead onto the factory even when it was no longer a site of processing, and other analysts have raised concerns about the ‘‘management’’ of the factory and the deﬂection of funds into political campaigns. These data suggest that the factory was carrying a wide array of costs that had little to do with the production and commercialization of Brazil nuts. Rather, the factory was a cash cow, and when its inefﬁciency was noted, the approach was simply to extract the only productive element from it: the processors.

6. CONCLUSION The number of alluring elements (sustainability, jobs for women, decentralization, cooperatives, forest conservation, etc.) made the project a sure sell. Coupled with the history of Chico Mendes and the rich rhetoric about the new forest development model, the castanha project was a foundation ofﬁcer’s dream. Unfortunately, what was to be a job-creating enterprise ended up destroying more jobs than it created, substituting a less productive for a more productive system, and exchanging adult labor for those of children as women withdrew their labor from the activity. The project foundered due to gender mythologies about extractive economies, deep

Extraction, Gender and Neoliberalism in the Western Amazon

283

misunderstanding of women’s role in rural production, and a careless adoption of neoliberal decentralization approaches to forest economies.

NOTES 1. Henry Pearson was the Editor of the indispensable journal of the rubber era ‘‘India Rubber News’’ and traveled throughout the Amazon. 2. O Liberal, the main newspaper of the state of Para, reported on February 12, 2003 that 212 quilombos detaining some 350,000 ha still exist today in the state of Para. 3. For example, the statements about the policy of alliances in the Third National meeting of CNS in 1992, emphasized alliances with unions, politicians and delegates who defend the interests of CNS, global workers movements, popular movements and the women’s movement. 4. Initially, the decentralization had hoped to widely contract to local households.

REFERENCES Agassiz, J. L. R. (1868). A journey in Brazil. Boston: Houghton-Mifﬂin. Almeida, M. (1992). Rubber tappers of the upper jurua river. Ph.D. dissertation, Cambridge University, Cambridge. Anderson, A. (1988). Use and management of native forests dominated by the Acai palm. Advances in Economic Botany, 6, 144–154. Anderson, A., & Ioris, E. (1992). Valuing the rain forest: Economic activities by forest extractivists in the Amazon estuary. Human Ecology, 20, 337–369. Anderson, A., Balick, M., & May, P. (1991). The subsidy from nature. New York: New York Botanical Garden. Barham, B., & Coomes, O. (1994). Reinterpreting the Amazon rubber boom. Latin American Research Review, 29(2), 73–109. Browder, J. (1992). Limits to extractivism. Bioscience, 42, 174–181. Browder, J., & Godfrey, B. (1997). Rainforest cities. New York: Columbia University Press. Brown, K., & Rosendo, S. (2000). Environmentalists, rubber tappers and empowerment: The politics and economics of extractive reserves. Development and Change, 31(1), 201–227. Bunker, S. (1984). Modes of extraction, unequal exchange, and the progressive underdevelopment of an extreme periphery: The Brazilian Amazon, 1600–1980. American Journal of Sociology, 89, 1017–1064. Bunker, S. (1985). Underdeveloping the Amazon: Extraction, unequal exchange and the failure of the modern state. Urbana: University of Illinois Press. CAEX (Cooperativa Agro-Extractivista Xapuri). (1993). Annual Report. Xapuri: Brazil. Campbell, C. (1996). Out on the front lines but still struggling for voice. In: D. Rocheleau, B. Thomas-Slayter & E. Wangari (Eds), Feminist Political Ecology (pp. 27–62). New York: Routledge. Cardoso, C. (2002). Extractive reserves in the Brazilian Amazon. Burlington: Ashgate Press.

284

SUSANNA HECHT

Casement, R. (1911). Inferno in a paradise. Daily News. In: Angus Mitchell (Ed.), Sir Roger Casement’s heart of darkness: The 1911 documents (pp. 534–535). Dublin: Irish Manuscripts Commission, Government of Ireland. Casement, R. (1912). The Putumayo Indians. Contemporary Review, 102, 317–328. CIDA (Canadian International Development Agency). (1992). Levantamento Socio Economico da Reserva Extractivista Chico Mendes e Proyectos de Assentamentos Extractivistas da Regiao do Vale do Acre. Rio Branco: CNS/FUNTAC/CIDA. Clay, J. (1997). The impact of palm heart harvesting in the Amazon estuary. In: C. Freese (Ed.), Harvesting wild species: Implications for biodiversity conservation (pp. 283–314). Baltimore: Johns Hopkins University Press. Cleary, D. (1998). Lost altogether to the Civilized World. Comparative Studies in Society and History, 40(1), 109–135. CNS (Conselho Nacional de Seringueiros). (1988). Manfesto de Povos da Floresta. Brasilia: CNS. CNS (Conselho Nacional dos Seringuieros). (1988). Relatorio Socio-Economico e Cadastro da Reserva Extractivista da Chico Mendes. Rio Branco: CNS. Da Cunha, E. (1906). A margins da historia. Rio de Janiero: Lello Brasiliera. Dean, W. (1987). Brazil and the struggle for rubber. Cambridge: Cambridge University press. De Rego, J. F. (1999). Amazonia: De extrativismo a new extrativismo. Ciencia Hoje, 25(146), 62–65. Deere, C. D. (2001). Empowering women: Land and property rights in Latin America. Pittsburgh: University of Pittsburgh Press. Deere, C. D., & Leo´n, M. (1987). Rural women and state policy: Feminist perspectives on Latin American agricultural development. Boulder: Westview. ECOTEC. (1993). Projecto Castanha. Recife: ECOTEC. Edwards, H. E. (1847). The River Amazon. London: Murray. Escobal, J., & Aldana, U. (2003). Are nontimber forest products the antidote to rainforest degradation? Brazil nut extraction in the Madre de Dios. World Development, 31(11), 1873–1887. Facao, E. (1906). Album do Rio Acre. Belem: E. Falcao. Ferreira, C. A. (1993). ECOTEC Report. Porto Velho, Rondonia. Freese, C. (Ed.) (1997). Harvesting wild species: Implications for biodiversity conservation. Baltimore: Johns Hopkins Press. Gomes, C. (2001). Dynamics of land use in an Amazonian Extractive Reserve: The case of Chico Mendes extractive reserve. MA thesis, University of Florida, Gainesville, FL. Hardenburg, W. E. (1905). The Putomayo. London: T.F. Unwin. Hecht, S. B. (1992). Extractive communities, biodiversity and gender Issues in Amazonia. In: L. Borkenhagen & A. Abromowitz (Eds), Proceedings of an International Conference on Women and Biodiversity. Washington, DC: WRI. Hecht, S. B. (2005). Forests ﬁelds and family: Gender and production in extractive reserves. Development and Change, in press. Hecht, S. B., Anderson, A., & May, P. (1988). The subsidy from nature. Human Organization, 47(1), 25–35. Hecht, S. B., & Cockburn, A. (1989). The fate of the forest. London: Verso. Homma, A. (1993). Extractivismo vegetal na Amazonia: Limites e Opportunidades. Brasilia: Embrapa. IDE (International Development Enterprise). (1991). Proposal for Decentralized Processing.

Extraction, Gender and Neoliberalism in the Western Amazon

285

Kainer, K., & Duryea, D. (1992). Tapping women’s knowledge. Economic Botany, 46(4), 408–425. Kainer, K., Schmink, M., Leite, A., & Fadell, M. (2003). Experiments in forest-based development in Western Amazonia. Society and Natural Resources, 16, 869–886. Nepstad, D., & Schwartzman, S. (1992). Non timber products from tropical forests. Advances in Economic Botany (Vol. 9). The Bronx: New York Botanical Garden. Nugent, S. (1994). Amazon Caboclo Society. London: Berg. Nugent, S. (2002). Whither o Campesinato? Historical peasantries of the Brazilian Amazon. Journal of Peasant Studies, 22(11), 162–189. Padoch, C. (1992). Marketing of non-timber products in the western Amazon. Advances in Economic Botany, 9, 115–129. Padoch, C., Ayers, M., & Pinedo, M. (Eds) (1999). Varzea. The Bronx: New York Botanical Garden. Pearson, H. (1911). In the rubber country of the Amazon. New York: India Rubber World. Rafﬂes, H. (2002). In Amazonia. Princeton: Princeton University Press. Rafﬂes, H., & Winklerprins, A. (2003). Further reﬂections on Amazonian environmental history: Transformations of rivers and streams. Latin American Research Review. Rangel, A. (1927). Inferno Verde. Tours, Arrault. Rocha de Almeida, C. (1996). Imagems negras, espelhos broncos: Um estudo das mulheres negras no ﬁm do XIX em Belem. In: M. L. Alvares & M. A. D’Incao (Eds), A Mulher Existe? (pp. 27–40). Belem: Museu Goeldi. Salafsky, N., Dugelby, B., & Terborgh, J. (1993). Can extractive reserves save the rain forest? Conservation Biology, 7(1), 39–52. Schroth, G., Coutinho, P., Moraes, V. H. F., & Albernaz, A. L. (2003). Rubber agroforests at the Tapajoz river: Environmentally benign land use systems in an old forest frontier region. Agricultural systems and Environment, 97(1/3), 151–165. Scott, J. (1976). The moral economy of the peasant. New Haven, CT: Yale University Press. Scott, J. (1997). Weapons of the weak. New Haven: Yale University Press. Simonian, L. (1996). Mulheres seringueiras na Amazonia: Uma vida silenciada. In: M. L. Alvares & M. A. D’Incao (Eds), A Mulher Existe? (pp. 97–117). Belem: Museu Goeldi. Stanﬁeld, M. (1998). Red rubber, bleeding trees: Violence slavery and empire in the Northwest Amazon 1850–1933. Albuquerque: University of New Mexico Press. Taussig, M. (1987). Shamanism, colonialism and the wildman. Chicago: University of Chicago Press. Thapa, K., Bilborrow, R., & Murphy, L. (1996). Deforestation, land use and women’s agricultural activities in the Ecuadorian Amazon. World Development, 24(8), 1317–1332. Tocantins, L. (1979). A formacao historica de Acre. Rio de Janeiro: Civilizacao Brasileira. Trinidade, J. R. (1996). Mulheres de Ma Vida: Infeis e disordeiras em Belem 1890–1905. In: M. L. Alvares & M. A. D’Incao (Eds), A Mulher Existe? (pp. 41–51). Belem: Museu Goeldi. UCLA/IEA/FUNTAC. (1995). Survey of Cachoeira and Sao Luis Remanso. Weinstein, B. (1983). The Amazon rubber boom. Palo Alto: Stanford University Press. Wolff, C. (1999). Mulheres da Floresta. Sao Paulo: Huicitec.

MATERIAL PROCESS AND INDUSTRIAL ARCHITECTURE: INNOVATION ON THE CUBAN SUGAR FRONTIER, 1818–1857 Dale Tomich ABSTRACT While scholars have commonly inquired into how capital structures the material world, far less attention has been paid to how the material world has structured the historical relations of the capitalist world economy. This chapter is concerned with the expansion of Caribbean sugar industry in the world economic conjuncture of the first half of the nineteenth century. It examines the relation of the material requirements of sugar production, regional geography, and productive space. The ability of planters in particular locations to respond to world economic conditions was subject to material and spatial constraints. Increased output and technological innovation were dependent on the creation of new productive spaces – including both the formation of new commodity frontiers and the reconstitution the sugar plantation – that conformed to the changing requirements of sugar manufacture. Thus, the spatial and material conditions of staple production shaped the pattern of accumulation and political economic development. Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 287–307 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10013-4

287

288

DALE TOMICH

1. INTRODUCTION While scholars have commonly inquired into how capital structures the material world, far less attention has been paid to how the material world has structured the historical relations of the capitalist world economy. Authors writing on the expansion of the world economy during the nineteenth century, such as Karl Polanyi (1957), Giovanni Arrighi (1994), Immanuel Wallerstein (1984, 1989), and David Harvey (1982, esp. 373–412; 2001, pp. 237–266, 284–311), have largely focused on the political, commercial, and ﬁnancial aspects of the formation of British hegemony. The material expansion of the world economy – the greater volume and diversity of goods in circulation, hence the extension and intensiﬁcation of material production and their consequences – has been far less central to their inquiries. This chapter is concerned with the expansion of Caribbean sugar industry in the world economic conjuncture of the ﬁrst half of the nineteenth century. It examines the relation of the material requirements of sugar production, regional geography, and productive space. The ability of planters in particular locations to respond to world economic conditions was subject to material and spatial constraints. Increased output and technological innovation were dependent on the creation of new productive spaces – including both the formation of new commodity frontiers (Moore, 2000) and the reconstitution the sugar plantation – that conformed to the changing requirements of sugar manufacture. Thus, the spatial and material conditions of staple production shaped the pattern of accumulation and political economic development. World sugar production experienced one of the most profound cycles of expansion and transformation in its history during the ﬁrst half of the nineteenth century. Even as sugar gave way to cotton as the leading commodity in world trade, production increased at an accelerating rate during this period. New sugar producing regions emerged and old ones declined. The appearance of beet sugar and new varieties of cane sugar and the development of new milling and reﬁning technologies altered production processes. Labor was reorganized globally as slavery was abolished in some places and intensiﬁed in others while new modes of labor control such as indenture and sharecropping were instituted (Tomich, 1990, pp. 14–32). Perhaps the most dramatic development in this process was the unprecedented level and rate of growth of production achieved by the Cuban sugar industry. By 1827, Cuba emerged as the world’s leading sugar producer, and its output doubled each decade from the 1820s until the 1860s. The Cuban sugar industry’s rise was achieved through the incorporation of new lands

Material Process and Industrial Architecture

289

for sugar cultivation, the massive redeployment of slave labor, and the successful adoption of new reﬁning and milling technologies. This recombination of land, labor, and technology in Cuba increasingly transformed sugarmaking from an artisanal to an industrial process dependent upon modern science and technology and reconﬁgured the Cuban ingenio (sugar mill) as a productive enterprise. The transformation of the material processes and social organization of sugar production in Cuba created a profound rupture in the pattern of historical development of the world sugar industry and imposed on it new spatial scales and temporal rhythms (Tomich, 2004, pp. 75–94). This transformation of the Caribbean sugar industry was inextricably part of the processes reorganizing the capitalist world economy. Eric Hobsbawm (1969, pp. 134–153) has argued that industrialization and British domination of world trade during the nineteenth century divided the world between an industrialized core and a food and raw material producing periphery. Implicit in this division is the redefinition of the relation between agriculture and industry on a world scale. This redefinition is nowhere more evident than in the case of tropical agricultural staples. Agriculture is less susceptible to human intervention than industry. The cultivation of sugar, like other tropical staples, is subject to natural constraints that create spatial and temporal limits to production (Bunker, 1994). Sugar can only be grown in certain places. The scale of production depends upon location, the extent of cultivable lands, soil fertility, climate, the duration of the crop cycle, and the nature of processing. A hectare of land will only yield so many kilos of cane, and so many kilos of cane will only yield so much sucrose. Chemical or mechanical techniques may be employed to manipulate these constraints in order to increase productivity, but they cannot be transcended in any absolute sense. A determined amount of sugar can only be produced in a given space in a determined time. Consequently, in agriculture, there are natural limits on expansion of output and heightened productivity. The turnover time of capital invested in the production of sugar and other agricultural staples is subject to the temporal–spatial constraints presented by location, the ﬁxed growing season, the scale of production, and the rate of yield. Under such conditions, the acceleration and intensiﬁcation of the division of labor can most readily be achieved by the creation of new, more extensive zones of staple production. However, while such new zones produce staples on a greater scale, the relation of bulk, value, and distance alter the conditions of production (Bunker, 1994). New sources of supply are often located farther from markets than older ones, but in any case with the increased scale of production, a greater mass of material must be transported

290

DALE TOMICH

across longer distances. (This problem is accentuated in the case of commodities like sugar because the material must be transported immediately after the harvest season to prevent deterioration of the product. Transportation requirements are thus more difﬁcult and the costs greater because of the problem of unused capacity. Transportation demand is highly seasonal and outgoing cargoes are much greater than incoming cargoes [Innis, 1995, pp. 24–34, 139–154].) Thus, new production zones cannot simply replicate the old zones. Locational rigidity and the nature of the crop often result in transportation costs accounting for a greater proportion of the price of the staple (Bunker, 1994). The material and spatial expansion of staple production thus creates systemic pressure to reduce the costs of production and transport and to increase the productivity of labor. It is frequently accompanied by attempts to achieve economies of scale and to adapt new technologies of production and transport that can only be fully realized on the extended space of the commodity frontier. Consequently, the geographical expansion and increased scale of staple production necessarily entails a process of spatial and temporal differentiation that transforms the material and spatial division of labor (Tomich, 2004, pp. 95–136). This chapter emphasizes the importance of nineteenth-century Cuba as a sugar frontier (Moore, 2000) for the restructuring of production and adoption of new technologies as well as for the spatial reorganization of the ingenio itself. It thus seeks to disclose the importance of spatial organization in accounting for the unprecedented development of the Cuban sugar industry and the transformation of world sugar production. Many authors (for example Braverman, 1974; Gorz, 1978; Buroway, 1979; Coriat, 1979) have discussed technological innovation and work place organization in terms of labor discipline and exploitation of the labor force. While the social control of labor is certainly fundamental to technological change and workplace organization (Tomich, 1990, pp. 214–258), the focus here is on the ways that the material processes of sugar production are spatially organized, and, in turn, the ways that the spatial conﬁguration of the plantation impedes or facilitates the adoption of new technologies. (Of course, the reconstitution of productive space also creates new temporal rhythms of production.) The chapter thus calls attention to the spatially and temporally uneven development of Caribbean sugar production. This interpretation of the spatial and technological reorganization of the sugar plantation is based on an analysis of visual sources. William Clark’s lithographs, published in his Ten Views in the Island of Antiqua (London, 1823), depict all the processes of sugar making before the introduction of mechanization and steam power. Los Ingenios, published jointly by Eduardo

Material Process and Industrial Architecture

291

Laplante and Justo Cantero in 1857, documents the Cuban new sugar zone at the moment of its transformation. More particularly, the lithographs made by Laplante for the book visually represent the spatial articulation of material processes of production and the architecture of the slave plantation on the Cuban sugar frontier. Laplante’s images were made in the region around Havana, Matanzas, and Trinidad in the midst of the nineteenthcentury sugar boom. According to Manuel Moreno Fraginals, the preeminent historian of the Cuban sugar industry, Los Ingenios offers extremely valuable information on the largest sugar mills in Cuba in the 1850s, and the plates are beyond reproach from a technical point of view because the meticulous attention to detail with which the machinery is reproduced (Fraginals, 1978, III, pp. 189–190).

2. SUGAR MAKING AS MATERIAL PROCESS As is well known, the physical and chemical properties of sugar cane are such that sugar production requires a series of coordinated agricultural, mechanical, and physical–chemical operations. Each step in the process – planting, cultivating, harvesting, grinding, evaporation, and crystallization – depends upon the one preceding, and they all must be performed in the proper sequence for the ﬁnal product to be obtained. Each step entails a physical transformation of the material from plant to liquid to crystallized solid (hence a change in form and mass of the material to be treated). Further, the physical characteristics of sugar cane impose a temporal condition on the manufacturing process. Each cane stalk must be cut when it is ripe and converted into sugar within hours after it is cut or the quantity and quality of the ﬁnal product deteriorates. The sucrose content of the plant forms the natural limit of production. The varieties of types and grades of sugar differ from one another only in chemical purity and degree of crystallization. The more perfectly and completely the sucrose contained in the plant is extracted and processed, the greater is the quantity and quality of the sugar produced. The more technically perfect the process, the more homogenous and pure the product. Thus, from the perspective of material production, the history of sugar may be understood as the history of the development of the means to more fully extract and convert the sucrose contained in the plant into sugar. These material conditions of sugar production impart a distinctive temporal rhythm to the labor process and shape the industrial architecture of the sugar plantation. The interdependence of the steps in the manufacturing

292

DALE TOMICH

process implies not only a qualitative relation but also a quantitative relation between the various sectors of production. The amount of cane planted and harvested, the capacity of the mill to grind the cane, and the ability of the reﬁnery to convert a given amount of cane syrup into crystallized sugar within a given period of time must all be synchronized with one another for production to be successful. Large-scale commercial production further requires that these operations be performed simultaneously. This entails division of labor, specialization of workers, coordination of activities within each sector and between sectors, and concern for quantiﬁcation and measurement, speed, efﬁciency, and economy of time. Sugar manufacture is, then, a continuous process, and its constituent steps are necessarily consecutive in time and contiguous in space. The material interdependence of these processes establishes parameters for the organization of production and innovation. Planting more cane, building a more powerful mill, or increasing the capacity of the boiling house does not necessarily increase output or productivity unless analogous changes are made in the other sectors. Each step in the sequence – the relation of cane ﬁelds to mill, of mill to boiling house, and of boiling house to reﬁnery – represents a potential constraint in the transformation of the production process. The integration of these diverse operations into a continuous process creates a series of potential thresholds and bottlenecks in sugar manufacturing. Improvements or increases in output can be introduced in any of the sectors until they reach a threshold that puts pressure on other sectors and creates a bottleneck. Changes in technique cannot be introduced randomly at any point in the sequence, and no single innovation, for example the introduction of steam-power, will necessarily transform the manufacturing process. Rather, to be effective, technological changes must be implemented under conditions that maintain the proportional relation among sectors. From this perspective, we can see why innovation is likely to be discontinuous and that there are diverse solutions to the problems of sugar manufacture. Each sector of production represents a potential bottleneck. Change in any particular sector represents a partial transformation of the overall process that creates new conditions and limits with different potential consequences the other steps in the process. There is no technical necessity for adopting any particular technique. Rather, what techniques are adopted and what resources are allocated depend on a variety of geographical, technical, economic, social, and political conditions (Tomich, 1990, p. 139). The same change may have different consequences in different circumstances. Nonetheless, despite the contingent character of innovation, the

Material Process and Industrial Architecture

293

characteristics of sugar and the material interdependence of the production process present a system of constraints within which sugar technology and plantation organization historically develop. They govern the proportional relation among the various sectors and the spatial distribution of productive activities and allow us to comprehend the social–economic and technical– spatial transformations of sugar manufacture even as its particular material and technical content changes.

3. THE MODEL SUGAR PLANTATION, 1750–1815 The second half of the eighteenth century was the golden age of sugar production in the French and British Caribbean. Sugar was the leading commodity in international trade, and the islands of the West Indies were its leading producers. Avalle’s idealized plantation (see Image 1), drawn in Saint Domingue in 1796, represents the optimal development of the Caribbean sugar plantation before the introduction of steam-power, machine production, and the application of modern chemistry and physics to sugar manufacture (Watts, 1987, pp. 382–447). It ﬁxes a social–spatial model ordered by the functional requirements of sugar manufacture. Each of the phases of the agricultural-manufacturing process is integrated into a single productive unit. The interdependence of ﬁeld and factory shaped the scale and location of plantation activities. In this model, the ﬁelds are organized in relation to the capacity of the mill and boiling house and indicate the rationalization of economic space. The cane ﬁelds are divided up into regular sections that are calculated to yield a determined amount of cane that can be processed by the mill and boiling house in a given amount of time. Further, plantings were manipulated so that cane in different sections would ripen sequentially over months. In this way, it could be harvested and processed when sucrose content was at its peak and continuity of peak production could be maintained throughout the harvest season. The mill and boiling house are located in the center of the ﬁelds in order to facilitate transportation of the cut cane by ox-cart. Thus, cane cutting is synchronized with milling and reﬁning both in order to secure the maximum amount of sugar possible and to optimally utilize the productive capacity of the mill and reﬁnery and hence secure the maximum return on the capital invested in them (Watts, 1987, pp. 382–447; Tomich, 1990, pp. 139–188). Such a plantation typically produced from 150 to 300 metric tons of sugar annually. (According to Manuel Moreno Fraginals, the vertical mill was

294

DALE TOMICH

Image 1. Avalle’s (1796) Idealized Sugar Plantation. Source: Watts (1987, p. 390). Redrawn from Noel Deer, The History of Sugar, vol. 2 (London: Chapman & Hall, 1950).

Material Process and Industrial Architecture

295

only capable of producing 170 tons of sugar no matter what power source was used [Fraginals, 1978, I, p. 203]. Therefore, we may presume that plantations that produced 300 tons had two mills. This practice was not uncommon on large plantations, though most had only one mill.) It possessed 300 Ha of land, of which 120 Ha were planted in sugar, and had a population of 180–200 slaves. Typically, about half of the slaves were employed in sugar production. The plantation’s size and productive capacity were limited by the capacity of milling technology, based on wind, water, or animal power, and the open-kettle-reﬁning process (Tomich, 1990, pp. 139–188). The type of mill predominantly in use on such plantations consisted of three vertically mounted wooden rollers that were powered by animal, wind, or water traction (see Image 2). By the beginning of the nineteenth century, steam-power was also applied to supply the motive force. The three-cylinder vertical mill provided a relatively cheap and simple means to grind cane. Its

Image 2. A Mill-Yard. Source: William Clark, Ten Views in the Island of Antiqua, in which are represented the Process of Sugar-Making, and the Employment of the Negroes in the Field, BoilingHouse, and Distillery (London, 1823). Reproduced with the permission from the Caribbean Foundation (London).

296

DALE TOMICH

adaptability to various power sources allowed Caribbean planters to adapt their operations to a variety of circumstances and freed them from dependence on watercourses for the location of their mills. The rollers were usually made of wood, although by the end of the century they were often covered with iron shells to increase their durability and hardness. The three cylinders made it easier to pass the cane stalks through the mill more than once and increased the quantity of juice that could be extracted from a given amount of cane. Further, the vertical arrangement of the cylinders made the mill easier to turn. However, the mills were often poorly constructed and maintained. They were subject to uneven pressure and wear, and they were easily put out of alignment and their wooden parts worn out or broken. Even in the best of circumstances, such mills did not exert sufﬁcient pressure to efﬁciently extract the juice from the cane. It was difﬁcult to regulate the speed of the mill, and to adjust the pressure exerted by the rollers. If the rollers were too far apart, not enough juice was extracted; if they were too close, it was difﬁcult to turn the mill. Indeed, with the widespread adoption of the richer but thick and woody Otaiti cane in the 1780s, all-wood mills could not extract the juice and their shafts and gears frequently broke. Perhaps most importantly, the vertical placement of the rollers meant that only the lower portion of the grinding surface was utilized. No matter what power source was used, including steam-power, this type of mill could only grind 600 l of juice per hour because only a portion of the grinding surface was utilized (Tomich, 1990, pp. 150–171). (Thus, the introduction of Otaiti cane created higher yield of juice from the cane – as well as providing a source of fuel for the boilers – but its potential effects could not be fully realized without transforming the mill.) The problems of this type of mill were partially alleviated by the gradual adoption of metal parts: ﬁrst iron shells on the grinding cylinders, then metal gears, and ﬁnally by the 1820s, the introduction of all metal construction. A new type of horizontal mill also appears to have been developed in Jamaica as early as 1754 and was used in Saint Domingue in the 1790s. In this type of mill, the three rollers were arranged in an isosceles triangle. The full grinding surface could be utilized, and each cane stalk was automatically ground twice as it passed through the mechanism. A spring and weight mechanism allowed the rollers to yield when an irregular piece of cane passed through the mill, and even pressure was exerted on the cane stalks. By the 1820s all metal steam-powered horizontal mills began to appear throughout the Caribbean (Tomich, 1990, pp. 150–171). They were effective innovations on existing estates in old sugar regions because they more fully extracted the juice from the cane and provided more raw material for the reﬁnery. However, to fully

Material Process and Industrial Architecture

297

take advantage of them required increasing the area cultivated. Such increases could only be achieved with great difﬁculty, if at all, in established sugar zones. Of course, the greater quantity of juice extracted from the cane, also raised the question of the capacity of the reﬁnery to process it. As we shall see, these technical possibilities could be fully exploited in a sugar frontier region such as Cuba, where it invited a number of solutions to the reorganization of the production process. However, they had limited impact in the established sugar zones (Tomich, 2004, pp. 75–136). After the juice was extracted from the cane it passed through an open gutter from the mill to the reﬁnery. There it underwent a series of operations intended to remove physical and chemical impurities before being converted into sugar. In the boiling house, the juice was clariﬁed, evaporated, and progressively concentrated in a sequence of successive boiling and skimming operations carried out in a series of kettles of decreasing size (see Image 3).

Image 3. A Boiling-House. Source: William Clark, Ten Views in the Island of Antiqua, in which are represented the Process of Sugar-Making, and the Employment of the Negroes in the Field, BoilingHouse, and Distillery (London, 1823). Reproduced with the permission from the Caribbean Foundation (London).

298

DALE TOMICH

The juice was continuously evaporated and passed from one kettle to the next until the syrup was ready to crystallize. When the sugar was ready to ‘‘strike’’ it was ladled into cooling vats and then barrels or clay pots so that the boiled down liquid could crystallize into raw sugar. The barrels or pots were then taken to the curing house where the crystals were separated from the molasses. The separation of sugar from molasses was a slow and difﬁcult process that could take 3, 4 weeks or more. The presence of molasses with other impurities impeded crystallization and reduced the quality of sugar. It generally resulted in the production of semi-reﬁned brown or moscovado sugar while the high proportion of molasses that was a by-product of this inefﬁcient process made rum manufacture an important secondary activity. The capacity of the reﬁnery depended upon the boiling surface of the cane juice and was determined by the number of kettles. It was increased by simply adding more kettles. Multiplication of the kettles increased the scale and continuity of the process. It developed reﬁning as a large-scale continuous manufacturing process and enhanced the necessity of coordinating the movement of material and labor throughout the reﬁning process. There was no standard arrangement of kettles. The number, size, and disposition of the kettles varied from plantation to plantation, often due to the preferences of the master sugar maker. Four or ﬁve kettles were commonly used in a single set or train, but sometimes there were as many as six or seven. The kettles were divided between those in which the syrup was concentrated and those in which the sugar was cooked or struck. They decreased in size from beginning to end of the process. As the syrup boiled, impurities bubbled to the surface and were continuously skimmed off with paddles. When the syrup was ready for the next step, it was ladled into the next kettle in the train until it was ready to strike. Open-kettle boiling was an imperfect method of sugar making. Throughout the reﬁning process, the cane juice was exposed to open air, heat, humidity, dirt, and various other impurities that prejudiced the quantity and quality of the ﬁnal product. The very operation of skimming off the impurities with a paddle mixed them back into the juice, while ladling the syrup from one kettle to another inevitably mixed sugar at different stages of concentration in the same kettle. Subjective judgment decided each step of the process, and the craft kill of the sugar master (often a slave) was crucial to success of the operation (Tomich, 1990, pp. 163–188). By the second half of the eighteenth century, the kettles were arranged in what was known as a Jamaica or English train and heated by ﬁre from a single furnace. The heat passed through a ﬂue that ran beneath the kettles. The Jamaica train simpliﬁed work in the furnace and improved fuel

Material Process and Industrial Architecture

299

economy both because of the single ﬁre and because it permitted cane stalks or bagasse to be used as fuel. Fuel conservation was particularly important because of widespread deforestation created by the sugar industry. However, more often than not, the design and construction of furnace was inefﬁcient. It was difﬁcult to regulate heat even in the best of circumstances. Further, the kettles themselves were not uniform. They responded to the heat differently, and it was difﬁcult to coordinate the operations in each kettle with one another. Nonetheless, the process had to remain continuous. The kettles were constantly exposed to heat and could overheat and crack if left empty. The process was imprecise and it was difﬁcult to coordinate operations from one kettle to the next. Too much juice meant delays in the reﬁning process, which could cause the syrup to ferment. Too little juice could carmelize the sugar and damage the kettles (Tomich, 1990, pp. 163–171). The type of plantation represented by Avalle’s ‘‘idealized model’’ predominated in the Caribbean from the 1720s until the 1830s. Throughout this period various technical innovations were made within its framework (Watts, 1987, pp. 282–447). However, the interdependence of the different sectors of production inhibited major breakthroughs in productivity or technique. Once established on a particular technical–spatial scale, it was difﬁcult to recombine factors of production on a new scale. Changes in any one sector were checked by its dependence on the other sectors. Paradoxically, changes within the prevailing framework integrated more tightly their interdependence and had the effect of establishing a relative equilibrium among them. The technological transformation of the Caribbean sugar industry required new spatial conditions.

4. CUBA: TECHNOLOGICAL INNOVATION AND THE SUGAR FRONTIER The expansion of the world sugar market during the nineteenth century transformed the industrial organization of the sugar mill and plantation architecture. Nowhere was this more apparent than in the extensive space of the Cuban sugar frontier. The Cuban sugar industry expanded rapidly between 1790 and 1820s. By the end of the decade of the 1820s it emerged as the world’s largest sugar producer, and sugar production there increased at an accelerating rate into the 1860s. The growth of the Cuban sugar industry rested not simply on the construction of more plantations on more land, but

300

DALE TOMICH

on the radical reconﬁguration of the plantation and the incorporation of new sugar technologies in ways that were not possible in the old sugarproducing centers (Tomich, 2004, pp. 95–136). Laplante’s lithographs allow us to see the formation of this new productive space of the Cuban sugar mill. In the colonies of the British and French Caribbean geography and transportation costs generally conﬁned sugar plantations to the coastal lowlands. Plantation zones in the old colonies had been established since at least the beginning of the eighteenth century, if not before, and there was little room for reorganization of productive units. By the same token Cuba’s broad, fertile plains and ideal conditions for sugar cultivation could not be fully exploited without cheap overland transportation capable of opening the interior of the island to the bulky product. The early growth of the Cuban sugar plantations was conﬁned to the region around Havana and Matanzas, where access to the ports was available. However the development of the railroad in 1827 allowed not only the geographical expansion of Cuban sugar production, but also new combinations of land, slave labor, and sugar technology on larger scales and within new technical divisions of labor. Eduardo Laplante’s lithographs document the spatial and technological restructuring of the Cuban sugar mill. The Ingenio Trinidad (Image 4) was built in 1833, the period that Cuba emerged as the world’s leading sugar producer. The property of Esteban Santa Cruz Oviedo, it was located on the Sabanilla railroad line in the province of Matanzas. It possessed about 550 Ha of land when it was founded. Then it expanded to 725 Ha by 1857. Of these 604 Ha were planted in sugar (including 94 Ha that were rented from a neighboring property). It had a slave population of 1,000, of which 300 were ‘‘criollos.’’ In 1855, it produced 1,667 metric tons of sugar. (The size of the slave population was out of proportion with the need for labor on Trinidad. According to Justo Cantero, who supplied the text for Los Ingenios, Santa Cruz was the most famous ‘‘criollero’’ on the island and employed the surplus slaves as day laborers [Laplante & Cantero, 1857; Marrero, 1984, VIII]). The Trinidad mill represents a ﬁvefold increase in acreage planted and a 10-fold increase in the amount of sugar produced over the plantation represented in Avalle’s model. The single significant technical innovation that distinguishes Trinidad from its predecessor is the steam-powered mill. Trinidad had two steam-powered mills of English manufacture, one with an 8 Hp engine and the other with 15 Hp engine (Laplante & Cantero, 1857). From their output we can presume that they were all metal horizontal mills. The steam-powered horizontal mill used all of grinding surface, and the

Material Process and Industrial Architecture

301

Image 4. Ingenio Trinidad. Source: Laplante and Cantero (1857). Reproduced from Cantero and Laplante, Los Ingenios de Cuba, Selection and texts by Marrero (1984).

triangular arrangement of rollers automatically crushed each stalk twice. It could grind more cane more quickly, and it could squeeze more juice out of the cane than the three cylinder vertical mill discussed above. The horizontal steam-powered mill could be an effective innovation on older plantations of the type depicted by Avalle because it permitted a higher yield of cane juice from a given quantity of cane, and hence obtain more sugar from the same amount of cane. However, these plantations were unable to take advantage of its full effects if they were unable to plant more cane or process greater amounts of juice. In contrast, because fertile land with ideal conditions for sugar was abundant and available in Matanzas, plantations like Trinidad could increase the acreage they planted in accordance with the capacity of the mill. Innovations that were not possible in old sugar zones were possible on the sugar frontier. However, in the 1830s, no new reﬁning technologies were available. On Trinidad and plantations like it, the number of Jamaica trains was increased to take advantage of improvements made possible by steam mill. The multiplication of the existing reﬁning technology solved the problem of processing greater quantities of cane juice. The illustration of Trinidad above

302

DALE TOMICH

documents this adaptation. It shows seven small chimneys and one large chimney on the boiling house. Each small chimney is for a Jamaica train while the large one is for the steam engines. Further, each of the seven Jamaica trains has two clariﬁers, a device invented in 1778 to more effectively remove impurities from the cane juice before boiling in order to simplify and improve reﬁning (Laplante & Cantero, 1857). This arrangement made it possible to realize the full potential of the horizontal steam mill and to improve both the quantity and quality of the ﬁnal product. However, it did not overcome the existing problems of openkettle boiling. Indeed, it may well have increased them. The simultaneous operation of multiple Jamaica trains and greater quantity of syrup to be processed required more skilled workers, more complex coordination of workers and material on a larger scale, greater fuel consumption, larger buildings, and greater transportation capacity. At the same time, it did not alleviate the problems of controlling the temperature in the boiling kettles, of maintaining the continuous ﬂow of cane syrup through the reﬁning process, and of the relatively poor quantity of sugar and high molasses content of sugar associated with the Jamaica train. Nonetheless, it would be misleading to regard Trinidad as an example of incomplete modernization. Rather, Trinidad and other ingenios in its class provided an original solution to problems presented by horizontal steam mill and represent a distinct path of development. They achieved economies of scale and increased productivity that enabled them to surpass their competitors in the old sugar islands and dominate world sugar production by the 1830s. Trinidad’s new conﬁguration of land, labor, and technology created its own combination of thresholds and bottlenecks throughout the production sequence. Within these parameters, mills in its class remained viable as major centers of sugar production into the 1860s even as newer mills and technologies offered new solutions to the problems presented by them. From this perspective, the technological evolution of sugar production in Cuba may be seen to follow a nonlinear path of development. The integration of more extensive cane cultivation with increased capacity of the horizontal steam mill and the multiplication of Jamaica trains at Trinidad calls attention to the importance of the new productive space of the Cuban sugar frontier as the condition for innovation. The possibility of extensive cane cultivation allowed Trinidad to take full advantage of the productive potential of the horizontal steam mill. This, in turn depended upon access to new territory in the interior of the island and cheap transport to distant ports of the massive quantities of sugar produced in the new zones. Cuba’s dominance over world sugar production and the dramatic

Material Process and Industrial Architecture

303

technological progress of its sugar industry rested upon the continuous availability of land where the factors of production could be recombined in new proportions as technological innovations became available. The railroad was key to this integration (Zanetti & Garcı´ a, 1998). The image of the boiling house of the Flor de Cuba mill (see Image 5) suggests the ongoing reorganization of productive space on the sugar frontier and its significance for the development of the Cuban sugar industry. The Arrieta family founded the Flor De Cuba sugar mill in 1838. It was located in Ca´rdenas near a trunk line of the Ju´caro Railroad line. The image of the reﬁnery (casa de calderas) documents the solution to the problems of open-kettle boiling. The spherical devices at the center of the image are two multiple effect vacuum pans produced by the French ﬁrm of Charles Derosne (Laplante & Cantero, 1857). This apparatus boiled the cane syrup in a vacuum using compressed steam. It allowed the heat to be precisely controlled. Because boiling was done in a vacuum the temperature could be kept low. The juice was not exposed to impurities and humidity in the atmosphere as in open-kettle boiling and there was no carmelization. The vacuum pan used evaporated cane juice to provide steam heat. The steam from the ﬁrst kettle heated the second kettle that is at lower temperature, and so on. This multiple effect greatly reduced the need for water and fuel.

Image 5. Ingenio Flor de Cuba (Casa de Calderas). Source: Laplante and Cantero (1857). Reproduced from Cantero and Laplante, Los Ingenios de Cuba, Selection and texts by Marrero (1984).

304

DALE TOMICH

The vacuum pan reduced the amount of molasses and other by-products produced in the manufacturing process. It produced more sugar from the same quantity of juice and higher quality sugar than the open-kettle system. However, the vacuum pan represented a much larger investment than the Jamaica train, and it was only profitable if it was employed on a large scale. Consequently, it could only be successfully adopted if production in the other sectors was also increased in proportion to the capacity of the vacuum pan. Flor de Cuba represented another jump in the scale of production. It possessed 1,248 Ha of land, of which 805 were cultivated in cane in 1860 and employed 409 slaves and 170 Chinese indentured workers (Laplante & Cantero, 1857; Marrero, 1984, XVIII). The lithograph of the Flor de Cuba reﬁnery depicts the transformations of the sugar manufacturing into a continuous and fully mechanized process. At the right are two 9-m-diameter grinding mills powered by a 100 Hp steamengine powers. These were mills specially designed by Pablo de Arrieta to turn slowly and exert greater pressure on the cane stalks. They could process a much greater quantity of cane, and they extracted 72% of the juice from the cane instead of the 55% extracted by conventional mills. After the juice was squeezed from the cane stalks it was pass through carbon ﬁlters and then the clariﬁers arranged in long rows to remove impurities from the juice and facilitate reﬁning. Note the quantitative increase in the number of clariﬁers over the Trinidad mill. The juice was next transferred to the vacuum pan (at the center of the picture) for reﬁning. It then passed to one of the eight centrifuges (at the far left of the image) that used centrifugal force to effectively separate molasses from sugar crystals. The volume of material, whether in the form of cane syrup or sugar, was much greater than in previous types of reﬁnery. It had to be moved quickly and efﬁciently from sector of the reﬁnery to another in order to ensure the overall integration of the process. Carriers for both cane stalks and bagasse (crushed stalks) facilitated handling these materials during the operation of the mill. The reﬁnery had a system of pumps and gravity feeds to move the syrup in its liquid form as well as internal iron rails and carts to move the sugar. In addition, gaslights provided adequate lighting in the large structure and allowed the reﬁning process to continue into the night. In 1860, Flor de Cuba produced 3,086 metric tons of sugar, 85% of which was white sugar (Laplante & Cantero, 1857). Flor de Cuba was a model of scale and efﬁciency. It marks the total transformation of sugar making from an artisanal process based on craft skill to an industrial process subject to precise measurement and control based on the systematic application of physics and chemistry to production.

Material Process and Industrial Architecture

305

Although it responds to the bottlenecks created by the mechanized mill (i.e. Trinidad), Flor de Cuba is not the product of the linear evolution of sugar manufacturing technology. While subject to the same general material constraints of cane production as the earlier mills, it elaborates a new set of solutions to the problems of sugar manufacturing within a new spatial and technological framework. The creation of new productive space on a larger scale, including the spatial reorganization of the mill, permitted the successful adoption of technical innovations. Flor de Cuba increased the output of each sector of production. It cultivated more sugar, extracted more juice from the cane, and converted the juice into more and higher quality sugar than its predecessors. To achieve these results it altered the proportional relations between ﬁelds grinding mill and reﬁnery. The qualitative and quantitative integration of these processes, in turn, produced its own distinctive thresholds and bottlenecks.

5. CONCLUSION By emphasizing the relation of material process, space, and technological innovation, this chapter examines the speciﬁcity of Cuba as the new and dominant zone of world sugar production during the ﬁrst half of the nineteenth-century. In so doing, it reconstructs the particular local contexts that impart meaning and analytical utility to such universal categories as ‘‘industry,’’ ‘‘plantation,’’ or ‘‘(slave) labor.’’ This approach allows us to more adequately comprehend the complexity of global – local relations and to avoid the pitfalls of ‘‘violent abstractions’’ (Sayer, 1987) and ‘‘categorical thinking’’ that have too often typiﬁed thinking about ‘‘development’’ or ‘‘globalization.’’ Each of the types of sugar mill discussed in this chapter is subject to the general material constraints of sugar cane production. Nonetheless, the particular spatial and technical conﬁguration of each of them presents a different set of solutions to the problems of qualitatively and quantitatively integrating the production process. While each type of mill responds to the shortcomings of its predecessors, the sequence in which they develop is not one of linear evolution. Rather, we may think of the material conditions of sugar production as providing a grammar of innovation – a set of rules or conditions that must be addressed but which permit a variety of strategies and solutions. There is no single technically determined solution to the expansion of the sugar industry. A variety of techniques and combinations of techniques are possible with significant effect on quantity and quantity of

306

DALE TOMICH

sugar produced. There was no necessity for adopting any particular technique; rather, selection of particular techniques depended upon a multiplicity of factors that varied from estate to estate and sugar zone to sugar zone. At the same time, each of these technical solutions organizes productive space in specific ways and is only viable on a determined spatial scale. The technical and social problems that arose in the course of each innovation required new spatial and architectural solutions. Land, labor, and technology have distinct meanings in each different spatial–temporal conﬁguration. Attention to material processes helps us to differentiate one conﬁguration from another and to specify the development of the sugar plantation. A sugar frontier, Cuba broke the previously established ratio between land, labor, and reﬁning technology. On the one hand, the railroad continuously opened new land to sugar cultivation, and, on the other hand, creation of new estates in the frontier zone reconﬁgured the spatial–technical organization of sugar production. With each technical innovation, Cubans could recombine land and labor on a new scale in ways that were not possible in older sugar colonies. Constant innovation went hand in hand with the continuing reconﬁguration of land, labor, and productive technology and resulted in an unprecedented expansion of the sugar frontier. The fundamental innovation of Cuban ingenio is the reorganization of material processes and social space. Its development discloses the spatial logic of the material expansion of the world economy.

REFERENCES Arrighi, G. (1994). The long twentieth century: Money, power, and the origins of our times. London: Verso Books. Braverman, H. (1974). Labor and monopoly capital. New York: Monthly Review Press. Bunker, S. G. (1994). Regional development theory and the subordination of extractive peripheries. In: A. D. Kincaid & A. Portes (Eds), Comparative national development: Society and economy in the new global order (pp. 112–142). Chapel Hill: University of North Carolina Press. Buroway, M. (1979). Manufacturing consent. Chicago: University of Chicago Press. Coriat, B. (1979). L’Atelier et le chronome`tre. Paris: Christian Bourgois. Fraginals, M. M. (1978). El Ingenio. Complejo econo´mico social cubano del azu´car, 3 vols. Havana: Editorial de Ciencias Sociales. Gorz, A. (Ed.) (1978). The division of labour. Sussex, England: Harvester Press. Harvey, D. (1982). The limits to capital. London: Basil Blackwell Publisher. Harvey, D. (2001). Spaces of capital: Towards a critical geography. New York: Routledge. Hobsbawm, E. J. (1969). Industry and Empire: From 1750 to the present day. Harmondsworth: Penguin Books.

Material Process and Industrial Architecture

307

Innis, H., 1995. The political implications of unused capacity in frontier economics; unused capacity as a factor in Canadian economic history. In: D. Drache (Ed.), Staples, markets and cultural change: selected essays (pp. 24–34; 139–154). Montreal: McGill-Queen’s University Press. Laplante, E., & Cantero, J. (1857). Los Ingenios; coleccio´n de vistas de los principales ingenios de azu´car de la isla de Cuba. Havana: Litografı´ a de L. Marquier. Marrero, L. (Ed.) (1984). Los Ingenios de Cuba. Barcelona: Levi Marrero. Moore, J. (2000). Sugar and the expansion of the modern world economy: Commodity frontiers, ecological transformation, and industrialization. Review, 23(3), 409–433. Polanyi, K. (1957). The great transformation: The political and economic origins of our time. Boston: Beacon Press. Sayer, D. (1987). The violence of abstraction. Oxford: Basil Blackwell. Tomich, D. (1990). Slavery in the circuit of sugar: Martinique and the world economy, 1830–1848. Baltimore: The Johns Hopkins University Press. Tomich, D. (2004). Through the prism of slavery: Labor, capital, and world economy. Boulder, CO: Rowman and Littleﬁeld Publishers. Wallerstein, I. (1984). The politics of the world-economy. The states, the movements and the civilizations. New York: Cambridge University Press. Wallerstein, I. (1989). The modern world-system, III. The second era of expansion of the capitalist world-economy. San Diego: Academic Press. Watts, D. (1987). The West Indies. Patterns of development, culture and environmental change since 1492. Cambridge: Cambridge University Press. Zanetti, O., & Garcı´ a, A. (1998). Sugar and railroads. A Cuban history, 1837–1959. Chapel Hill: University of North Carolina Press.

WORLD-SYSTEMS IN THE BIOGEOSPHERE: THREE THOUSAND YEARS OF URBANIZATION, EMPIRE FORMATION AND CLIMATE CHANGE Christopher Chase-Dunn, Alexis Alvarez and Daniel Pasciuti ABSTRACT This chapter investigates the ‘‘pulsations’’ of regional interaction networks (world-systems) in Afroeurasia over the past 3,000 years. The purpose is to determine the causes of a fascinating synchrony that emerged between East Asia and the distant West Asian/Mediterranean region, but did not involve the intermediate South Asian region. The hypothesized causes of this synchrony are climate change, epidemics, trade cycles, and the incursions of Central Asian steppe nomads. This chapter formulates a strategy of data gathering, system modeling, and hypothesis testing that can allow us to discover which of these causes were the most important in producing synchrony as the Afroeurasian worldsystem came into being. Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 311–331 r 2005 Published by Elsevier Ltd. ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10014-6

311

312

CHRISTOPHER CHASE-DUNN ET AL.

1. INTRODUCTION World-systems are human interaction networks that display oscillations of expansion and contraction, with occasional large expansions that bring formerly separate regional systems into systemic intercourse with one another. These waves of expansion, now called globalization, have, in the last two centuries, created a single integrated intercontinental political economy in which all national societies are strongly linked. This chapter investigates the ‘‘pulsations’’ of regional interaction networks (world-systems) in Afroeurasia over the past 3,000 years. The purpose is to determine the causes of a fascinating synchrony that emerged between East Asia and the distant West Asian/Mediterranean region, but did not involve the intermediate South Asian region. The hypothesized causes of this synchrony are climate change, epidemics, trade cycles, and the incursions of Central Asian steppe nomads. This chapter formulates a strategy of data gathering, system modeling, and hypothesis testing that can allow us to discover which of these causes were the most important in producing synchrony as the Afroeurasian worldsystem came into being. In Kurt Vonnegut’s The Sirens of Titan a traveler from another solar system has crash-landed on one of the moons of Jupiter and is using his last bit of fuel to beam forces onto the Earth in order to send a message home. His efforts induce the Central Asian steppe nomads to behave in a way that causes successive Chinese states to build the Great Wall in the form of a script that appears from space as a rescue plea. This trope of distant forces affecting human history is an ironic tool in the hand of the ﬁction-smith who pokes fun at us for our hapless intentions. World historians have hypothesized other powerful mechanisms by which macrosocial processes may have been shaped by exogenous forces. Since Ellsworth Huntington’s Climate and Civilization there has been a growing literature on how spatial and temporal variation in rainfall, temperature, prevailing winds, and episodic weather extremes have inﬂuenced the course of history. Archaeologists routinely invoke climate change as the explanation for social and cultural developments. As much more has been learned about the patterns of global weather these accounts have become more sophisticated. Bryan Fagan’s (1999) Floods, Famines and Emperors: El Nino and the Fate of Civilizations is the most recent and compelling version. But instead of painting the humans as inert victims of powerful forces, Fagan argues that climate change has acted as the critical spur that pushed people to invent and implement radical new ways of interacting with nature and with one another.

World-Systems in the Biogeosphere

313

Mike Davis’s (2001) Late Victorian Holocausts: El Nino Famines and the Making of the Third World depicts how droughts caused by El Ninos in the nineteenth century interacted with the rapid integration of peripheral regions into global markets in a context of colonialism and neocolonialism to bring about unprecedented huge famines and epidemic disease fatalities in Brazil, India, China and the Philippines. There is also an important literature about how human action may affect the climate. Much of this is focused on anthropogenic global warming in the twentieth century, but there is also a literature on how deforestation, irrigation building and land-use patterns have affected local weather (e.g. Chew, 2001). And a growing research tradition on urban ecology has discovered the phenomenon of the ‘‘urban heat island,’’ an example of anthropogenic effects on the local weather (Gallo, n.d.). World-systems are human interaction networks that display oscillations of expansion and contraction, with occasional very large expansions that bring regional systems into contact with one another. These waves of expansion, now called globalization, have, in the last two centuries, created a single integrated intercontinental political economy in which all national societies are strongly linked. This chapter investigates the ‘‘pulsations’’ of regional interaction networks (world-systems) in Afroeurasia over the past 3,000 years. The purpose is to determine the causes of a fascinating synchrony that emerged between East Asia and the distant West Asian/Mediterranean region, but did not involve the intermediate South Asian region. The hypothesized causes of this synchrony are climate change, epidemics, trade cycles, and the incursions of Central Asian steppe nomads. This chapter formulates a strategy of data acquisition, system modeling, and hypothesis testing that can allow us to discover which of these causes were the most important in producing synchrony as the Afroeurasian worldsystem came into being.1 One limitation of some regional analyses has been the tendency to deﬁne regions in terms of homogeneous attributes, either natural or social. Thus comparative civilizationists have tended to focus on the core cultural characteristics that are embodied in religions or world-views and to construct lists of culturally deﬁned civilizations that then become the ‘‘cases’’ for the study of social change. Another approach that deﬁnes regions as areas with homogeneous characteristics is the ‘‘culture area’’ perspective developed by Carl Sauer and his colleagues (e.g. Wissler, 1927). This project gathered information on all sorts of cultural attributes – languages, architectural styles, technologies of production, kinship structures, etc. – and used these to designate bounded and adjacent ‘‘culture areas.’’

314

CHRISTOPHER CHASE-DUNN ET AL.

A major problem with both the civilizationist and cultural area approaches is the assumption that homogeneity is a good approach to bounding social systems that are evolving. Heterogeneity rather than homogeneity has long been an important aspect of human social systems. The effort to bound systems as homogeneous regions obscures this important fact. Spatial distributions of homogeneous characteristics do not bound separate social systems. Indeed, social heterogeneity is often produced by interaction, as in the case of core/periphery differentiation. Even sophisticated approaches that examine distributions of spatial characteristics statistically must make quite arbitrary choices in order to specify regional boundaries on this basis (e.g. Burton et al., 1996). The world-systems approach focuses instead on human interaction networks, and so it is able to deﬁne its units of analysis as systemic combinations of very different kinds of societies. This makes it possible to study multicultural systems and core/periphery relations as cases that can display dynamics of social evolution.2 The relationship between natural regions and human interaction networks is an important focus of theory and research. Cultural ecology has stressed the important ways in which local ecological factors conditioned sociocultural institutions and modes of living. This was an especially compelling perspective for understanding small-scale systems in which people were mainly interacting with adjacent neighbors not very far away. But this kind of local ecological determinism is much less compelling when worldsystems get larger because long-distance interaction networks and the development of larger scale technologies enable people to impose socially constructed logics on local ecologies. Some social evolutionists have interpreted this to mean that social institutions have become progressively less ecologically determined (e.g. Lenski, Lenski, & Nolan, 1995). But what has happened instead is that the spatial scale of ecological constraints have grown to the point where they are operating globally rather than locally (Chase-Dunn & Hall, 1998a, 1998b).

2. SPATIALLY BOUNDING WORLD-SYSTEMS The world-systems perspective emerged as a theoretical approach for modeling and interpreting the expansion and deepening of the European system as it engulfed the globe over the 500 years (Wallerstein, 1974; Arrighi, 1994; Chase-Dunn, 1998). The idea of a core/periphery hierarchy composed of ‘‘advanced’’ economically developed and powerful states dominating and

World-Systems in the Biogeosphere

315

exploiting ‘‘less developed’’ peripheral regions has been a central concept in the world-systems perspective. In the last decade the world-systems approach has been extended to the analysis of earlier and smaller intersocietal systems. Andre Gunder Frank and Barry Gills (1993) have argued that the contemporary world system is a continuation of a 5,000-year-old system that emerged with the ﬁrst states in Mesopotamia. Chase-Dunn and Hall (1997) have modiﬁed the basic world-systems concepts to make them useful for a comparative study of very different kinds of systems. They include very small intergroup networks composed of sedentary foragers (e.g. ChaseDunn & Mann, 1998), as well as larger regional systems containing chiefdoms, early states, agrarian empires and the contemporary global political economy in their scope of comparison. The comparative world-systems perspective is designed to be general enough to allow comparisons between quite different systems. Chase-Dunn and Hall (1997) deﬁne world-systems as important networks of interaction that impinge upon a local society and condition social reproduction and social change. They note that different kinds of interaction often have distinct spatial characteristics and degrees of importance in different sorts of systems. And they hold that the question of the degree of systemic interaction between two locales is prior to the question of core/periphery relations. Indeed they make the existence of core/periphery relations an empirical question in each case, rather than an assumed characteristic of all world-systems. Spatially bounding world-systems necessarily must proceed from a localecentric beginning rather than from a whole-system focus. This is because all human societies, even nomadic hunter-gatherers, interact importantly with neighboring societies. Thus if we consider all indirect interactions to be of systemic importance (even very indirect ones) then there has been a single global world-system since humankind spread to all the continents. But interaction networks, while they were always intersocietal, have not always been global in the sense that actions in one region had major and relatively quick effects on distant regions. When transportation and communications were over short distances the world-systems that affected people were small. Thus it is necessary to use the notion of ‘‘fall-off’’ of effects over space to bound the networks of interaction that importantly impinge upon any focal locale. The world-system of which any locality is a part includes those peoples whose actions in production, communication, warfare, alliance and trade have a large and interactive impact on that locality. It is also important to distinguish between endogenous systemic interaction processes and exogenous impacts that may importantly change a system but are not part

316

CHRISTOPHER CHASE-DUNN ET AL.

of that system. So maize diffused from Mesoamerica to Eastern North America, but that need not mean that the two areas were part of the same world-system. Or a virulent microparasite might contact a population with no developed immunity and ravage that population. But such an event does not necessarily mean that the region from which the microparasite came and the region it penetrated are parts of a single interactive social system. Interactions must be two-way and regularized to be socially systemic. One-shot deals do not make a world-system. Chase-Dunn and Hall (1997) note that in most intersocietal systems there are several important networks of different spatial scales that impinge upon any particular locale:

Information Networks (INs) Prestige Goods Networks (PGNs) Political/Military Networks (PMNs), and Bulk Goods Networks (BGNs).

The largest networks are those in which information travels. Information is light and it travels a long way, even in systems based on down-the-line interaction.3 These are termed Information Networks (INs). A usually somewhat smaller interaction network is based on the exchange of prestige goods or luxuries that have a high value/weight ratio. Such goods travel far, even in down-the-line systems. These are called Prestige Goods Networks (PGNs). The next largest interaction net is composed of polities that are allying or making war with one another. These are called Political/Military Networks (PMNs). And the smallest networks are those based on a division of labor in the production of basic everyday necessities such as food and raw materials. These are Bulk Goods Networks (BGNs). Fig. 1 illustrates how these interaction networks are spatially related in many world-systems. The ﬁrst question for any focal locale is about the nature and spatial characteristics of its links with the above four interaction nets. This is prior to any consideration of core/periphery position because one region must be linked to another by systemic interaction in order for consideration of core/ periphery relations to be relevant. The spatial characteristics of these networks clearly depend on many things – the costs of transportation and communications, and whether or not interaction is only with neighbors or there are regularized long-distance trips being made. But these factors affect all kinds of interaction and so the relative size of networks is expected to approximate what is shown in Fig. 1. As an educated guess we would suppose that fall-off in the PMN generally occurs after two or three indirect links. Suppose group A is ﬁghting and

World-Systems in the Biogeosphere

Fig. 1.

317

Nested Interaction Networks.

allying with its immediate neighbors and with the immediate neighbors of its neighbors. So its direct links extend to the neighbors of the neighbors. But how many indirect links will involve actions that will importantly affect this original group? The number of indirect links that bound a PMN are probably either two or three. As polities get larger and interactions occur over greater distances each indirect link extends much farther across space. But the point of important fall-off will usually be after either two or three indirect links. Chase-Dunn and Hall (1997) divide the conceptualization of core/periphery relations into two analytically separate aspects: core/periphery differentiation, and core/periphery hierarchy. Core/periphery differentiation exists when two societies are in systemic interaction with one another and one of these has higher population density and/or greater complexity than the other. The second aspect, core/periphery hierarchy, exists when one society dominates or exploits another. These two aspects often go together because a society with greater population density/ complexity usually has more power than a society with less of these, and so can effectively dominate/exploit the less powerful neighbor. But there are important instances of reversal (e.g. the less dense, less complex Central Asian steppe nomads exploited agrarian China) and so this analytical separation is necessary so that the actual relations can be determined in each

318

CHRISTOPHER CHASE-DUNN ET AL.

Fig. 2.

Core/Periphery Hierarchy.

case.4 The question of core/periphery relations needs to be asked at each level of interaction designated above. It is more difﬁcult to project power over long distances, and so one would not expect to ﬁnd strong core/periphery hierarchies at the level of Information or Prestige Goods Networks. Fig. 2 illustrates a core/periphery hierarchy. Core/periphery hierarchies are important in processes of social evolution because semiperipheral societies, those that are intermediate between core regions and peripheral hinterlands, are fertile locations for institutional innovations and frequently are the key actors that transform the developmental logic of world-systems. Chase-Dunn and Hall, (1997, Chapter 5) call this ‘‘semiperipheral development.’’ Semiperipheral marcher chiefdoms conquer more senior and older core chiefdoms to form larger and more centralized complex chiefdoms, as do the much better known semiperipheral marcher states (e.g. Chin China, Assyria, Achaemenid Persia, Alexandrian Macedonia, Rome, Islamic Arabia, and the Ottoman Empire). Semiperipheral capitalist city-states (the Phoenicians, the Italian city-states, the Hanse cities, Malakka) were the agents of commercialization in the interstices of the tributary empires. In the modern world-system it has been the semiperipheral and capitalist Dutch republic, England and the United States of America that have risen to hegemony and further globalized the organization of the world economy. Semiperipheral development is still an important pattern in the twentieth and twenty-ﬁrst centuries (Chase-Dunn & Boswell, 2002).

World-Systems in the Biogeosphere

Fig. 3.

319

Chronograph of PMNs (adapted from Wilkinson, 1987).

Using the conceptual apparatus for spatially bounding world-systems outlined above we can construct spatio-temporal chronographs for how the interaction networks of the human population changed their spatial scales to eventuate in the single global political economy of today. Fig. 3 uses PMNs as the unit of analysis to show how a ‘‘Central’’ PMN, composed of the merging of the Mesopotamian and Egyptian PMNs in about 1500 BCE, eventually incorporated all the other PMNs into itself.

3. WORLD-SYSTEM CYCLES: RISE-AND-FALL AND PULSATIONS Comparative research reveals that all world-systems exhibit cyclical processes of change. There are two major cyclical phenomena: the rise and fall of large polities, and pulsations in the spatial extent and intensity of trade

320

CHRISTOPHER CHASE-DUNN ET AL.

networks. ‘‘Rise and fall’’ corresponds to changes in the centralization of political/military power in a set of polities – an ‘‘international’’ system. It is a question of the relative size of, and distribution of, power across a set of interacting polities. The term ‘‘cycling’’ has been used to describe this phenomenon as it operates among chiefdoms (Anderson, 1994). All world-systems in which there are hierarchical polities experience a cycle in which relatively larger polities grow in power and size and then decline. This applies to interchiefdom systems as well as interstate systems, to systems composed of empires, and to the modern rise and fall of hegemonic core powers (e.g. Britain and the United States). Though very egalitarian and small-scale systems such as the sedentary foragers of Northern California (Chase-Dunn, & Mann, 1998) do not display a cycle of rise and fall, they do experience pulsations. All systems, including even very small and egalitarian ones, exhibit cyclical expansions and contractions in the spatial extent and intensity of exchange networks. We call this sequence of trade expansion and contraction pulsation. Different kinds of trade (especially bulk goods trade vs. prestige goods trade) usually have different spatial characteristics. It is also possible that different sorts of trade exhibit different temporal sequences of expansion and contraction. It should be an empirical question in each case as to whether or not changes in the volume of exchange correspond to changes in its spatial extent. In the modern global system, large trade networks cannot get spatially larger because they are already global in extent.5 But they can get denser and more intense relative to smaller networks of exchange. A good part of what has been called globalization is simply the intensiﬁcation of larger interaction networks relative to the intensity of smaller ones. This kind of integration is often understood to be an upward trend that has attained its greatest peak in recent decades of so-called global capitalism. But research on trade and investment shows that there have been two recent waves of integration, one in the last half of the nineteenth century and the most recent since World War II (Chase-Dunn, Kawano, & Brewer, 2000). The simplest hypothesis regarding the temporal relationships between rise-and-fall and pulsation is that they occur in tandem. Whether or not this is so, and how it might differ in distinct types of world-systems, is a set of problems that are amenable to empirical research. Chase-Dunn and Hall (1997) have contended that the causal processes of rise and fall differ depending on the predominant mode of accumulation. One big difference between the rise and fall of empires and the rise and fall of modern hegemons is in the degree of centralization achieved within the

World-Systems in the Biogeosphere

321

core. Tributary systems alternate back and forth between a structure of multiple and competing core states on the one hand and core-wide (or nearly core-wide) empires on the other. The modern interstate system experiences the rise and fall of hegemons, but these never take over the other core states to form a core-wide empire. This is the case because modern hegemons are pursuing a capitalist, rather than a tributary form of accumulation. Analogously, rise and fall works somewhat differently in interchiefdom systems because the institutions that facilitate the extraction of resources from distant groups are less fully developed in chiefdom systems. David G. Anderson’s (1994) study of the rise and fall of Mississippian chiefdoms in the Savannah River valley provides an excellent and comprehensive review of the anthropological and sociological literature about what Anderson calls ‘‘cycling,’’ the processes by which a chiefly polity extended control over adjacent chiefdoms and erected a two-tiered hierarchy of administration over the tops of local communities. At a later point these regionally centralized chiefly polities disintegrated back toward a system of smaller and less hierarchical polities. Chiefs relied more completely on hierarchical kinship relations, control of ritual hierarchies, and control of prestige goods imports than do the rulers of true states. These chiefly techniques of power are all highly dependent on normative integration and ideological consensus. States developed specialized organizations for extracting resources that chiefdoms lacked – standing armies and bureaucracies. And states and empires in the tributary world-systems were more dependent on the projection of armed force over great distances than modern hegemonic core states have been. The development of commodity production and mechanisms of ﬁnancial control, as well as further development of bureaucratic techniques of power, have allowed modern hegemons to extract resources from far-away places with much less overhead cost. The development of techniques of power have made core/periphery relations ever more important for competition among core powers and have altered the way in which the rise-and-fall process works in other respects. Chase-Dunn and Hall (1997, Chapter 6) argued that population growth in interaction with the environment, and changes in productive technology and social structure produce social evolution that is marked by cycles and periodic jumps. This is because each world-system oscillates around a central tendency (mean) due both to internal instabilities and environmental ﬂuctuations. Occasionally, on one of the upswings, people solve systemic problems in a new way that allows substantial expansion. We want to explain expansions, evolutionary changes in systemic logic, and collapses. That is the point of comparing world-systems.

322

CHRISTOPHER CHASE-DUNN ET AL.

The multiscalar regional method of bounding world-systems as nested interaction networks outlined above is complimentary with a multiscalar temporal analysis of the kind suggested by Fernand Braudel’s work. Temporal depth, the longue duree, needs to be combined with analyses of shortrun and middle-run processes to fully understand social change. The shallow presentism of most social science and contemporary culture needs to be denounced at every opportunity. A strong case for the very longue duree is made by Jared Diamond’s (1997) study of original zoological and botanical wealth. The geographical distribution of those species that could be easily and profitably domesticated explains a huge portion of the variance regarding which world-systems expanded and incorporated other world-systems thousands of years hence. Diamond also contends that the diffusion of domesticated plant and animal species occurs much more quickly in the latitudinal dimension (East/West) than in the longitudinal dimension (North/South), and so this explains why domesticated species spread so quickly to Europe and East Asia from West Asia, while the spread south into Africa was much slower, and the North/ South orientation of the American continents made diffusion much slower than in the Old World Island of Eurasia. The diagram below depicts the coming together of the East Asian and the West Asian/Mediterranean systems. Both the PGNs and the PMNs are shown, as are the pulsations and rise-and-fall sequences. The PGNs linked intermittently and then joined. The Mongol conquerors linked the PMNs briefly in the thirteenth century, but the Eastern and Western PMNs were not permanently linked until the Europeans and Americans established Asian treaty ports in the nineteenth century (Fig. 4).

4. SYNCHRONIZATION OF EMPIRES, CITIES AND DEMOGRAPHIC WAVES Earlier studies have used data on both city sizes and the territorial sizes of empires to examine different regional interaction systems and the hypothesis that regions distant from one another were experiencing synchronous cycles of growth and decline (e.g. Chase-Dunn & Willard, 1993; Chase-Dunn, Manning, & Hall, 2000; Chase-Dunn & Manning, 2002). Frederick Teggart’s (1939) path-breaking world historical study of temporal correlations between events on the edges of the Roman and Han Empires argued the thesis that incursions by Central Asian steppe nomads were the key to East/West

World-Systems in the Biogeosphere

Fig. 4.

323

East/West Pulsations and Merger.

synchrony. An early study of city-size distributions in Afroeurasia (Chase-Dunn & Willard, 1993; see also Chase-Dunn & Hall, 1997, pp. 222–223) found an apparent synchrony between changes in city-size distributions and the growth of large cities in East Asia and West Asia–North Africa over a period of 2,000 years. That led us to examine data on the territorial sizes of empires for similar synchrony, which we found (ChaseDunn, Manning, & Hall, 1999). Chase-Dunn & Manning (2002) have reexamined the city-size data using constant regions6 rather than PMNs to see if the East/West synchronous city growth hypothesis holds when the units that are compared are somewhat different. Their results conﬁrm the existence of East/West city growth synchrony. Here we present a new analysis of East/West synchrony that uses overall population estimates compiled by McEvedy and Jones (1975). They note a synchrony in periods of regional demographic growth and decline during the late ﬁrst millennium BCE and during the ﬁrst millennium CE between East Asia and the Mediterranean. Interestingly, McEvedy and Jones (1975, pp. 345–346) reject the idea that climate change may have caused this synchrony in favor of a hypothesis of parallel and connected technological and

324

CHRISTOPHER CHASE-DUNN ET AL.

organizational change. We have computed the partial correlations, controlling for year to remove the trend, of population levels from 1000 BCE to 1800 CE among three regions. We stop at 1800 CE because the trend becomes exponential after that and would drown out earlier middle range variations. What we want to know is whether or not the middle term ups and downs, what we have called growth/decline phases, are synchronous or not. We examine four regions: East Asia, South Asia, West Asia/Mediterranean and Europe.7 These are the same constant regions that Chase-Dunn and Manning (2002) used to study the synchrony of city growth/decline phases. Table 1 shows the partial correlation coefﬁcients of population change estimates for four Old World regions. These have been detrended in two ways in order to look for synchronous growth-decline phases across regions. We eliminate the years after 1,800 CE when most of the regions were undergoing geometric growth rates. And we compute the inter-regional correlations controlling for year, which should take out the long-term trend. The results in Table 1 are somewhat surprising. There are statistically significant partial correlations among all the regions despite our efforts to take out the long-term trend. The correlation between East Asia and the West Asian/Mediterranean region is higher than that for either city or empire size cross-regional partial correlations (0.81), but it is not as high as some of the other coefﬁcients in Table 1. Curiously, the correlations between Europe and both East Asia and South Asia are very high (0.95, 0.92). The lowest correlation is between West Asia and South Asia (0.60). And the correlation between Europe and the West Asia/Mediterranean region is relatively low despite that these two ‘‘regions’’ overlap geographically (see Note 7). It is possible that these high partial correlations are partly due to the rather coarse temporal resolution of the population estimates that we have extracted from graphs produced by McEvedy and Jones (1975). Our data set Table 1. Inter-Regional Partial Correlations of Population Change Controlling for Year, 1,000 BCE–1,800CE. (Population Estimates from McEvedy & Jones, 1975).

West Asia/Mediterranean East Asia South Asia Europe

West Asia/Mediterranean

East Asia

South Asia

Europe

1

0.81 (26) 1

0.60 (26) 0.88 (26) 1

0.79 (26) 0.95 (26) 0.92 (26) 1

World-Systems in the Biogeosphere

325

is organized in one hundred year intervals, a temporal resolution that smooths out most of the growth/decline ﬂuctuations we are trying to study. Unfortunately, McEvedy and Jones do not present enough detail about the evidence they used to produce their graphs. We are looking into the possibility that this material may be obtained. Fig. 5 presents the demographic data in graphical form for the same four regions. Examination of Fig. 5 shows both the long-term trends and the shorterterm variations, though these have been smoothed by the low temporal resolution just discussed. What we see is a long hump that starts slowly in 1000 BCE and winds back down to a low point around 600 CE in all the regions except South Asia. In South Asia the slump does not appear. This is the East/West synchrony noted by McEvedy and Jones. After about 600 CE all the regions go up again, but then the patterns partly diverge. The East Asian rise is early and steeper. All the regions except South Asia display a partly synchronous decline after the twelfth century. East Asia has another

Fig. 5.

Regional Population Growth (From McEvedy and Jones, 1975).

326

CHRISTOPHER CHASE-DUNN ET AL.

decline in the seventeenth century and this is also a period of slow growth in Europe and decline in West Asia, but South Asia continues to grow in this period. The West Asian/Mediterranean region does not partake in the rapid population growth that sweeps the other regions after the ﬁfteenth century.

5. THE MORAN EFFECT IN POPULATION ECOLOGY The temporal aspects of climate change cycles lead easily to hypotheses about how these may be causes of certain cyclical (or at least sequential) phenomena in human affairs. And this is especially the case when cycles in distant regions appear to come into synchrony. Population ecologists have long studied the phenomenon of increases and then decreases in the population densities of plant and animal species. They model population dynamics of species within adjacent and distant ‘‘patches,’’ explaining how predator–prey relationships, food availability, and migration affect the cycles of population growth and decline. Moran’s (1953) study of the population cycles of the Canadian lynx led him to formulate what has become known as the ‘‘Moran effect’’ – the idea that synchronized exogenous shocks to local oscillating systems will cause them to come into synchrony even when the exogenous shocks do not themselves display much periodicity (Ranta, Kaitala, Lindstrom, & Helle, 1997; Ranta, Kaitala, & Lindstrom, 1999). Population ecologists usually have climate change in mind as the most likely source of exogenous shocks. The important implications of the Moran effect for our problem of the causes of synchrony are that any exogenous shock can bring oscillating systems into synchrony even if the temporal features of the exogenous variable are completely different from the temporality of the local oscillating systems. A meteor impact could reset local systems and put them into synchrony. Turchin and Hall (2002) also point out that the empirical study of synchrony requires exact measurement and ﬁne temporality, and also many oscillations and many different cases of oscillating systems in order to disentangle different plausible causes of synchrony. These are daunting requisites for our single case of East/West synchrony. Comparable other instances of distant systems that come into weak contact with one another can be found. Within the Old World, the Mesopotamian and Egyptian core regions were interacting with one another by means of prestige goods exchange from about 3000 BCE until their PMNs merged in 1500 BCE. Chase-Dunn and Hall (2001) have already examined this case for synchronicity and have not found it, though the data on Bronze Age city and empire sizes are crude with regard to temporality and accuracy. It is also possible to study the

World-Systems in the Biogeosphere

327

temporality of rise and fall and oscillations in the New World. Chase-Dunn and Hall (1998b) and Stephen Kowalewski (2002) have not found synchrony between distant systems in the new world, though much more systematic and comparable research needs to be done before ﬁrm conclusions are possible. The Moran effect implies that synchrony occurs easily because a single exogenous impact that resets systems with similar endogenous oscillations will bring them into synchrony. But if this is true we would expect to ﬁnd more synchrony than we have found up to now. Population ecology also usually ﬁnds greater synchrony in patches that are close to one another than in those that are more distant (Ranta et al., 1999), but this is not what we ﬁnd in Afroeurasia. The South Asian system, intermediate between East and West, seems to be marching to its own drummer.

6. MODELING CLIMATE CHANGE EFFECTS ON POPULATION Patrick Galloway (1986) models the way in which climate change can affect human population growth. He argues that it was climate change that caused the synchrony of demographic cycles noted by McEvedy and Jones (1975). Galloway’s model is depicted in Fig. 6. Galloway’s model is entirely plausible and could easily be amended to include affects on city growth and empire-formation. But in order for this model to account for synchrony across regions, the changes in temperature (and other climatological variables) would need to also be synchronous, or else there would have to at least be an initial strong climatological shift that affects all the regions during the same period. The only way to sort this out is to obtain indicators of climate change in or near to the regions we are studying in the relevant time periods. Knowing about the climate change record in Greenland will not settle the question, because despite global teleconnections, climate change is ultimately local. Our effort to gather the relevant climate change data has only just gotten underway.

7. A COMPREHENSIVE MODEL OF THE CAUSES OF INTER-REGIONAL SYNCHRONY We can propose a comprehensive model of all the plausible causes of East/ West synchrony. The purpose of complex causal modeling is to allow us to

328

CHRISTOPHER CHASE-DUNN ET AL.

Fig. 6.

Galloway’s (1986) Climate and Population Model.

discover the relative strengths of different causal mechanisms by examining the logic implications of causal relations and the parameters that are hypothesized to be operant. Fig. 7 displays a complex causal model that contains all the hypothesized effects that result in the East/West synchrony discussed above. This model can be translated into a complex system of structural equations and estimated parameters for these can allow us to examine the conditions under which causation can lead to synchrony. We plan to combine this theoretical exercise with a campaign to improve our empirical knowledge of the population sizes of cities and the territorial sizes of empires and climate change over the past 3,000 years (e.g. Pasciuti & Chase-Dunn, 2002). By approaching the problem from the angles of both induction and deduction we hope to be able to estimate the relative strengths of the different hypothesized cause of East/West synchrony. And the outcome should be a better understanding of the way in which human systems have interacted with biological and geological processes in world history.

NOTES 1. Some social scientists erroneously assume that GIS (Geographical Information Systems) data structures are restricted to the mapping of attributes that are stationary in space and that GIS is useless for studying things that move. Geographers are

World-Systems in the Biogeosphere

Fig. 7.

329

Comprehensive East/West Synchrony Model.

now developing GIS techniques based on vectors for mapping prevailing winds, but also for studying migration (Tobler, 1995, n.d.). 2. The notion of ‘‘interaction spheres’’ developed by archaeologist Joseph Caldwell (1964) is another approach that recognizes that diversity has long been an important characteristic of human systems. 3. Down-the-line trade passes goods from group to group. 4. Kradin (2002) argues that pastoral peoples mimic the political organization of societies they are adjacent to and so Central Asian steppe empires were, in their external aspects, similar to the agrarian empires from whom they successfully managed to extract surplus product. This is a fascinating instance of a peripheral society that managed to exploit the core. 5. If we manage to get through several sticky wickets looming in the 21st century the human system will probably expand into the solar system, and so ‘‘globalization’’ will continue to be spatially expansive. 6. The earlier research on cities had used political–military networks (PMNs) as the units of analysis following the method of bounding world-systems as interaction networks. When we use PMNs the Central System expands spatially over time. Chase-Dunn and Manning (2002) reanalyzed the city data using constant regions (Near East Europe, East Asia, and South Asia) that do not change over time. 7. The West Asia/Mediterranean region includes the whole Mediterranean littoral so as to include the whole interactive city system that originated in West Asia and spread to the Mediterranean with Etruscan, Greek, and Phoenician migration and the emergence of the Latin cities. Thus, Europe and the West Asian/Mediterranean region are geographically overlapping one another.

330

CHRISTOPHER CHASE-DUNN ET AL.

REFERENCES Anderson, D. G. (1994). The Savannah River chiefdoms: Political change in the late prehistoric southeast. Tuscaloosa, AL: University of Alabama Press. Arrighi, G. (1994). The Long twentieth century: Money, power and the origins of our times. London: Verso. Barry, G. (1993). Low intensity democracy. Boulder, CO: Pluto Press. Burton, M., Moore, C. C., Whiting, J. W. M., & Romney, A. K. (1996). Regions based on social structure. Current Anthropology, 37(1), 87–123. Caldwell, J. R. (1964). Interaction spheres in prehistory. Hopewellian Studies, 12(6), 133–156 Springﬁeld, IL: Illinois State Museum Scientific Papers. Chase-Dunn, C. (1998). Global formation: Structures of the world-economy. Lanham, MD: Rowman and Littleﬁeld. Chase-Dunn, C., & Boswell, T. (2002). Transnational social movements and democratic socialist parties in the semiperiphery. Presented at the meetings of the California Sociological Association, Riverside, CA. October 19. http://irows.ucr.edu/papers/csa02/ csa02.htm Chase-Dunn, C., & Hall, T. D. (1997). Rise and demise: Comparing world-systems. Boulder, CO: Westview Press. Chase-Dunn, C., & Hall, T. D. (1998a). Ecological degradation and the evolution of worldsystems. Journal of World-Systems Research, 3, 403–431 http://jwsr.ucr.edu/archive/ vol3/v3n3a3.php. Chase-Dunn, C., & Hall, T. D. (1998b). World-systems in North America: Networks, rise and fall and pulsations of trade in stateless systems. American Indian Culture and Research Journal, 22(1), 23–72. Chase-Dunn, C., & Hall, T. D. (2001). City and empire growth/decline sequences in ancient Mesopotamian and Egyptian world-systems. Presented at the annual meetings of the International Studies Association, Chicago, February 24. Chase-Dunn, C., Kawano, Y., & Brewer, B. (2000). Trajectories of globalization since 1800: cycles of world-system integration. American Sociological Review, 65, 77–95. Chase-Dunn, C., & Mann, K. M. (1998). The Wintu and their neighbors: A very small worldsystem in Northern California. Tucson: University of Arizona Press. Chase-Dunn, C., Manning, S., & Hall, T. D. (2000). Rise and fall: East-West synchrony and indic exceptionalism reexamined. Social Science History, 24(4), 727–754. Chase-Dunn, C., & Manning, E. S. (2002). City systems and world-systems: Four millennia of city growth an decline. Cross-Cultural Research, 36(4), 379–398. Chase-Dunn, C., & Willard, A. (1993). Systems of cities and world-systems: Settlement size hierarchies and cycles of political centralization, 2000 BC–1988AD. A paper presented at the International Studies Association meeting, March 24–27, Acapulco. http:// www.irows.ucr.edu/papers/irows5/irows5.htm Chew, S. C. (2001). World ecological degradation: Accumulation, urbanization and deforestation, 3000 BC–AD 2000. Walnut Creek, CA: Altamira. Davis, M. (2001). Late Victorian holocausts: El Nino famines and the making of the third world. New York: Verso. Diamond, J. (1997). Guns germs, and steel. New York: W.W. Norton and Co. Fagan, B. (1999). Floods, famines and emperors: El Nino and the fate of civilizations. New York: Basic Books.

World-Systems in the Biogeosphere

331

Gallo, K. P., Ower, T. (n.d.). Identiﬁcation of urban heat islands using remotely sensed data: A multi sensor approach. Remote Sensing Core Curriculum, Volume 4. http:// research.umbc.edu/tbenja1/gallo/gallo.html Galloway, P. R. (1986). Long-term ﬂuctuations in climate and population in the preindustrial era. Population and Development Review, 12(1), 1–24. Kradin, N. N. (2002). Nomadism, evolution and world-systems: Pastoral societies in theories of historical development. Journal of World-Systems Research, 8(3), 368–388. Lenski, G., Nolan, P., & Lenski, J. (1995). Human Societies: An introduction to macrosociology (7th ed.). New York: McGraw-Hill. McEvedy, C., & Jones, R. (1975). Atlas of world population history. New York: Penguin. Moran, P. A. P. (1953). The statistical analysis of the Canadian lynx cycle. II Synchronization and meteorology. Australian Journal of Zoology, 1, 291–298. Pasciuti, D., & Chase-Dunn, C. (2002). Estimating the population sizes of cities. http:// irows.ucr.edu/research/citemp/estcit/estcit.htm Ranta, E., Kaitala, V., Lindstrom, J., & Helle, E. (1997). The Moran effect and synchrony in population dynamics. OIKOS, 78, 136–142. Ranta, E., Kaitala, V., & Lindstrom, J. (1999). Spatially autocorrelated disturbances and patterns in population synchrony. Proceedings of the Royal Society, London, Series B, 266, 1851–1856. Teggart, F. J. (1939). Rome and china: A study of correlations in historical events. Berkeley: University of California Press. Tobler, W. (1995). Migration: Ravenstein, Thornthwaite, and beyond. Urban Geography, 16(4), 327–343. Tobler, W. (n.d.). The care and feeding of vector ﬁelds. PowerPoint Presentation, University of California, Santa Barbara. Turchin, P., & Hall, T. D. (2002). Spatial synchrony among and within world-systems: Insights from theoretical ecology. Journal of World-Systems Research, 9(1), 37–64. Wallerstein, I. (1974). The modern world-system. New York: Academic Press. Wilkinson, D. (1987). Central civilization. Comparative Civilizations Review, 17, 31–59. Wissler, C. (1927). The culture area concept in social anthropology. American Journal of Sociology, 32(6), 881–891.

COFFEE, REVOLUTION, AND DEMOCRACY IN CENTRAL AMERICA$ Jeffery M. Paige ABSTRACT The chapter draws on historical evidence from Central America to test two of the most influential theories of the development of democracy: (1) structural theories derived from the work of Barrington Moore and (2) theories of the ‘‘political economy of democratic transitions.’’ The Central American evidence confirms Moore’s theory in regard to the anti-democratic role of landed elites, but not the democratic role of the bourgeoisie. Contrary to some structural theories, the industrial working class was also not important in the development of democracy in Central America. Nor does the Central American evidence fit the political economy of democratic transitions model of negotiated or imposed ‘‘transitions from above.’’ A new model, termed the route to democracy through socialist revolution from below is proposed to account for the Central American evidence and the implications of the model are explored for the development of democracy generally.

$

The research on which this chapter is based was supported by grants from the National Science Foundation (SES 8920899), the Fulbright Program, and the Ofﬁce of the Vice President for Research and the LSA Faculty Assistance Fund of the University of Michigan.

Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 333–352 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10015-8

333

334

JEFFERY M. PAIGE

1. INTRODUCTION The Central American societies of Costa Rica, El Salvador, and Nicaragua in the revolutionary decade of the 1980s provide an unusual laboratory for the study of the determinants of dictatorship and democracy and the transition from authoritarian to democratic regimes. Costa Rica had (and still has) the longest-lived stable democracy in Latin America with contested elections and a universal franchise since 1948. El Salvador, by contrast, had the longest authoritarian tradition in Latin America. A series of military dictatorships controlled the country almost continuously from 1932 to the mid-1980s. In Nicaragua, the 1979 Sandinista revolution ended the personal dictatorship of the Somoza family and installed what was for a decade the only revolutionary socialist regime on the Latin American mainland. The revolution also initiated a gradual transition to democracy that culminated in the Sandinistas’ electoral defeat in 1990. Despite the profound political differences among the three societies in the 1980s, by 1992 when the Salvadoran civil war ended, the political systems of all three had converged on a common, if tentative, model of representative democracy. Furthermore, to the great surprise of many, a decade after the electoral defeat of the Sandinistas and the Salvadoran peace treaty these three societies remain viable democracies. Salvador has successfully conducted two contested elections under close international supervision and is approaching a third. Nicaragua has already completed three election cycles under similar circumstances, including the Sandinista defeat in 1990. Neither country has suffered a breakdown in the constitutional order despite continuing high levels of criminal violence in El Salvador and continued struggles over fundamental property rights in Nicaragua. In both countries revolutionary organizations have made a successful transition to well-organized opposition parties, although since the Sandinista’s defeat in 1990 there has yet to be a successful turn-over of power. When the putative ‘‘transition to democracy’’ began in Latin America in the 1980s, few countries could have seemed less-promising candidates for successful ‘‘transitions’’ than revolutionary Nicaragua or war-torn El Salvador. Not only were they consumed by political conﬂict, but they were among the most backward and least industrialized regions in Latin America. Two decades later the much-touted ‘‘transition’’ in Latin America is in deep trouble. In the Andes, Venezuela, once considered the leading example of democracy in South America, has just survived an abortive coup and is tottering on the verge of civil war. Colombia remains convulsed by civil war; Ecuador and Bolivia have both seen the ouster of presidents by protests

Coffee, Revolution, and Democracy in Central America

335

from increasingly militant and effective indigenous groups; and Peru has seen the notorious auto-golpe by Alberto Fujimori and later his disgrace and exile. Only in Mexico, the Southern Cone, and, surprisingly, Central America has the democratic transition seemed to have achieved some degree of stability. The remarkable transitions to democracy in Central America and their even more astonishing survival for more than a decade make the region an ideal site to test and even confound existing theories of both the structural and conjunctural determinants of democratic transitions. Furthermore, both democracy and revolution in Central America are intimately tied to the export of a raw material, coffee, that was the focus of one of Stephen Bunker’s (1987) most inﬂuential works – Peasants Against the State. In Uganda, conﬂicts over coffee contributed to Idi Amin’s dictatorship; in Central America they led to both revolution and democracy. Central America demonstrates not only the immense power of raw materials in shaping the fate of nations but also the power of history in determining that fate, just as Stephen Bunker has always argued. These historical outcomes in turn represent a profound challenge to the two major theoretical traditions in the study of democracy.

2. THEORIES OF DEMOCRACY: MOORE AND THE TRANSITION SCHOOL Two theoretical traditions have dominated the study of democracy: the structural and largely sociological tradition associated in particular with the work of Barrington Moore Jr. (1966) and the so-called ‘‘transitions school’’ associated with the work of Guillermo O’Donnell, Philip Schmitter, and Laurence Whitehead (O’Donnell & Schmitter, 1986; O’Donnell, Schmitter, & Whitehead, 1986). Although the theoretical traditions differ profoundly, both employ the same minimalist procedural definition of democracy emphasizing (1) the extension of popular suffrage, and (2) guarantees of rights of association and other basic civil liberties that make this suffrage meaningful (see for example the definitions of Moore, 1966, p. 414; O’Donnell & Schmitter, 1986, p. 12). In general these definitions are parallel and often directly based on Robert Dahl’s (1971, pp. 1–16) notion of ‘‘polyarchy’’ which treats suffrage and civil liberties as deﬁnitive. Although Dahl’s definition has been widely adopted, his terminology has not. The term democracy will therefore be used here, as it has been in much of the democratization

336

JEFFERY M. PAIGE

literature, to refer to the restricted definition of democracy implied in Dahl’s ‘‘polyarchy.’’ Such a definition excludes the social welfare dimension of democratic rights included in Marshall’s (1950) ‘‘social citizenship’’ as well as wider issues of social and economic equity. These issues, as will be demonstrated, are of critical importance in the viability of even procedural democracy in Central America. The most inﬂuential structural theory of democracy remains that of Barrington Moore Jr. as described in his Social Origins of Dictatorship and Democracy (1966). Moore’s theory has inﬂuenced numerous case studies of particular countries and regions (Abraham, 1986; Billings, 1979; O’Donnell, 1973; Parsa, 1989; Stokes, 1987; Tilton, 1974; Wiener, 1978) and shaped much sociological debate on the determinants of democracy and revolution (Skocpol, 1973, 1979; Stephens, 1989; Wiener, 1976). Moore argued that democracy is a product of a ‘‘bourgeois revolution’’ against a backward landed aristocracy (‘‘no bourgeoisie, no democracy,’’ p. 418). The survival of a powerful landed aristocracy dependent on coercive forms of labor organization (what Moore calls ‘‘labor repressive agriculture’’) into the capitalist period is a barrier to democracy in Moore’s model. A ‘‘revolutionary break with the past’’ is required to break the power of the labor repressive agriculturalists and this can only be accomplished by a bourgeois revolution that clears out the ‘‘feudal underbrush.’’ Without such a revolutionary breakthrough, a powerful landed aristocracy will tend to form a conservative coalition with a weak bourgeoisie crushing democracy. The encyclopaedic review of the development of democracy in Europe, Latin America, and the Caribbean by Rueschemeyer, Stephens, and Stephens in their Capitalist Development and Democracy (1992) ﬁnds support for Moore’s general argument while dissenting from some of its specifics. They found that Moore’s hypothesis on the anti-democratic and authoritarian role of a powerful landed class ‘‘bore the test of repeated examinations across the countries studied’’ (p. 270). They argue, however, that the bourgeoisie, agrarian or industrial, has seldom pushed for the extension of the franchise to popular classes. Instead, industrial working classes, often allied with middle class groups, have been the primary forces in the full extension of the franchise to most adult males. As Rueschemeyer, Stephens, and Stephens conclude, capitalism and democracy are related because ‘‘capitalist development weakens the landed upper class and strengthens the working class’’ (p. 271). The substantial literature on the recent wave of ‘‘democratic transitions’’ (see Diamond, Linz, & Lipset, 1995, pp. 1–66; Haggard & Kaufman, 1995, pp. 3–20; Mainwaring, O’Donnell, & Valenzuela, 1992, pp. 294–341 for

Coffee, Revolution, and Democracy in Central America

337

reviews of this literature) has developed in large part in isolation from the structural tradition represented by Moore. Although initially this tradition emphasized the short-run and often conjunctural factors stressed by O’Donnell, Schmitter and Whitehead in Transitions from Authoritarian Rule (1986), more recent work such as Haggard and Kaufman’s (1995) comparative study The Political Economy of Democratic Transitions has begun to emphasize issues of underlying structure. Haggard and Kaufman develop an explicit model of transitions, but many elements of their approach are also found in the work of Caravozzi (1992), Cumings (1989), Frieden (1991), Schamis, (1991), Seidman (1994), and Whitehead (1992). The political economy of democratic transitions approach emphasizes a sequence of structural factors and structural contradictions, which under the pressure of economic crises lead to a transition to democracy. First, a ‘‘bureaucratic authoritarian’’ regime destroys, sometimes physically, the left, leaving it a free hand to manage the transition. The regime comes under pressure from an economic crisis that it is unable to manage. The economic crisis divides the military from its natural allies in the business community, leaving it increasingly isolated. Strikes, demonstrations and other forms of popular protest erupting in response to the economic crisis further weaken the military and widen the gap with its allies in the business community. Faced with the loss of legitimacy, withdrawal of support by the business community, popular protest, and continuing economic uncertainty, the military withdraws and turns the government over to a civilian and usually democratic government in a pattern that Whitehead (1992, p. 158) terms ‘‘democracy by default.’’ The resulting transition is either negotiated by the military or imposed by them. It is, in Terry Karl’s (1990, p. 9) words, a ‘‘transition from above’’ that does not involve popular participation, except indirectly through mass protest. The transitions paradigm was ﬁrst developed to describe transitions to democracy in southern Europe (Spain and Portugal), the Southern Cone of Latin America (Argentina, Uruguay, Chile), and other middle-income semiperipheral societies. As James Dunkerly (1994) and Susanne Jonas (1989) point out, it fails completely to account for the rise of democracy in the poor, peripheral societies of Central America. There agro-export-led agricultural intensiﬁcation, not import substitution industrialization, was the principal development strategy. The praetorian military dictatorship and narrow paternalistic regimes of Central America have little in common with the ‘‘bureaucratic authoritarianism’’ of the transitions paradigm, and the Central American militaries surrendered power only after extended civil war, not by ‘‘default.’’ Nor was the armed Central American left

338

JEFFERY M. PAIGE

disarticulated by political repression, at least not until it tried to assume a democratic rule after democracy had been established. Economic crisis and mass protests were not enough to split the military from its supporters in the business community; this occurred only after armed insurrection began in Nicaragua, after revolution nearly succeeded in El Salvador, and after the 1948 civil war in Costa Rica. Most importantly, democracy in Central America was a result of revolutionary pressures from below, not negotiated transitions from above. The transitions paradigm is no help in understanding developments in Central America. The fate of the Moore model in the region is considerably more complex and requires much more detailed consideration.

3. THE MOORE MODEL AND CENTRAL AMERICA As Lowell Gudmundson (n.d., p. 1) has pointed out, the Moore theory has achieved such currency in the study of Central American politics that his framework has been used not only in the classic works of Cardoso (1975), Torres-Rivas (1975, 1981; Torres-Rivas & Flores, 1989) and Pe´rez-Brignoli (Cardoso and Pe´rez-Brignoli, 1977), but also in more recent works by Baloyra (1983), Paige (1987, 1997), Weeks (1986), Williams (1994), Winson (1989), Yashar (1992), and Mahoney (2001). The contrast between the authoritarian North (El Salvador and, especially, Guatemala) and the democratic South (Costa Rica) has been viewed, after Moore, as a consequence of the greater importance of labor repressive forms of export agriculture in the North and the greater importance of small holders and agrarian capitalists in the South. The conventional wisdom in the study of Central America, then, tends to support Moore. Nevertheless, there are significant problems with the Moore model as applied to Central America. First, as was long ago pointed out by the great Central American comparativist Edelberto Torres-Rivas (1975, pp. 27–29), the clear division between landed aristocracy and industrial bourgeoisie implied by Moore’s thesis is lacking in Central America, as it is in fact in much of Latin America (Zeitlin & Ratcliff, 1984, pp. 150–152). As a consequence of the dependent, peripheral role of these societies in the world economy, capitalism develops in the countryside, not the city, based on agro-exports, not manufacturing. As a result export agriculture (including labor repressive forms), agro-industrial and other manufacturing activities, and mercantile and ﬁnancial activities tend to be concentrated in the same social groups and even in the same dynastic families.

Coffee, Revolution, and Democracy in Central America

339

In the Central American societies (except Honduras), the dominance of a single agro-export commodity, coffee, reinforced the interdependence of the agrarian and agro-industrial fractions of the elite. Since high-quality Central American coffee requires elaborate processing in industrial installations before it can be exported, the Central American agro-export elite is simultaneously an agrarian elite dependent on labor intensive agriculture and an agro-industrial elite dependent on capital intensive processing (Paige, 1987, 1997). The laborintensive agriculture is more important in the North and processing in the South, but both activities are found throughout the region (Cardoso, 1975; Williams, 1994). Furthermore, successful coffee processors are usually also coffee exporters, and processors and exporters typically provided the initial capital for the Central American banking system. So the elite is also mercantile and ﬁnancial as well as agrarian and agro-industrial. As Torres-Rivas (1982) has described it, the Central American elite is a ‘‘three footed beast’’ with one foot in agriculture, one in agro-industry, and one in ﬁnance. For the Moore model to apply to Central America, the agro-industrial fraction of the elite would somehow have to separate itself from the agrarian fraction and then wage a revolutionary offensive against them. Even if the agro-industrial fraction could somehow separate itself, however, it is not clear that a bourgeoisie, agrarian or industrial, would be willing to lead the struggle for the full development of democracy. In recent Central American history, the agro-industrial fraction of the elite has not on its own been able to separate itself from the agrarian fraction, and even in Costa Rica, where the landed fraction was weakest, the dominant agro-industrial coffee processors never led a democratic revolution against the remnants of agrarian power. Just as Moore argued, in Central America labor repressive agriculturalists have been bastions of reaction and authoritarianism. But the assault against these bastions has not been led by a bourgeoisie, agrarian or industrial. Instead it has been led by socialist and social democratic revolutionaries.

3.1. Democracy through Socialist Revolution: The Case of El Salvador El Salvador is the clearest case of both the strength of the alliance between the landed and agro-industrial fractions of the elite and the revolutionary pressures from below necessary to break it. The division between what Italo Lo´pez Vallecillos (1979) has termed the ‘‘agro-ﬁnancial’’ and the ‘‘agroindustrial-ﬁnancial’’ fractions of the Salvadoran elite has become what Enrique Baloyra (1982, p. 28) calls ‘‘the consensus of scholarly opinion on politics in El Salvador.’’ As Lo´pez Vallecillos (1979, p. 588) describes the

340

JEFFERY M. PAIGE

division, the agro-industrial fraction ‘‘opposes any attempt to transform the rigid framework of land concentration and low salaries in its devotion to the plantation economy that is the basis of its income and profits.’’ The agroindustrial fraction, in contrast, ‘‘tries to introduce changes in the economic systemyand opts for less authoritarian political forms within the framework of liberal democracy, representative, but restricted and controlled.’’ Oscillation between these two elite strategies has long characterized Salvadoran politics, and in the 1920s, the temporary dominance of the more liberal fraction gave El Salvador the reputation of being, according to Ralph Lee Woodward (1985, p. 171), ‘‘the most progressive of the Central American republics.’’ During times of crisis like the 1930s and the early 1980s, however, the positions of the two fractions converged to support the antidemocratic policies of the ‘‘agro-ﬁnancial fraction’’ that gave El Salvador an international reputation for ferocious authoritarianism. In El Salvador the wealth and power of both the agrarian and agroindustrial fractions of the elite rested on the production, processing, and export of coffee. The coffee export economy created both economic interdependence between the two fractions of the elite and a possible point of cleavage between the landowning producers on the one hand, and the agroindustrial coffee processors on the other. By the 1960s this division was sufﬁciently pronounced to occasion the founding of a new organization to represent the interest of the processor-exporters. The Asociacio´n Salvadoren˜a de Beneﬁciadores y Exportadores de Cafe´ (ABECAFE) was established as an alternative to the traditional organization of Salvadoran coffee producers, the Asociacio´n Salvadoren˜a de Cafe´ (ASCAFE, often known in El Salvador as simply the Cafetalera) (Pe´rez, Mena, Lozo, & Magan˜a, 1975, p. 24). These developments were reﬂected politically in a growing division within the conservative party of the elite, the Alianza Republicana Nacionalista (ARENA), and in growing conﬂict within and between ABECAFE and the Cafetalera. At the date of its ofﬁcial founding in 1981, ARENA was a coalition of the backers of ultra-right wing terrorist and former head of military intelligence Roberto D’Aubuisson and the Alianza Productiva (Productive Alliance), itself a coalition of conservative industrialists and businessmen (Baires Martı´ nez, n.d. [1990?]). Although the party contested elections in 1982 and 1985, it was initially ambivalent about democracy. More conservative elements of the party talked openly of murdering up to 100,000 people to restore the pre1979 status quo (Montgomery, 1995, p. 157). At the extreme right of the party Orlando de Sola, scion of a prominent coffee family, announced that the tens of thousands killed by death squads in 1980–1981 were ‘‘communist stooges who deserved to die’’ (New York Times, 11 August 1989, p. A4).

Coffee, Revolution, and Democracy in Central America

341

ARENA’s reliance on violence and opposition to democracy limited its electoral appeal even in conservative El Salvador. After electoral defeats in 1984 and 1985 D’Aubuisson began a conscious effort to transform the party’s image and constituency. A critical element in D’Aubuisson’s effort to construct a ‘‘new ARENA’’ was the recruitment of politically moderate (at least by Salvadoran standards) coffee processor and businessman Alfredo Cristiani to the party in 1984 and his selection with D’Aubuisson’s blessing as its presidential candidate in 1989 (Miles & Ostertag, 1989). Cristiani, whose family ﬁrm was the 10th largest coffee processor in El Salvador and who had served as president of ABECAFE, was a key ﬁgure among the immigrant coffee processors and other agro-industrialists who represented the less regressive fraction of the Salvadoran elite. He was also a key ﬁgure in a new business sector organization founded in 1983, FUSADES (Fundacio´n Salvadoren˜a para el Desarrollo Econo´mico y Social or Salvadoran Foundation for Economic and Social Development). It was the Cristiani faction that led El Salvador to a negotiated settlement of the war and an eventual transition to democracy. The big losers in the transition were the hard line agrarians and their allies in the military. Cristiani refused to reverse land reform that had already taken place as a result of decrees of the revolutionary junta in 1980. Full implementation of the 1980 reforms (as modiﬁed by the ARENA dominated legislature in 1982), as well as additional land distribution, was required as a condition for the peace settlement (Spence & Vickers, 1994; Wood, 1993). Similarly, Cristiani agreed to purge military ofﬁcers with the worst human rights abuse records, including most of the army high command, significantly reducing the power of hard line elements within the military (Spence & Vickers, 1994, pp. 14–17). Finally, the peace accords committed ARENA to politically incorporate the demobilized FMLN rebels, whose mass murder D’Aubuisson had demanded only a decade before. The case of El Salvador reveals both the strengths and weaknesses of the Moore model. Moore is undoubtedly right about the role of labor-repressive agriculture and the agrarian elite in blocking democracy. They were an intransigent anti-democratic force in El Salvador for most of this century. On the other hand, the agro-industrial bourgeoisie was never able to force a democratic opening on its own, even if it had wanted to. It took a massive civil war led by one of the most effective guerrilla armies in Latin America to force even a compromise solution and a tentative opening for democracy. It was socialist revolution from below, not bourgeois revolution from above, that reduced the power of the landed elite and their military allies and opened the way for democracy. The principal beneﬁciaries, however, were

342

JEFFERY M. PAIGE

not the socialist revolutionaries but the ‘‘new ARENA’’ of the agroindustrialists, who in 1994 decisively won the ﬁrst democratic election after the peace settlement (and the ﬁrst to involve participation by the left since 1931) and have remained in power ever since. Ironically, the principal outcome of more than a decade of revolutionary war was the triumph of democracy and neo-liberalism, not socialism or even social democracy.

3.2. The Absent Bourgeois Revolution in Costa Rica As Lowell Gudmundson has pointed out (n.d., p. 11), the Costa Rican case would seem to lend considerable support to Moore’s theory: ‘‘a commercially based [coffee] processing group constitutes what amounts to a primarily non-landed elite, a working ‘bourgeoisie’ for all practical purposes, and the development of petty bourgeois groups [i.e. small coffee growers] both allies and adversaries in political and economic terms, leads eventually to competitive electoral regimes.’’ There is, as Gudmundson also points out (n.d., pp. 11–12), one glaring inadequacy in this account. The ‘‘working bourgeois’’ of coffee processors did not back a ‘‘bourgeois revolution’’ in favor of democracy and the extension of the franchise. Instead, as Gudmundson and an increasing number of scholars argue (Rojas Bolan˜os, 1989; Schifter, 1983; Winson, 1989), they backed counter-revolution against a regime that resulted from the extension of the franchise to the working classes in the Costa Rican civil war of 1948. Although Costa Rica had been a limited democracy since 1889, only after the civil war was a democracy incorporating the popular classes ﬁnally established. The Costa Rican civil war pitted against one another two improbable coalitions. On the one side was aristocratic reformer Rafael Angel Caldero´n Guardia who in 1940 had initiated an ambitious reform program that created the extensive Costa Rican welfare state and, significantly, for the ﬁrst time incorporated the working classes and urban poor into Costa Rican democracy. Caldero´n and his reforms were supported by an unlikely partnership between the Costa Rican Communist party and progressive elements of the Catholic church. Opposing the Caldero´n/Communist/Catholic coalition was an equally unlikely alliance of coffee landowners and processors and a group of middle class social reformers led by the father of modern Costa Rica, Jose´ Figueres. The agrarian bourgeoisie of coffee processors opposed both Caldero´n’s reforms and his extension of the franchise to the popular classes and backed armed insurrection to enforce their views.

Coffee, Revolution, and Democracy in Central America

343

Ironically, Figueres, the winner in the civil war, embraced the reforms of his defeated opponents, even though they had been a principal cause of the war. The reforms of Figueres himself, particularly his nationalization of the banking system, largely controlled by coffee processor capital, were in some respects more radical than those of Caldero´n. Betrayed, as they saw it, by Figueres, their ally in the civil war, the coffee landowners and processors found themselves in the political wilderness. Ironically, as Gudmundson points out (n.d., p. 12), Figueres emerged as the structural ally, if the insurrectionary opponent, of the Communists and Caldero´n. Although Figueres purged communism from Costa Rican life, his armed rebellion institutionalized the reforms of Caldero´n and the Communists. Figueres also accepted the political incorporation of the popular classes begun under Caldero´n and, to prevent a restoration by the displaced conservatives of the coffee elite, abolished the army. It is difﬁcult to see in these events much role for the ‘‘working bourgeois’’ of coffee processors in the transition to democracy in Costa Rica. Certainly they led no revolutionary assault against a bastion of agrarian authoritarian reaction. In the end it was their exclusion from power, ﬁrst by Caldero´n and then by Figueres and his armed rebels, that opened the way for social democracy in Costa Rica. The elite was clearly unwilling to accept the results of the transformation of Costa Rica carried out by Caldero´n through entirely democratic means between 1940 and 1948 and welcomed Figueres’ armed insurrection. In the end it was the consequences of armed rebellion, this time by social democrats, that made possible the transition to democracy by politically isolating the coffee landowners and processors and abolishing the armed forces. The conventional interpretation of Costa Rica as supportive of the Moore model overlooks the actual pattern of alliances in the civil war. Social democratic revolution, structurally allied with Social Christian reformism and Communism, not bourgeois revolution, brought democracy to Costa Rica.

3.3. Bourgeois Revolutionaries in a Socialist Revolution: The Case of Nicaragua In Nicaragua too there was no ‘‘bourgeois revolution’’ against a backward landed aristocracy leading to democracy. The 1979 Sandinista revolution was led by a party with a decidedly heterodox but nonetheless recognizably Marxist–Leninist ideology (Gilbert, 1988, pp. 22–27; Hodges, 1986, pp. 173–179) and the social base of the revolution was among the urban informal sector and students who made up the largest proportion

344

JEFFERY M. PAIGE

of participants in the urban insurrections of 1978–1979 (Vilas, 1986, pp. 108–122). Nonetheless virtually all observers of the revolution (Booth, 1985; Black, 1981; Gilbert, 1988; Lo´pez, Soto, Barrios, & Serres, 1980; Vilas, 1986), agree that the bourgeoisie played an important and possibly decisive role in its eventual triumph. The bourgeoisie may have hoped for democracy as interviews with many of them after the revolution indicate (Gilbert, 1988, pp. 105–127; Paige, 1989, 1997), but they backed a socialist, not a bourgeois, revolution in the hopes of getting it. Once again the revolutionary break with the authoritarian past was provided by revolutionary socialists, not an insurgent bourgeoisie. In fact the Nicaraguan bourgeoisie was divided by the revolution and a substantial fraction stayed with Somoza. Most observers agree that the most important business groups surrounding the bank holding companies that dominated the prerevolutionary Nicaraguan economy continued to support Somoza, as did a majority of the large coffee and cotton growers that made up the core of the Nicaraguan agro-export economy (Black, 1981, p. 65; Gilbert, 1988, p. 107; Lo´pez et al., 1980, pp. 85–93). Nevertheless, a substantial minority of agro-exporters and agro-industrialists, termed the ‘‘middle bourgeoisie’’ in the literature on the revolution, supported the Sandinistas and stayed in Nicaragua after the revolution (Gilbert, 1988, p. 85, 106; Paige, 1989, 1997; Spaulding, 1991). Most business interests, even those opposed to Somoza, however, preferred retaining the basic political structure without the dynasty (Booth, 1985, p. 102). The revolution itself eliminated that political structure, including the Nicaraguan National Guard, which had functioned as a kind of praetorian guard for the Somoza dynasty (Millet, 1977). Revolutionary land reform eliminated many of the largest landowners. Since the reforms were directed at allies of Somoza and those not using their lands efﬁciently, the traditional landed elite was hard hit by the revolution. Although many properties have been returned by the Chamorro government, a substantial number have not, so that the political and social consequences of the revolution in the countryside have not been entirely reversed (Spaulding, 1992, pp. 23–26; Luciak, 1995, pp. 10–12). In Nicaragua, then, the socialist revolutionaries played the role that Moore assigns to the bourgeois – clearing out the feudal underbrush and opening the way for democracy. Although Sandinista attitudes about electoral as opposed to popular democracy were complex and variable (Gilbert, 1988, p. 34; Luciak, n.d., pp. 21–29), in the end they supported an election in 1984 that international observers regarded as reasonably fair (Latin American Studies Association, 1985) and any doubts about their commitment to elections were removed by their defeat in the freely contested elections of 1990.

Coffee, Revolution, and Democracy in Central America

345

3.4. The Limits of the Moore Model in Central America Despite the differences among Nicaragua, Costa Rica, and El Salvador, their transitions to democracy share some fundamental characteristics. In all three cases, just as Moore argued, a revolutionary break with the landed elite and authoritarian rule is necessary for the development of democratic institutions. However, in contrast to Moore’s theory, in none of the three cases was this revolutionary break primarily a result of a bourgeois offensive against the landed elite. Instead the revolutionary pressure was provided by socialist or social democratic revolutionaries who broke the power of the landed elite (even in Costa Rica where it was never strong to begin with) and destroyed the authoritarian institutions of the old order. In Costa Rica, and Nicaragua the old military structure was liquidated; in El Salvador it was significantly modiﬁed. In El Salvador and Nicaragua land reform also liquidated the landed elite; in Costa Rica it was excluded from power by Figueres. Institutional changes necessary for democracy in Moore’s model were in fact made. But not by the bourgeoisie. In Central America it was the revolutionary mobilization of the informal and semi-proletarian sectors by socialist revolutionaries, not the gradual incorporation of the working class by union and social democratic parties, that made democracy possible. The cases of El Salvador, Costa Rica, and Nicaragua provide support for two factors emphasized by Moore – the importance of a revolutionary break with the past and the anti-democratic role of landed elites, especially those dependent on labor repression. They do not provide support for his theories in regard to the role of the bourgeoisie. These cases, however, also do not provide support for theories like those of Reuschemeyer et al. emphasizing the role of the industrial working class in the gradual extension of the democratic franchise. Central America then departs in significant respects from models of the development of democracy in the Moore structural tradition.

4. CONCLUSIONS: REVOLUTION AND DEMOCRACY The Central American experience stands out as a distinct form of transition to democracy – through socialist revolution from below. The entirely unexpected and, from the point of view of the left, unintended consequence of the failed socialist revolutions in Central America in the 1980s and in Costa Rica in the 1940s was the triumph of electoral democracy. Despite the disparate experience of the three countries, some elements of the

346

JEFFERY M. PAIGE

revolutionary road to democracy are common in all three cases. First, an armed revolution by the left splits the agro-industrial and agrarian fractions of the elite. Socialist or social democratic revolutionaries mobilize semiproletarian and informal sector workers, not industrial workers. Second, intervention or the threat of intervention by the United States blocks or undermines a victory by the left. Nevertheless, the authoritarian structures of the old regime are dismantled or weakened under the threat of revolutionary armed force. The power of the landed elite is then reduced either through agrarian reform as in El Salvador and Nicaragua or through political isolation as in Costa Rica. The army of the old regime is either destroyed (Nicaragua, Costa Rica) or weakened (El Salvador). The left then assumes a generally subordinate role in an electoral democracy dominated by the agro-industrial bourgeoisie as in the Chamorro regime in Nicaragua, the Cristiani government in El Salvador, or the Figueres regime and its successors in Costa Rica. The transitions paradigm, unlike the Moore model, did not postulate any such ‘‘revolutionary break with the past.’’ On the contrary, the absence of a revolutionary upsurge from below was seen as one of the preconditions for a successful transition. In ‘‘transitions from above’’ the institutional structure of the old regime, most significantly the armed forces, is left intact. Similarly the class structure remains unaltered with traditional economic and social elites continuing to dominate post-transition politics, since the model postulates the destruction of the left under the preceding ‘‘bureaucratic authoritarian regime.’’ It is precisely the destruction of the old order, including the armed forces and the landed elite, that has made democracy possible in Costa Rica and established a shaky democratic regime in Nicaragua. The presence of even a purged armed force in El Salvador has been regarded by some observers (Bland, 1992, p. 177; Spence & Vickers, 1994, p. 16) as a possible threat to democracy there. In his discussion of the ‘‘democratic route to modern society,’’ Moore considers the absence of a ‘‘revolutionary break with the past’’ in his Indian case to be a cause of the ‘‘extraordinary difﬁculties Liberal democracy faces in that country.’’ The same might be said for the Andean nations today and possibly for the Southern Cone, despite the apparent stability of democracy there. Even in Brazil, where the ﬁrst victory of a working class president in the country’s history would seem to conﬁrm a successful transition, the inequalities inherited from colonial landownership patterns remain an explosive issue. In Argentina the economic and social consequences of a return to an agro-export economy based on large land holdings has generated increasing instability. In Chile, where the reforms of the Frei-Allende era eroded landed power, the military retains a special constitutional status.

Coffee, Revolution, and Democracy in Central America

347

The transitions paradigm may in fact not be examining a ‘‘transition to democracy’’ at all. This possibility should not lead to an examination of a putative process of ‘‘transition’’ back to authoritarian rule similar to the brief ﬂurry of theorizing along this line after the rise of authoritarian regimes in Latin America in the 1970s (Collier, 1979; Malloy, 1977). Rather it suggests that renewed attention be given to the structural determinants of democracy and the need for a revolutionary break with the past. Paradoxically, the prospects for democracy in Central America may be better than they are in many other places where ‘‘transitions to democracy’’ involve no such sharp break with the past. The example of Costa Rica, arguably Latin America’s only stable democracy, is instructive. The abolition of the armed forces in 1948 and Figueres’ isolation of the coffee elite at approximately the same time, as well as the incorporation of the left purged of radical inﬂuences into the political process, created the preconditions for democracy. These conditions have been met in few other places where ‘‘transitions to democracy’’ are said to be taking place. The development of the Costa Rican welfare state under both Caldero´n and Figueres points to another important structural determinant of democracy – reductions in extremes of wealth and poverty. Costa Rica has indexes of health and education that place it at levels comparable to some developing countries, even though its per capita income is barely more than $4,000. Although substantial inequalities of wealth exist, the grinding poverty characteristic of much of the Third World and the rest of Central America is notably absent. Few observers think that democracy can exist with large inequalities of wealth. The case of Costa Rica would seem to support this assertion. Little has been done about inequalities of wealth in most of the other cases considered by the ‘‘transitions’’ school. Indeed, the introduction of neo-liberal economic policies has been associated with a dramatic increase in poverty and inequality in Latin America (Jonas & McCaughan, 1994, p. 1; Rosenthal, 1989). The prospects for democracy therefore must be guarded for the region, as they must be in Nicaragua and El Salvador, where neo-liberal economic programs have also led to dramatic increases in poverty and inequality (Dunkerly, 1994, p. 17). The evidence from Central America does not imply that the only way that democracy can occur in poor countries is through socialist revolution. It does suggest, however, that some other means may have to be found to accomplish what revolution accomplished in Central America. First, some means must be found to eliminate the autonomous power of the military. As long as this force continues to exist, it represents a potential threat to democracy. Second, the power of elite groups, particularly landlords, linked to

348

JEFFERY M. PAIGE

the military must be weakened or destroyed. Third, some way of introducing social welfare measures to reduce inequality must be found to insure that conﬂicts over poverty and unequal distribution do not undermine democratic institutions. The political incorporation of the left as happened in Costa Rica is one way that such a reduction can occur, but it is by no means a guarantee, as the increasing inequality and reduction in social welfare measures in El Salvador and Nicaragua, despite left representation in the legislature, indicate. The contrast between the long-lived, stable democracy of Costa Rica and the conﬂict-ridden although constitutionally stable democracies of El Salvador and Nicaragua indicates that to ignore issues of social and economic equity in theories of democracy is unlikely to lead to successful long-term prediction. In part the difﬁculty lies with minimalist procedural definitions of democracy and the neglect of social citizenship inherent in both the structural and transitions theories. Although logically there is no reason why economic and social equity could not be included as an independent variable accounting for procedural democracy, in practice minimalist definitions have shifted the focus away from these issues. In developed industrial societies, successful democracies have almost invariably been associated with the extension of the welfare state. Where this is not the case, democracy has been associated with political violence and instability. In raw material dependent societies like the Central American coffee economies the welfare state is even more important. Of the Central American coffee (and banana) exporters only Costa Rica has such a welfare state. It is also the longest lived democracy in Latin America.

REFERENCES Abraham, D. (1986). The collapse of the Wiemar Republic: Political economy and crisis. New York: Holmes and Meier. Baires Martı´ nez, Y. (N.d. (1990?)). Origenes y formacio´n del partido ARENA. Unpublished manuscript, San Jose´, Costa Rica. Baloyra, E. (1982). El Salvador in transition. Chapel Hill: University of North Carolina Press. Baloyra, E. (1983). Reactionary despotism in Central America. Journal of Latin American Studies, 15, 295–319. Billings, D. B. (1979). Planters and the making of the New South: Class, politics, and develo pment in North Carolina, 1865–1900. Chapel Hill: University of North Carolina Press. Black, G. (1981). Triumph of the people: The Sandinista revolution in Nicaragua. London: Zed.

Coffee, Revolution, and Democracy in Central America

349

Bland, G. (1992). Assessing the transition to democracy. In: J. S. Tulchin (Ed.), Is there a transition to democracy in El Salvador? (pp. 163–206). Boulder: Lynne Rienner. Booth, J. A. (1985) (2nd ed.). The end and the beginning: The Nicaraguan revolution. Boulder: Westview. Bunker, S. (1987). Peasants against the state: The politics of market control in Bugisu, Uganda, 1900–1983. Urbana: University of Illinois Press. Caravozzi, M. (1992). Beyond transitions to democracy in Latin America. Journal of Latin American Studies, 24, 665–684. Cardoso, C. F. S. (1975). Historia econo´mica del cafe´ en Centroamerica (siglo XIX). Estudios Sociales Centroamericanos (enero-abril), 4, 9–55. Cardoso, C. F. S., & Pe´rez-Brignoli, H. (1977). Centro America y la economı´a occidental (1520–1930). San Jose´: Editorial Universitaria Centroamericana. Collier, D. (Ed.) (1979). The new authoritarianism in Latin America. Princeton: Princeton University Press. Cumings, B. (1989). The abortive Abertura: South Korea in the light of Latin American experience. New Left Review, 173, 5–32. Dahl, R. (1971). Polyarchy. New Haven: Yale University Press. Diamond, L., Linz, J. J., & Lipset, S. M. (1995). Politics in developing countries: Comparing experiences with democracy. Boulder: Lynne Rienner. Dunkerly, J. (1994). The pacification of Central America: Political change in the Isthmus, 1987–1993. London: Verso. Frieden, J. (1991). Debt, development and democracy. Princeton: Princeton University Press. Gilbert, D. (1988). Sandinistas: The party and the revolution. New York: Basil Blackwell. Gudmundson, L. (n.d.) Central American Agarian History and the Barrington Moore Thesis. Unpublished paper, University of Oklahoma. Haggard, S., & Kaufman, R. R. (1995). The political economy of democratic transitions. Princeton: Princeton University Press. Hodges, D. C. (1986). Intellectual foundations of the Nicaraguan revolution. Austin: University of Texas. Jonas, S. (1989). Elections and transitions: The Guatemalan and Nicaraguan Cases. In: J. A. Booth & M. A. Seligson (Eds), Elections and democracy in Central America (pp. 126–157). Chapel Hill: University of North Carolina. Jonas, S., & McCaughan, E. J. (1994). Introduction: The quest for social transformation in Latin America. In: S. Jonas & E. J. McCaughan (Eds), Latin America faces the twentyfirst century. Reconstructing a social justice agenda. Boulder: Westview. Karl, T. L. (1990). Dilemmas of democratization in Latin America. Comparative Politics, 23, 1–21. Latin American Studies Association. (1985). Report of the Latin American Studies Association delegation to observe the Nicaraguan general election of November 4, 1984. LASA Forum (Winter). Lo´pez, J. C., Soto, O. N., Barrios, C. F. C., & Serres, P. (1980). La caida del Somocismo y la lucha Sandinista en Nicaragua. Ciudad Universitaria Rodrigo Facio, Costa Rica: Editorial Universitaria Centroamericana. Lo´pez Vallecillos, I. (1979). Fuerzas sociales y cambio social en El Salvador. Estudios Centroamericanos, 34, 369–370, 557–590. Luciak, I. A. (n.d.) The political economy of transition. Unpublished manuscript. Department of Political Science, Virginia Polytechnic Institute and State University.

350

JEFFERY M. PAIGE

Luciak, I. A. (1995). The challenge of peace: A comparative study of El Salvador and Nicaragua. Unpublished manuscript. Department of Political Science, Virginia Polytechnic Institute and State University. Mahoney, J. (2001). The legacies of liberalism: Path dependence and political regimes in Central America. Baltimore: Johns Hopkins University Press. Mainwaring, S., O’Donnell, G., & Valenzuela, J. S. (1992). Issues in democratic consolidation: The new South American democracies in comparative pespective. Notre Dame: University of Notre Dame Press. Malloy, J. M. (Ed.) (1977). Authoritarianism and corporatism in Latin America. Pittsburgh: University of Pittsburgh Press. Marshall, T. H. (1950). Citizenship and social class. Cambridge: Cambridge University Press. Miles, S., & Ostertag, B. (1989). D’Aubuisson’s new ARENA. NACLA Report on the Americas, 23, 14–39. Millet, R. (1977). Guardians of the dynasty. Maryknoll, NY: Orbis. Montgomery, T. S. (1995) (2nd ed.). Revolution in El Salvador: From civil strife to civil peace. Boulder: Westview. Moore, B., Jr. (1966). Social origins of dictatorship and democracy: Lord and peasant in the making of the modern world. Boston: Beacon. O’Donnell, G. (1973). Modernization and bureaucratic-authoritarianism: Studies in South American politics. Berkeley: Institute for International Studies. O’Donnell, G., & Schmitter, P. C. (1986). Transitions from authoritarian rule: Tentative conclusions about uncertain democracies. Baltimore: Johns Hopkins University Press. O’Donnell, G., Schmitter, P. C., & Whitehead, L. (Eds) (1986). Transitions from authoritarian rule: Prospects for democracy. Baltimore: Johns Hopkins. Paige, J. M. (1987). Coffee and politics in Central America. In: R. Tardanico (Ed.), Crises in the Caribbean Basin (pp. 141–190). Newbury Park: SAGE. Paige, J. M. (1989). Revolution and the agrarian bourgeoisie in Nicaragua. In: T. Boswell (Ed.), War and revolution in the world-system (pp. 99–128). Westport, CT: Greenwood. Paige, J. M. (1997). Coffee and power: Revolution and the rise of democracy in Central America. Harvard: Harvard University Press. Parsa, M. (1989). Social origins of the Iranian revolution. New Brunswick, NJ: Rutgers University Press. Pe´rez, B., Mena, A. M., Lozo, E. N., & Magan˜a, E. M. (1975). Investigacio´n sobre la crisis del cafe´ en el an˜o 1974: beneficiadores. San Salvador: Universidad Centromericana Jose´ Simeo´n Can˜as. Rojas Bolan˜os, M. (1989). Lucha social y guerra civil en Costa Rica 1940–1948. San Jose´: Editorial Porvenir. Rosenthal, G. (1989). Latin American and Caribbean developments in the 1980s and the outlook for the future. CEPAL Review, 39, 7–17. Rueschemeyer, D., Stephens, E. H., & Stephens, J. D. (1992). Capitalist development and democracy. Chicago: University of Chicago Press. Schamis, H. E. (1991). Reconceptualizing Latin American authoritarianism in the 1970s: From bureaucratic authoritarianism to neoconservatism. Comparative Politics, (January), 201–220. Schifter, J. (1983). La democracia en Costa Rica como producto de la neutralizacio´n de classes. In: C. Zelaya, O. Aguliar, D. Camacho, R. Cerdas, & J. Schifter (Eds), Democracia en

Coffee, Revolution, and Democracy in Central America

351

Costa Rica? cinco opiniones pole´micas (2nd ed., pp. 183–258). San Jose´: Editorial Universidad Estatal a Distancia. Seidman, G. W. (1994). Manufacturing militance: Workers’ movements in Brazil and South Africa. Berkeley: University of California Press. Skocpol, T. (1973). A critical view of Barrington Moore’s social origins of dictatorship and democracy. Politics and Society, 12(2), 1–34. Skocpol, T. (1979). States and social revolutions. Cambridge: Cambridge University Press. Spaulding, R. (1991). Capitalists and revolution: State-private sector relations in revolutionary Nicaragua (1979–1990). Paper presented at the Latin American Studies Association Meetings, Washington, DC, April 4–6. Spaulding, R. (1992). From revolution to neoliberalism: Private sector ambivalence in postrevolutionary Nicaragua (1990–1992). Paper presented at the Latin American Studies Association Meetings, Los Angeles, September 24–27. Spence, J., & Vickers, G. (1994). A negotiated revolution? A two year progress report on the Salvadoran peace accords. Cambridge, MA: Hemisphere Initiatives. Stephens, J. (1989). Democratic transition and breakdown in Western Europe. American Journal of Sociology, 94, 1019–1077. Stokes, G. (1987). The social origins of Eastern European politics. Eastern European Politics and Society, 1, 30–74. Tilton, T. (1974). The social origins of liberal democracy: The Swedish case. American Political Science Review, 69(2), 561–571. Torres-Rivas, E. (1975). Sı´ ntesis histo´rica del proceso po´litico. In: E. Torres-Rivas, R. K. Gert, E. Lizano, R. Menjı´ var & S. Ramı´ rez (Eds), Centroamerica Hoy (pp. 9–118). Mexico, DF: Siglo Veintiuno. Torres-Rivas, E. (1981). Crisis del poder en Centroamerica. San Jose´: Editorial Universitaria Centroamericana. Torres-Rivas, E. (1982). State making and revolution in Central America. Lecture given at the Center for Research on Social Organization of the University of Michigan, November 19. Torres-Rivas, E., & Flores, J. L. (1989). Sociology of developing societies: Historical bases of insurgency in Central America. In: J. L. Flora & E. Torres-Rivas (Eds), Central America: Sociology of developing societies (pp. 32–55). New York: Monthly Review. Vilas, C. M. (1986). The Sandinista revolution. National liberation and social transformation in Central America. New York: Monthly Review. Weeks, J. (1986). An interpretation of the Central American crisis. Latin American Research Review, 21(3), 31–53. Whitehead, L. (1992). The alternatives to ‘‘Liberal Democracy:’’ A Latin American perspective. Political Studies, 40, 146–158. Wiener, J. M. (1976). Review of reviews: The social origins of dictatorship and democracy. History and Theory, 15(2), 146–175. Wiener, J. M. (1978). Social origins of the new South: Alabama 1860–1885. Baton Rouge: Louisiana State University Press. Williams, R. (1994). States and social evolution: Coffee and the rise of national governments in Central America. Chapel Hill: University of North Carolina Press. Winson, A. (1989). Coffee and democracy in modern Costa Rica. New York: St. Martin’s Press. Wood, E. J. (1993). Implementing the peace in El Salvador: Agrarian property rights in the aftermath of the civil war. Rural social relations, democracy, and development in El

352

JEFFERY M. PAIGE

Salvador: Agrarian politics and the negotiated resolution of the Civil War in El Salvador. Unpublished manuscript. Woodward, R. L. (1985). Central America: A nation divided (2nd ed.). Oxford: Oxford University Press. Yashar, D. (1992). Demanding democracy: Reform and reaction in Costa Rica and Guatemala. Ph.D. dissertation, University of California, Berkeley. Zeitlin, M., & Ratcliff, R. (1984). Landlords and capitalists: The dominant class of Chile. Princeton: Princeton University Press.

PEASANTS, PLANTERS, AND THE PREDATORY STATE: EXPORT DIVERSIFICATION IN THE DOMINICAN REPUBLIC, 1970–2000 Andrew Schrank ABSTRACT What are the social and ecological roots of export diversification in the developing world? On the one hand, I attribute the growth of nontraditional, manufactured exports from the Dominican Republic to the traditional agro-export elite’s use of free trade zones to offset the consequences of urban biased, import-substituting industrialization in the 1970s, and thereby portray diversification as an incremental response to government predation rather than a coherent product of government planning. On the other hand, I hold that the nature, timing, and location of the nontraditional export supply response have necessarily been circumscribed by preexisting social and ecological circumstances, and thereby underscore the structural impediments to similar diversification efforts elsewhere in the developing world. My findings are of both theoretical relevance and policy import, for they serve to underscore the limitations to the regnant neoliberal development orthodoxy as well as the available sociological alternatives.

Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 353–371 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10016-X

353

354

ANDREW SCHRANK

1. INTRODUCTION Stephen Bunker’s approach to the political economy of the Third World cuts against the grain. While a number of his contemporaries attribute crossnational variation in economic performance to cross-national variation in the state’s willingness and ability to facilitate intersectoral capital mobility, and therefore use national-level case studies and cross-national comparisons to challenge the ‘‘myth of the powerless state’’ (Weiss, 1998; see also, e.g., Evans, 1995), Bunker attributes cross-national variation in the state’s willingness or ability to facilitate intersectoral capital mobility – for better or worse – to cross-national variation in social and ecological organization, and therefore uses subnational case studies and comparisons to ‘‘take local power bases, and their evolution through struggles against the state, into account’’ (Bunker, 1987, p. 16). In other words, he treats public policy as an endogenous, rather than an exogenous, variable. Which approach is superior? While a cross-national research strategy may bear fruit in the newly industrializing countries (NICs) of East Asia and Latin America, where political authority has long been consolidated and industrial development is already well underway, it would be ill-advised in their impoverished, poorly governed Third World counterparts, where the coherence of the state apparatus cannot be taken for granted and so-called development regimes are more likely to reﬂect than to guide the preferences of competing agents and organizations. In fact, Ronald Dore has called the very notion of a ‘‘development strategy’’ into question by asking whether ‘‘the mixture of forecasts and intentions and assumptions about probable causal sequences held by policy makers when they take economic decisions’’ is indeed self-conscious and coherent (Dore, 1990, p. 353; see also Haggard, 1990, p. 23), and his warning would appear to merit particularly careful consideration in the less developed countries (LDCs) of Africa, Asia, and Latin America. Nevertheless, the regnant approaches to the political economy of the Third World – whether orthodox or critical in orientation – offer no alternative to coherent, national-level accounts of social and economic performance, and therefore leave us unable to explain anything but chaos in the LDCs. Consider, for example, the case of the contemporary Dominican Republic. While the rapidly growing island nation has at long last evolved from a traditional agro-export economy into a relatively competitive export manufacturing platform, and has therefore been characterized as a ‘‘textbook case’’ of structural adjustment by representatives of the international ﬁnancial institutions (IFIs) (Giustiniani & Young, 2001, p. 351), it features

Peasants, Planters, and Predatory State

355

the highest average tariffs in Latin America, the highest barriers to entry in the world, an all but unparalleled level of political and bureaucratic corruption, and a decidedly ‘‘predatory’’ state – and the alleged ‘‘Dominican miracle’’ (Colitt, 1999; see also Oppenheimer, 1999; Velasco, 2002) is therefore difﬁcult to reconcile with either the regnant neoclassical orthodoxy or a critical, state-centered alternative (Lundahl & Vedovato, 1989; Pellerano, 1991).1 What would a superior approach to the political economy of the Third World look like? I hope to answer the question by examining the social and ecological roots of export diversiﬁcation in the Dominican Republic. On the one hand, I attribute the growth of nontraditional, manufactured exports to the traditional agro-export elite’s use of free trade zones (FTZs) to offset the consequences of urban biased, import-substituting industrial development in the 1970s, and therefore portray diversiﬁcation as an incremental response to government predation rather than a predictable consequence of either government ‘‘midwifery’’ or ‘‘husbandry’’ (Evans, 1995, pp. 13–14) or neoliberal market reform (Williamson, 1990). On the other hand, I hold that the nature, timing, and location of the nontraditional export (NTX) supply response have necessarily been circumscribed by preexisting social and ecological circumstances. While the southeastern sugar towns of La Romana and San Pedro have traditionally been dominated by foreign sugar corporations and their executives, and have therefore lured foreign investors toward their FTZs, the tobacco-, cocoa-, and coffee-growing Cibao Valley has traditionally been dominated by an indigenous commercial class (ICC) of traders, ﬁnanciers, and processors, and has therefore engendered more dynamic, domesticated FTZs.2 In other words, I hold that the Dominican Republic’s diversiﬁcation effort is not only an outgrowth of ‘‘struggles against the state,’’ as Bunker would most certainly have anticipated, but that it has in many ways been circumscribed by the regionally distinctive social and ecological characteristics of the country’s traditional agricultural exports and exporters: foreign sugar planters, native-born peasants, and the capitalists, merchants, and middlemen who buy, sell, ﬁnance, and process their tobacco, coffee, and cocoa. My ﬁndings are of both theoretical relevance and policy import, for they serve to underscore the limitations to the regnant neoliberal development orthodoxy as well as the available sociological alternatives. While the IFIs frequently portray the Dominican Republic as a potentially replicable neoliberal success story (Giustiniani & Young, 2001; see also Oppenheimer, 1999; Lozada, 2003), the evidence I bring to bear suggests that it is neither neoliberal, replicable, nor in any meaningful sense a success story. In fact,

356

ANDREW SCHRANK

the Dominican Republic’s greatest achievement, the domestication of the northern Cibao Valley’s FTZs, is not even being replicated in the rest of the country, let alone the rest of the world – a telling fact which would appear to call the international donor community’s faith in both the Dominican Republic’s achievements and their potential transferability into question. I have divided the remainder of my chapter into three principal sections. First, I provide the necessary historical background on the Dominican Republic’s diversiﬁcation experience. Second, I compare the relatively successful diversiﬁcation efforts of the country’s two principal agro-export regions – the smallholding north and the plantation southeast – to the less successful experiences of the subsistence-oriented southwest, the industrial and political capital of Santo Domingo. And, ﬁnally, I discuss my ﬁndings and their implications for sociological theory, development policymaking, and future research.

2. THE SOCIAL AND ECOLOGICAL ROOTS OF EXPORT DIVERSIFICATION Stephen Bunker’s approach to the political economy of development is in many ways the product of two distinct ﬁeldwork experiences: his dissertation research on ‘‘the politics of market control’’ in Bugisu, Uganda; and his subsequent inquiry into rural development and ‘‘the failure of the modern state’’ in the Brazilian Amazon (see Bunker, 1981, 1985, 1987). According to Bunker, the two settings have little in common. While the fertile soils of eastern Uganda play host to an autochthonous peasantry cultivating coffee (and to a lesser extent cotton) on family sized farms, a dense middle class of administrators, civil servants, and traders, and a relatively egalitarian distribution of income, power, and status, the Amazon’s tropical lowlands feature a growing population of migratory workers and farmers, a largely alien administrative elite, and a decidedly inequitable distribution of income, power, and status. Nevertheless, the relationship between the revealed differences in social and political circumstances and the underlying differences in material conditions (i.e., climate, ecology, topography, geography, and demography) has in many ways, according to Bunker, been mediated by the organization of agricultural or rural activity in both regions. Labor-intensive peasant or family farming is associated with a graduated class structure, participatory politics, and local control in Uganda. And capital intensive or extractive

Peasants, Planters, and Predatory State

357

activities (i.e., large-scale rubber tapping, mining, logging, and ranching) are associated with bipolar class structures, exclusionary politics, and external control in the Amazon (see Bunker (1981) for an explicit comparison). While Bunker’s framework has been applied to a variety of different settings in a number of different countries, it is perhaps surprisingly relevant to the twentieth century history of the Dominican Republic. After all, the Dominican Republic’s two principal agro-export regions, the northern Cibao Valley and the southeastern coastal plain, bear a striking – albeit markedly incomplete and perhaps diminishing – resemblance to eastern Uganda and the Amazon. On the one hand, the fertile Cibao Valley has traditionally played host to both peasant smallholders and the merchants, traders, and middlemen who buy, sell, ﬁnance, and process their tobacco, coffee, and cocoa, and it therefore retains a broad resemblance to Bugisu. On the other hand, the southeastern coastal plain has played host to both North American sugar corporations and the Hatitian and Dominican workers who toil in their ﬁelds and factories, and it therefore shares a number of characteristics – i.e., external control, semi-servile and migrant labor, deeply rooted inequality – with the Amazon (see, e.g., Baud, 1987; Hoetink, 1982; Lozano, 1985; Moya Pons, 1995). Table 1 offers a brief summary of the key differences between the two regions through the middle of the twentieth century. Table 1. Agro-export Regions in the Dominican Republic – Early to Mid-twentieth Century. Variable Climate Topography Population density Soil fertility Principal crops Agrarian organization Capital intensity Dominant class

North (Cibao Valley)

Subordinate class

Mild, temperate Hilly to mountainous Relatively high High Tobacco, coffee, cocoa Family farms Low Larger merchants, traders, ﬁnanciers Middlemen and larger farmers Peasants and family farmers

Principal cities

Santiago

Middle class

Source: Schrank (forthcoming).

Southeast Hot, tropical Flat plains Relatively low Medium to low Sugar Plantations High Foreign planters and agribusiness Insignificant Mill workers and migrant laborers La Romana, San Pedro

358

ANDREW SCHRANK

The Dominican story is complicated, however, by the mid-twentieth century accession of General Rafael Leonidas Trujillo. After all, ‘‘El Jefe,’’ as the dictator was known, treated the Dominican Republic as his personal ﬁefdom from 1930, when he assumed power in a coup d’etat, until 1961, when he was assassinated. He nationalized 12 of the Dominican Republic’s 16 sugar plantations, assumed control over 35 percent of the country’s arable land, and used the available agricultural surplus to fuel and take advantage of an aggressive import-substituting industrialization (ISI) effort in the capital of Santo Domingo (Moya Pons, 1995). Consequently, by the time of his assassination in 1961 the ruthless dictator had broken the back of the plantation sector, increased the size and authority of the state, and transformed the coastal city of Santo Domingo from ‘‘an old administrative and commercial capital into a thriving industrial and commercial center that attracted increasing waves of immigrants from the interior of the country’’ (Moya Pons, 1995, p. 363). While Trujillo’s successors would intensify the process of ISI in Santo Domingo, and thereby accelerate the growth of urban primacy in the country as a whole, they would simultaneously offer manufacturers who would agree to export the entirety of their output from legally and geographically circumscribed FTZs with a wide variety of tax and tariff incentives, and thereby open the door to export-led industrialization (ELI) as well. In other words, they would carve islands of free trade, or FTZs, out of the otherwise heavily protected Dominican economy and thereby reconcile the eventual export of import-dependent manufactured goods like textiles, clothing, and footwear with the maintenance of import-substitution. Nevertheless, the earliest FTZs are best portrayed not as part of a coherent central government effort to foster export diversiﬁcation, but as part of an incremental secondary city effort to arrest the growth of Santo Domingo and the pernicious consequences of urban primacy (Schrank, 2003). Gulf & Western of the Americas, the North American owner of the Dominican Republic’s largest sugar estate, established the country’s ﬁrst free zone in La Romana in 1969. The Industrial Promotion Corporation (Corporacio´n de Fomento Industrial, CFI), for the most part a marginal government agency dedicated to the encouragement of industrial development, established a second zone – at the request of municipal elites – in the nearby sugar town of San Pedro in 1973. The Association for Development, Inc. (Asociacio´n para el Desarrollo, Inc., APEDI), a private development agency founded by northern boosters, established the country’s third FTZ in the agro-commercial center of Santiago in 1974. And by 1980 the three zones played host to 71 ﬁrms, employed more than 15,000 workers, exported

Peasants, Planters, and Predatory State

359

more than US$100 million worth of manufactured goods, and claimed at least part of the responsibility for the hinterland’s unanticipated social and demographic revival (see CNZF, 1999, 2002; see also Tejada, 2003). The NTX sector would grow in size, scope, and import in the late 1980s, however, for the Latin American debt crisis would enervate the domestic market, lower the real wage rate, and thereby increase both the desirability and the accessibility of markets and consumers overseas (see Table 2). In fact, the deeply indebted Dominicans would respond to pressure from their foreign creditors not only by building a host of new free zones in cities like La Vega, Puerto Plata, San Cristobal, and Santo Domingo, but also by liberalizing the dollar to peso exchange rate – and thereby lowering their real wage rate – in 1985. The FTZs would therefore go on to attract more ﬁrms in the next half decade than they had in their ﬁrst 15 years of existence combined. The expansion would continue into the 1990s, and by the end of the decade more than 80 percent of the Dominican Republic’s exports – and almost all of the country’s nontraditional, manufactured exports – would originate in the FTZs. But the theoretical and policy implications of the Dominican Republic’s NTX experience are nonetheless unclear. While mainstream economists tend to attribute the NTX boom to the growth of competition and integration in the free zones (see, e.g., Willmore, 1993, 1995a, b; World Bank, 1995, 1999; Warden, 2000; Panzer & Soto, 2001), and therefore endorse the rapid liberalization of the rest of the Dominican economy, their revisionist critics acknowledge the allegedly valuable legacy of a prior round of import-substituting industrialization, and therefore counsel caution. For example, Thomas Klak and Jamie Rulli hold that the Dominicans ‘‘developed the infrastructure and ties to foreign capital necessary for export industrialization’’ not by embracing foreign competition Table 2.

FTZ Growth in the Dominican Republic – 1980–2000.

Factor No. of FTZs No. of ﬁrms No. of workers Value of exports (million US$) FTZ exports % total exports Dominican ﬁrms % all ﬁrms

1980

1985

1990

1995

2000

3 71 16,440 117 11 NA

3 136 30,902 215 23 11

25 331 130,045 839 53 27

33 469 165,571 2,907 79 29

46 481 195,262 4,771 86 35

Source: Schrank (forthcoming). Data: Services Group (1991); Matthews (1995); and CNZF (1996, 1999, 2001, 2002).

360

ANDREW SCHRANK

outright, but by pursuing ‘‘a relatively extensive ISI policy based on foreign investment between 1966 and 1976’’ (Klak & Rulli 1993, p. 141; Hillcoat & Quenan 1991, p. 215). Thus, the Dominican Republic’s admittedly limited achievements inspire a return to a pregnant question ﬁrst asked of the East Asian NICs in the 1980s. Is the ISI era, to paraphrase Robert Wade, best portrayed as an unfortunate accident or a necessary prelude to export-led industrialization (Wade, 1988, p. 151; see also Glick & Moreno, 1997)? Are the Dominican Republic’s export-oriented manufacturers the offspring of ISI or products of the FTZs? And, ﬁnally, what are the implications for the generality or potential transferability of the so-called Dominican model of export diversiﬁcation? I plan to answer the aforementioned questions by unpacking the social and geographic underpinnings of the Dominican Republic’s NTX drive in the next section. While I ﬁnd that the country’s most successful nontraditional exporters are indeed products of the free zones rather than descendants of ISI, and thereby call the revisionist account into question, I hold that they are products of particular free zones in a single region with unique characteristics – and that their achievements are therefore unlikely to be replicated in the rest of the Dominican Republic let alone in the rest of the Third World.

3. COMPARATIVE REGIONAL PERFORMANCE The two accounts of NTX growth I have identiﬁed – orthodox and critical – have spatial as well as sociological implications. While mainstream analysts treat the Dominican Republic’s nontraditional exporters as utility maximizing ‘‘global scanners’’ (Vernon, 1979), and therefore expect their location decisions to reﬂect labor, property, and transportation cost considerations, their revisionist critics treat nontraditional exporters as products of ISI, and therefore expect their location decisions to reﬂect the country’s centralized industrial legacy (see, e.g., Krugman (1996) and Markusen (1996) for contrasting views on development strategies and industrial location). Which perspective is more accurate? Table 3 cross-classiﬁes the 481 ﬁrms in the Dominican Republic’s FTZs in 2000 by origin (i.e., foreign or local) and location and counsels against a straightforward revisionist interpretation. After all, Santo Domingo is home to approximately 75 percent of the Dominican Republic’s import-competing industrial establishments (Moya Pons, 1992, p. 410) and nonetheless features no more than 20 percent of

Peasants, Planters, and Predatory State

Table 3. Origin

361

FTZ Manufacturers by Region and National Origin – 2000. Southeast

North

Santo Domingo

Southwest

Total

Foreign Local

139 21

110 116

63 28

3 1

315 166

Total

160

226

91

4

481

Source: Schrank (forthcoming). Data: CNZF, Encuesta Octubre 2000, dated February 8, 2001. The w2 statistic is 61.3798 with three degrees of freedom; the differences across the regions are significant at po0.01. I have used the National Planning Ofﬁce (Oficina Nacional de Planificacio´n, or ONAPLAN) regional classiﬁcation system rather than the administrative categories used by the CNZF. While ONAPLAN’s system is slightly different from the CNZF system, it is substantive rather than simply administrative and is consistent with the government data invoked below. Neither the substantive nor the statistical results are affected by choice of classiﬁcation system.

their counterparts in the free zones – and, perhaps surprisingly, a slightly smaller percentage of the country’s locally owned free zone enterprises. The capital’s failure to export is not for lack of effort. The government not only built or approved the establishment of seven different FTZs in the city in the 1980s and 1990s, but also offered traditionally import-competing manufacturers the right to take advantage of the free zone regime’s tax and tariff exemptions without relocating their operations to a physically circumscribed FTZ by simply declaring their already existing enterprises ‘‘special free zones.’’ But Santo Domingo’s FTZs are nonetheless sparsely populated. The owners of the special free zones have for the most part declared bankruptcy or returned to the ISI sector. And their traditionally import-competing neighbors have therefore learned to avoid world markets entirely (Schrank, forthcoming). What accounts for the capital’s unanticipated inability to take advantage of the Dominican Republic’s free zone regime? The city’s traditional manufacturers, according to interviews undertaken by the author in 1998 and 2001, are trapped in a vicious circle of inefﬁciency, ignorance, and isolation. On the one hand, they are unable to meet foreign price, quality, and delivery standards, and they have therefore had difﬁculty ﬁnding and retaining buyers overseas. On the other they have had trouble learning how to meet foreign price, quality, and delivery standards, for the buyers in question are the principal sources of the relevant information (Keesing & Lall, 1992; Schmitz & Knorringa, 2000). Thus, the capitalen˜os are at long last paying a high price for their half century of protection and isolation from world markets.

362

ANDREW SCHRANK

While the capital’s travails would appear to call the revisionist interpretation of the Dominican Republic’s diversiﬁcation effort into question, they are not necessarily compatible with an orthodox, factor cost approach either. After all, the orthodox approach expects utility maximizing investors to search for low cost production sites. But the capital’s FTZs offer their tenants the lowest transportation costs and the second lowest unskilled labor costs in the country but are nonetheless sparsely populated. And the northern Cibao Valley’s zones offer their tenants high labor and transportation costs but are nonetheless densely populated (see Table 4).3 What, then, are the underpinnings of northern success? Unlike Santo Domingo, the northern agro-commercial center of Santiago traditionally embraced international markets and competition, and the entrepreneurs and boosters who built the city’s ﬁrst FTZ in the 1970s thereby engendered a positive feedback loop of foreign investment, local learning, and NTX growth. Consequently, the northern free zones are dominated not by the foreign investors who populate their counterparts in the rest of the country, but by cibaen˜o investors who began their manufacturing careers as either partners of foreign investors or managers or engineers in the free zones, learned how to make quality products at competitive prices from their foreign associates and employers, and therefore had little difﬁculty meeting the

Table 4. Region

Unskilled Labor (pesos/ week)

Factor Costs by Region – 2000. Skilled Labor (pesos/week)

Distance to Port (kilometers)

No. of EPZs

Southeast North Santo Domingo Southwest

675 (100) 751 (156) 699 (82)

1,527 (446) 1,704 (484) 1,788 (404)

31 (26) 68 (25) 15 (6)

15 21 7

700 (141)

2,009 (1,638)

70 (92)

2

Average

716 (130)

1,671 (518)

48 (34)

45

Source: Schrank (forthcoming). Data: CNZF (2001). The cell values are regional averages; the parenthesized values are standard errors. The interregional wage differences are not statistically significant; the interregional distance differentials are statistically significant at pp0.001 under either one way analysis of variance or a nonparametric Kruskal–Wallis test; however, the table includes data from all but one of the free zones operating in 2000 and should therefore be treated as more or less representative of the full population of FTZs rather than a sample. The north is missing one observation on skilled labor and one observation on distance; the southeast is missing two observations on distance; and the capital is missing one observation on distance.

Peasants, Planters, and Predatory State

363

price, quality, and delivery standards imposed by northern buyers when they established their own ‘‘breakaway’’ operations in the 1980s and 1990s (Jacobs, 1984, p. 101). Indeed, the long-term contrast between Santo Domingo and the Cibao Valley could not be more striking. While the traditionally import-competing capital descended into a vicious circle of inefﬁciency and ignorance, and thereby garnered a debilitating reputation for high costs and shoddy workmanship, the traditionally agro-commercial Cibao Valley ascended into a virtuous circle of efﬁciency, education, and export growth, and thereby engendered a number of decidedly competitive, vertically integrated, and locally owned exporters who are all but cornering the – admittedly fragile (see, e.g., Schrank, 2004) – market, for textiles, clothing, and footwear originating in the Dominican Republic (see Dahle, 2000; Gonza´lez, 2003; Schrank, 2003). The Cibao Valley’s achievements are not obviously replicable, however, for the southeastern cities of La Romana and San Pedro built their FTZs before Santiago, have lower costs than the cibaen˜os, lure more foreign investors than any other region in the Dominican Republic, and have nonetheless failed to engender an equally dynamic class of indigenous – or even foreign – exporters. What is the key difference between the northern Cibao Valley and the southeastern coastal plain? I would trace the principal difference to the traditional system of agrarian organization: peasant versus plantation. While the export of tobacco, coffee, and cocoa grown by family farmers requires a dense, knowledgeable class of traders, bankers, and processors, and therefore encourages local control, indigenous capital formation, and the growth of a potentially entrepreneurial middle class, the export of sugar grown on capital intensive plantations requires large ﬁxed investments by oligopolistic northern ﬁrms, and therefore militates against the emergence of an upper or even middle class likely to invest in FTZs down the road (see Beckford, 1972; Paige, 1975).4 The consequences would be clear by the early 1980s. While Gulf & Western and to a lesser extent the CFI had lured foreign investors to La Romana and San Pedro, and had thereby created ‘‘dependent’’ FTZs in the southeast, the northern elite had not only encouraged but had actively contributed to indigenous as well as foreign investment in Santiago, and had thereby created ‘‘domesticated’’ free zones in the Cibao Valley. Thus, Gerold Janka, who carried out the ﬁrst systematic study of the free zone sector in 1986, found that the Santiago zone already played host to more than twice as many indigenous investors as the La Romana and San Pedro zones combined (Janka, 1986, p. 96; see also Me´ndez de Crouch, 1988, p. 7).

364

ANDREW SCHRANK

The point is not that all of the locally owned ﬁrms in the northern FTZs are owned by members of the traditional Santiago elite. A number of the most successful enterprises are owned by upwardly mobile entrepreneurs who learned how to make and export textiles, clothing, and footwear by working as managers and engineers in the free zones in the 1970s and 1980s. But members of the traditional elite made an essential contribution to the Cibao Valley’s success, for they not only established a number of the region’s more prosperous ﬁrms but also offered their parvenu co-regionalists economic and political support including low-cost loans, equity partnerships, education, and training programs, and the establishment of the Santiago free zone in the ﬁrst place. They did so, I would argue, for instrumental reasons as well as altruistic ones, for the tax free FTZs promised not only to revitalize their regional economy – and to thereby salvage their extensive ﬁxed investments in property and immobile assets – but to shelter their individual earnings from the hands of the predatory state. In short, the evidence that I have adduced suggests that the Dominican Republic’s ongoing diversiﬁcation effort is a product of neither ISI nor market opening per se but of ‘‘struggles against the state’’ which have played out differently in different regions of the country. Fig. 1 offers a bivariate representation of the overarching argument. The key variables are the depth of world market integration in the ISI era and the strength of the indigenous capitalist class. The former accounts for the desire or willingness to encourage export-oriented manufacturing – especially in the pre-debt crisis era – and therefore holds the key to the origins and popularity of the different zones. The latter accounts for the ability to exploit the new international division of labor and therefore speaks to the variation in the degree of domestication of the zones (cf. Table 3). For example, the northern Cibao Valley features a tradition of world market integration and an indigenous commercial class (Rosario, 1997), and it has therefore been particularly successful in the NTX era. The northern Strength of indigenous capitalist class

Degree of world market integration

Low High

Fig. 1.

Low

High

Southwest Depopulated FTZs Southeast Dependent FTZs

Santo Domingo Disappointing FTZs North/Cibao Valley Domesticated FTZs

Bivariate Explanation of Regional Outcomes in the Dominican Republic. Source: Modiﬁed from Schrank (forthcoming).

Peasants, Planters, and Predatory State

365

elite has not only lured foreign investors to Santiago but has also encouraged joint ventures, human and physical capital formation, and indigenous investment, and the Cibao Valley’s FTZs have therefore been ‘‘domesticated.’’ The southeastern plantation zone, on the other hand, features a tradition of world market integration and a foreign dominant class, and it has therefore been characterized as ‘‘dependent.’’ In other words, the foreign investors who dominate La Romana and San Pedro have found neither local partners nor an upwardly mobile middle class, and the southeastern FTZs have therefore continued to rely upon foreign transplants. Santo Domingo features a large class of import-competing manufacturers and little experience in world markets, and the capital’s free zones are therefore neither particularly popular nor especially domesticated. They are, I would argue, ‘‘disappointing.’’ And the southwest features a tradition of peasant subsistence agriculture and isolation from world markets, and the southwestern free zones are therefore home to neither foreign nor domestic investors. They are, on the contrary, ‘‘depopulated.’’

4. DISCUSSION The new international division of labor allegedly calls forth winners and losers in the Third World. While the winners build new sources of comparative advantage, and are therefore identiﬁed with nontraditional – and in particular manufactured – export growth, the losers are unable to reconstruct their comparative advantages, and are therefore identiﬁed with deindustrialization and the revival of traditional agricultural and mineral exports (Bunker, 1991; Portes, 1997; Evans, 1998). The East Asian NICs represent the characteristic winners. And the least industrialized countries of Asia, Africa, and Latin America represent the characteristic losers. What differentiates the winners from the losers? The question is almost invariably answered by way of reference to development strategies. The developing country’s success stories, according to most observers, have pursued ‘‘the right policies,’’ whatever they may be, and similarly positioned LDCs can replicate their achievements by following their leads. Nevertheless, the policy-centric approach suffers from at least three different errors. First, it suggests that so-called development strategies are far more coherent than they really are. The invocation of the noun ‘‘strategy,’’ according to Stephan Haggard, ‘‘implies a purposiveness of state action that

366

ANDREW SCHRANK

may not exist; imputation of a central design requires caution. Strategies emerge by default, trial-and-error, and compromise; they take years to crystallize; and they are often plagued by internal inconsistencies’’ (Haggard, 1990, p. 23). Second, it overlooks methodological difﬁculties fostered by the sequential – as well as incremental – nature of development policy making. Most countries move from so-called strategy to the so-called strategy over time, and it is therefore difﬁcult to assign credit or blame for particular outcomes with any degree of certainty. And, third, it ignores subnational variations in performance which are not obviously attributable to industrial policy choices at the national level. The limits of the current approach are readily illuminated by the growing debate over the Dominican Republic’s recent growth performance. A decade ago, for example, Alain Touraine warned against the Dominican Republic’s impending ‘‘incorporation into the North American zone as a marginal entity’’ (Touraine, 1994, p. 58). The island nation featured a predatory state, a feeble economy, and a bleak future. And, yet, by the late 1990s the Dominican Republic had been transformed into the darling of the international donor community – classiﬁed as a ‘‘miracle’’ and treated as a model to be imitated throughout the Third World (see, e.g., Oppenheimer, 1999; Velasco, 2002; Lozada, 2003). What happened? The key industrial policies have not changed. The FTZs are more than 30 years old. The exchange rate was liberalized more than two decades ago. And the Dominican Republic retains the highest tariffs in Latin America, the highest barriers to entry in the world, and a reputation for pronounced legislative and administrative corruption. But the country’s rapid export and economic growth rates – the highest in Latin America not only over the past decade but over the last four decades (see Sachs & Vial, 2002, p. 24) – are ultimately reconcilable with neither Dominican history nor contemporary development theory. So Dominican history has to be rewritten to accommodate the theory: we therefore read about a coherent development strategy (albeit one which differed depending on the source), a transparent state, and a bright Dominican future.5 Such optimism, I have argued, is neither necessary nor warranted. It is rendered unnecessary by a superior approach which acknowledges variation within as well as across development regimes, treats economic performance outcomes as consequences of social conﬂict as well as coherent government action, and thereby reconciles investment, exports, and growth in the short run with structural continuity in the long run. And it is rendered unwarranted by the many impending constraints on the Dominican Republic’s economic performance (e.g., human resource shortfalls, infrastructural

Peasants, Planters, and Predatory State

367

bottlenecks, and institutional limits imposed by the still predatory state) – constraints which are apparent to the Cibao Valley’s nontraditional exporters, by the way, but not to their admirers in the IFIs (see, e.g., Pen˜a Brito, 2003). What, then, are the broader implications of the Dominican Republic’s diversiﬁcation experience? On the one hand, the neoliberals have a point. A small, island nation like the Dominican Republic is likely to pay a high price for isolation from the world economy over the long run, and it is therefore not surprising that import-competing Santo Domingo and the traditionally isolated southwest – unlike the traditionally export-oriented north and southeast – are having difﬁculty adjusting to world market competition. On the other hand, the neoliberals are overly optimistic, for the Dominican experience suggests that the requisites of successful ELI are not only extraordinarily restrictive but go beyond tractable policy variables like the exchange rate or the tariff level to include deeply entrenched, structural features of the economy and society. Thus, the growth of indigenous investment in the FTZs indicates not, contra the regnant neoclassical orthodoxy, that Dominican ﬁrms ‘‘are capable of competing in the international marketplace without protection, if given a ‘level playing ﬁeld’’’ (Warden, 2000, p. 8; see also Willmore, 1995b, p. 534; World Bank, 1995, p. 1) but that cibaen˜o ﬁrms are capable of competing in the international marketplace with, and perhaps only with, the backing of a powerful regional elite. The rest of the country’s growing reliance upon foreign investment, on the one hand, or outright failure to adjust to world market competition, on the other, suggests that the real beneﬁciaries of the new international division of labor are likely to be found not in the LDCs but in the core of the world economy – a lesson Stephen Bunker has underscored time and time again.

NOTES 1. See Djankov, La Porta, Lopez de Silanes and Shleifer (2002) for cross-national survey data on barriers to entry. See the World Bank (2000) for data on average tariffs. And see Transparency International (2002) for data on corruption. 2. According to Ernest J. Wilson, an indigenous commercial class ‘‘is based in trade, commerce, control of property, and only very marginally in industry’’ and therefore tends to favor ‘‘more laissez-faire, less interventionist policies’’ (Wilson, 1990, pp. 405–406). The wealth and authority of the Cibao Valley ICC are rooted, as Jeffrey Paige would have argued, in ‘‘the control of processing machinery, storage facilities, transportation, and ﬁnance capital’’ (1975, p. 13).

368

ANDREW SCHRANK

3. Schrank (2003) offers a multivariate assessment of the factor cost approach. 4. Note that I am not attributing the southeast’s persistent ‘‘dependency’’ to the existence of plantations per se but to the legacy of foreign-ownership which undermined local capital formation in the early to mid-twentieth century. My argument, therefore, has more in common with Beckford’s approach (1972, esp. p. 43, note 18) than with Michael Shafer’s more recent intervention (Shafer, 1994). Shafer’s case for the inﬂexibility of plantation systems in general – but not Beckford’s focus on the principal locus of plantation ownership and control – would appear to be gainsaid by the Franco-Mauritian sugar elite’s recent entry into export manufacturing (see Wellisz & Saw, 1993). The key difference between the Mauritian and Dominican diversiﬁcation experiences, I would argue, lies in the organization of the plantation sector prior to diversiﬁcation: local ownership in Mauritius versus foreign ownership in the Dominican Republic. 5. See, for example, LatinFinance (2002) on the allegedly ‘‘transparent’’ Dominican economy (cf. Sobrevela, 2003).

ACKNOWLEDGEMENTS I would like to thank the Institute for International Education for funding, the National Council of Free Zones for data, my respondents for their patience, honesty, and trust, and Stephen Bunker for his unwavering support.

REFERENCES Baud, M. (1987). The origins of capitalist agriculture in the Dominican Republic. Latin American Research Review, 21(2), 136–158. Beckford, G. (1972). Persistent poverty: Underdevelopment in plantation economies of the Third World. Oxford: Oxford University Press. Page references from 1983 edition published by Zed Books. Bunker, S. (1981). Class, status, and the small farmer: Rural development programs and the advance of capitalism in Uganda and Brazil. Latin American Perspectives, 8, 89–107. Bunker, S. (1985). Underdeveloping the Amazon: Extraction, unequal exchange, and the failure of the modern state. Chicago: University of Chicago Press. Bunker, S. (1987). Peasants against the state: The politics of market control in Bugisu, Uganda, 1900–1983. Urbana: University of Illinois Press. Bunker, S. (1991). The social impact of the 1980s crises. In: W. Baer, et al. (Eds), Latin America: The crisis of the eighties and the opportunities of the nineties (pp. 205–223). Champaign: Bureau of Economic and Business Research. Colitt, R. (1999). Sell-offs help Dominican economy excel. Financial Times, December 2. Consejo Nacional de Zonas Francas, (CNZF). (Various years). Informe Estadı´stico. Santo Domingo: CNZF.

Peasants, Planters, and Predatory State

369

Dahle, C. (2000). The new fabric of success. Fast Company, 35(June), 252–262. Djankov, S., La Porta, R., Lopez de Silanes, F., & Shleifer, A. (2002). The regulation of entry. Quarterly Journal of Economics, 117, 1–37. Dore, R. (1990). Reﬂections on culture and social change. In: G. Gerefﬁ & D. Wyman (Eds), Manufacturing miracles: Paths of industrialization in Latin American and East Asia (pp. 353–367). Princeton: Princeton University Press. Evans, P. (1995). Embedded autonomy: States and industrial transformation. Princeton: Princeton University Press. Evans, P. (1998). Transnational corporations and Third World states: From the old internationalization to the new. In: R. Kozul-Wright & R. Rowthorn (Eds), Transnational corporations and the global economy (pp. 195–224). Helsinki: WIDER. Giustiniani, A., & Young, P. (2001). Dominican Republic makes dramatic progress in boosting economic growth and reducing poverty. IMF Survey, 30(21), 351–353. Glick, R., & Moreno, R. (1997). The East Asian miracle: Growth because of government intervention and protection or in spite of It? Business Economics, 32, 20–26. Gonza´lez, S. (2003). Jo´venes empresarios de Santiago se convierten en motor de desarrollo. Hoy 1 abril. Haggard, S. (1990). Pathways from the periphery: The politics of growth in the newly industrializing countries. Ithaca: Cornell University Press. Hillcoat, G., & Quenan, C. (1991). International restructuring and re-specialization in the Caribbean. Caribbean Studies, 24(1–2), 191–223. Hoetink, H. (1982). The Dominican people: Notes for a historical sociology, S. K. Ault (trans.), Baltimore: Johns Hopkins University Press. Jacobs, J. (1984). Cities and the wealth of nations: Principles of economic life. New York: Random House. Janka, G. (1986). Estudio de las zonas francas industriales de La Romana, San Pedro de Macorı´s, Santiago y Puerto Plata. Santo Domingo, CFI. Keesing, D., & Lall, S. (1992). Marketing manufactured exports from developing countries: Learning sequences and public support. In: G. K. Helleiner (Ed.), Trade policy, industrialization, and development (pp. 176–193). Oxford: Clarendon. Klak, T., & Rulli, J. (1993). Regimes of accumulation, the Caribbean basin initiative, and export processing zones: Scales of inﬂuence on Caribbean development. In: E. Goetz & S. Clarke (Eds), The new localism: Comparative urban politics in a global era (pp. 117–150). Newbury Park: Sage. Krugman, P. (1996). Urban concentration: The role of increasing returns and transport costs. International Regional Science Review, 19(1/2), 5–30. LatinFinance. (2002). Bolstering local business. LatinFinance, 151(November), 23–24. Lozada, C. (2003). Think again: Latin America. Foreign Policy, 135, 18–23. Lozano, W. (1985). Proletarianizacio´n y Campesinado en el Capitalismo Agroexportador. Santo Domingo: INTEC. Lundahl, M., & Vedovato, C. (1989). The state and economic development in Haiti and the Dominican Republic. Scandinavian Economic History Review, 37(3), 39–59. Markusen, A. (1996). Interaction between regional and industrial policies: Evidence from four countries. International Regional Science Review, 19(1/2), 49–77. Matthews, D. (1995). Caracterı´ sticas operacionales de las empresas localizadas en las zonas francas de exportacio´n de la Repu´blica Dominicana. Caribbean Studies, 28(2), 255–284.

370

ANDREW SCHRANK

Me´ndez de Crouch, J. (1988). El boom de las zonas francas. El Industrial, 232, 5–8. Moya Pons, F. (1992). Empresarios en conflicto: Polı´ticas de industrializacio´n y sustitucio´n de importaciones en la Repu´blica Dominicana. Santo Domingo: Fondo para el Avance de las Ciencias Sociales. Moya Pons, F. (1995). The Dominican Republic: A national history. New Rochelle: Hispaniola Books. Oppenheimer, A. (1999). Dominican Republic seen as role model. Miami Herald, November 28. Paige, J. (1975). Agrarian revolution: Social movements and export agriculture in the underdeveloped world. New York: Free Press. Panzer, J., & Soto, R. (2001). Capital accumulation, total factor productivity, and growth. In: P. Young (Ed.), The Dominican Republic: Stabilization, reform, and growth (pp. 37–46). Washington, DC: IMF. Pellerano, F. (1991). Schumpeter y el Estado Dominican: del Estado Depredador al Estado Empresarial. In: F. Pellerano (Ed.), Apertura y Reformas Estructurales: El Desafio Dominicano (pp. 213–219). Santo Domingo: Siglo XXI. Pen˜a Brito, A. (2003). Entrevista: ‘‘El futuro esta´ en la diversiﬁcacio´n.’’ El Caribe 27 abril. Portes, A. (1997). Neoliberalism and the sociology of development: Emerging trends and unanticipated facts. Population and Development Review, 23, 229–259. Rosario, E. (1997). La oligarquı´a de Santiago. Santo Domingo: Central. Sachs, J., & Vial, J. (2002). Can Latin America compete? In: K. Schwab, et al. (Eds), Latin American competitiveness report 2001–2002 (pp. 10–28). New York: Oxford University Press. Schmitz, H., & Knorringa, P. (2000). Learning from global buyers. Journal of Development Studies, 37(2), 177–205. Schrank, A. (2003). Luring, learning, and lobbying: The limits to capital mobility in the Dominican Republic. Studies in Comparative International Development, 37, 89–116. Schrank, A. (2004). Ready-to-wear development? Foreign investment, technology transfer, and learning-by-watching in the apparel trade. Social Forces, 83, 123–156. Schrank, A. (forthcoming). Entrepreneurship, export diversiﬁcation, and economic reform: The birth of a ‘‘Developmental Community’’ in the Dominican Republic. Comparative Politics. Services Group. (1991). Rationalization of free zone policies in the Dominican Republic. Arlington: Services Group. Shafer, M. (1994). Winners and losers: How sectors shape the developmental prospects of states. Ithaca: Cornell University Press. Sobrevela, V. (2003). 90%: hay corrupcio´n; 65%: no se combate. Hoy 21 febrero. Tejada, V.M. (2003). La ‘‘nueva economı´ a’’ empuja la migracio´n interna. El Caribe 7 febrero. Touraine, A. (1994). From the mobilising state to democratic politics. In: C. Bradford (Ed.), Redefining the state in Latin America (pp. 45–65). Paris: OECD. Transparency International. (2002). Corruption perceptions index 2002. Berlin: Transparency International. Velasco, A. (2002). Dependency theory. Foreign Policy, 133, 44–45. Vernon, R. (1979). The product cycle hypothesis in a new international environment. Oxford Bulletin of Economics and Statistics, 41, 255–267. Wade, R. (1988). The role of government in overcoming market failure: Taiwan, Republic of Korea, and Japan. In: H. Hughes (Ed.), Achieving Industrialization in East Asia (pp. 129–163). Cambridge: Cambridge University Press.

Peasants, Planters, and Predatory State

371

Warden, S. (2000). Assessing export platforms: The case of the Dominican Republic. Harvard/ CAER II Discussion Paper no. 77, Cambridge. Weiss, L. (1998). The myth of the powerless state. Ithaca: Cornell University Press. Wellisz, S., & Saw, P. L. S. (1993). Mauritius. In: R. Findlay & S. Wellisz (Eds), The political economy of poverty, equity, and growth: Five small, open economies (pp. 219–255). Oxford: Oxford University Press/World Bank. Williamson, J. (1990). Latin American adjustment: How much has happened? Washington, DC: Institute for International Economics. Willmore, L. (1993). Export processing in the Dominican Republic: Ownership, linkages and transfer of technology. Port of Spain: ECLAC. Willmore, L. (1995a). Dominican Republic: Domestic policies that affect its manufacturing sector. Port of Spain: ECLAC. Willmore, L. (1995b). Export processing zones in the Dominican Republic: A comment on Kaplinsky. World Development, 23(3), 529–535. Wilson, E. (1990). Strategies of state control of the economy: Nationalization and indigenization in Africa. Comparative Politics, 22(4), 401–419. World Bank. (1995). The Dominican Republic. Growth with equity: An agenda for reform. Washington, DC: World Bank. World Bank. (1999). Memorandum of the president of the international bank for reconstruction and development to the executive directors on a country assistance strategy of the world bank Group for the Dominican Republic. Washington, DC: World Bank. World Bank. (2000). Data on import tariffs and NTBs. Accessed May 27, 2003. http:// www1.worldbank.org/wbiep/trade/data/TR_Data.html

SELLING THE RIVER: GENDERED EXPERIENCES OF RESOURCE EXTRACTION AND DEVELOPMENT IN LESOTHO Yvonne A. Braun ABSTRACT As the small Southern African country of Lesotho grapples with implementing one of the five largest dam-development projects in the world today, the local people impacted confront the challenges of resettlement, loss of means of production, and changed access to natural resources. This chapter reveals the ways in which the Lesotho Highlands Water Project (LHWP) serves to reorganize and commodify rural resources for the benefit of the nation-state in gendered ways. Based on interviews of rural households conducted during 13 months of ethnographic fieldwork in Lesotho, this chapter analyzes impacted people’s experiences of the gendered implications of the extraction and sale of water from the rural highlands of Lesotho to South Africa. This case study documents the gendered environmental and social impacts of the LHWP on local communities, and illuminates the everyday lived experiences of the affected people as they effectively subsidize this international project with their environmental resources, labor, money, and, in some cases, their nutritional status. Nature, Raw Materials, and Political Economy Research in Rural Sociology and Development, Volume 10, 373–396 Copyright r 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1057-1922/doi:10.1016/S1057-1922(05)10017-1

373

374

YVONNE A. BRAUN

1. INTRODUCTION As a region, Southern Africa in the post-Apartheid era is a considerable social and economic force globally and within Africa. While the Republic of South Africa (RSA) is more industrialized than most parts of Africa, it also encompasses some of the greatest socio-economic disparities as a result of past racial, social, and economic policies. As it struggles with new democratic principles, it must also struggle to develop its natural resources economically and sustainably. Lesotho, landlocked within South Africa, has an inextricable and, historically, dependent relationship with its larger, more economically dominant neighbor. Not having other natural resources, Lesotho does have water – the resource South Africa needs the most in its development efforts. As the small Southern African country of Lesotho grapples with implementing one of the ﬁve largest dam-development projects in the world today, the local people impacted confront the challenges of resettlement, loss of means of production, and changed access to natural resources. This chapter reveals the ways in which the Lesotho Highlands Water Project (LHWP) serves to reorganize and commodify rural resources for the beneﬁt of the nation-state in gendered ways. Based on interviews of rural households conducted during 13 months of ethnographic ﬁeldwork in Lesotho, this chapter analyzes impacted people’s experiences of the gendered implications of the extraction and sale of water from the rural highlands of Lesotho to South Africa. As the multi-dam LHWP is constructed in the highland areas to generate revenues for the state through the sale of water, it covers the wild and commonly managed resources used by rural populations that currently bring little or no beneﬁt to the state. These same rural households absorb the losses of such resources and subsidize the LHWP as they struggle to replace the lost pieces of their livelihoods. This case study documents the gendered environmental and social impacts of the LHWP on local communities, and illuminates the everyday lived experiences of the affected people as they effectively subsidize this international project with their environmental resources, labor, money, and, in some cases, their nutritional status in gendered ways. I provide some ethnographic context in which to understand the importance of these natural resources and the role of development in Lesotho in the ﬁrst section of this chapter. Then I offer a brief discussion of the LHWP and the history of development more generally in Lesotho, and particularly how it has been gendered. I discuss what we know about the use of large

Selling the River

375

dams as development, and why the Government of Lesotho chose this particular form of development. In the second section, I briefly discuss my methods and research design informing this research. Lastly, I present a discussion of how three particular natural resources utilized by the rural poor are reorganized for the beneﬁt of the state in the process of the implementation of LHWP, and how the experiences of this reorganization had particular gendered consequences.

2. LESOTHO Landlocked within South Africa as a result of colonial wars and politics, the Kingdom of Lesotho has historically been a labor reserve enclave economy for the gold and diamond mines of South Africa. The system of oscillating male migration1 between these mines and Lesotho is shaped by, and has shaped, the historical pattern of relations between the two countries (Epprecht, 2000; Gordon, 1981; Murray, 1981; Nkomo, 1990). Lesotho became an independent country in 1966, but it is still largely dependent upon its dominant neighbor (Bardill & Cobbe, 1985; Eldredge, 1993; Epprecht, 2000; Hassan, 2002; Sechaba Consultants, 1995; United Nations, 2003). Almost everything in Lesotho is an import from South Africa, including most electricity. The almost 2 million people in Lesotho overwhelmingly live in rural areas (84%) (Hassan, 2002; Tshabalala & Turner, 1989). The components of the ‘‘household production system’’ for these highland areas involve a variety of strategies for income and survival, including wage labor, small-scale commercial activities such as selling produce or goods, cropping agriculture, and livestock management (POE, 1989), as well as brewing beer and making handicrafts. Many rural households in Lesotho also collect wild vegetables and herbal medicines from the natural environment. These are collective resources used on a frequent basis to supplement household resources. In some of the most remote and poorest communities in my study, people reported spending 5–15 h a week collecting wild vegetables for food prior to the LHWP (Author’s interviews, 2001). These respondents considered their access to these vegetables very good and relied on them as significant food resources, not as luxuries. Other communities had seen changes in their access to wild vegetables and medicines in years prior to the LHWP. Some plants ﬂuctuated in abundance either due to environmental changes or over-collection.

376

YVONNE A. BRAUN

These strategies are part of a gendered village economy in rural Lesotho (Ashton, 1967; Eldredge, 1993; Ferguson, 1985, 1990; Gay, 1980; Gordon, 1981; Hassan, 2002; Letuka, Matashane, & Morolong, 1997; Murray, 1981). Women are usually the farmers, involved in the informal economy and the reproduction of the household, carrying water and collecting fuelwood, and possibly raising poultry and small animals. Men, on the other hand, are involved in some aspects of farming, raising livestock, and/or wage labor. Many men are migrant workers in South Africa (roughly 13%, and providing 17% of Gross National Product (GNP)) or elsewhere in Lesotho (Tshabalala & Turner, 1989), often making women the de facto heads of household. Male migration has been an historical strategy for households to get wage labor; however it needs to be understood in balance with a consistent reliance on agriculture and other strategies (Gordon, 1996; Nkomo, 1990). The nature of the topography in Lesotho (its plentiful rivers and highland areas providing natural gravitational ﬂow) inﬂuenced the sites of the dams and tunneling infrastructure that created the largest material losses within the LHWP (described in more detail below). The highlands areas chosen for construction (Katse, ‘Muela, Mohale) contain some of the most remote and poorest communities within Lesotho, with some of the highest rates of unemployment and destitution (Sechaba Consultants, 1994a, p. 59; Tshabalala & Turner, 1989, p. 6). At the beginning of the LHWP, 60% of households in both areas of Katse and ‘Muela fell below the average income for each area and are considered ‘‘very poor’’ (POE, 1991, p. 25; Tshabalala & Turner, 1989, p. 9). With one of the highest HIV/AIDS prevalence rates in the world, Lesotho’s rural households face additional challenges of access to health care, increased levels of poverty, loss of eligible working adults, increased need for caretaking labor, increased number of orphans, and the emotional toll of high levels of sickness and death. Even as households diversify their subsistence strategies through agriculture and wage labor, Lesotho as a whole remains a net importer of food (mostly from South Africa). Male migration to the RSA mines peaked in the late 1970s (Gordon, 1996), and since the democratization of South Africa, Basotho men have increasingly been remitted from the South African mines without replacement employment. It is within this context that the Lesotho government of the mid-1980s chose a national development strategy, the LHWP that involved the extraction and sale of rural, natural water resources for the generation of country-level revenues.

Selling the River

377

3. LESOTHO HIGHLANDS WATER PROJECT The LHWP is a multi-dam2 development project between the governments of Lesotho and South Africa.3 Based on a treaty signed in 1986, each government formed their own development authority (Lesotho Highlands Development Authority (LHDA), in Lesotho) and a joint commission (Joint Permanent Technical Commission (JPTC)) to oversee the planning and implementation of the $8 billion project. This 30-year project was funded by various sources such as the World Bank, the African Development Bank, the European Community, and several European funding agencies.4 Their agreement outlines two major objectives. The ﬁrst objective is to deliver water from Lesotho’s Senqu River and its tributaries to the Gauteng industrial region of South Africa. In return, South Africa was estimated to pay approximately $55 million in royalties to Lesotho each year; however, recent reports show that Lesotho has received closer to $18 million in average annual revenues (Hassan, 2002; United Nations, 2003). The second objective of the project is to create a hydroelectric power station that will allow Lesotho to create electricity domestically.5 An important documented obligation of the project is for its implementation to not worsen the current standards of living6 of the project-affected people (LHDA, 1986). Due to the remoteness and ruggedness of the terrain, a vast amount of infrastructure was needed to reach the river basin dam sites in the mountains. Construction of infrastructure and camp facilities for foreign engineers7 began in 1987 (Detter, Gunnewig, & Seiffert, 1994; Sechaba Consultants, 1994b; Transformation Resource Centre, 1995). In 1993, Katse was the ﬁrst site to begin dam construction because its reservoir is the largest holding tank for the water being delivered to South Africa (through ‘Muela). The 185 m arch dam began impoundment in October 1996, and at the time of this ﬁeldwork in 2000–2001, the Katse portion of the project was mostly ﬁnished – the dam was built, the basin inundated, and the infrastructure complete. The local people impacted were engaged in interaction with, and possibly receiving compensation from, the LHDA for almost 10 years at this time. ‘Muela, the second dam of Phase 1A, was almost ﬁnished. The construction of the smaller 55-m-high dam was complete, and the catchment area (the reservoir) underwent impoundment in 1999. Tunnels from Katse and tunnels to South Africa connect at the tailpond site in ‘Muela, thus creating an extensive geographical area that is impacted. Mohale Dam and its subsidiary infrastructure were still under construction during my ﬁeldwork

378

YVONNE A. BRAUN

(2000–2001), but the ﬁrst phase of resettlement was in process and many Mohale communities had been impacted by the project for years. As the LHWP and its terms are negotiated between the governments of South Africa and Lesotho, there are inevitable consequences for the people impacted directly by its construction and implementation. People were impacted by the construction of the dams themselves; the infrastructure for the dam and tunnel system through Lesotho to South Africa; the construction of roads through mountains; and the construction of employee ‘‘camps’’ due to the duration of LHWP.

4. DEVELOPMENT IN LESOTHO Lesotho, as a ‘‘least developed country’’ (Ferguson, 1990; Hassan, 2002; United Nations, 2003), has a long history of externally funded development projects, but none quite as extensive as the LHWP.8 The history of development projects in Lesotho and more generally is to ‘‘fail’’ in regards to their stated purposes. Ferguson’s (1990) critical analysis of development in Lesotho reveals the ways that the conceptual apparatus of the development industry constructs hierarchical bodies of knowledge and invokes categories that reproduce the need to justify successive development interventions. With the LHWP, the investment in and presence of the development industry in Lesotho has significantly increased and has even been institutionalized within the LHDA. That this occurs despite the critiques of ‘‘development failure’’ supports Ferguson’s illustration of the ways in which the ‘‘need’’ for development projects becomes reproduced despite their ineffectiveness, and possibly because of the unauthored effects that these projects enable in the process. As documented elsewhere (Braun, 2005), the prospect of large dams as a development strategy for Lesotho has periodically been voiced since the 1950s (by colonial developers), but only in the late 1970s and early 1980s did it begin to be seen as viable by the military government of the Jonathan administration and interested parties in the apartheid South African government. For the South African Ministry of Water Affairs, the challenge was to identify an extensive water source for the prospering commercial and industrial sectors of the Gauteng region (including Johannesburg). In contrast, World Bank country reports and national development planners in Lesotho painted a picture of a country that had almost no other development options. The majority rural population was engaged in primarily subsistence-oriented agriculture. Lesotho was considered to not have a

Selling the River

379

viable market because exported crops could not compete with those of South Africa and Lesotho continued to be a net importer of food. Male labor migration rates to the mines of South Africa peaked during this period (see Santho & Sejanamane, 1991; Tsikoane, 1991). Within the context of the historical dominance of South Africa’s relationship with Lesotho, the convergence of the needs of de Klerk’s apartheid administration to secure its prospering industrial center, and the desire of Jonathan’s military administration to increase revenues and secure its status in relation to internal politics and to South Africa, seems to have set the stage for Lesotho’s participation in the LHWP (Bardill & Cobbe, 1985; Nkomo, 1990; Santho & Sejanamane, 1991; Tsikoane, 1991).

4.1. Large Dams as Development This particular form of ‘‘development’’ involves massive investments of external funds, demands a centralization of authority in both the state and an implementing development body (LHDA), the resettlement and dislocation of marginalized rural communities, and mechanisms in which the natural resources of rural areas are reorganized under the control of and toward the beneﬁt of the state (McMichael, 2000), and in this case, the larger regional hegemon of South Africa (Bunker & Ciccantell, 1999; McMichael, 2000). Since the LHWP, Lesotho has become more indebted to international development institutions and other states, and consistently strives to reorganize socially and economically to meet standards for credit (George & Sabelli, 1994). In this process, the Government of Lesotho restructures access to rural resources to meet the macro-economic indicators of the Western market model as it simultaneously marginalizes and restricts current socio-cultural and economic values and uses of those same resources (McMichael, 2000). As I have documented elsewhere (Braun, 2005), choosing large dams for development involves significant social and environmental impacts (Braun, 2000, 2003; Cernea, 1988, 1990; Chambers, 1969; Clark, Colson, Lee, & Scudder, 1995; Colson, 1971; Fahim, 1981; Fast & Berkes, 1994; Goldsmith & Hildyard, 1984; Goodland, 1997; Hart, 1980; Horowitz, 1991; McCully, 1996; Peterson, 1954; Roder, 1994; Scudder, 1985, 1988, 1993, 1997; World Commission on Dams (WCD), 2000). These consequences include resettlement and dislocation; loss of means of production; changes in river basin ecology; socio-economic, socio-cultural, and psycho-physiological effects; greater risks of impoverishment; changes in community, household, gender

380

YVONNE A. BRAUN

relations; and changed access to natural resources. As revealed by the report of the World Commission on Dams (2000), the consequences of dam-development projects have been increasingly criticized for the devastating costs – in contrast to the minimal beneﬁcial results – absorbed locally by those directly affected. Historically, dams are not new. What is unique about the current stage in the construction of dams is that there are multi-lateral institutions and multi-national corporations that propose, negotiate, fund, and set the terms for the feasibility of such projects with national governments across the globe (but mostly in the ‘‘Third World’’). As Fahim (1981) reminds us, dams are inherently social in that they are always proposed for social beneﬁt, and most often this is equated with economic growth. Many large dam projects are multi-purpose, with the national agenda for economic growth and improved standards of living as the primary goals (planned to be achieved through hydroelectric generation, irrigation, ﬂood control, and projects ﬂowing from these changes). Development programs are secondary aspects, often without participation and having the effect of incorporating rural people into a wider economy (Colson & Scudder, 1988; Scudder, 1997). In the case of large-scale dam projects, which are almost always constructed in rural areas, and especially in cases like the LHWP where the extraction of water takes place for the generation of country-level revenues, rich resources such as river basins and ﬂowing water are restructured to provide for the beneﬁt of urban areas. This urban bias toward the appropriation of rural resources for national economic growth supports the shift toward industrial markets that are signs of growth at the national level, and in the World Bank country reports. These country reports often serve as the basis for loan decisions and categorize the status of countries in the global order. Following Ferguson (1990), funding development such as the LHWP permits the international development industry access to greater surveillance and control of ‘‘developing countries,’’ and internally, the increased centralization of the state as it gains access to the surveillance and control of rural areas, their resources, and residents. Rural residents affected by such large dam projects experience changes in their relation to their environment and its resource base. These changes occur either directly (through loss of access to resources and loss of land) or indirectly (through policies that change the nature of relations to resources or division of labor). While these changes are differentially distributed and absorbed across socio-cultural and socio-economic categories such as gender, class, and age, the general pattern reveals ‘‘that large-scale water

Selling the River

381

resource development projects unnecessarily have lowered the living standards of millions of local people’’ (Scudder, 1997, p. 45).

4.2. Dams and Development Most development programs are typically justiﬁed as poverty-reducing measures and, in the case of dam-infrastructure projects, as mitigative measures for those impacted. While the World Bank has declared that any subsidized economic project must also be a development project (Cernea, 1988), the programs that compose the ‘‘development’’ portion of the LHWP generally are secondary measures that aim to serve the dual purpose of fulﬁlling the World Bank standards and the LHWP’s treaty obligations to ensure that the standards of living of the project-affected people are not lowered.9 The development and governmental institutions that design and implement these types of projects frame the project-affected people not as intended beneﬁciaries, but rather as a part of the cost–beneﬁt analysis to be negotiated and minimized (Goodland, 1997). The struggle between the governments of South Africa and Lesotho to minimize their respective costs or to pass on costs to the other, including the costs of compensation programs, underlies the decision-making processes regarding planning and implementation (POE, 1989, p. 11). The impacted people then become commodiﬁed as they become a price to be negotiated, to be reduced and minimized, within an ideology of capitalist accumulation, and for the justiﬁcation of increased state formation (Bunker, 1985). This institutional arrangement and ideology of development practice prioritizes a technocratic interpretation of value and process of meaning through subordination of other socio-cultural meanings and values. Kardam (1997) and other scholars writing on institutions such as the World Bank reveal how, as institutions, they have the goal of seeing that their loans are used effectively and efﬁciently. This creates an emphasis on the economic returns of their investments, and less so on the social ramiﬁcations of those same investments. As this study documents, the people impacted incur a variety of social and environmental costs. There are material losses of agricultural and grazing land, houses, fuelwood and medicinal plants. There are psychological costs of insecurity, increased pressure on resources, increased social conﬂict (including increasing gender inequality), uncertainty, decreased resiliency and control, powerlessness, and fear. Losses of land and other resources are also

382

YVONNE A. BRAUN

associated with psycho-physiological stress and increased rates of morbidity among ‘‘development refugees’’ (Scudder, 1979, 1997).

4.3. Gender and Development Generally, these losses disproportionately burden the rural poor, and create especially intense pressures on women (Elson & Pearson, 1980; Sen & Grown, 1987; Mohanty, Russo, & Torres, 1991; Beneria & Feldman, 1992). Contextualizing the LHWP as part of this larger pattern of development processes, the rural poor are in fact burdened with a disproportionate amount of the losses from development projects, and arguably receive an inadequate amount of the beneﬁts (POE, 1993, p. 2; 1994, p. 1; 1995, p. 17; 1997). As policies structure the relations around resources, the definition of rights to certain domains can effectively (re)create or intensify asymmetrical social relations. Gender is a useful axis for analysis as it highlights the differential access to resources and knowledge, as well as the gendered rights to control and regulate that access the actions of others (Rocheleau, Thomas-Slayter, & Wangari 1996, p. 10). The delineation of rights often becomes deﬁned and legitimated by rights to certain spaces and domains. As Rocheleau, Thomas-Slayter and Wangari reveal, the ‘‘command over space and gendered rights of control and access are sources of social and political power’’ (1996, p. 253) that serve to justify certain groups’ maintenance of power, and thus others’ exclusion from those sources of power. As development projects structure access to resources and control over domains, they (re)create and impact gender relations whether explicitly targeting women or remaining ‘‘gender blind.’’ In their landmark study on women in development, Sen and Grown (1987) examined the importance of women for the reproduction of the household, and focused primarily on the ‘‘food–fuel–water’’ crisis (1987, p. 51), revealing how development policies typically burdened women by making their responsibilities more difﬁcult to fulﬁll. I use a ‘‘Gender and Development’’ (GAD) approach, in contrast to ‘‘Women in Development’’ (WID) and ‘‘Women and Development’’ (WAD) approaches. Much of the WID literature, in the style of Boserup (1970), revealed how women have been systematically ignored in development issues and analyses, and attempted to bring women into the picture by showing their important roles in society (see Elson & Pearson, 1980; Rosaldo & Lamphere, 1974; Sen & Grown, 1987; Weiner, 1976). ‘‘Women

Selling the River

383

and Development’’ builds on the WID approach by revealing the importance of women in society, but goes beyond WID by centering their focus on women as potential resources for development and change (see Beneria & Feldman, 1992; Leacock & Safa, 1986; Lockwood, 1992; Sharma, 1985; Thomas-Slayter & Rocheleau, 1994). The GAD approach incorporates, but also diverges from, these two approaches. While recognizing women’s roles in society and their importance and resource potential, the GAD approach attempts to not only focus on the differential experience of women, but of gender itself (Gordon, 1996; Kabeer, 1994; Momsen & Kinnaird, 1993). This approach starts with the premise that all categories are constructed and experienced relationally, and so GAD analyses typically focus on gender relations, as does this study. In particular, the current geo-political topography of wealth and power disadvantages the rural poor of the highlands in Lesotho in gendered ways. As their government increasingly prioritizes the commercial uses of resources and the reorganization of rural resources toward the beneﬁt of the state and urban focused development, the rural households undergo serious disruption to their livelihoods as they absorb the economic, ecological, and social costs of their resources being restructured. In this process, the rural impacted people effectively subsidize this international development project as they attempt to replace their lost resources using strategies of increased labor allocation, increased purchasing or a greater reliance on cash, or are left to live without these resources. In this chapter, I examine the ways in which the burdens of this subsidization are distributed in gendered ways.

5. RESEARCH METHODS This chapter draws from two periods of ﬁeldwork in Lesotho. In 1997, I conducted 25 semi-structured interviews with impacted people and 13 semi-structured interviews with development ofﬁcials, consultants, and local non-governmental organization (NGO) workers over 3 months. The interviews were conducted in English with a translator as necessary. In 2000–2001, I conducted 13 months of ﬁeldwork funded by the National Science Foundation and the US Fulbright Student Fellowship. During my latest ﬁeldwork, I worked as a Research Associate at the National University of Lesotho (NUL). Throughout this period, I lived and worked in all three areas of the LHWP with households directly affected by the dam project. I hired six teams of research assistants – including directly affected people – to conduct two waves of surveys and semi-structured interviews at

384

YVONNE A. BRAUN

all three dam sites of the LHWP (once in winter and once in summer). The villages in my sample were chosen and stratiﬁed according to impact and socio-economic variables obtained from development authorities and consultants, and households were chosen randomly. We surveyed approximately 40 new households at each site in both waves; however, in the second wave we also revisited about 25% of the households from the ﬁrst wave. In total, 263 households participated in both waves. Surveys and interviews were completed for each household, and interviews were recorded when possible. All surveys and interviews in the villages were conducted in Sesotho, and later transcribed and translated into English. Additionally, I conducted approximately 13 semi-structured interviews with development ofﬁcials in English. This chapter is based solely on the qualitative data collected during my ﬁeldwork, and the analysis draws from approximately half of the data currently available.

6. GENDERED EXPERIENCES OF RESOURCE EXTRACTION AND DEVELOPMENT The nature and extensiveness of the LHWP demands the appropriation of such collectively owned resources as river basins and pastoral lands for infrastructure, buildings, and dams. Through the LWHP, the Government of Lesotho gains access to natural resources previously seen as not having value or as underutilized. Commodiﬁcation of water for national revenues results in the ﬂooding of the best farm and grazing lands in the country and ﬂooding of the wild vegetables and herbs so essential to the diets of the rural poor households. In this process, competing values of reorganized natural resources are rendered irrelevant. As development agencies use projects to structure and deﬁne rights and relations to resources, they create programs designed around ameliorating certain losses, acknowledging specific expectations and impacts (to the exclusion of other losses, expectations, and impacts). In large dam projects, project-affected people often subsidize the project with their agricultural and grazing lands, gardens, trees, river valleys and water sources, homes and burial grounds, as well as absorbing multiple social changes in their communities. Compensation packages that serve as mitigation for these losses depend on the sets of values determined by the development authorities and on their successful implementation of those policies. For example, the compensation plans for Katse and ‘Muela, the ﬁrst two dam areas of the LHWP, were designed and implemented under the same

Selling the River

385

project phase (1A). After local and international pressure, the compensation plans for these areas were retroactively changed. In the new plan, compensation for losses included: lump sum cash payments for small plots of land and annual deliveries of maize and beans for larger agricultural plots, with some receiving annuities in perpetuity. Private fruit and wood trees were supposed to be replaced with saplings, but there are conﬂicting reports of this happening systematically or at all. There was no compensation for common resources except for small amounts of fodder for animal owners for 5 years. Resettlement for Katse households (there was no ofﬁcial resettlement at ‘Muela) was restricted to moving within the same basin. At the third dam area, Mohale, the compensation plan was part of the second phase (1B) of the LHWP and was designed more comprehensively. This package includes annuities in perpetuity for agricultural land and assets of certain sizes. The resettlement program at Mohale was significantly different from Katse in that it allowed resettlers to choose among options for where to be resettled. In particular, resettlers could choose to stay in the larger Mohale area, to relocate to villages near Mohale (the lowlands), or to relocate to peri-urban areas near the capital, Maseru.

6.1. Compensation and Value The development industry privileges economic evaluation over socio-culturally informed rural means of production. The pattern of compensation – what gets compensated, what does not – reveals and reﬂects the assumptions of value that support these determinations. Development authorities compensate losses seen as having ‘‘value.’’ In the LHWP, this has meant that privately owned agricultural ﬁelds, privately owned gardens, fruit trees, and houses receive compensation as losses. In contrast, commonly held and managed natural resources and non-cultivated natural resources are not compensated in a systematic manner, despite being one of the largest categories of loss in the LHWP. But obviously, just because there is no plan of compensation does not mean there is no value in these resources. This type of valuation hides the ways in which communities ‘‘work’’ to maintain these common resources and ignores the everyday values of commonly managed natural resources to rural households. These losses of access for households are very real, and have the most extreme effects on the poorest households. Current values of the reorganized natural resources are rendered invisible as households are not compensated for the losses of common resources. Impacted people subsidize the project as they incur extra

386

YVONNE A. BRAUN

costs to replace the lost access to these resources, and the costs of that absorption are differentiated by gender. Not surprisingly, natural resources are a significant component of the means of production for rural families. Lost or decreased access to those resources, without compensation, creates pressure to replace those resources with the following consequences: reallocation of labor; increased reliance on cash; exacerbation of existing inequalities; and challenging existing social relationships and ties. In the following sections, I outline the types of impact that affected households as a result of their changed access to the common resources during the LHWP and how these were gendered in particular ways.

6.2. Reorganization of Resources Like most rural areas in the developing world, rivers and river basins serve multiple purposes for rural households, especially the poor. Some common uses of the rivers are: water collection for household use; livestock watering; irrigation; washing clothes; bathing; relatively easy crossing; and household water consumption, if there are no other sources such as springs and streams. River basins additionally provide: the best farmland; winter grazing; stones for building houses, kraals, and other structures; soils; tree planting; forest maintenance; collection of wild vegetables and herbs; collection of grasses for basketry (making baskets, mats, and household utensils like sifters) for household use or for selling; and thatching grasses for houses for private use or for selling. Some households lost access to resources as a result of the ﬁlling of the dam that created a reservoir over their previously available resources. One woman in Katse describes the differences in access prior to the LHWP and after: It has changedybecause most of our resources we used to ﬁnd them at the river here. Soyas the dam has covered thingsygrass and others are no longer available, indeed they are no longer there because we would ﬁnd them [where the dam is] before, most of them. (2K11LM009, 2001)

One woman in the ‘Muela area described the ways in which the transformation of ‘‘their river’’ to a reservoir affected her access to these everyday resources: At other places like that one down there at the damythere are herbs – they were available there, that I used to dig up from there, nowythat place is covered up there. I

Selling the River

387

no more get those medicines, they are found in different places. There is an herb that grows – it is known – I mean a particular herb will be found at a particular placeythey don’t just grow – I mean they chose where to grow. I no more get them. We collected wood. We dug up soils and we painted like the Basotho peopleythose soils are not there now, they are ﬁnishedythe gully (donga) – because of the [LHWP] and the dam has covered up those soils, it has concealed the gullies we used to dig up those soils from. (M11LD010, 2001)

Another woman in Katse ﬁnds the loss of these resources affecting the economic livelihood of her household: We would get Loli [from the river], making baskets and sifters. Now it has closed our things, which we knew that when we have made them, we would go to the lowlands and they would buy them [and] we would make a living.ySo now we don’t know what we [will] doy[to] make a living. (2K8NM004, 2001)

Access to resources also changed due to increased pressure on existing resources through migration or loss of some access. Some sources were lost altogether from changes in ecology. Some households lost access through resettlement. In the latter case, resettlers had the stress of going through resettlement and then the potential additional burdens of losing access to some natural resources without compensation. For example, one woman resettled in Mohale noted how the resources at her place of origin were not available at all in her new place: We get nothing, there is really nothing, even the medicines we were used to digging up ourselves, to make ourselves feel better, are not available hereywe live by going to the chemist/pharmacy to buy medicines. (MO17LM004, 2001)

In her statement, she reveals the ways in which this burden of lost resources has health and economic implications. The stress of resettlement makes these resettlers especially vulnerable to health risks, and the decreased access to medicines exacerbates these risks. The necessity to now buy medicines, if possible at all, creates extra ﬁnancial burdens on households already likely to be in poverty. The increased reliance on cash was a theme that resonated throughout my interviews and had differential implications for men and women. There were similar causes and effects for the loss of wild vegetables and for sources of water. As part of a resettled household in Mohale, this woman speaks about the ways in which the resettled people in her village were not compensated for some of the natural resources that were significant components of their livelihoods prior to resettlement: They had promised they would feed these people, and give fodder to cattle. These are all the things they did not do – natural resources which include the vegetables we survived

388

YVONNE A. BRAUN

by, are not – they are things we have not received, as to how we are going to get those vegetables [we don’t know]. (MO19MM009, 2001)

She goes on to conclude that their access has changed significantly in that there are no replacement resources for some of their losses, and that has implications for their allocation of labor as she questions the effects of development more generally: I can’t see whereydevelopment is – you see because here there are no resources like those we had where we were before. The things we use here are different from those I had beforeybecause the rivers here are faryI was much nearer to water sources there, water was thus nearer to me, from springsynow here there is no chance to go there and any other place for those rivers. (MO19MM009, 2001)

As the above quotes reveal, LHWP impacted people endure changed access to natural resources with patterns of economic, health, and labor consequences. In the next section, I analytically separate these consequences to highlight the ways in which these patterns are gendered in particular ways, and then use the example of pastoral lands to reveal the intersection of these gendered patterns for impacted people in the rural highlands in Lesotho.

6.3. Nutritional Costs Women in rural Lesotho are responsible for the feeding of their families, and one of the ways in which they do so is by collecting wild vegetables. For many this creates a low-cost (time and energy only), very nutritious food source for their families. Lost access to wild vegetables creates both economic and nutritional consequences for poor households that are unable to allocate other resources toward replacing these nutritious foods and therefore must go without them. If there are little or no replacement foods for lost wild vegetables, the nutritional costs of losing those foods in the diet are differentiated by gender. Wage laborers in the household traditionally receive food, including vegetables, ﬁrst. Women, as reproducers of the household, cook the meals but eat last. Women in rural Lesotho are more likely to be Vitamin A deﬁcient and nutritionally at risk, and so a loss of these wild vegetables is a more significant loss for them than for others (Hassan, 2002; Letuka, Matashane, & Morolong, 1997; Sechaba Consultants, 1994a; UNICEF, 1994). Nutritional costs are also differentiated by other existing inequalities. Poorer households are less able to replace lost resources with purchased substitutes and poorer households may also be less ﬂexible with the labor

Selling the River

389

allocation of household members. By their very nature, poorer households also have less room for decreases in nutritional status. In Lesotho, women are more likely to be in poverty and widows are especially vulnerable to being in poverty and food insecure. The intersection of poverty and gender serves to render poor women as the most vulnerable to the nutritional consequences of the loss of these wild food sources to the LHWP.

6.4. Allocation of Labor Even as this project is about water itself as it sells and transfers it to South Africa, some local people lose access to water at the source: [The dam] has brought [change] because now our springs have vanished, which we used to draw water from, which were near, vanished. yWe are leading a hard life. (2K8NM004, 2001)

In particular, the change of the river to a reservoir changed the ways some people could use local water sources. Women and children, especially girls, absorb the additional labor costs of collection of water after the LHWP. Women and girls must walk longer distances to ﬁnd potable water for household use. In Katse, the loss of local springs increased water collection distance by 2–4 times, and by up to 2.5 miles roundtrip. Most households reported collecting water 2–3 times a day, so these are significant increases in daily labor needs. Women and children also absorb the extra labor costs involved in the collection of replacement vegetables. Prior to the LHWP, households in the Katse and ‘Muela areas reported spending 5–15 h a week collecting wild vegetables. After the LHWP, some villages in these areas lost all or partial access to these wild vegetables which compose an important nutrient source for rural diets. While some endured increased pressure on the remaining sources of wild vegetables, many walked further to ﬁnd new sources of vegetables. Residents in Katse and ‘Muela reported spending up to 12–24 h a week collecting wild vegetables after being impacted by the LHWP.

6.5. Increased Reliance on Cash Men and women shoulder the increased labor costs associated with a greater reliance on cash in different ways. As poor households lose these natural resources, they face the increased burden of purchasing replacement

390

YVONNE A. BRAUN

resources. For many, this necessitates an increased pressure to reallocate more labor toward cash-generating activities in order to meet these needs. Such strategies of increased labor allocation toward wage labor tend to have direct implications for men, as they are often the wage laborers in the household. Extraordinarily high rates of unemployment in the dam site areas suggest the difﬁculties of such strategies. In some cases, particularly at Katse, women ﬁnding jobs as domestic servants for LHWP development professionals and in prostitution shouldered some of this burden directly (Braun, 2000). The potential long-term costs of such strategies as prostitution, such as the contraction of HIV/AIDS, STDs, and stigmatization, are yet to be seen for these women and girls. Basotho women also absorb the difﬁculties of tightening budgets as their households endure this increased pressure on existing ﬁnancial and labor resources. As rural impacted women struggle to balance a tightening budget and meet household needs – school fees, clothes, food, rent, farming – women often reallocate their labor toward collection and away from other activities. Women negotiate the reallocation of monies toward replacement resources and away from other costs. For many women, this means making challenging decisions or allocating more of their own labor toward replacement activities.

6.6. Loss of Pastoral Lands In addition to the resources of the river and river basins, the appropriation of pastoral lands for infrastructure, buildings, and dams created the largest material losses during the ﬁrst phases of the LHWP. I use the example of decreased or lost access to pastoral lands to demonstrate the ways in which these consequences and inequalities intersect in gendered ways. Pastoral lands in Lesotho are ‘‘managed commons’’ for cattle, sheep, and goats. These common lands are an economic resource as well as environmental resource for rural households. Common pastures serve as a direct economic resource for those with livestock. However, for those without their own animals, these common pastures also serve as a means of economic opportunity to be employed as caretakers of other people’s livestock. Lost or diminished access to pastoral lands has similar effects as previous resources: changes in allocation of labor, increased reliance on cash, and exacerbation of existing inequalities. With the loss of pastoral lands, households with animals need to reallocate labor to ﬁnd replacement resources,

Selling the River

391

use money to buy feed, or pay caretakers to take livestock longer distances to ﬁnd available grazing lands. The effects of lost or diminished access to pastoral lands exacerbates existing inequalities. Poorer households absorb greater costs of decreased access and the reallocation of more labor toward ﬁnding pastoral lands is the hardest on poor households. The poor, not surprisingly, have less ﬂexibility to choose other options; they are less likely to be able to purchase replacements for lost resources or to employ others for the increased work to replace those resources. The poorest households may actually beneﬁt in the short term from the loss of some of these natural resources. The poor without animals of their own may ﬁnd increased opportunities for employment as caretakers of other people’s animals. Typically, these jobs are delegated to young boys of these households who will travel with the animals in a seasonal grazing pattern. However, these strategies may serve to reproduce a gendered cycle of poverty for some of these young boys most directly impacted. As local grazing lands decrease, boys employed as herdboys will travel farther away from home during school age years (ages 6–16). Longer careers as herdboys lead to lower lifetime educational achievement and these boys are more likely to remain in poverty as adults (Sechaba Consultants, 1991, 1994a; UNICEF, 1994). Such gendered consequences enable the generational reproduction of poverty and reveal some of the ways in which the impacted people of the LHWP subsidize this international development project. In particular, impoverished households are more likely to rely on these common resources and are more likely to be unable to replace them in other ways when lost to the LHWP, exacerbating existing economic and nutritional disadvantages among households. The changed access to resources intersects with existing inequalities in communities in particular gendered ways. Ironically, the construction of the LHWP has the potential to create environmental consequences that exacerbate the existing poverty of particular households and people, rather than serving to mitigate their poverty as the ofﬁcial project goals state.

7. CONCLUSION The current geo-political topography of wealth and power disadvantages the rural poor of the highlands in Lesotho in gendered ways. As their government increasingly prioritizes the commercial uses of resources and the

392

YVONNE A. BRAUN

reorganization of rural resources toward the beneﬁt of the state and urban focused development, the rural households undergo serious disruption to their livelihoods as they absorb the economic, ecological, and social costs of their resources being restructured. In this process, the rural impacted people effectively subsidize this international development project as they attempt to replace their lost resources using gendered strategies of increased labor allocation, increased purchasing or a greater reliance on cash, or are left to go without these resources. As the Government of Lesotho chooses large dams as a development strategy, the local people impacted confront the challenges of resettlement, loss of means of production, and changed access to natural resources. In the process of investing in and implementing large dams as a means of ‘‘development,’’ the Government of Lesotho appropriates the common rural natural resources, such as wild vegetables and herbs, water, and pastoral lands, and restructures access away from rural populations, and toward the beneﬁt of the nation-state in the accumulation of national revenues. The affected people directly absorb the loss of access to these resources in gendered ways, and absorb the direct and indirect economic consequences that result. These consequences are differentially distributed among the rural poor and often serve to exacerbate existing inequalities. Some households need to purchase replacement foods for wild vegetables (if possible), and allocate more time, labor, and money toward locating the same or similar wild vegetables, water sources, and pastoral lands. In so doing, the men and women impacted by the LHWP effectively subsidize this international project with their environmental resources, labor, money, and, in some cases, their nutritional status.

NOTES 1. ‘‘Oscillating male migration’’ is Murray’s term (1981). It refers to the cyclical movement of migrant Basotho male miners to South African mines for 8–11 months of the year, and then their return to Lesotho for the remaining months of the year. 2. In anticipation of South Africa’s increasing water needs, the water delivery scheme will be composed of ﬁve dams and various tunnels constructed in four phases over a period of 30 years (1987–2017). The ﬁrst phase is covered in the 1986 treaty; however, the subsequent three phases require a new contract. The absence of a binding document, as well as potential evidence that the amount of water available was highly overestimated, is allowing for much controversy and debate as to whether the latter two phases will be completed.

Selling the River

393

3. For a political economy interpretation of how the LHWP contradicts the Government of Lesotho’s documented national development plans (see Tsikoane, 1991). 4. LHWP background information comes from the compilation of reading project documents, personal communication with development ofﬁcials, and through contact with the Highlands Church Action Group. 5. The small number of households electriﬁed prior to the LHWP imported electricity from South Africa. Preliminary ﬁgures suggest that even after the operation of the ‘Muela hydropower station, the cost of setting up electricity has prohibited new consumers and the small proportion that had electricity prior to the LHWP continue to import it from a South African company, since it remains more cost effective to do so. 6. ‘‘Current’’ meaning the time of the ﬁrst intervention at each site. See Panel of Environmental Experts (POE), 1991. 7. Camp facilities are employee housing areas. 8. For a small country with such minimal economic resources and questionable institutional capacity for a project of this extent, it is quite surprising that Lesotho was deemed ‘‘eligible’’ for receiving World Bank funding. At the time of the LHWP agreements, South Africa was under apartheid rule with full sanctions against aid of this type. Lesotho was made the proxy receiver of the loans despite their ‘‘ineligibility’’ (for a more detailed discussion see Tsikoane, 1991). 9. See the LHDA Order of 1986, Article XII for the original conditions. Currently, there is a lawsuit initiated by the Highlands Church Action Group (HCAG) against the development authorities, and one in the beginning process. They allege that the authorities have violated the Constitution of Lesotho in their acquisition of and compensation for land. The outcome of these lawsuits have important implications for local politics, as many question to what extent the development authorities are their own governing body and outside state regulation.

REFERENCES Ashton, H. (1967). The Basuto: A social study of traditional and modern Lesotho. London: Oxford University Press. Bardill, J. E., & Cobbe, J. H. (1985). Lesotho: Dilemmas of dependence in Southern Africa. Boulder: Westview Press. Beneria, L., & Feldman, S. (1992). Unequal burden: Economic Crises, persistent poverty, and Women’s work. Boulder: Westview Press. Boserup, E. (1970). Woman’s role in economic development. New York: St. Martins Press. Braun, Y. (2000). Lesotho’s white gold: The gendered effects of the Lesotho Highlands Water Project. MA thesis, Program in Social Relations, University of California, Irvine. Braun, Y. (2003). Investigating the social impacts of large dams. Braun, Y. (2005). Large dams as development: Restructuring access to natural resources in Lesotho. In: A.K. Jorgenson & E.L. Kick (Eds), Globalization and the environment.

394

YVONNE A. BRAUN

Bunker, S. G. (1985). Underdeveloping the Amazon: Extraction, unequal exchange, and the failure of the modern state. Urbana: University of Illinois Press. Bunker, S. G., & Ciccantell, P. S. (1999). Economic ascent and the global environment: Worldsystems theory and the new historical materialism. In: W. L. Goldfrank, D. Goodman & A. Szasz (Eds), Ecology and the world-system. Westport, CT: Greenwood Press. Cernea, M. M. (1988). Involuntary resettlement in development projects: Policy guidelines in World Bank-financed projects. World Bank Technical Paper 180, Washington, DC. Cernea, M. M. (1990). Poverty risks from population displacement in water resources development. Cambridge, MA: Harvard Institute for International Development. Chambers, R. (1969). Settlement schemes in tropical Africa: A study of organizations and development. New York: Praeger. Clark, S., Colson, E., Lee, J., & Scudder, T. (1995). Ten thousand Tonga: A longitudinal anthropological study of Southern Zambia, 1956–1991. Population Studies, 49, 91–109. Colson, E. (1971). The social consequences of resettlement: The impact of the Kariba resettlement upon the Gwembe Tonga. Manchester: University of Zambia, Manchester University Press for Institute for African Studies. Colson, E., & Scudder, T. (1988). For prayer and profit: The ritual, economic, and social importance of beer in Gwembe District, Zambia, 1950–1982. Stanford: Stanford University Press. Detter, A., Gunnewig, M., & Seiffert, B. (1994). Transformations caused by Lesotho Highlands Water Project Phase 1-A with focus on the Ha Lejone and Ha Mensel in Katse local catchment. Maseru, Lesotho. Eldredge, E. (1993). A South African kingdom: The pursuit of security in nineteenth-century Lesotho. Cambridge: Cambridge University Press. Elson, D., & Pearson, R. (1980). The latest phase of the internationalisation of capital and its implications for women in the third world. Institute of Development Studies at the University of Sussex: Brighton. Epprecht, M. (2000). ‘This matter of women is getting very bad’: Gender, development and politics in colonial Lesotho. Pietermaritzburg: University of Natal Press. Fahim, H. M. (1981). Dams, people, and development: The Aswan high dam case. New York: Pergamon Press. Fast, H. B., & Berkes, F. (1994). Native land use, traditional knowledge and the subsistence economy in the Hudson Bay bioregion. Ottawa: Hudson Bay Programme. Ferguson, J. (1985). Bovine mystique: Power, property, and livestock in rural Lesotho. Man, 20, 647–674. Ferguson, J. (1990). The anti-politics machine. Minneapolis: Minnesota University Press. Gay, J. (1980). Basotho women’s options: A study of marital careers in rural Lesotho. Ph.D. thesis, University of Cambridge. George, S., & Sabelli, F. (1994). Faith and credit: The World Bank’s secular empire. Boulder: Westview Press. Goldsmith, E., & Hildyard, N. (1984). The social and environmental impacts of large dams. Wadebridge: Wadebridge Ecological Centre, Cornwall, UK. Goodland, R. J. (1997). Environmental sustainability in the hydro industry: Disaggregating the debate. In: T. Dorcey, A. Steiner, M. C. Acreman & B. Orlando (Eds), Large dams: Learning from the past. Gland, Switzerland: IUCN. Gordon, A. A. (1996). Transforming capitalism and patriarchy: Gender and development in Africa. Boulder: Lynne Rienner Press.

Selling the River

395

Gordon, E. (1981). An analysis of the impact of labour migration on the lives of women in Lesotho. In: N. Nelson (Ed.), African women in the development process. London: Cass. Hart, D. (1980). The Volta river project: A case study in politics and technology. Edinburgh: Edinburgh University Press. Hassan, F. M. A. (2002). Lesotho – development in a challenging environment: A joint World Bank–African development bank evaluation. African Development Bank and World Bank, Washington, DC. Horowitz, M. M. (1991). Victims upstream and down. Journal of Refugee Studies, 2, 164–181. Kardam, N. (1997). The adaptability of international development agencies: The response of the World Bank to women in development. In: K. Staudt (Ed.), Women, international development, and politics: The bureaucratic mire. Philadelphia, PA: Temple. Kabeer, N. (1994). Reversed realities: Gender hierarchies in development thought. London: Verso Press. Leacock, E., & Safa, H. I. (1986). Women’s Work: Development and the division of labor by gender. Massachusetts: Bergin & Garvey. Lockwood, M. (1992). Engendering adjustment or adjusting gender?: Some new approaches to women and development in Africa. Institute of Development Studies, University of Sussex, Brighton. Lesotho Highlands Development Authority (LHDA). (1986). LHDA Order of 1986. Lesotho Government Gazette Extraordinary, Maseru, Lesotho. Letuka, P., Matashane, K., & Morolong, B. L. (1997). Beyond inequalities: Women in Lesotho. Maseru, Lesotho: Southern African Research and Documentation Centre. McCully, P. (1996). Silenced rivers: The ecology and politics of large dams. London: Zed Books. McMichael, P. (2000). Development and social change: A global perspective. Thousand Oaks: Pine Forge Press. Mohanty, C. T., Russo, A., & Torres, L. (1991). Third world women and the politics of feminism. Bloomington: Indiana University Press. Momsen, J., & Kinnaird, V. (1993). Different places, different voices: Gender and development in Africa, Asia, and Latin America. London: Routledge. Murray, C. (1981). Families divided: The impact of migrant labour in Lesotho. New York: Cambridge University Press. Nkomo, S. (1990). Confrontation and the challenge for independent development: The case of Lesotho. In: R. E. Mazur (Ed.), Breaking the links: Development theory and practice in Southern Africa. Trenton: Africa World Press, Inc. Panel of Environmental Experts (POE). (1989). Report to LHDA and the World Bank. Panel of Environmental Experts (POE). (1993). Report to LHDA and the World Bank. Panel of Environmental Experts (POE). (1994). Report to LHDA and the World Bank. Panel of Environmental Experts (POE). (1995). Report to LHDA and the World Bank. Panel of Environmental Experts (POE). (1997). Report to LHDA and the World Bank. Peterson, E. (1954). Big dam foolishness: The problem of modern flood control and water storage. New York: Devin-Adair. Rocheleau, D., Thomas-Slayter, B., & Wangari, E. (1996). Feminist political ecology: Global issues and local experiences. New York: Routledge. Roder, W. (1994). Human adjustment to the Kainji reservoir in Nigeria: An assessment of the economic and environmental consequences of a major man-made lake in Africa. Lanham, NY: University Press of America.

396

YVONNE A. BRAUN

Rosaldo, M., & Lamphere, L. (1974). Women, culture, and Society. Stanford: Stanford University Press. Santho, S., & Sejanamane, M. (1991). Southern Africa after apartheid: Prospects for the inner periphery in the 1990’s. Harare: SAPES Trust. Scudder, T. (1979). Expected impacts of compulsory relocation on Navajos with special emphasis on relocation from the Former Joint Use Area Required by Public Law 93–531. The Navajo Nation and the Institute for Development Anthropology, Binghamton. Scudder, T. (1985). What it means to be dammed: The anthropology of large-scale development projects in the tropics and sub-tropics. Engineering and Science, 54, 9–15. Scudder, T. (1988). Overview of the African experience with river basin development: Achievements to date, the role of traditions, and strategies for the future. Institute of Development Anthropology, Binghamton. Scudder, T. (1993). Development induced relocation and refugee studies: Thirty-seven years of change and continuity among Zambia’s Gwembe Tonga. Journal of Refugee Studies, 62, 123–152. Scudder, T. (1997). Social impacts of large dams. In: T. Dorcey, A. Steiner, M. C. Acreman & B. Orlando (Eds), Large dams: Learning from the past. Gland, Switzerland: IUCN. Sechaba Consultants. (1991). Poverty in Lesotho 1991: A mapping exercise. Maseru, Lesotho: Government of Lesotho. Sechaba Consultants. (1994a). Poverty in Lesotho 1994: A mapping exercise. Government of Lesotho Working Document. Maseru, Lesotho: Government of Lesotho. Sechaba Consultants. (1994b). Sechaba interviews. Maseru, Lesotho. Sechaba Consultants. (1995). Lesotho’s long journey. Maseru, Lesotho: Sechaba Consultants. Sen, G., & Grown, C. (1987). Development, crises and alternative visions: Third world women’s perspectives. New York: Monthly Review Press. Sharma, K. (1985). Women and development: Gender concerns. Center for Women’s Development Studies, New Delhi. Thomas-Slayter, B., & Rocheleau, D. (1994). Essential Connections: Linking gender to effective natural resource management and sustainable development. Women in International Development, Michigan State University. Transformation Resource Centre. (1995). Work for justice. Maseru, Lesotho: Transformation Resource Centre. Tshabalala, M., & Turner, S. D. (1989). 1988 Socio-economic census of the Lesotho Highlands Water Project Phase 1A Areas (Vol. I), Main Report. Maseru, Lesotho: LHDA. Tsikoane, T. (1991). Towards a redeﬁned role of the Lesotho Highlands Water Project in the post-Apartheid South Africa. In: S. Santho & M. Sejanamane (Eds), Southern Africa after apartheid: Prospects for the inner periphery in the 1990’s. Harare: SAPES Trust. UNICEF. (1994). The situation of children and women in Lesotho. Maseru, Lesotho: Government of Lesotho. United Nations. (2003). Investment policy review: Lesotho. New York: United Nations. Weiner, A. (1976). Women of value, men of renown: New perspectives in Trobriand Exchange. Austin: University of Texas Press. World Commission on Dams (WCD). (2000). Dams and development: A new framework for decision-making: The report of the World Commission on dams. London: Earthscan Publications.